[pypy-commit] pypy win64 test: merge default
ctismer
noreply at buildbot.pypy.org
Thu Nov 3 16:11:28 CET 2011
Author: Christian Tismer <tismer at stackless.com>
Branch: win64 test
Changeset: r48702:19ea93d6b3ae
Date: 2011-11-03 16:02 +0100
http://bitbucket.org/pypy/pypy/changeset/19ea93d6b3ae/
Log: merge default
diff too long, truncating to 10000 out of 87222 lines
diff --git a/.hgtags b/.hgtags
--- a/.hgtags
+++ b/.hgtags
@@ -1,1 +1,3 @@
b590cf6de4190623aad9aa698694c22e614d67b9 release-1.5
+b48df0bf4e75b81d98f19ce89d4a7dc3e1dab5e5 benchmarked
+d8ac7d23d3ec5f9a0fa1264972f74a010dbfd07f release-1.6
diff --git a/LICENSE b/LICENSE
--- a/LICENSE
+++ b/LICENSE
@@ -37,22 +37,22 @@
Armin Rigo
Maciej Fijalkowski
Carl Friedrich Bolz
+ Antonio Cuni
Amaury Forgeot d'Arc
- Antonio Cuni
Samuele Pedroni
Michael Hudson
Holger Krekel
+ Benjamin Peterson
Christian Tismer
- Benjamin Peterson
+ Hakan Ardo
+ Alex Gaynor
Eric van Riet Paap
- Anders Chrigström
- Håkan Ardö
+ Anders Chrigstrom
+ David Schneider
Richard Emslie
Dan Villiom Podlaski Christiansen
Alexander Schremmer
- Alex Gaynor
- David Schneider
- Aurelién Campeas
+ Aurelien Campeas
Anders Lehmann
Camillo Bruni
Niklaus Haldimann
@@ -63,16 +63,17 @@
Bartosz Skowron
Jakub Gustak
Guido Wesdorp
+ Daniel Roberts
Adrien Di Mascio
Laura Creighton
Ludovic Aubry
Niko Matsakis
- Daniel Roberts
Jason Creighton
- Jacob Hallén
+ Jacob Hallen
Alex Martelli
Anders Hammarquist
Jan de Mooij
+ Wim Lavrijsen
Stephan Diehl
Michael Foord
Stefan Schwarzer
@@ -83,9 +84,13 @@
Alexandre Fayolle
Marius Gedminas
Simon Burton
+ Justin Peel
Jean-Paul Calderone
John Witulski
+ Lukas Diekmann
+ holger krekel
Wim Lavrijsen
+ Dario Bertini
Andreas Stührk
Jean-Philippe St. Pierre
Guido van Rossum
@@ -97,15 +102,16 @@
Georg Brandl
Gerald Klix
Wanja Saatkamp
+ Ronny Pfannschmidt
Boris Feigin
Oscar Nierstrasz
- Dario Bertini
David Malcolm
Eugene Oden
Henry Mason
+ Sven Hager
Lukas Renggli
+ Ilya Osadchiy
Guenter Jantzen
- Ronny Pfannschmidt
Bert Freudenberg
Amit Regmi
Ben Young
@@ -122,8 +128,8 @@
Jared Grubb
Karl Bartel
Gabriel Lavoie
+ Victor Stinner
Brian Dorsey
- Victor Stinner
Stuart Williams
Toby Watson
Antoine Pitrou
@@ -134,19 +140,23 @@
Jonathan David Riehl
Elmo Mäntynen
Anders Qvist
- Beatrice Düring
+ Beatrice During
Alexander Sedov
+ Timo Paulssen
+ Corbin Simpson
Vincent Legoll
+ Romain Guillebert
Alan McIntyre
- Romain Guillebert
Alex Perry
Jens-Uwe Mager
+ Simon Cross
Dan Stromberg
- Lukas Diekmann
+ Guillebert Romain
Carl Meyer
Pieter Zieschang
Alejandro J. Cura
Sylvain Thenault
+ Christoph Gerum
Travis Francis Athougies
Henrik Vendelbo
Lutz Paelike
@@ -157,6 +167,7 @@
Miguel de Val Borro
Ignas Mikalajunas
Artur Lisiecki
+ Philip Jenvey
Joshua Gilbert
Godefroid Chappelle
Yusei Tahara
@@ -165,26 +176,31 @@
Gustavo Niemeyer
William Leslie
Akira Li
- Kristján Valur Jónsson
+ Kristjan Valur Jonsson
Bobby Impollonia
+ Michael Hudson-Doyle
Andrew Thompson
Anders Sigfridsson
+ Floris Bruynooghe
Jacek Generowicz
Dan Colish
- Sven Hager
Zooko Wilcox-O Hearn
+ Dan Villiom Podlaski Christiansen
Anders Hammarquist
+ Chris Lambacher
Dinu Gherman
Dan Colish
+ Brett Cannon
Daniel Neuhäuser
Michael Chermside
Konrad Delong
Anna Ravencroft
Greg Price
Armin Ronacher
+ Christian Muirhead
Jim Baker
- Philip Jenvey
Rodrigo Araújo
+ Romain Guillebert
Heinrich-Heine University, Germany
Open End AB (formerly AB Strakt), Sweden
diff --git a/ctypes_configure/configure.py b/ctypes_configure/configure.py
--- a/ctypes_configure/configure.py
+++ b/ctypes_configure/configure.py
@@ -559,7 +559,9 @@
C_HEADER = """
#include <stdio.h>
#include <stddef.h> /* for offsetof() */
-#include <stdint.h> /* FreeBSD: for uint64_t */
+#ifndef _WIN32
+# include <stdint.h> /* FreeBSD: for uint64_t */
+#endif
void dump(char* key, int value) {
printf("%s: %d\\n", key, value);
diff --git a/ctypes_configure/stdoutcapture.py b/ctypes_configure/stdoutcapture.py
--- a/ctypes_configure/stdoutcapture.py
+++ b/ctypes_configure/stdoutcapture.py
@@ -15,6 +15,15 @@
not hasattr(os, 'fdopen')):
self.dummy = 1
else:
+ try:
+ self.tmpout = os.tmpfile()
+ if mixed_out_err:
+ self.tmperr = self.tmpout
+ else:
+ self.tmperr = os.tmpfile()
+ except OSError: # bah? on at least one Windows box
+ self.dummy = 1
+ return
self.dummy = 0
# make new stdout/stderr files if needed
self.localoutfd = os.dup(1)
@@ -29,11 +38,6 @@
sys.stderr = os.fdopen(self.localerrfd, 'w', 0)
else:
self.saved_stderr = None
- self.tmpout = os.tmpfile()
- if mixed_out_err:
- self.tmperr = self.tmpout
- else:
- self.tmperr = os.tmpfile()
os.dup2(self.tmpout.fileno(), 1)
os.dup2(self.tmperr.fileno(), 2)
diff --git a/dotviewer/graphparse.py b/dotviewer/graphparse.py
--- a/dotviewer/graphparse.py
+++ b/dotviewer/graphparse.py
@@ -36,48 +36,45 @@
print >> sys.stderr, "Warning: could not guess file type, using 'dot'"
return 'unknown'
-def dot2plain(content, contenttype, use_codespeak=False):
- if contenttype == 'plain':
- # already a .plain file
- return content
+def dot2plain_graphviz(content, contenttype, use_codespeak=False):
+ if contenttype != 'neato':
+ cmdline = 'dot -Tplain'
+ else:
+ cmdline = 'neato -Tplain'
+ #print >> sys.stderr, '* running:', cmdline
+ close_fds = sys.platform != 'win32'
+ p = subprocess.Popen(cmdline, shell=True, close_fds=close_fds,
+ stdin=subprocess.PIPE, stdout=subprocess.PIPE)
+ (child_in, child_out) = (p.stdin, p.stdout)
+ try:
+ import thread
+ except ImportError:
+ bkgndwrite(child_in, content)
+ else:
+ thread.start_new_thread(bkgndwrite, (child_in, content))
+ plaincontent = child_out.read()
+ child_out.close()
+ if not plaincontent: # 'dot' is likely not installed
+ raise PlainParseError("no result from running 'dot'")
+ return plaincontent
- if not use_codespeak:
- if contenttype != 'neato':
- cmdline = 'dot -Tplain'
- else:
- cmdline = 'neato -Tplain'
- #print >> sys.stderr, '* running:', cmdline
- close_fds = sys.platform != 'win32'
- p = subprocess.Popen(cmdline, shell=True, close_fds=close_fds,
- stdin=subprocess.PIPE, stdout=subprocess.PIPE)
- (child_in, child_out) = (p.stdin, p.stdout)
- try:
- import thread
- except ImportError:
- bkgndwrite(child_in, content)
- else:
- thread.start_new_thread(bkgndwrite, (child_in, content))
- plaincontent = child_out.read()
- child_out.close()
- if not plaincontent: # 'dot' is likely not installed
- raise PlainParseError("no result from running 'dot'")
- else:
- import urllib
- request = urllib.urlencode({'dot': content})
- url = 'http://codespeak.net/pypy/convertdot.cgi'
- print >> sys.stderr, '* posting:', url
- g = urllib.urlopen(url, data=request)
- result = []
- while True:
- data = g.read(16384)
- if not data:
- break
- result.append(data)
- g.close()
- plaincontent = ''.join(result)
- # very simple-minded way to give a somewhat better error message
- if plaincontent.startswith('<body'):
- raise Exception("the dot on codespeak has very likely crashed")
+def dot2plain_codespeak(content, contenttype):
+ import urllib
+ request = urllib.urlencode({'dot': content})
+ url = 'http://codespeak.net/pypy/convertdot.cgi'
+ print >> sys.stderr, '* posting:', url
+ g = urllib.urlopen(url, data=request)
+ result = []
+ while True:
+ data = g.read(16384)
+ if not data:
+ break
+ result.append(data)
+ g.close()
+ plaincontent = ''.join(result)
+ # very simple-minded way to give a somewhat better error message
+ if plaincontent.startswith('<body'):
+ raise Exception("the dot on codespeak has very likely crashed")
return plaincontent
def bkgndwrite(f, data):
@@ -148,10 +145,13 @@
def parse_dot(graph_id, content, links={}, fixedfont=False):
contenttype = guess_type(content)
- try:
- plaincontent = dot2plain(content, contenttype, use_codespeak=False)
- return list(parse_plain(graph_id, plaincontent, links, fixedfont))
- except PlainParseError:
- # failed, retry via codespeak
- plaincontent = dot2plain(content, contenttype, use_codespeak=True)
- return list(parse_plain(graph_id, plaincontent, links, fixedfont))
+ if contenttype == 'plain':
+ plaincontent = content
+ else:
+ try:
+ plaincontent = dot2plain_graphviz(content, contenttype)
+ except PlainParseError, e:
+ print e
+ # failed, retry via codespeak
+ plaincontent = dot2plain_codespeak(content, contenttype)
+ return list(parse_plain(graph_id, plaincontent, links, fixedfont))
diff --git a/lib-python/2.7/ssl.py b/lib-python/2.7/ssl.py
--- a/lib-python/2.7/ssl.py
+++ b/lib-python/2.7/ssl.py
@@ -62,7 +62,6 @@
from _ssl import OPENSSL_VERSION_NUMBER, OPENSSL_VERSION_INFO, OPENSSL_VERSION
from _ssl import SSLError
from _ssl import CERT_NONE, CERT_OPTIONAL, CERT_REQUIRED
-from _ssl import PROTOCOL_SSLv2, PROTOCOL_SSLv3, PROTOCOL_SSLv23, PROTOCOL_TLSv1
from _ssl import RAND_status, RAND_egd, RAND_add
from _ssl import \
SSL_ERROR_ZERO_RETURN, \
@@ -74,6 +73,18 @@
SSL_ERROR_WANT_CONNECT, \
SSL_ERROR_EOF, \
SSL_ERROR_INVALID_ERROR_CODE
+from _ssl import PROTOCOL_SSLv3, PROTOCOL_SSLv23, PROTOCOL_TLSv1
+_PROTOCOL_NAMES = {
+ PROTOCOL_TLSv1: "TLSv1",
+ PROTOCOL_SSLv23: "SSLv23",
+ PROTOCOL_SSLv3: "SSLv3",
+}
+try:
+ from _ssl import PROTOCOL_SSLv2
+except ImportError:
+ pass
+else:
+ _PROTOCOL_NAMES[PROTOCOL_SSLv2] = "SSLv2"
from socket import socket, _fileobject, _delegate_methods, error as socket_error
from socket import getnameinfo as _getnameinfo
@@ -408,16 +419,7 @@
return DER_cert_to_PEM_cert(dercert)
def get_protocol_name(protocol_code):
- if protocol_code == PROTOCOL_TLSv1:
- return "TLSv1"
- elif protocol_code == PROTOCOL_SSLv23:
- return "SSLv23"
- elif protocol_code == PROTOCOL_SSLv2:
- return "SSLv2"
- elif protocol_code == PROTOCOL_SSLv3:
- return "SSLv3"
- else:
- return "<unknown>"
+ return _PROTOCOL_NAMES.get(protocol_code, '<unknown>')
# a replacement for the old socket.ssl function
diff --git a/lib-python/2.7/test/test_ssl.py b/lib-python/2.7/test/test_ssl.py
--- a/lib-python/2.7/test/test_ssl.py
+++ b/lib-python/2.7/test/test_ssl.py
@@ -58,32 +58,35 @@
# Issue #9415: Ubuntu hijacks their OpenSSL and forcefully disables SSLv2
def skip_if_broken_ubuntu_ssl(func):
- # We need to access the lower-level wrapper in order to create an
- # implicit SSL context without trying to connect or listen.
- try:
- import _ssl
- except ImportError:
- # The returned function won't get executed, just ignore the error
- pass
- @functools.wraps(func)
- def f(*args, **kwargs):
+ if hasattr(ssl, 'PROTOCOL_SSLv2'):
+ # We need to access the lower-level wrapper in order to create an
+ # implicit SSL context without trying to connect or listen.
try:
- s = socket.socket(socket.AF_INET)
- _ssl.sslwrap(s._sock, 0, None, None,
- ssl.CERT_NONE, ssl.PROTOCOL_SSLv2, None, None)
- except ssl.SSLError as e:
- if (ssl.OPENSSL_VERSION_INFO == (0, 9, 8, 15, 15) and
- platform.linux_distribution() == ('debian', 'squeeze/sid', '')
- and 'Invalid SSL protocol variant specified' in str(e)):
- raise unittest.SkipTest("Patched Ubuntu OpenSSL breaks behaviour")
- return func(*args, **kwargs)
- return f
+ import _ssl
+ except ImportError:
+ # The returned function won't get executed, just ignore the error
+ pass
+ @functools.wraps(func)
+ def f(*args, **kwargs):
+ try:
+ s = socket.socket(socket.AF_INET)
+ _ssl.sslwrap(s._sock, 0, None, None,
+ ssl.CERT_NONE, ssl.PROTOCOL_SSLv2, None, None)
+ except ssl.SSLError as e:
+ if (ssl.OPENSSL_VERSION_INFO == (0, 9, 8, 15, 15) and
+ platform.linux_distribution() == ('debian', 'squeeze/sid', '')
+ and 'Invalid SSL protocol variant specified' in str(e)):
+ raise unittest.SkipTest("Patched Ubuntu OpenSSL breaks behaviour")
+ return func(*args, **kwargs)
+ return f
+ else:
+ return func
class BasicSocketTests(unittest.TestCase):
def test_constants(self):
- ssl.PROTOCOL_SSLv2
+ #ssl.PROTOCOL_SSLv2
ssl.PROTOCOL_SSLv23
ssl.PROTOCOL_SSLv3
ssl.PROTOCOL_TLSv1
@@ -964,7 +967,8 @@
try_protocol_combo(ssl.PROTOCOL_SSLv3, ssl.PROTOCOL_SSLv3, True)
try_protocol_combo(ssl.PROTOCOL_SSLv3, ssl.PROTOCOL_SSLv3, True, ssl.CERT_OPTIONAL)
try_protocol_combo(ssl.PROTOCOL_SSLv3, ssl.PROTOCOL_SSLv3, True, ssl.CERT_REQUIRED)
- try_protocol_combo(ssl.PROTOCOL_SSLv3, ssl.PROTOCOL_SSLv2, False)
+ if hasattr(ssl, 'PROTOCOL_SSLv2'):
+ try_protocol_combo(ssl.PROTOCOL_SSLv3, ssl.PROTOCOL_SSLv2, False)
try_protocol_combo(ssl.PROTOCOL_SSLv3, ssl.PROTOCOL_SSLv23, False)
try_protocol_combo(ssl.PROTOCOL_SSLv3, ssl.PROTOCOL_TLSv1, False)
@@ -976,7 +980,8 @@
try_protocol_combo(ssl.PROTOCOL_TLSv1, ssl.PROTOCOL_TLSv1, True)
try_protocol_combo(ssl.PROTOCOL_TLSv1, ssl.PROTOCOL_TLSv1, True, ssl.CERT_OPTIONAL)
try_protocol_combo(ssl.PROTOCOL_TLSv1, ssl.PROTOCOL_TLSv1, True, ssl.CERT_REQUIRED)
- try_protocol_combo(ssl.PROTOCOL_TLSv1, ssl.PROTOCOL_SSLv2, False)
+ if hasattr(ssl, 'PROTOCOL_SSLv2'):
+ try_protocol_combo(ssl.PROTOCOL_TLSv1, ssl.PROTOCOL_SSLv2, False)
try_protocol_combo(ssl.PROTOCOL_TLSv1, ssl.PROTOCOL_SSLv3, False)
try_protocol_combo(ssl.PROTOCOL_TLSv1, ssl.PROTOCOL_SSLv23, False)
diff --git a/lib-python/conftest.py b/lib-python/conftest.py
--- a/lib-python/conftest.py
+++ b/lib-python/conftest.py
@@ -154,18 +154,18 @@
RegrTest('test_cmd.py'),
RegrTest('test_cmd_line_script.py'),
RegrTest('test_codeccallbacks.py', core=True),
- RegrTest('test_codecencodings_cn.py'),
- RegrTest('test_codecencodings_hk.py'),
- RegrTest('test_codecencodings_jp.py'),
- RegrTest('test_codecencodings_kr.py'),
- RegrTest('test_codecencodings_tw.py'),
+ RegrTest('test_codecencodings_cn.py', usemodules='_multibytecodec'),
+ RegrTest('test_codecencodings_hk.py', usemodules='_multibytecodec'),
+ RegrTest('test_codecencodings_jp.py', usemodules='_multibytecodec'),
+ RegrTest('test_codecencodings_kr.py', usemodules='_multibytecodec'),
+ RegrTest('test_codecencodings_tw.py', usemodules='_multibytecodec'),
- RegrTest('test_codecmaps_cn.py'),
- RegrTest('test_codecmaps_hk.py'),
- RegrTest('test_codecmaps_jp.py'),
- RegrTest('test_codecmaps_kr.py'),
- RegrTest('test_codecmaps_tw.py'),
- RegrTest('test_codecs.py', core=True),
+ RegrTest('test_codecmaps_cn.py', usemodules='_multibytecodec'),
+ RegrTest('test_codecmaps_hk.py', usemodules='_multibytecodec'),
+ RegrTest('test_codecmaps_jp.py', usemodules='_multibytecodec'),
+ RegrTest('test_codecmaps_kr.py', usemodules='_multibytecodec'),
+ RegrTest('test_codecmaps_tw.py', usemodules='_multibytecodec'),
+ RegrTest('test_codecs.py', core=True, usemodules='_multibytecodec'),
RegrTest('test_codeop.py', core=True),
RegrTest('test_coercion.py', core=True),
RegrTest('test_collections.py'),
@@ -314,10 +314,10 @@
RegrTest('test_mmap.py'),
RegrTest('test_module.py', core=True),
RegrTest('test_modulefinder.py'),
- RegrTest('test_multibytecodec.py'),
+ RegrTest('test_multibytecodec.py', usemodules='_multibytecodec'),
RegrTest('test_multibytecodec_support.py', skip="not a test"),
RegrTest('test_multifile.py'),
- RegrTest('test_multiprocessing.py', skip='FIXME leaves subprocesses'),
+ RegrTest('test_multiprocessing.py', skip="FIXME leaves subprocesses"),
RegrTest('test_mutants.py', core="possibly"),
RegrTest('test_mutex.py'),
RegrTest('test_netrc.py'),
@@ -359,7 +359,7 @@
RegrTest('test_property.py', core=True),
RegrTest('test_pstats.py'),
RegrTest('test_pty.py', skip="unsupported extension module"),
- RegrTest('test_pwd.py', skip=skip_win32),
+ RegrTest('test_pwd.py', usemodules="pwd", skip=skip_win32),
RegrTest('test_py3kwarn.py'),
RegrTest('test_pyclbr.py'),
RegrTest('test_pydoc.py'),
diff --git a/lib-python/modified-2.7/ctypes/__init__.py b/lib-python/modified-2.7/ctypes/__init__.py
--- a/lib-python/modified-2.7/ctypes/__init__.py
+++ b/lib-python/modified-2.7/ctypes/__init__.py
@@ -489,9 +489,12 @@
_flags_ = _FUNCFLAG_CDECL | _FUNCFLAG_PYTHONAPI
return CFunctionType
-_cast = PYFUNCTYPE(py_object, c_void_p, py_object, py_object)(_cast_addr)
def cast(obj, typ):
- return _cast(obj, obj, typ)
+ try:
+ c_void_p.from_param(obj)
+ except TypeError, e:
+ raise ArgumentError(str(e))
+ return _cast_addr(obj, obj, typ)
_string_at = PYFUNCTYPE(py_object, c_void_p, c_int)(_string_at_addr)
def string_at(ptr, size=-1):
diff --git a/lib-python/modified-2.7/ctypes/util.py b/lib-python/modified-2.7/ctypes/util.py
--- a/lib-python/modified-2.7/ctypes/util.py
+++ b/lib-python/modified-2.7/ctypes/util.py
@@ -72,8 +72,8 @@
return name
if os.name == "posix" and sys.platform == "darwin":
- from ctypes.macholib.dyld import dyld_find as _dyld_find
def find_library(name):
+ from ctypes.macholib.dyld import dyld_find as _dyld_find
possible = ['lib%s.dylib' % name,
'%s.dylib' % name,
'%s.framework/%s' % (name, name)]
diff --git a/lib-python/modified-2.7/distutils/sysconfig_pypy.py b/lib-python/modified-2.7/distutils/sysconfig_pypy.py
--- a/lib-python/modified-2.7/distutils/sysconfig_pypy.py
+++ b/lib-python/modified-2.7/distutils/sysconfig_pypy.py
@@ -116,6 +116,12 @@
if compiler.compiler_type == "unix":
compiler.compiler_so.extend(['-fPIC', '-Wimplicit'])
compiler.shared_lib_extension = get_config_var('SO')
+ if "CFLAGS" in os.environ:
+ cflags = os.environ["CFLAGS"]
+ compiler.compiler.append(cflags)
+ compiler.compiler_so.append(cflags)
+ compiler.linker_so.append(cflags)
+
from sysconfig_cpython import (
parse_makefile, _variable_rx, expand_makefile_vars)
diff --git a/lib-python/modified-2.7/distutils/unixccompiler.py b/lib-python/modified-2.7/distutils/unixccompiler.py
--- a/lib-python/modified-2.7/distutils/unixccompiler.py
+++ b/lib-python/modified-2.7/distutils/unixccompiler.py
@@ -324,7 +324,7 @@
# On OSX users can specify an alternate SDK using
# '-isysroot', calculate the SDK root if it is specified
# (and use it further on)
- cflags = sysconfig.get_config_var('CFLAGS')
+ cflags = sysconfig.get_config_var('CFLAGS') or ''
m = re.search(r'-isysroot\s+(\S+)', cflags)
if m is None:
sysroot = '/'
diff --git a/lib-python/modified-2.7/heapq.py b/lib-python/modified-2.7/heapq.py
new file mode 100644
--- /dev/null
+++ b/lib-python/modified-2.7/heapq.py
@@ -0,0 +1,442 @@
+# -*- 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
+all k, counting elements from 0. For the sake of comparison,
+non-existing elements are considered to be infinite. The interesting
+property of a heap is that a[0] is always its smallest element.
+
+Usage:
+
+heap = [] # creates an empty heap
+heappush(heap, item) # pushes a new item on the heap
+item = heappop(heap) # pops the smallest item from the heap
+item = heap[0] # smallest item on the heap without popping it
+heapify(x) # transforms list into a heap, in-place, in linear time
+item = heapreplace(heap, item) # pops and returns smallest item, and adds
+ # new item; the heap size is unchanged
+
+Our API differs from textbook heap algorithms as follows:
+
+- We use 0-based indexing. This makes the relationship between the
+ index for a node and the indexes for its children slightly less
+ obvious, but is more suitable since Python uses 0-based indexing.
+
+- Our heappop() method returns the smallest item, not the largest.
+
+These two make it possible to view the heap as a regular Python list
+without surprises: heap[0] is the smallest item, and heap.sort()
+maintains the heap invariant!
+"""
+
+# Original code by Kevin O'Connor, augmented by Tim Peters and Raymond Hettinger
+
+__about__ = """Heap queues
+
+[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,
+non-existing elements are considered to be infinite. The interesting
+property of a heap is that a[0] is always its smallest element.
+
+The strange invariant above is meant to be an efficient memory
+representation for a tournament. The numbers below are `k', not a[k]:
+
+ 0
+
+ 1 2
+
+ 3 4 5 6
+
+ 7 8 9 10 11 12 13 14
+
+ 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30
+
+
+In the tree above, each cell `k' is topping `2*k+1' and `2*k+2'. In
+an usual binary tournament we see in sports, each cell is the winner
+over the two cells it tops, and we can trace the winner down the tree
+to see all opponents s/he had. However, in many computer applications
+of such tournaments, we do not need to trace the history of a winner.
+To be more memory efficient, when a winner is promoted, we try to
+replace it by something else at a lower level, and the rule becomes
+that a cell and the two cells it tops contain three different items,
+but the top cell "wins" over the two topped cells.
+
+If this heap invariant is protected at all time, index 0 is clearly
+the overall winner. The simplest algorithmic way to remove it and
+find the "next" winner is to move some loser (let's say cell 30 in the
+diagram above) into the 0 position, and then percolate this new 0 down
+the tree, exchanging values, until the invariant is re-established.
+This is clearly logarithmic on the total number of items in the tree.
+By iterating over all items, you get an O(n ln n) sort.
+
+A nice feature of this sort is that you can efficiently insert new
+items while the sort is going on, provided that the inserted items are
+not "better" than the last 0'th element you extracted. This is
+especially useful in simulation contexts, where the tree holds all
+incoming events, and the "win" condition means the smallest scheduled
+time. When an event schedule other events for execution, they are
+scheduled into the future, so they can easily go into the heap. So, a
+heap is a good structure for implementing schedulers (this is what I
+used for my MIDI sequencer :-).
+
+Various structures for implementing schedulers have been extensively
+studied, and heaps are good for this, as they are reasonably speedy,
+the speed is almost constant, and the worst case is not much different
+than the average case. However, there are other representations which
+are more efficient overall, yet the worst cases might be terrible.
+
+Heaps are also very useful in big disk sorts. You most probably all
+know that a big sort implies producing "runs" (which are pre-sorted
+sequences, which size is usually related to the amount of CPU memory),
+followed by a merging passes for these runs, which merging is often
+very cleverly organised[1]. It is very important that the initial
+sort produces the longest runs possible. Tournaments are a good way
+to that. If, using all the memory available to hold a tournament, you
+replace and percolate items that happen to fit the current run, you'll
+produce runs which are twice the size of the memory for random input,
+and much better for input fuzzily ordered.
+
+Moreover, if you output the 0'th item on disk and get an input which
+may not fit in the current tournament (because the value "wins" over
+the last output value), it cannot fit in the heap, so the size of the
+heap decreases. The freed memory could be cleverly reused immediately
+for progressively building a second heap, which grows at exactly the
+same rate the first heap is melting. When the first heap completely
+vanishes, you switch heaps and start a new run. Clever and quite
+effective!
+
+In a word, heaps are useful memory structures to know. I use them in
+a few applications, and I think it is good to keep a `heap' module
+around. :-)
+
+--------------------
+[1] The disk balancing algorithms which are current, nowadays, are
+more annoying than clever, and this is a consequence of the seeking
+capabilities of the disks. On devices which cannot seek, like big
+tape drives, the story was quite different, and one had to be very
+clever to ensure (far in advance) that each tape movement will be the
+most effective possible (that is, will best participate at
+"progressing" the merge). Some tapes were even able to read
+backwards, and this was also used to avoid the rewinding time.
+Believe me, real good tape sorts were quite spectacular to watch!
+From all times, sorting has always been a Great Art! :-)
+"""
+
+__all__ = ['heappush', 'heappop', 'heapify', 'heapreplace', 'merge',
+ 'nlargest', 'nsmallest', 'heappushpop']
+
+from itertools import islice, repeat, count, imap, izip, tee, chain
+from operator import itemgetter
+import bisect
+
+def heappush(heap, item):
+ """Push item onto heap, maintaining the heap invariant."""
+ heap.append(item)
+ _siftdown(heap, 0, len(heap)-1)
+
+def heappop(heap):
+ """Pop the smallest item off the heap, maintaining the heap invariant."""
+ lastelt = heap.pop() # raises appropriate IndexError if heap is empty
+ if heap:
+ returnitem = heap[0]
+ heap[0] = lastelt
+ _siftup(heap, 0)
+ else:
+ returnitem = lastelt
+ return returnitem
+
+def heapreplace(heap, item):
+ """Pop and return the current smallest value, and add the new item.
+
+ This is more efficient than heappop() followed by heappush(), and can be
+ more appropriate when using a fixed-size heap. Note that the value
+ returned may be larger than item! That constrains reasonable uses of
+ this routine unless written as part of a conditional replacement:
+
+ if item > heap[0]:
+ item = heapreplace(heap, item)
+ """
+ returnitem = heap[0] # raises appropriate IndexError if heap is empty
+ heap[0] = item
+ _siftup(heap, 0)
+ return returnitem
+
+def heappushpop(heap, item):
+ """Fast version of a heappush followed by a heappop."""
+ if heap and heap[0] < item:
+ item, heap[0] = heap[0], item
+ _siftup(heap, 0)
+ return item
+
+def heapify(x):
+ """Transform list into a heap, in-place, in O(len(heap)) time."""
+ n = len(x)
+ # Transform bottom-up. The largest index there's any point to looking at
+ # is the largest with a child index in-range, so must have 2*i + 1 < n,
+ # or i < (n-1)/2. If n is even = 2*j, this is (2*j-1)/2 = j-1/2 so
+ # j-1 is the largest, which is n//2 - 1. If n is odd = 2*j+1, this is
+ # (2*j+1-1)/2 = j so j-1 is the largest, and that's again n//2-1.
+ for i in reversed(xrange(n//2)):
+ _siftup(x, i)
+
+def nlargest(n, iterable):
+ """Find the n largest elements in a dataset.
+
+ Equivalent to: sorted(iterable, reverse=True)[:n]
+ """
+ if n < 0: # for consistency with the c impl
+ return []
+ it = iter(iterable)
+ result = list(islice(it, n))
+ if not result:
+ return result
+ heapify(result)
+ _heappushpop = heappushpop
+ for elem in it:
+ _heappushpop(result, elem)
+ result.sort(reverse=True)
+ return result
+
+def nsmallest(n, iterable):
+ """Find the n smallest elements in a dataset.
+
+ Equivalent to: sorted(iterable)[:n]
+ """
+ if n < 0: # for consistency with the c impl
+ return []
+ if hasattr(iterable, '__len__') and n * 10 <= len(iterable):
+ # For smaller values of n, the bisect method is faster than a minheap.
+ # It is also memory efficient, consuming only n elements of space.
+ it = iter(iterable)
+ result = sorted(islice(it, 0, n))
+ if not result:
+ return result
+ insort = bisect.insort
+ pop = result.pop
+ los = result[-1] # los --> Largest of the nsmallest
+ for elem in it:
+ if los <= elem:
+ continue
+ insort(result, elem)
+ pop()
+ los = result[-1]
+ return result
+ # An alternative approach manifests the whole iterable in memory but
+ # saves comparisons by heapifying all at once. Also, saves time
+ # over bisect.insort() which has O(n) data movement time for every
+ # insertion. Finding the n smallest of an m length iterable requires
+ # O(m) + O(n log m) comparisons.
+ h = list(iterable)
+ heapify(h)
+ return map(heappop, repeat(h, min(n, len(h))))
+
+# 'heap' is a heap at all indices >= startpos, except possibly for pos. pos
+# is the index of a leaf with a possibly out-of-order value. Restore the
+# heap invariant.
+def _siftdown(heap, startpos, pos):
+ newitem = heap[pos]
+ # Follow the path to the root, moving parents down until finding a place
+ # newitem fits.
+ while pos > startpos:
+ parentpos = (pos - 1) >> 1
+ parent = heap[parentpos]
+ if newitem < parent:
+ heap[pos] = parent
+ pos = parentpos
+ continue
+ break
+ heap[pos] = newitem
+
+# The child indices of heap index pos are already heaps, and we want to make
+# a heap at index pos too. We do this by bubbling the smaller child of
+# pos up (and so on with that child's children, etc) until hitting a leaf,
+# then using _siftdown to move the oddball originally at index pos into place.
+#
+# We *could* break out of the loop as soon as we find a pos where newitem <=
+# both its children, but turns out that's not a good idea, and despite that
+# many books write the algorithm that way. During a heap pop, the last array
+# element is sifted in, and that tends to be large, so that comparing it
+# against values starting from the root usually doesn't pay (= usually doesn't
+# get us out of the loop early). See Knuth, Volume 3, where this is
+# explained and quantified in an exercise.
+#
+# Cutting the # of comparisons is important, since these routines have no
+# way to extract "the priority" from an array element, so that intelligence
+# is likely to be hiding in custom __cmp__ methods, or in array elements
+# storing (priority, record) tuples. Comparisons are thus potentially
+# expensive.
+#
+# On random arrays of length 1000, making this change cut the number of
+# comparisons made by heapify() a little, and those made by exhaustive
+# heappop() a lot, in accord with theory. Here are typical results from 3
+# runs (3 just to demonstrate how small the variance is):
+#
+# Compares needed by heapify Compares needed by 1000 heappops
+# -------------------------- --------------------------------
+# 1837 cut to 1663 14996 cut to 8680
+# 1855 cut to 1659 14966 cut to 8678
+# 1847 cut to 1660 15024 cut to 8703
+#
+# Building the heap by using heappush() 1000 times instead required
+# 2198, 2148, and 2219 compares: heapify() is more efficient, when
+# you can use it.
+#
+# The total compares needed by list.sort() on the same lists were 8627,
+# 8627, and 8632 (this should be compared to the sum of heapify() and
+# heappop() compares): list.sort() is (unsurprisingly!) more efficient
+# for sorting.
+
+def _siftup(heap, pos):
+ endpos = len(heap)
+ startpos = pos
+ newitem = heap[pos]
+ # Bubble up the smaller child until hitting a leaf.
+ childpos = 2*pos + 1 # leftmost child position
+ while childpos < endpos:
+ # Set childpos to index of smaller child.
+ rightpos = childpos + 1
+ if rightpos < endpos and not heap[childpos] < heap[rightpos]:
+ childpos = rightpos
+ # Move the smaller child up.
+ heap[pos] = heap[childpos]
+ pos = childpos
+ childpos = 2*pos + 1
+ # The leaf at pos is empty now. Put newitem there, and bubble it up
+ # to its final resting place (by sifting its parents down).
+ heap[pos] = newitem
+ _siftdown(heap, startpos, pos)
+
+# If available, use C implementation
+try:
+ from _heapq import *
+except ImportError:
+ pass
+
+def merge(*iterables):
+ '''Merge multiple sorted inputs into a single sorted output.
+
+ Similar to sorted(itertools.chain(*iterables)) but returns a generator,
+ does not pull the data into memory all at once, and assumes that each of
+ the input streams is already sorted (smallest to largest).
+
+ >>> list(merge([1,3,5,7], [0,2,4,8], [5,10,15,20], [], [25]))
+ [0, 1, 2, 3, 4, 5, 5, 7, 8, 10, 15, 20, 25]
+
+ '''
+ _heappop, _heapreplace, _StopIteration = heappop, heapreplace, StopIteration
+
+ h = []
+ h_append = h.append
+ for itnum, it in enumerate(map(iter, iterables)):
+ try:
+ next = it.next
+ h_append([next(), itnum, next])
+ except _StopIteration:
+ pass
+ heapify(h)
+
+ while 1:
+ try:
+ while 1:
+ v, itnum, next = s = h[0] # raises IndexError when h is empty
+ yield v
+ s[0] = next() # raises StopIteration when exhausted
+ _heapreplace(h, s) # restore heap condition
+ except _StopIteration:
+ _heappop(h) # remove empty iterator
+ except IndexError:
+ return
+
+# Extend the implementations of nsmallest and nlargest to use a key= argument
+_nsmallest = nsmallest
+def nsmallest(n, iterable, key=None):
+ """Find the n smallest elements in a dataset.
+
+ Equivalent to: sorted(iterable, key=key)[:n]
+ """
+ # Short-cut for n==1 is to use min() when len(iterable)>0
+ if n == 1:
+ it = iter(iterable)
+ head = list(islice(it, 1))
+ if not head:
+ return []
+ if key is None:
+ return [min(chain(head, it))]
+ return [min(chain(head, it), key=key)]
+
+ # When n>=size, it's faster to use sort()
+ try:
+ size = len(iterable)
+ except (TypeError, AttributeError):
+ pass
+ else:
+ if n >= size:
+ return sorted(iterable, key=key)[:n]
+
+ # When key is none, use simpler decoration
+ if key is None:
+ it = izip(iterable, count()) # decorate
+ result = _nsmallest(n, it)
+ return map(itemgetter(0), result) # undecorate
+
+ # General case, slowest method
+ in1, in2 = tee(iterable)
+ it = izip(imap(key, in1), count(), in2) # decorate
+ result = _nsmallest(n, it)
+ return map(itemgetter(2), result) # undecorate
+
+_nlargest = nlargest
+def nlargest(n, iterable, key=None):
+ """Find the n largest elements in a dataset.
+
+ Equivalent to: sorted(iterable, key=key, reverse=True)[:n]
+ """
+
+ # Short-cut for n==1 is to use max() when len(iterable)>0
+ if n == 1:
+ it = iter(iterable)
+ head = list(islice(it, 1))
+ if not head:
+ return []
+ if key is None:
+ return [max(chain(head, it))]
+ return [max(chain(head, it), key=key)]
+
+ # When n>=size, it's faster to use sort()
+ try:
+ size = len(iterable)
+ except (TypeError, AttributeError):
+ pass
+ else:
+ if n >= size:
+ return sorted(iterable, key=key, reverse=True)[:n]
+
+ # When key is none, use simpler decoration
+ if key is None:
+ it = izip(iterable, count(0,-1)) # decorate
+ result = _nlargest(n, it)
+ return map(itemgetter(0), result) # undecorate
+
+ # General case, slowest method
+ in1, in2 = tee(iterable)
+ it = izip(imap(key, in1), count(0,-1), in2) # decorate
+ result = _nlargest(n, it)
+ return map(itemgetter(2), result) # undecorate
+
+if __name__ == "__main__":
+ # Simple sanity test
+ heap = []
+ data = [1, 3, 5, 7, 9, 2, 4, 6, 8, 0]
+ for item in data:
+ heappush(heap, item)
+ sort = []
+ while heap:
+ sort.append(heappop(heap))
+ print sort
+
+ import doctest
+ doctest.testmod()
diff --git a/lib-python/modified-2.7/httplib.py b/lib-python/modified-2.7/httplib.py
new file mode 100644
--- /dev/null
+++ b/lib-python/modified-2.7/httplib.py
@@ -0,0 +1,1377 @@
+"""HTTP/1.1 client library
+
+<intro stuff goes here>
+<other stuff, too>
+
+HTTPConnection goes through a number of "states", which define when a client
+may legally make another request or fetch the response for a particular
+request. This diagram details these state transitions:
+
+ (null)
+ |
+ | HTTPConnection()
+ v
+ Idle
+ |
+ | putrequest()
+ v
+ Request-started
+ |
+ | ( putheader() )* endheaders()
+ v
+ Request-sent
+ |
+ | response = getresponse()
+ v
+ Unread-response [Response-headers-read]
+ |\____________________
+ | |
+ | response.read() | putrequest()
+ v v
+ Idle Req-started-unread-response
+ ______/|
+ / |
+ response.read() | | ( putheader() )* endheaders()
+ v v
+ Request-started Req-sent-unread-response
+ |
+ | response.read()
+ v
+ Request-sent
+
+This diagram presents the following rules:
+ -- a second request may not be started until {response-headers-read}
+ -- a response [object] cannot be retrieved until {request-sent}
+ -- there is no differentiation between an unread response body and a
+ partially read response body
+
+Note: this enforcement is applied by the HTTPConnection class. The
+ HTTPResponse class does not enforce this state machine, which
+ implies sophisticated clients may accelerate the request/response
+ pipeline. Caution should be taken, though: accelerating the states
+ beyond the above pattern may imply knowledge of the server's
+ connection-close behavior for certain requests. For example, it
+ is impossible to tell whether the server will close the connection
+ UNTIL the response headers have been read; this means that further
+ requests cannot be placed into the pipeline until it is known that
+ the server will NOT be closing the connection.
+
+Logical State __state __response
+------------- ------- ----------
+Idle _CS_IDLE None
+Request-started _CS_REQ_STARTED None
+Request-sent _CS_REQ_SENT None
+Unread-response _CS_IDLE <response_class>
+Req-started-unread-response _CS_REQ_STARTED <response_class>
+Req-sent-unread-response _CS_REQ_SENT <response_class>
+"""
+
+from array import array
+import os
+import socket
+from sys import py3kwarning
+from urlparse import urlsplit
+import warnings
+with warnings.catch_warnings():
+ if py3kwarning:
+ warnings.filterwarnings("ignore", ".*mimetools has been removed",
+ DeprecationWarning)
+ import mimetools
+
+try:
+ from cStringIO import StringIO
+except ImportError:
+ from StringIO import StringIO
+
+__all__ = ["HTTP", "HTTPResponse", "HTTPConnection",
+ "HTTPException", "NotConnected", "UnknownProtocol",
+ "UnknownTransferEncoding", "UnimplementedFileMode",
+ "IncompleteRead", "InvalidURL", "ImproperConnectionState",
+ "CannotSendRequest", "CannotSendHeader", "ResponseNotReady",
+ "BadStatusLine", "error", "responses"]
+
+HTTP_PORT = 80
+HTTPS_PORT = 443
+
+_UNKNOWN = 'UNKNOWN'
+
+# connection states
+_CS_IDLE = 'Idle'
+_CS_REQ_STARTED = 'Request-started'
+_CS_REQ_SENT = 'Request-sent'
+
+# status codes
+# informational
+CONTINUE = 100
+SWITCHING_PROTOCOLS = 101
+PROCESSING = 102
+
+# successful
+OK = 200
+CREATED = 201
+ACCEPTED = 202
+NON_AUTHORITATIVE_INFORMATION = 203
+NO_CONTENT = 204
+RESET_CONTENT = 205
+PARTIAL_CONTENT = 206
+MULTI_STATUS = 207
+IM_USED = 226
+
+# redirection
+MULTIPLE_CHOICES = 300
+MOVED_PERMANENTLY = 301
+FOUND = 302
+SEE_OTHER = 303
+NOT_MODIFIED = 304
+USE_PROXY = 305
+TEMPORARY_REDIRECT = 307
+
+# client error
+BAD_REQUEST = 400
+UNAUTHORIZED = 401
+PAYMENT_REQUIRED = 402
+FORBIDDEN = 403
+NOT_FOUND = 404
+METHOD_NOT_ALLOWED = 405
+NOT_ACCEPTABLE = 406
+PROXY_AUTHENTICATION_REQUIRED = 407
+REQUEST_TIMEOUT = 408
+CONFLICT = 409
+GONE = 410
+LENGTH_REQUIRED = 411
+PRECONDITION_FAILED = 412
+REQUEST_ENTITY_TOO_LARGE = 413
+REQUEST_URI_TOO_LONG = 414
+UNSUPPORTED_MEDIA_TYPE = 415
+REQUESTED_RANGE_NOT_SATISFIABLE = 416
+EXPECTATION_FAILED = 417
+UNPROCESSABLE_ENTITY = 422
+LOCKED = 423
+FAILED_DEPENDENCY = 424
+UPGRADE_REQUIRED = 426
+
+# server error
+INTERNAL_SERVER_ERROR = 500
+NOT_IMPLEMENTED = 501
+BAD_GATEWAY = 502
+SERVICE_UNAVAILABLE = 503
+GATEWAY_TIMEOUT = 504
+HTTP_VERSION_NOT_SUPPORTED = 505
+INSUFFICIENT_STORAGE = 507
+NOT_EXTENDED = 510
+
+# Mapping status codes to official W3C names
+responses = {
+ 100: 'Continue',
+ 101: 'Switching Protocols',
+
+ 200: 'OK',
+ 201: 'Created',
+ 202: 'Accepted',
+ 203: 'Non-Authoritative Information',
+ 204: 'No Content',
+ 205: 'Reset Content',
+ 206: 'Partial Content',
+
+ 300: 'Multiple Choices',
+ 301: 'Moved Permanently',
+ 302: 'Found',
+ 303: 'See Other',
+ 304: 'Not Modified',
+ 305: 'Use Proxy',
+ 306: '(Unused)',
+ 307: 'Temporary Redirect',
+
+ 400: 'Bad Request',
+ 401: 'Unauthorized',
+ 402: 'Payment Required',
+ 403: 'Forbidden',
+ 404: 'Not Found',
+ 405: 'Method Not Allowed',
+ 406: 'Not Acceptable',
+ 407: 'Proxy Authentication Required',
+ 408: 'Request Timeout',
+ 409: 'Conflict',
+ 410: 'Gone',
+ 411: 'Length Required',
+ 412: 'Precondition Failed',
+ 413: 'Request Entity Too Large',
+ 414: 'Request-URI Too Long',
+ 415: 'Unsupported Media Type',
+ 416: 'Requested Range Not Satisfiable',
+ 417: 'Expectation Failed',
+
+ 500: 'Internal Server Error',
+ 501: 'Not Implemented',
+ 502: 'Bad Gateway',
+ 503: 'Service Unavailable',
+ 504: 'Gateway Timeout',
+ 505: 'HTTP Version Not Supported',
+}
+
+# maximal amount of data to read at one time in _safe_read
+MAXAMOUNT = 1048576
+
+class HTTPMessage(mimetools.Message):
+
+ def addheader(self, key, value):
+ """Add header for field key handling repeats."""
+ prev = self.dict.get(key)
+ if prev is None:
+ self.dict[key] = value
+ else:
+ combined = ", ".join((prev, value))
+ self.dict[key] = combined
+
+ def addcontinue(self, key, more):
+ """Add more field data from a continuation line."""
+ prev = self.dict[key]
+ self.dict[key] = prev + "\n " + more
+
+ def readheaders(self):
+ """Read header lines.
+
+ Read header lines up to the entirely blank line that terminates them.
+ The (normally blank) line that ends the headers is skipped, but not
+ included in the returned list. If a non-header line ends the headers,
+ (which is an error), an attempt is made to backspace over it; it is
+ never included in the returned list.
+
+ The variable self.status is set to the empty string if all went well,
+ otherwise it is an error message. The variable self.headers is a
+ completely uninterpreted list of lines contained in the header (so
+ printing them will reproduce the header exactly as it appears in the
+ file).
+
+ If multiple header fields with the same name occur, they are combined
+ according to the rules in RFC 2616 sec 4.2:
+
+ Appending each subsequent field-value to the first, each separated
+ by a comma. The order in which header fields with the same field-name
+ are received is significant to the interpretation of the combined
+ field value.
+ """
+ # XXX The implementation overrides the readheaders() method of
+ # rfc822.Message. The base class design isn't amenable to
+ # customized behavior here so the method here is a copy of the
+ # base class code with a few small changes.
+
+ self.dict = {}
+ self.unixfrom = ''
+ self.headers = hlist = []
+ self.status = ''
+ headerseen = ""
+ firstline = 1
+ startofline = unread = tell = None
+ if hasattr(self.fp, 'unread'):
+ unread = self.fp.unread
+ elif self.seekable:
+ tell = self.fp.tell
+ while True:
+ if tell:
+ try:
+ startofline = tell()
+ except IOError:
+ startofline = tell = None
+ self.seekable = 0
+ line = self.fp.readline()
+ if not line:
+ self.status = 'EOF in headers'
+ break
+ # Skip unix From name time lines
+ if firstline and line.startswith('From '):
+ self.unixfrom = self.unixfrom + line
+ continue
+ firstline = 0
+ if headerseen and line[0] in ' \t':
+ # XXX Not sure if continuation lines are handled properly
+ # for http and/or for repeating headers
+ # It's a continuation line.
+ hlist.append(line)
+ self.addcontinue(headerseen, line.strip())
+ continue
+ elif self.iscomment(line):
+ # It's a comment. Ignore it.
+ continue
+ elif self.islast(line):
+ # Note! No pushback here! The delimiter line gets eaten.
+ break
+ headerseen = self.isheader(line)
+ if headerseen:
+ # It's a legal header line, save it.
+ hlist.append(line)
+ self.addheader(headerseen, line[len(headerseen)+1:].strip())
+ continue
+ else:
+ # It's not a header line; throw it back and stop here.
+ if not self.dict:
+ self.status = 'No headers'
+ else:
+ self.status = 'Non-header line where header expected'
+ # Try to undo the read.
+ if unread:
+ unread(line)
+ elif tell:
+ self.fp.seek(startofline)
+ else:
+ self.status = self.status + '; bad seek'
+ break
+
+class HTTPResponse:
+
+ # strict: If true, raise BadStatusLine if the status line can't be
+ # parsed as a valid HTTP/1.0 or 1.1 status line. By default it is
+ # false because it prevents clients from talking to HTTP/0.9
+ # servers. Note that a response with a sufficiently corrupted
+ # status line will look like an HTTP/0.9 response.
+
+ # See RFC 2616 sec 19.6 and RFC 1945 sec 6 for details.
+
+ def __init__(self, sock, debuglevel=0, strict=0, method=None, buffering=False):
+ if buffering:
+ # The caller won't be using any sock.recv() calls, so buffering
+ # is fine and recommended for performance.
+ self.fp = sock.makefile('rb')
+ else:
+ # The buffer size is specified as zero, because the headers of
+ # the response are read with readline(). If the reads were
+ # buffered the readline() calls could consume some of the
+ # response, which make be read via a recv() on the underlying
+ # socket.
+ self.fp = sock.makefile('rb', 0)
+ self.debuglevel = debuglevel
+ self.strict = strict
+ self._method = method
+
+ self.msg = None
+
+ # from the Status-Line of the response
+ self.version = _UNKNOWN # HTTP-Version
+ self.status = _UNKNOWN # Status-Code
+ self.reason = _UNKNOWN # Reason-Phrase
+
+ self.chunked = _UNKNOWN # is "chunked" being used?
+ self.chunk_left = _UNKNOWN # bytes left to read in current chunk
+ self.length = _UNKNOWN # number of bytes left in response
+ self.will_close = _UNKNOWN # conn will close at end of response
+
+ def _read_status(self):
+ # Initialize with Simple-Response defaults
+ line = self.fp.readline()
+ if self.debuglevel > 0:
+ print "reply:", repr(line)
+ if not line:
+ # Presumably, the server closed the connection before
+ # sending a valid response.
+ raise BadStatusLine(line)
+ try:
+ [version, status, reason] = line.split(None, 2)
+ except ValueError:
+ try:
+ [version, status] = line.split(None, 1)
+ reason = ""
+ except ValueError:
+ # empty version will cause next test to fail and status
+ # will be treated as 0.9 response.
+ version = ""
+ if not version.startswith('HTTP/'):
+ if self.strict:
+ self.close()
+ raise BadStatusLine(line)
+ else:
+ # assume it's a Simple-Response from an 0.9 server
+ self.fp = LineAndFileWrapper(line, self.fp)
+ return "HTTP/0.9", 200, ""
+
+ # The status code is a three-digit number
+ try:
+ status = int(status)
+ if status < 100 or status > 999:
+ raise BadStatusLine(line)
+ except ValueError:
+ raise BadStatusLine(line)
+ return version, status, reason
+
+ def begin(self):
+ if self.msg is not None:
+ # we've already started reading the response
+ return
+
+ # read until we get a non-100 response
+ while True:
+ version, status, reason = self._read_status()
+ if status != CONTINUE:
+ break
+ # skip the header from the 100 response
+ while True:
+ skip = self.fp.readline().strip()
+ if not skip:
+ break
+ if self.debuglevel > 0:
+ print "header:", skip
+
+ self.status = status
+ self.reason = reason.strip()
+ if version == 'HTTP/1.0':
+ self.version = 10
+ elif version.startswith('HTTP/1.'):
+ self.version = 11 # use HTTP/1.1 code for HTTP/1.x where x>=1
+ elif version == 'HTTP/0.9':
+ self.version = 9
+ else:
+ raise UnknownProtocol(version)
+
+ if self.version == 9:
+ self.length = None
+ self.chunked = 0
+ self.will_close = 1
+ self.msg = HTTPMessage(StringIO())
+ return
+
+ self.msg = HTTPMessage(self.fp, 0)
+ if self.debuglevel > 0:
+ for hdr in self.msg.headers:
+ print "header:", hdr,
+
+ # don't let the msg keep an fp
+ self.msg.fp = None
+
+ # are we using the chunked-style of transfer encoding?
+ tr_enc = self.msg.getheader('transfer-encoding')
+ if tr_enc and tr_enc.lower() == "chunked":
+ self.chunked = 1
+ self.chunk_left = None
+ else:
+ self.chunked = 0
+
+ # will the connection close at the end of the response?
+ self.will_close = self._check_close()
+
+ # do we have a Content-Length?
+ # NOTE: RFC 2616, S4.4, #3 says we ignore this if tr_enc is "chunked"
+ length = self.msg.getheader('content-length')
+ if length and not self.chunked:
+ try:
+ self.length = int(length)
+ except ValueError:
+ self.length = None
+ else:
+ if self.length < 0: # ignore nonsensical negative lengths
+ self.length = None
+ else:
+ self.length = None
+
+ # does the body have a fixed length? (of zero)
+ if (status == NO_CONTENT or status == NOT_MODIFIED or
+ 100 <= status < 200 or # 1xx codes
+ self._method == 'HEAD'):
+ self.length = 0
+
+ # if the connection remains open, and we aren't using chunked, and
+ # a content-length was not provided, then assume that the connection
+ # WILL close.
+ if not self.will_close and \
+ not self.chunked and \
+ self.length is None:
+ self.will_close = 1
+
+ def _check_close(self):
+ conn = self.msg.getheader('connection')
+ if self.version == 11:
+ # An HTTP/1.1 proxy is assumed to stay open unless
+ # explicitly closed.
+ conn = self.msg.getheader('connection')
+ if conn and "close" in conn.lower():
+ return True
+ return False
+
+ # Some HTTP/1.0 implementations have support for persistent
+ # connections, using rules different than HTTP/1.1.
+
+ # For older HTTP, Keep-Alive indicates persistent connection.
+ if self.msg.getheader('keep-alive'):
+ return False
+
+ # At least Akamai returns a "Connection: Keep-Alive" header,
+ # which was supposed to be sent by the client.
+ if conn and "keep-alive" in conn.lower():
+ return False
+
+ # Proxy-Connection is a netscape hack.
+ pconn = self.msg.getheader('proxy-connection')
+ if pconn and "keep-alive" in pconn.lower():
+ return False
+
+ # otherwise, assume it will close
+ return True
+
+ def close(self):
+ if self.fp:
+ self.fp.close()
+ self.fp = None
+
+ def isclosed(self):
+ # NOTE: it is possible that we will not ever call self.close(). This
+ # case occurs when will_close is TRUE, length is None, and we
+ # read up to the last byte, but NOT past it.
+ #
+ # IMPLIES: if will_close is FALSE, then self.close() will ALWAYS be
+ # called, meaning self.isclosed() is meaningful.
+ return self.fp is None
+
+ # XXX It would be nice to have readline and __iter__ for this, too.
+
+ def read(self, amt=None):
+ if self.fp is None:
+ return ''
+
+ if self._method == 'HEAD':
+ self.close()
+ return ''
+
+ if self.chunked:
+ return self._read_chunked(amt)
+
+ if amt is None:
+ # unbounded read
+ if self.length is None:
+ s = self.fp.read()
+ else:
+ s = self._safe_read(self.length)
+ self.length = 0
+ self.close() # we read everything
+ return s
+
+ if self.length is not None:
+ if amt > self.length:
+ # clip the read to the "end of response"
+ amt = self.length
+
+ # we do not use _safe_read() here because this may be a .will_close
+ # connection, and the user is reading more bytes than will be provided
+ # (for example, reading in 1k chunks)
+ s = self.fp.read(amt)
+ if self.length is not None:
+ self.length -= len(s)
+ if not self.length:
+ self.close()
+ return s
+
+ def _read_chunked(self, amt):
+ assert self.chunked != _UNKNOWN
+ chunk_left = self.chunk_left
+ value = []
+ while True:
+ if chunk_left is None:
+ line = self.fp.readline()
+ i = line.find(';')
+ if i >= 0:
+ line = line[:i] # strip chunk-extensions
+ try:
+ chunk_left = int(line, 16)
+ except ValueError:
+ # close the connection as protocol synchronisation is
+ # probably lost
+ self.close()
+ raise IncompleteRead(''.join(value))
+ if chunk_left == 0:
+ break
+ if amt is None:
+ value.append(self._safe_read(chunk_left))
+ elif amt < chunk_left:
+ value.append(self._safe_read(amt))
+ self.chunk_left = chunk_left - amt
+ return ''.join(value)
+ elif amt == chunk_left:
+ value.append(self._safe_read(amt))
+ self._safe_read(2) # toss the CRLF at the end of the chunk
+ self.chunk_left = None
+ return ''.join(value)
+ else:
+ value.append(self._safe_read(chunk_left))
+ amt -= chunk_left
+
+ # we read the whole chunk, get another
+ self._safe_read(2) # toss the CRLF at the end of the chunk
+ chunk_left = None
+
+ # read and discard trailer up to the CRLF terminator
+ ### note: we shouldn't have any trailers!
+ while True:
+ line = self.fp.readline()
+ if not line:
+ # a vanishingly small number of sites EOF without
+ # sending the trailer
+ break
+ if line == '\r\n':
+ break
+
+ # we read everything; close the "file"
+ self.close()
+
+ return ''.join(value)
+
+ def _safe_read(self, amt):
+ """Read the number of bytes requested, compensating for partial reads.
+
+ Normally, we have a blocking socket, but a read() can be interrupted
+ by a signal (resulting in a partial read).
+
+ Note that we cannot distinguish between EOF and an interrupt when zero
+ bytes have been read. IncompleteRead() will be raised in this
+ situation.
+
+ This function should be used when <amt> bytes "should" be present for
+ reading. If the bytes are truly not available (due to EOF), then the
+ IncompleteRead exception can be used to detect the problem.
+ """
+ # NOTE(gps): As of svn r74426 socket._fileobject.read(x) will never
+ # return less than x bytes unless EOF is encountered. It now handles
+ # signal interruptions (socket.error EINTR) internally. This code
+ # never caught that exception anyways. It seems largely pointless.
+ # self.fp.read(amt) will work fine.
+ s = []
+ while amt > 0:
+ chunk = self.fp.read(min(amt, MAXAMOUNT))
+ if not chunk:
+ raise IncompleteRead(''.join(s), amt)
+ s.append(chunk)
+ amt -= len(chunk)
+ return ''.join(s)
+
+ def fileno(self):
+ return self.fp.fileno()
+
+ def getheader(self, name, default=None):
+ if self.msg is None:
+ raise ResponseNotReady()
+ return self.msg.getheader(name, default)
+
+ def getheaders(self):
+ """Return list of (header, value) tuples."""
+ if self.msg is None:
+ raise ResponseNotReady()
+ return self.msg.items()
+
+
+class HTTPConnection:
+
+ _http_vsn = 11
+ _http_vsn_str = 'HTTP/1.1'
+
+ response_class = HTTPResponse
+ default_port = HTTP_PORT
+ auto_open = 1
+ debuglevel = 0
+ strict = 0
+
+ def __init__(self, host, port=None, strict=None,
+ timeout=socket._GLOBAL_DEFAULT_TIMEOUT, source_address=None):
+ self.timeout = timeout
+ self.source_address = source_address
+ self.sock = None
+ self._buffer = []
+ self.__response = None
+ self.__state = _CS_IDLE
+ self._method = None
+ self._tunnel_host = None
+ self._tunnel_port = None
+ self._tunnel_headers = {}
+
+ self._set_hostport(host, port)
+ if strict is not None:
+ self.strict = strict
+
+ def set_tunnel(self, host, port=None, headers=None):
+ """ Sets up the host and the port for the HTTP CONNECT Tunnelling.
+
+ The headers argument should be a mapping of extra HTTP headers
+ to send with the CONNECT request.
+ """
+ self._tunnel_host = host
+ self._tunnel_port = port
+ if headers:
+ self._tunnel_headers = headers
+ else:
+ self._tunnel_headers.clear()
+
+ def _set_hostport(self, host, port):
+ if port is None:
+ i = host.rfind(':')
+ j = host.rfind(']') # ipv6 addresses have [...]
+ if i > j:
+ try:
+ port = int(host[i+1:])
+ except ValueError:
+ raise InvalidURL("nonnumeric port: '%s'" % host[i+1:])
+ host = host[:i]
+ else:
+ port = self.default_port
+ if host and host[0] == '[' and host[-1] == ']':
+ host = host[1:-1]
+ self.host = host
+ self.port = port
+
+ def set_debuglevel(self, level):
+ self.debuglevel = level
+
+ def _tunnel(self):
+ self._set_hostport(self._tunnel_host, self._tunnel_port)
+ self.send("CONNECT %s:%d HTTP/1.0\r\n" % (self.host, self.port))
+ for header, value in self._tunnel_headers.iteritems():
+ self.send("%s: %s\r\n" % (header, value))
+ self.send("\r\n")
+ response = self.response_class(self.sock, strict = self.strict,
+ method = self._method)
+ (version, code, message) = response._read_status()
+
+ if code != 200:
+ self.close()
+ raise socket.error("Tunnel connection failed: %d %s" % (code,
+ message.strip()))
+ while True:
+ line = response.fp.readline()
+ if line == '\r\n': break
+
+
+ def connect(self):
+ """Connect to the host and port specified in __init__."""
+ self.sock = socket.create_connection((self.host,self.port),
+ self.timeout, self.source_address)
+
+ if self._tunnel_host:
+ self._tunnel()
+
+ def close(self):
+ """Close the connection to the HTTP server."""
+ if self.sock:
+ self.sock.close() # close it manually... there may be other refs
+ self.sock = None
+ if self.__response:
+ self.__response.close()
+ self.__response = None
+ self.__state = _CS_IDLE
+
+ def send(self, data):
+ """Send `data' to the server."""
+ if self.sock is None:
+ if self.auto_open:
+ self.connect()
+ else:
+ raise NotConnected()
+
+ if self.debuglevel > 0:
+ print "send:", repr(data)
+ blocksize = 8192
+ if hasattr(data,'read') and not isinstance(data, array):
+ if self.debuglevel > 0: print "sendIng a read()able"
+ datablock = data.read(blocksize)
+ while datablock:
+ self.sock.sendall(datablock)
+ datablock = data.read(blocksize)
+ else:
+ self.sock.sendall(data)
+
+ def _output(self, s):
+ """Add a line of output to the current request buffer.
+
+ Assumes that the line does *not* end with \\r\\n.
+ """
+ self._buffer.append(s)
+
+ def _send_output(self, message_body=None):
+ """Send the currently buffered request and clear the buffer.
+
+ Appends an extra \\r\\n to the buffer.
+ A message_body may be specified, to be appended to the request.
+ """
+ self._buffer.extend(("", ""))
+ msg = "\r\n".join(self._buffer)
+ del self._buffer[:]
+ # If msg and message_body are sent in a single send() call,
+ # it will avoid performance problems caused by the interaction
+ # between delayed ack and the Nagle algorithim.
+ if isinstance(message_body, str):
+ msg += message_body
+ message_body = None
+ self.send(msg)
+ if message_body is not None:
+ #message_body was not a string (i.e. it is a file) and
+ #we must run the risk of Nagle
+ self.send(message_body)
+
+ def putrequest(self, method, url, skip_host=0, skip_accept_encoding=0):
+ """Send a request to the server.
+
+ `method' specifies an HTTP request method, e.g. 'GET'.
+ `url' specifies the object being requested, e.g. '/index.html'.
+ `skip_host' if True does not add automatically a 'Host:' header
+ `skip_accept_encoding' if True does not add automatically an
+ 'Accept-Encoding:' header
+ """
+
+ # if a prior response has been completed, then forget about it.
+ if self.__response and self.__response.isclosed():
+ self.__response = None
+
+
+ # in certain cases, we cannot issue another request on this connection.
+ # this occurs when:
+ # 1) we are in the process of sending a request. (_CS_REQ_STARTED)
+ # 2) a response to a previous request has signalled that it is going
+ # to close the connection upon completion.
+ # 3) the headers for the previous response have not been read, thus
+ # we cannot determine whether point (2) is true. (_CS_REQ_SENT)
+ #
+ # if there is no prior response, then we can request at will.
+ #
+ # if point (2) is true, then we will have passed the socket to the
+ # response (effectively meaning, "there is no prior response"), and
+ # will open a new one when a new request is made.
+ #
+ # Note: if a prior response exists, then we *can* start a new request.
+ # We are not allowed to begin fetching the response to this new
+ # request, however, until that prior response is complete.
+ #
+ if self.__state == _CS_IDLE:
+ self.__state = _CS_REQ_STARTED
+ else:
+ raise CannotSendRequest()
+
+ # Save the method we use, we need it later in the response phase
+ self._method = method
+ if not url:
+ url = '/'
+ hdr = '%s %s %s' % (method, url, self._http_vsn_str)
+
+ self._output(hdr)
+
+ if self._http_vsn == 11:
+ # Issue some standard headers for better HTTP/1.1 compliance
+
+ if not skip_host:
+ # this header is issued *only* for HTTP/1.1
+ # connections. more specifically, this means it is
+ # only issued when the client uses the new
+ # HTTPConnection() class. backwards-compat clients
+ # will be using HTTP/1.0 and those clients may be
+ # issuing this header themselves. we should NOT issue
+ # it twice; some web servers (such as Apache) barf
+ # when they see two Host: headers
+
+ # If we need a non-standard port,include it in the
+ # header. If the request is going through a proxy,
+ # but the host of the actual URL, not the host of the
+ # proxy.
+
+ netloc = ''
+ if url.startswith('http'):
+ nil, netloc, nil, nil, nil = urlsplit(url)
+
+ if netloc:
+ try:
+ netloc_enc = netloc.encode("ascii")
+ except UnicodeEncodeError:
+ netloc_enc = netloc.encode("idna")
+ self.putheader('Host', netloc_enc)
+ else:
+ try:
+ host_enc = self.host.encode("ascii")
+ except UnicodeEncodeError:
+ host_enc = self.host.encode("idna")
+ # Wrap the IPv6 Host Header with [] (RFC 2732)
+ if host_enc.find(':') >= 0:
+ host_enc = "[" + host_enc + "]"
+ if self.port == self.default_port:
+ self.putheader('Host', host_enc)
+ else:
+ self.putheader('Host', "%s:%s" % (host_enc, self.port))
+
+ # note: we are assuming that clients will not attempt to set these
+ # headers since *this* library must deal with the
+ # consequences. this also means that when the supporting
+ # libraries are updated to recognize other forms, then this
+ # code should be changed (removed or updated).
+
+ # we only want a Content-Encoding of "identity" since we don't
+ # support encodings such as x-gzip or x-deflate.
+ if not skip_accept_encoding:
+ self.putheader('Accept-Encoding', 'identity')
+
+ # we can accept "chunked" Transfer-Encodings, but no others
+ # NOTE: no TE header implies *only* "chunked"
+ #self.putheader('TE', 'chunked')
+
+ # if TE is supplied in the header, then it must appear in a
+ # Connection header.
+ #self.putheader('Connection', 'TE')
+
+ else:
+ # For HTTP/1.0, the server will assume "not chunked"
+ pass
+
+ def putheader(self, header, *values):
+ """Send a request header line to the server.
+
+ For example: h.putheader('Accept', 'text/html')
+ """
+ if self.__state != _CS_REQ_STARTED:
+ raise CannotSendHeader()
+
+ hdr = '%s: %s' % (header, '\r\n\t'.join([str(v) for v in values]))
+ self._output(hdr)
+
+ def endheaders(self, message_body=None):
+ """Indicate that the last header line has been sent to the server.
+
+ This method sends the request to the server. The optional
+ message_body argument can be used to pass message body
+ associated with the request. The message body will be sent in
+ the same packet as the message headers if possible. The
+ message_body should be a string.
+ """
+ if self.__state == _CS_REQ_STARTED:
+ self.__state = _CS_REQ_SENT
+ else:
+ raise CannotSendHeader()
+ self._send_output(message_body)
+
+ def request(self, method, url, body=None, headers={}):
+ """Send a complete request to the server."""
+ self._send_request(method, url, body, headers)
+
+ def _set_content_length(self, body):
+ # Set the content-length based on the body.
+ thelen = None
+ try:
+ thelen = str(len(body))
+ except TypeError, te:
+ # If this is a file-like object, try to
+ # fstat its file descriptor
+ try:
+ thelen = str(os.fstat(body.fileno()).st_size)
+ except (AttributeError, OSError):
+ # Don't send a length if this failed
+ if self.debuglevel > 0: print "Cannot stat!!"
+
+ if thelen is not None:
+ self.putheader('Content-Length', thelen)
+
+ def _send_request(self, method, url, body, 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:
+ skips['skip_host'] = 1
+ if 'accept-encoding' in header_names:
+ skips['skip_accept_encoding'] = 1
+
+ self.putrequest(method, url, **skips)
+
+ if body and ('content-length' not in header_names):
+ self._set_content_length(body)
+ for hdr, value in headers.iteritems():
+ self.putheader(hdr, value)
+ self.endheaders(body)
+
+ def getresponse(self, buffering=False):
+ "Get the response from the server."
+
+ # if a prior response has been completed, then forget about it.
+ if self.__response and self.__response.isclosed():
+ self.__response = None
+
+ #
+ # if a prior response exists, then it must be completed (otherwise, we
+ # cannot read this response's header to determine the connection-close
+ # behavior)
+ #
+ # note: if a prior response existed, but was connection-close, then the
+ # socket and response were made independent of this HTTPConnection
+ # object since a new request requires that we open a whole new
+ # connection
+ #
+ # this means the prior response had one of two states:
+ # 1) will_close: this connection was reset and the prior socket and
+ # response operate independently
+ # 2) persistent: the response was retained and we await its
+ # isclosed() status to become true.
+ #
+ if self.__state != _CS_REQ_SENT or self.__response:
+ raise ResponseNotReady()
+
+ args = (self.sock,)
+ kwds = {"strict":self.strict, "method":self._method}
+ if self.debuglevel > 0:
+ args += (self.debuglevel,)
+ if buffering:
+ #only add this keyword if non-default, for compatibility with
+ #other response_classes.
+ kwds["buffering"] = True;
+ response = self.response_class(*args, **kwds)
+
+ try:
+ response.begin()
+ except:
+ response.close()
+ raise
+ assert response.will_close != _UNKNOWN
+ self.__state = _CS_IDLE
+
+ if response.will_close:
+ # this effectively passes the connection to the response
+ self.close()
+ else:
+ # remember this, so we can tell when it is complete
+ self.__response = response
+
+ return response
+
+
+class HTTP:
+ "Compatibility class with httplib.py from 1.5."
+
+ _http_vsn = 10
+ _http_vsn_str = 'HTTP/1.0'
+
+ debuglevel = 0
+
+ _connection_class = HTTPConnection
+
+ def __init__(self, host='', port=None, strict=None):
+ "Provide a default host, since the superclass requires one."
+
+ # some joker passed 0 explicitly, meaning default port
+ if port == 0:
+ port = None
+
+ # Note that we may pass an empty string as the host; this will throw
+ # an error when we attempt to connect. Presumably, the client code
+ # will call connect before then, with a proper host.
+ self._setup(self._connection_class(host, port, strict))
+
+ def _setup(self, conn):
+ self._conn = conn
+
+ # set up delegation to flesh out interface
+ self.send = conn.send
+ self.putrequest = conn.putrequest
+ self.putheader = conn.putheader
+ self.endheaders = conn.endheaders
+ self.set_debuglevel = conn.set_debuglevel
+
+ conn._http_vsn = self._http_vsn
+ conn._http_vsn_str = self._http_vsn_str
+
+ self.file = None
+
+ def connect(self, host=None, port=None):
+ "Accept arguments to set the host/port, since the superclass doesn't."
+
+ if host is not None:
+ self._conn._set_hostport(host, port)
+ self._conn.connect()
+
+ def getfile(self):
+ "Provide a getfile, since the superclass' does not use this concept."
+ return self.file
+
+ def getreply(self, buffering=False):
+ """Compat definition since superclass does not define it.
+
+ Returns a tuple consisting of:
+ - server status code (e.g. '200' if all goes well)
+ - server "reason" corresponding to status code
+ - any RFC822 headers in the response from the server
+ """
+ try:
+ if not buffering:
+ response = self._conn.getresponse()
+ else:
+ #only add this keyword if non-default for compatibility
+ #with other connection classes
+ response = self._conn.getresponse(buffering)
+ except BadStatusLine, e:
+ ### hmm. if getresponse() ever closes the socket on a bad request,
+ ### then we are going to have problems with self.sock
+
+ ### should we keep this behavior? do people use it?
+ # keep the socket open (as a file), and return it
+ self.file = self._conn.sock.makefile('rb', 0)
+
+ # close our socket -- we want to restart after any protocol error
+ self.close()
+
+ self.headers = None
+ return -1, e.line, None
+
+ self.headers = response.msg
+ self.file = response.fp
+ return response.status, response.reason, response.msg
+
+ def close(self):
+ self._conn.close()
+
+ # note that self.file == response.fp, which gets closed by the
+ # superclass. just clear the object ref here.
+ ### hmm. messy. if status==-1, then self.file is owned by us.
+ ### well... we aren't explicitly closing, but losing this ref will
+ ### do it
+ self.file = None
+
+try:
+ import ssl
+except ImportError:
+ pass
+else:
+ class HTTPSConnection(HTTPConnection):
+ "This class allows communication via SSL."
+
+ default_port = HTTPS_PORT
+
+ def __init__(self, host, port=None, key_file=None, cert_file=None,
+ strict=None, timeout=socket._GLOBAL_DEFAULT_TIMEOUT,
+ source_address=None):
+ HTTPConnection.__init__(self, host, port, strict, timeout,
+ source_address)
+ self.key_file = key_file
+ self.cert_file = cert_file
+
+ def connect(self):
+ "Connect to a host on a given (SSL) port."
+
+ sock = socket.create_connection((self.host, self.port),
+ self.timeout, self.source_address)
+ if self._tunnel_host:
+ self.sock = sock
+ self._tunnel()
+ self.sock = ssl.wrap_socket(sock, self.key_file, self.cert_file)
+
+ __all__.append("HTTPSConnection")
+
+ class HTTPS(HTTP):
+ """Compatibility with 1.5 httplib interface
+
+ Python 1.5.2 did not have an HTTPS class, but it defined an
+ interface for sending http requests that is also useful for
+ https.
+ """
+
+ _connection_class = HTTPSConnection
+
+ def __init__(self, host='', port=None, key_file=None, cert_file=None,
+ strict=None):
+ # provide a default host, pass the X509 cert info
+
+ # urf. compensate for bad input.
+ if port == 0:
+ port = None
+ self._setup(self._connection_class(host, port, key_file,
+ cert_file, strict))
+
+ # we never actually use these for anything, but we keep them
+ # here for compatibility with post-1.5.2 CVS.
+ self.key_file = key_file
+ self.cert_file = cert_file
+
+
+ def FakeSocket (sock, sslobj):
+ warnings.warn("FakeSocket is deprecated, and won't be in 3.x. " +
+ "Use the result of ssl.wrap_socket() directly instead.",
+ DeprecationWarning, stacklevel=2)
+ return sslobj
+
+
+class HTTPException(Exception):
+ # Subclasses that define an __init__ must call Exception.__init__
+ # or define self.args. Otherwise, str() will fail.
+ pass
+
+class NotConnected(HTTPException):
+ pass
+
+class InvalidURL(HTTPException):
+ pass
+
+class UnknownProtocol(HTTPException):
+ def __init__(self, version):
+ self.args = version,
+ self.version = version
+
+class UnknownTransferEncoding(HTTPException):
+ pass
+
+class UnimplementedFileMode(HTTPException):
+ pass
+
+class IncompleteRead(HTTPException):
+ def __init__(self, partial, expected=None):
+ self.args = partial,
+ self.partial = partial
+ self.expected = expected
+ def __repr__(self):
+ if self.expected is not None:
+ e = ', %i more expected' % self.expected
+ else:
+ e = ''
+ return 'IncompleteRead(%i bytes read%s)' % (len(self.partial), e)
+ def __str__(self):
+ return repr(self)
+
+class ImproperConnectionState(HTTPException):
+ pass
+
+class CannotSendRequest(ImproperConnectionState):
+ pass
+
+class CannotSendHeader(ImproperConnectionState):
+ pass
+
+class ResponseNotReady(ImproperConnectionState):
+ pass
+
+class BadStatusLine(HTTPException):
+ def __init__(self, line):
+ if not line:
+ line = repr(line)
+ self.args = line,
+ self.line = line
+
+# for backwards compatibility
+error = HTTPException
+
+class LineAndFileWrapper:
+ """A limited file-like object for HTTP/0.9 responses."""
+
+ # The status-line parsing code calls readline(), which normally
+ # get the HTTP status line. For a 0.9 response, however, this is
+ # actually the first line of the body! Clients need to get a
+ # readable file object that contains that line.
+
+ def __init__(self, line, file):
+ self._line = line
+ self._file = file
+ self._line_consumed = 0
+ self._line_offset = 0
+ self._line_left = len(line)
+
+ def __getattr__(self, attr):
+ return getattr(self._file, attr)
+
+ def _done(self):
+ # called when the last byte is read from the line. After the
+ # call, all read methods are delegated to the underlying file
+ # object.
+ self._line_consumed = 1
+ self.read = self._file.read
+ self.readline = self._file.readline
+ self.readlines = self._file.readlines
+
+ def read(self, amt=None):
+ if self._line_consumed:
+ return self._file.read(amt)
+ assert self._line_left
+ if amt is None or amt > self._line_left:
+ s = self._line[self._line_offset:]
+ self._done()
+ if amt is None:
+ return s + self._file.read()
+ else:
+ return s + self._file.read(amt - len(s))
+ else:
+ assert amt <= self._line_left
+ i = self._line_offset
+ j = i + amt
+ s = self._line[i:j]
+ self._line_offset = j
+ self._line_left -= amt
+ if self._line_left == 0:
+ self._done()
+ return s
+
+ def readline(self):
+ if self._line_consumed:
+ return self._file.readline()
+ assert self._line_left
+ s = self._line[self._line_offset:]
+ self._done()
+ return s
+
+ def readlines(self, size=None):
+ if self._line_consumed:
+ return self._file.readlines(size)
+ assert self._line_left
+ L = [self._line[self._line_offset:]]
+ self._done()
+ if size is None:
+ return L + self._file.readlines()
+ else:
+ return L + self._file.readlines(size)
+
+def test():
+ """Test this module.
+
+ A hodge podge of tests collected here, because they have too many
+ external dependencies for the regular test suite.
+ """
+
+ import sys
+ import getopt
+ opts, args = getopt.getopt(sys.argv[1:], 'd')
+ dl = 0
+ for o, a in opts:
+ if o == '-d': dl = dl + 1
+ host = 'www.python.org'
+ selector = '/'
+ if args[0:]: host = args[0]
+ if args[1:]: selector = args[1]
+ h = HTTP()
+ h.set_debuglevel(dl)
+ h.connect(host)
+ h.putrequest('GET', selector)
+ h.endheaders()
+ status, reason, headers = h.getreply()
+ print 'status =', status
+ print 'reason =', reason
+ print "read", len(h.getfile().read())
+ print
+ if headers:
+ for header in headers.headers: print header.strip()
+ print
+
+ # minimal test that code to extract host from url works
+ class HTTP11(HTTP):
+ _http_vsn = 11
+ _http_vsn_str = 'HTTP/1.1'
+
+ h = HTTP11('www.python.org')
+ h.putrequest('GET', 'http://www.python.org/~jeremy/')
+ h.endheaders()
+ h.getreply()
+ h.close()
+
+ try:
+ import ssl
+ except ImportError:
+ pass
+ else:
+
+ for host, selector in (('sourceforge.net', '/projects/python'),
+ ):
+ print "https://%s%s" % (host, selector)
+ hs = HTTPS()
+ hs.set_debuglevel(dl)
+ hs.connect(host)
+ hs.putrequest('GET', selector)
+ hs.endheaders()
+ status, reason, headers = hs.getreply()
+ print 'status =', status
+ print 'reason =', reason
+ print "read", len(hs.getfile().read())
+ print
+ if headers:
+ for header in headers.headers: print header.strip()
+ print
+
+if __name__ == '__main__':
+ test()
diff --git a/lib-python/modified-2.7/json/encoder.py b/lib-python/modified-2.7/json/encoder.py
--- a/lib-python/modified-2.7/json/encoder.py
+++ b/lib-python/modified-2.7/json/encoder.py
@@ -2,14 +2,7 @@
"""
import re
-try:
- from _json import encode_basestring_ascii as c_encode_basestring_ascii
-except ImportError:
- c_encode_basestring_ascii = None
-try:
- from _json import make_encoder as c_make_encoder
-except ImportError:
- c_make_encoder = None
+from __pypy__.builders import StringBuilder, UnicodeBuilder
ESCAPE = re.compile(r'[\x00-\x1f\\"\b\f\n\r\t]')
ESCAPE_ASCII = re.compile(r'([\\"]|[^\ -~])')
@@ -24,8 +17,7 @@
'\t': '\\t',
}
for i in range(0x20):
- ESCAPE_DCT.setdefault(chr(i), '\\u{0:04x}'.format(i))
- #ESCAPE_DCT.setdefault(chr(i), '\\u%04x' % (i,))
+ ESCAPE_DCT.setdefault(chr(i), '\\u%04x' % (i,))
# Assume this produces an infinity on all machines (probably not guaranteed)
INFINITY = float('1e66666')
@@ -37,10 +29,9 @@
"""
def replace(match):
return ESCAPE_DCT[match.group(0)]
- return '"' + ESCAPE.sub(replace, s) + '"'
+ return ESCAPE.sub(replace, s)
-
-def py_encode_basestring_ascii(s):
+def encode_basestring_ascii(s):
"""Return an ASCII-only JSON representation of a Python string
"""
@@ -53,20 +44,18 @@
except KeyError:
n = ord(s)
if n < 0x10000:
- return '\\u{0:04x}'.format(n)
- #return '\\u%04x' % (n,)
+ return '\\u%04x' % (n,)
else:
# surrogate pair
n -= 0x10000
s1 = 0xd800 | ((n >> 10) & 0x3ff)
s2 = 0xdc00 | (n & 0x3ff)
- return '\\u{0:04x}\\u{1:04x}'.format(s1, s2)
- #return '\\u%04x\\u%04x' % (s1, s2)
- return '"' + str(ESCAPE_ASCII.sub(replace, s)) + '"'
-
-
-encode_basestring_ascii = (
- c_encode_basestring_ascii or py_encode_basestring_ascii)
+ return '\\u%04x\\u%04x' % (s1, s2)
+ if ESCAPE_ASCII.search(s):
+ return str(ESCAPE_ASCII.sub(replace, s))
+ return s
+py_encode_basestring_ascii = lambda s: '"' + encode_basestring_ascii(s) + '"'
+c_encode_basestring_ascii = None
class JSONEncoder(object):
"""Extensible JSON <http://json.org> encoder for Python data structures.
@@ -147,6 +136,17 @@
self.skipkeys = skipkeys
self.ensure_ascii = ensure_ascii
+ if ensure_ascii:
+ self.encoder = encode_basestring_ascii
+ else:
+ self.encoder = encode_basestring
+ if encoding != 'utf-8':
+ orig_encoder = self.encoder
+ def encoder(o):
+ if isinstance(o, str):
+ o = o.decode(encoding)
+ return orig_encoder(o)
+ self.encoder = encoder
self.check_circular = check_circular
self.allow_nan = allow_nan
self.sort_keys = sort_keys
@@ -184,24 +184,126 @@
'{"foo": ["bar", "baz"]}'
"""
- # This is for extremely simple cases and benchmarks.
+ if self.check_circular:
+ markers = {}
+ else:
+ markers = None
+ if self.ensure_ascii:
+ builder = StringBuilder()
+ else:
+ builder = UnicodeBuilder()
+ self._encode(o, markers, builder, 0)
+ return builder.build()
+
+ def _emit_indent(self, builder, _current_indent_level):
+ if self.indent is not None:
+ _current_indent_level += 1
+ newline_indent = '\n' + (' ' * (self.indent *
+ _current_indent_level))
+ separator = self.item_separator + newline_indent
+ builder.append(newline_indent)
+ else:
+ separator = self.item_separator
+ return separator, _current_indent_level
+
+ def _emit_unindent(self, builder, _current_indent_level):
+ if self.indent is not None:
+ builder.append('\n')
+ builder.append(' ' * (self.indent * (_current_indent_level - 1)))
+
+ def _encode(self, o, markers, builder, _current_indent_level):
if isinstance(o, basestring):
- if isinstance(o, str):
- _encoding = self.encoding
- if (_encoding is not None
- and not (_encoding == 'utf-8')):
- o = o.decode(_encoding)
- if self.ensure_ascii:
- return encode_basestring_ascii(o)
+ builder.append('"')
+ builder.append(self.encoder(o))
+ builder.append('"')
+ elif o is None:
+ builder.append('null')
+ elif o is True:
+ builder.append('true')
+ elif o is False:
+ builder.append('false')
+ elif isinstance(o, (int, long)):
+ builder.append(str(o))
+ elif isinstance(o, float):
+ builder.append(self._floatstr(o))
+ elif isinstance(o, (list, tuple)):
+ if not o:
+ builder.append('[]')
+ return
+ self._encode_list(o, markers, builder, _current_indent_level)
+ elif isinstance(o, dict):
+ if not o:
+ builder.append('{}')
+ return
+ self._encode_dict(o, markers, builder, _current_indent_level)
+ else:
+ self._mark_markers(markers, o)
+ res = self.default(o)
+ self._encode(res, markers, builder, _current_indent_level)
+ self._remove_markers(markers, o)
+ return res
+
+ def _encode_list(self, l, markers, builder, _current_indent_level):
+ self._mark_markers(markers, l)
+ builder.append('[')
+ first = True
+ separator, _current_indent_level = self._emit_indent(builder,
+ _current_indent_level)
+ for elem in l:
+ if first:
+ first = False
else:
- return encode_basestring(o)
- # This doesn't pass the iterator directly to ''.join() because the
- # exceptions aren't as detailed. The list call should be roughly
- # equivalent to the PySequence_Fast that ''.join() would do.
- chunks = self.iterencode(o, _one_shot=True)
- if not isinstance(chunks, (list, tuple)):
- chunks = list(chunks)
- return ''.join(chunks)
+ builder.append(separator)
+ self._encode(elem, markers, builder, _current_indent_level)
+ del elem # XXX grumble
+ self._emit_unindent(builder, _current_indent_level)
+ builder.append(']')
+ self._remove_markers(markers, l)
+
+ def _encode_dict(self, d, markers, builder, _current_indent_level):
+ self._mark_markers(markers, d)
+ first = True
+ builder.append('{')
+ separator, _current_indent_level = self._emit_indent(builder,
+ _current_indent_level)
+ if self.sort_keys:
+ items = sorted(d.items(), key=lambda kv: kv[0])
+ else:
+ items = d.iteritems()
+
+ for key, v in items:
+ if first:
+ first = False
+ else:
+ builder.append(separator)
+ if isinstance(key, basestring):
+ pass
+ # JavaScript is weakly typed for these, so it makes sense to
+ # also allow them. Many encoders seem to do something like this.
+ elif isinstance(key, float):
+ key = self._floatstr(key)
+ elif key is True:
+ key = 'true'
+ elif key is False:
+ key = 'false'
+ elif key is None:
+ key = 'null'
+ elif isinstance(key, (int, long)):
+ key = str(key)
+ elif self.skipkeys:
+ continue
+ else:
+ raise TypeError("key " + repr(key) + " is not a string")
+ builder.append('"')
+ builder.append(self.encoder(key))
+ builder.append('"')
+ builder.append(self.key_separator)
+ self._encode(v, markers, builder, _current_indent_level)
+ del key
+ del v # XXX grumble
+ self._emit_unindent(builder, _current_indent_level)
+ builder.append('}')
+ self._remove_markers(markers, d)
def iterencode(self, o, _one_shot=False):
"""Encode the given object and yield each string
@@ -217,86 +319,54 @@
markers = {}
else:
markers = None
- if self.ensure_ascii:
- _encoder = encode_basestring_ascii
+ return self._iterencode(o, markers, 0)
+
+ def _floatstr(self, o):
+ # Check for specials. Note that this type of test is processor
+ # and/or platform-specific, so do tests which don't depend on the
+ # internals.
+
+ if o != o:
+ text = 'NaN'
+ elif o == INFINITY:
+ text = 'Infinity'
+ elif o == -INFINITY:
+ text = '-Infinity'
else:
- _encoder = encode_basestring
- if self.encoding != 'utf-8':
- def _encoder(o, _orig_encoder=_encoder, _encoding=self.encoding):
- if isinstance(o, str):
- o = o.decode(_encoding)
- return _orig_encoder(o)
+ return FLOAT_REPR(o)
- def floatstr(o, allow_nan=self.allow_nan,
- _repr=FLOAT_REPR, _inf=INFINITY, _neginf=-INFINITY):
- # Check for specials. Note that this type of test is processor
- # and/or platform-specific, so do tests which don't depend on the
- # internals.
+ if not self.allow_nan:
+ raise ValueError(
+ "Out of range float values are not JSON compliant: " +
+ repr(o))
- if o != o:
- text = 'NaN'
- elif o == _inf:
- text = 'Infinity'
- elif o == _neginf:
- text = '-Infinity'
- else:
- return _repr(o)
+ return text
- if not allow_nan:
- raise ValueError(
- "Out of range float values are not JSON compliant: " +
- repr(o))
+ def _mark_markers(self, markers, o):
+ if markers is not None:
+ if id(o) in markers:
+ raise ValueError("Circular reference detected")
+ markers[id(o)] = None
- return text
+ def _remove_markers(self, markers, o):
+ if markers is not None:
+ del markers[id(o)]
-
- if (_one_shot and c_make_encoder is not None
- and not self.indent and not self.sort_keys):
- _iterencode = c_make_encoder(
- markers, self.default, _encoder, self.indent,
- self.key_separator, self.item_separator, self.sort_keys,
- self.skipkeys, self.allow_nan)
- else:
- _iterencode = _make_iterencode(
- markers, self.default, _encoder, self.indent, floatstr,
- self.key_separator, self.item_separator, self.sort_keys,
- self.skipkeys, _one_shot)
- return _iterencode(o, 0)
-
-def _make_iterencode(markers, _default, _encoder, _indent, _floatstr,
- _key_separator, _item_separator, _sort_keys, _skipkeys, _one_shot,
- ## HACK: hand-optimized bytecode; turn globals into locals
- ValueError=ValueError,
- basestring=basestring,
- dict=dict,
- float=float,
- id=id,
- int=int,
- isinstance=isinstance,
- list=list,
- long=long,
- str=str,
- tuple=tuple,
- ):
-
- def _iterencode_list(lst, _current_indent_level):
+ def _iterencode_list(self, lst, markers, _current_indent_level):
if not lst:
yield '[]'
return
- if markers is not None:
- markerid = id(lst)
- if markerid in markers:
- raise ValueError("Circular reference detected")
- markers[markerid] = lst
+ self._mark_markers(markers, lst)
buf = '['
- if _indent is not None:
+ if self.indent is not None:
_current_indent_level += 1
- newline_indent = '\n' + (' ' * (_indent * _current_indent_level))
- separator = _item_separator + newline_indent
+ newline_indent = '\n' + (' ' * (self.indent *
+ _current_indent_level))
+ separator = self.item_separator + newline_indent
buf += newline_indent
else:
newline_indent = None
- separator = _item_separator
+ separator = self.item_separator
first = True
for value in lst:
if first:
@@ -304,7 +374,7 @@
else:
buf = separator
if isinstance(value, basestring):
- yield buf + _encoder(value)
+ yield buf + '"' + self.encoder(value) + '"'
elif value is None:
yield buf + 'null'
elif value is True:
@@ -314,44 +384,43 @@
elif isinstance(value, (int, long)):
yield buf + str(value)
elif isinstance(value, float):
- yield buf + _floatstr(value)
+ yield buf + self._floatstr(value)
else:
yield buf
if isinstance(value, (list, tuple)):
- chunks = _iterencode_list(value, _current_indent_level)
+ chunks = self._iterencode_list(value, markers,
+ _current_indent_level)
elif isinstance(value, dict):
- chunks = _iterencode_dict(value, _current_indent_level)
+ chunks = self._iterencode_dict(value, markers,
+ _current_indent_level)
else:
- chunks = _iterencode(value, _current_indent_level)
+ chunks = self._iterencode(value, markers,
+ _current_indent_level)
for chunk in chunks:
yield chunk
if newline_indent is not None:
_current_indent_level -= 1
- yield '\n' + (' ' * (_indent * _current_indent_level))
+ yield '\n' + (' ' * (self.indent * _current_indent_level))
yield ']'
- if markers is not None:
- del markers[markerid]
+ self._remove_markers(markers, lst)
- def _iterencode_dict(dct, _current_indent_level):
+ def _iterencode_dict(self, dct, markers, _current_indent_level):
if not dct:
yield '{}'
return
- if markers is not None:
- markerid = id(dct)
- if markerid in markers:
- raise ValueError("Circular reference detected")
- markers[markerid] = dct
+ self._mark_markers(markers, dct)
yield '{'
- if _indent is not None:
+ if self.indent is not None:
_current_indent_level += 1
- newline_indent = '\n' + (' ' * (_indent * _current_indent_level))
- item_separator = _item_separator + newline_indent
+ newline_indent = '\n' + (' ' * (self.indent *
+ _current_indent_level))
+ item_separator = self.item_separator + newline_indent
yield newline_indent
else:
newline_indent = None
- item_separator = _item_separator
+ item_separator = self.item_separator
first = True
- if _sort_keys:
+ if self.sort_keys:
items = sorted(dct.items(), key=lambda kv: kv[0])
else:
items = dct.iteritems()
@@ -361,7 +430,7 @@
# JavaScript is weakly typed for these, so it makes sense to
# also allow them. Many encoders seem to do something like this.
elif isinstance(key, float):
- key = _floatstr(key)
+ key = self._floatstr(key)
elif key is True:
key = 'true'
elif key is False:
@@ -370,7 +439,7 @@
key = 'null'
elif isinstance(key, (int, long)):
key = str(key)
- elif _skipkeys:
+ elif self.skipkeys:
continue
else:
raise TypeError("key " + repr(key) + " is not a string")
@@ -378,10 +447,10 @@
first = False
else:
yield item_separator
- yield _encoder(key)
- yield _key_separator
+ yield '"' + self.encoder(key) + '"'
+ yield self.key_separator
if isinstance(value, basestring):
- yield _encoder(value)
+ yield '"' + self.encoder(value) + '"'
elif value is None:
yield 'null'
elif value is True:
@@ -391,26 +460,28 @@
elif isinstance(value, (int, long)):
yield str(value)
elif isinstance(value, float):
- yield _floatstr(value)
+ yield self._floatstr(value)
else:
if isinstance(value, (list, tuple)):
- chunks = _iterencode_list(value, _current_indent_level)
+ chunks = self._iterencode_list(value, markers,
+ _current_indent_level)
elif isinstance(value, dict):
- chunks = _iterencode_dict(value, _current_indent_level)
+ chunks = self._iterencode_dict(value, markers,
+ _current_indent_level)
else:
- chunks = _iterencode(value, _current_indent_level)
+ chunks = self._iterencode(value, markers,
+ _current_indent_level)
for chunk in chunks:
yield chunk
if newline_indent is not None:
_current_indent_level -= 1
- yield '\n' + (' ' * (_indent * _current_indent_level))
+ yield '\n' + (' ' * (self.indent * _current_indent_level))
yield '}'
- if markers is not None:
- del markers[markerid]
+ self._remove_markers(markers, dct)
- def _iterencode(o, _current_indent_level):
+ def _iterencode(self, o, markers, _current_indent_level):
if isinstance(o, basestring):
- yield _encoder(o)
+ yield '"' + self.encoder(o) + '"'
elif o is None:
yield 'null'
elif o is True:
@@ -420,23 +491,19 @@
elif isinstance(o, (int, long)):
yield str(o)
elif isinstance(o, float):
- yield _floatstr(o)
+ yield self._floatstr(o)
elif isinstance(o, (list, tuple)):
- for chunk in _iterencode_list(o, _current_indent_level):
+ for chunk in self._iterencode_list(o, markers,
+ _current_indent_level):
yield chunk
elif isinstance(o, dict):
- for chunk in _iterencode_dict(o, _current_indent_level):
+ for chunk in self._iterencode_dict(o, markers,
+ _current_indent_level):
yield chunk
else:
- if markers is not None:
- markerid = id(o)
- if markerid in markers:
- raise ValueError("Circular reference detected")
- markers[markerid] = o
- o = _default(o)
- for chunk in _iterencode(o, _current_indent_level):
+ self._mark_markers(markers, o)
+ obj = self.default(o)
+ for chunk in self._iterencode(obj, markers,
+ _current_indent_level):
yield chunk
- if markers is not None:
- del markers[markerid]
-
- return _iterencode
+ self._remove_markers(markers, o)
diff --git a/lib-python/modified-2.7/json/tests/test_unicode.py b/lib-python/modified-2.7/json/tests/test_unicode.py
--- a/lib-python/modified-2.7/json/tests/test_unicode.py
+++ b/lib-python/modified-2.7/json/tests/test_unicode.py
@@ -80,3 +80,9 @@
self.assertEqual(type(json.loads(u'["a"]')[0]), unicode)
# Issue 10038.
self.assertEqual(type(json.loads('"foo"')), unicode)
+
+ def test_encode_not_utf_8(self):
+ self.assertEqual(json.dumps('\xb1\xe6', encoding='iso8859-2'),
+ '"\\u0105\\u0107"')
+ self.assertEqual(json.dumps(['\xb1\xe6'], encoding='iso8859-2'),
+ '["\\u0105\\u0107"]')
diff --git a/lib-python/modified-2.7/sqlite3/test/regression.py b/lib-python/modified-2.7/sqlite3/test/regression.py
--- a/lib-python/modified-2.7/sqlite3/test/regression.py
+++ b/lib-python/modified-2.7/sqlite3/test/regression.py
@@ -274,6 +274,18 @@
cur.execute("UPDATE foo SET id = 3 WHERE id = 1")
self.assertEqual(cur.description, None)
+ def CheckStatementCache(self):
+ cur = self.con.cursor()
+ cur.execute("CREATE TABLE foo (id INTEGER)")
+ values = [(i,) for i in xrange(5)]
+ cur.executemany("INSERT INTO foo (id) VALUES (?)", values)
+
+ cur.execute("SELECT id FROM foo")
+ self.assertEqual(list(cur), values)
+ self.con.commit()
+ cur.execute("SELECT id FROM foo")
+ self.assertEqual(list(cur), values)
+
def suite():
regression_suite = unittest.makeSuite(RegressionTests, "Check")
return unittest.TestSuite((regression_suite,))
diff --git a/lib-python/modified-2.7/ssl.py b/lib-python/modified-2.7/ssl.py
--- a/lib-python/modified-2.7/ssl.py
+++ b/lib-python/modified-2.7/ssl.py
@@ -62,7 +62,6 @@
from _ssl import OPENSSL_VERSION_NUMBER, OPENSSL_VERSION_INFO, OPENSSL_VERSION
from _ssl import SSLError
from _ssl import CERT_NONE, CERT_OPTIONAL, CERT_REQUIRED
-from _ssl import PROTOCOL_SSLv2, PROTOCOL_SSLv3, PROTOCOL_SSLv23, PROTOCOL_TLSv1
from _ssl import RAND_status, RAND_egd, RAND_add
from _ssl import \
SSL_ERROR_ZERO_RETURN, \
@@ -74,6 +73,18 @@
SSL_ERROR_WANT_CONNECT, \
SSL_ERROR_EOF, \
SSL_ERROR_INVALID_ERROR_CODE
+from _ssl import PROTOCOL_SSLv3, PROTOCOL_SSLv23, PROTOCOL_TLSv1
+_PROTOCOL_NAMES = {
+ PROTOCOL_TLSv1: "TLSv1",
+ PROTOCOL_SSLv23: "SSLv23",
+ PROTOCOL_SSLv3: "SSLv3",
+}
+try:
+ from _ssl import PROTOCOL_SSLv2
+except ImportError:
+ pass
+else:
+ _PROTOCOL_NAMES[PROTOCOL_SSLv2] = "SSLv2"
from socket import socket, _fileobject, error as socket_error
from socket import getnameinfo as _getnameinfo
@@ -400,16 +411,7 @@
return DER_cert_to_PEM_cert(dercert)
def get_protocol_name(protocol_code):
- if protocol_code == PROTOCOL_TLSv1:
- return "TLSv1"
- elif protocol_code == PROTOCOL_SSLv23:
- return "SSLv23"
- elif protocol_code == PROTOCOL_SSLv2:
- return "SSLv2"
- elif protocol_code == PROTOCOL_SSLv3:
- return "SSLv3"
- else:
- return "<unknown>"
+ return _PROTOCOL_NAMES.get(protocol_code, '<unknown>')
# a replacement for the old socket.ssl function
diff --git a/lib-python/modified-2.7/test/regrtest.py b/lib-python/modified-2.7/test/regrtest.py
--- a/lib-python/modified-2.7/test/regrtest.py
+++ b/lib-python/modified-2.7/test/regrtest.py
@@ -1403,7 +1403,26 @@
test_zipimport
test_zlib
""",
- 'openbsd3':
+ 'openbsd4':
+ """
+ test_ascii_formatd
+ test_bsddb
+ test_bsddb3
+ test_ctypes
+ test_dl
+ test_epoll
+ test_gdbm
+ test_locale
+ test_normalization
+ test_ossaudiodev
+ test_pep277
+ test_tcl
+ test_tk
+ test_ttk_guionly
+ test_ttk_textonly
+ test_multiprocessing
+ """,
+ 'openbsd5':
"""
test_ascii_formatd
test_bsddb
diff --git a/lib-python/modified-2.7/test/test_array.py b/lib-python/modified-2.7/test/test_array.py
--- a/lib-python/modified-2.7/test/test_array.py
+++ b/lib-python/modified-2.7/test/test_array.py
@@ -295,9 +295,10 @@
)
b = array.array(self.badtypecode())
- self.assertRaises(TypeError, "a + b")
-
- self.assertRaises(TypeError, "a + 'bad'")
+ with self.assertRaises(TypeError):
+ a + b
+ with self.assertRaises(TypeError):
+ a + 'bad'
def test_iadd(self):
a = array.array(self.typecode, self.example[::-1])
@@ -316,9 +317,10 @@
)
b = array.array(self.badtypecode())
- self.assertRaises(TypeError, "a += b")
-
- self.assertRaises(TypeError, "a += 'bad'")
+ with self.assertRaises(TypeError):
+ a += b
+ with self.assertRaises(TypeError):
+ a += 'bad'
def test_mul(self):
a = 5*array.array(self.typecode, self.example)
@@ -345,7 +347,8 @@
array.array(self.typecode)
)
- self.assertRaises(TypeError, "a * 'bad'")
+ with self.assertRaises(TypeError):
+ a * 'bad'
def test_imul(self):
a = array.array(self.typecode, self.example)
@@ -374,7 +377,8 @@
a *= -1
self.assertEqual(a, array.array(self.typecode))
- self.assertRaises(TypeError, "a *= 'bad'")
+ with self.assertRaises(TypeError):
+ a *= 'bad'
def test_getitem(self):
a = array.array(self.typecode, self.example)
diff --git a/lib-python/modified-2.7/test/test_bz2.py b/lib-python/modified-2.7/test/test_bz2.py
--- a/lib-python/modified-2.7/test/test_bz2.py
+++ b/lib-python/modified-2.7/test/test_bz2.py
@@ -50,6 +50,7 @@
self.filename = TESTFN
def tearDown(self):
+ test_support.gc_collect()
if os.path.isfile(self.filename):
os.unlink(self.filename)
diff --git a/lib-python/modified-2.7/test/test_fcntl.py b/lib-python/modified-2.7/test/test_fcntl.py
new file mode 100644
--- /dev/null
+++ b/lib-python/modified-2.7/test/test_fcntl.py
@@ -0,0 +1,108 @@
+"""Test program for the fcntl C module.
+
+OS/2+EMX doesn't support the file locking operations.
+
+"""
+import os
+import struct
+import sys
+import unittest
+from test.test_support import (verbose, TESTFN, unlink, run_unittest,
+ import_module)
+
+# Skip test if no fnctl module.
+fcntl = import_module('fcntl')
+
+
+# TODO - Write tests for flock() and lockf().
+
+def get_lockdata():
+ if sys.platform.startswith('atheos'):
+ start_len = "qq"
+ else:
+ try:
+ os.O_LARGEFILE
+ except AttributeError:
+ start_len = "ll"
+ else:
+ start_len = "qq"
+
+ if sys.platform in ('netbsd1', 'netbsd2', 'netbsd3',
+ 'Darwin1.2', 'darwin',
+ 'freebsd2', 'freebsd3', 'freebsd4', 'freebsd5',
+ 'freebsd6', 'freebsd7', 'freebsd8',
+ 'bsdos2', 'bsdos3', 'bsdos4',
+ 'openbsd', 'openbsd2', 'openbsd3', 'openbsd4', 'openbsd5'):
+ if struct.calcsize('l') == 8:
+ off_t = 'l'
+ pid_t = 'i'
+ else:
+ off_t = 'lxxxx'
+ pid_t = 'l'
+ lockdata = struct.pack(off_t + off_t + pid_t + 'hh', 0, 0, 0,
+ fcntl.F_WRLCK, 0)
+ elif sys.platform in ['aix3', 'aix4', 'hp-uxB', 'unixware7']:
+ lockdata = struct.pack('hhlllii', fcntl.F_WRLCK, 0, 0, 0, 0, 0, 0)
+ elif sys.platform in ['os2emx']:
+ lockdata = None
+ else:
+ lockdata = struct.pack('hh'+start_len+'hh', fcntl.F_WRLCK, 0, 0, 0, 0, 0)
+ if lockdata:
+ if verbose:
+ print 'struct.pack: ', repr(lockdata)
+ return lockdata
+
+lockdata = get_lockdata()
+
+
+class TestFcntl(unittest.TestCase):
+
+ def setUp(self):
+ self.f = None
+
+ def tearDown(self):
+ if self.f and not self.f.closed:
+ self.f.close()
+ unlink(TESTFN)
+
+ def test_fcntl_fileno(self):
+ # the example from the library docs
+ self.f = open(TESTFN, 'w')
+ rv = fcntl.fcntl(self.f.fileno(), fcntl.F_SETFL, os.O_NONBLOCK)
+ if verbose:
+ print 'Status from fcntl with O_NONBLOCK: ', rv
+ if sys.platform not in ['os2emx']:
+ rv = fcntl.fcntl(self.f.fileno(), fcntl.F_SETLKW, lockdata)
+ if verbose:
+ print 'String from fcntl with F_SETLKW: ', repr(rv)
+ self.f.close()
+
+ def test_fcntl_file_descriptor(self):
+ # again, but pass the file rather than numeric descriptor
+ self.f = open(TESTFN, 'w')
+ rv = fcntl.fcntl(self.f, fcntl.F_SETFL, os.O_NONBLOCK)
+ if sys.platform not in ['os2emx']:
+ rv = fcntl.fcntl(self.f, fcntl.F_SETLKW, lockdata)
+ self.f.close()
+
+ def test_fcntl_64_bit(self):
+ # Issue #1309352: fcntl shouldn't fail when the third arg fits in a
+ # C 'long' but not in a C 'int'.
+ try:
+ cmd = fcntl.F_NOTIFY
+ # This flag is larger than 2**31 in 64-bit builds
+ flags = fcntl.DN_MULTISHOT
+ except AttributeError:
+ self.skipTest("F_NOTIFY or DN_MULTISHOT unavailable")
+ fd = os.open(os.path.dirname(os.path.abspath(TESTFN)), os.O_RDONLY)
+ try:
+ fcntl.fcntl(fd, cmd, flags)
+ finally:
+ os.close(fd)
+
+
+def test_main():
+ run_unittest(TestFcntl)
+
+if __name__ == '__main__':
+ test_main()
diff --git a/lib-python/modified-2.7/test/test_heapq.py b/lib-python/modified-2.7/test/test_heapq.py
--- a/lib-python/modified-2.7/test/test_heapq.py
+++ b/lib-python/modified-2.7/test/test_heapq.py
@@ -186,6 +186,11 @@
self.assertFalse(sys.modules['heapq'] is self.module)
self.assertTrue(hasattr(self.module.heapify, 'func_code'))
+ def test_islice_protection(self):
+ m = self.module
+ self.assertFalse(m.nsmallest(-1, [1]))
+ self.assertFalse(m.nlargest(-1, [1]))
+
class TestHeapC(TestHeap):
module = c_heapq
diff --git a/lib-python/modified-2.7/test/test_multibytecodec.py b/lib-python/modified-2.7/test/test_multibytecodec.py
--- a/lib-python/modified-2.7/test/test_multibytecodec.py
+++ b/lib-python/modified-2.7/test/test_multibytecodec.py
@@ -148,7 +148,8 @@
class Test_StreamReader(unittest.TestCase):
def test_bug1728403(self):
try:
- open(TESTFN, 'w').write('\xa1')
+ with open(TESTFN, 'w') as f:
+ f.write('\xa1')
f = codecs.open(TESTFN, encoding='cp949')
self.assertRaises(UnicodeDecodeError, f.read, 2)
finally:
diff --git a/lib-python/modified-2.7/test/test_multiprocessing.py b/lib-python/modified-2.7/test/test_multiprocessing.py
--- a/lib-python/modified-2.7/test/test_multiprocessing.py
+++ b/lib-python/modified-2.7/test/test_multiprocessing.py
@@ -510,7 +510,6 @@
p.join()
- @unittest.skipIf(os.name == 'posix', "PYPY: FIXME")
def test_qsize(self):
q = self.Queue()
try:
@@ -532,7 +531,6 @@
time.sleep(DELTA)
q.task_done()
- @unittest.skipIf(os.name == 'posix', "PYPY: FIXME")
def test_task_done(self):
queue = self.JoinableQueue()
@@ -1091,7 +1089,6 @@
class _TestPoolWorkerLifetime(BaseTestCase):
ALLOWED_TYPES = ('processes', )
- @unittest.skipIf(os.name == 'posix', "PYPY: FIXME")
def test_pool_worker_lifetime(self):
p = multiprocessing.Pool(3, maxtasksperchild=10)
self.assertEqual(3, len(p._pool))
@@ -1280,7 +1277,6 @@
queue = manager.get_queue()
queue.put('hello world')
- @unittest.skipIf(os.name == 'posix', "PYPY: FIXME")
def test_rapid_restart(self):
authkey = os.urandom(32)
manager = QueueManager(
@@ -1297,6 +1293,7 @@
queue = manager.get_queue()
self.assertEqual(queue.get(), 'hello world')
del queue
+ test_support.gc_collect()
manager.shutdown()
manager = QueueManager(
address=addr, authkey=authkey, serializer=SERIALIZER)
@@ -1573,7 +1570,6 @@
ALLOWED_TYPES = ('processes',)
- @unittest.skipIf(os.name == 'posix', "PYPY: FIXME")
def test_heap(self):
iterations = 5000
maxblocks = 50
diff --git a/lib-python/modified-2.7/test/test_ssl.py b/lib-python/modified-2.7/test/test_ssl.py
--- a/lib-python/modified-2.7/test/test_ssl.py
+++ b/lib-python/modified-2.7/test/test_ssl.py
@@ -58,32 +58,35 @@
# Issue #9415: Ubuntu hijacks their OpenSSL and forcefully disables SSLv2
def skip_if_broken_ubuntu_ssl(func):
- # We need to access the lower-level wrapper in order to create an
- # implicit SSL context without trying to connect or listen.
- try:
- import _ssl
- except ImportError:
- # The returned function won't get executed, just ignore the error
- pass
- @functools.wraps(func)
- def f(*args, **kwargs):
+ if hasattr(ssl, 'PROTOCOL_SSLv2'):
+ # We need to access the lower-level wrapper in order to create an
+ # implicit SSL context without trying to connect or listen.
try:
- s = socket.socket(socket.AF_INET)
- _ssl.sslwrap(s._sock, 0, None, None,
- ssl.CERT_NONE, ssl.PROTOCOL_SSLv2, None, None)
- except ssl.SSLError as e:
- if (ssl.OPENSSL_VERSION_INFO == (0, 9, 8, 15, 15) and
- platform.linux_distribution() == ('debian', 'squeeze/sid', '')
- and 'Invalid SSL protocol variant specified' in str(e)):
- raise unittest.SkipTest("Patched Ubuntu OpenSSL breaks behaviour")
- return func(*args, **kwargs)
- return f
+ import _ssl
+ except ImportError:
+ # The returned function won't get executed, just ignore the error
+ pass
+ @functools.wraps(func)
+ def f(*args, **kwargs):
+ try:
+ s = socket.socket(socket.AF_INET)
+ _ssl.sslwrap(s._sock, 0, None, None,
+ ssl.CERT_NONE, ssl.PROTOCOL_SSLv2, None, None)
+ except ssl.SSLError as e:
+ if (ssl.OPENSSL_VERSION_INFO == (0, 9, 8, 15, 15) and
+ platform.linux_distribution() == ('debian', 'squeeze/sid', '')
+ and 'Invalid SSL protocol variant specified' in str(e)):
+ raise unittest.SkipTest("Patched Ubuntu OpenSSL breaks behaviour")
+ return func(*args, **kwargs)
+ return f
+ else:
+ return func
class BasicSocketTests(unittest.TestCase):
def test_constants(self):
- ssl.PROTOCOL_SSLv2
+ #ssl.PROTOCOL_SSLv2
ssl.PROTOCOL_SSLv23
ssl.PROTOCOL_SSLv3
ssl.PROTOCOL_TLSv1
@@ -966,7 +969,8 @@
try_protocol_combo(ssl.PROTOCOL_SSLv3, ssl.PROTOCOL_SSLv3, True)
try_protocol_combo(ssl.PROTOCOL_SSLv3, ssl.PROTOCOL_SSLv3, True, ssl.CERT_OPTIONAL)
try_protocol_combo(ssl.PROTOCOL_SSLv3, ssl.PROTOCOL_SSLv3, True, ssl.CERT_REQUIRED)
- try_protocol_combo(ssl.PROTOCOL_SSLv3, ssl.PROTOCOL_SSLv2, False)
+ if hasattr(ssl, 'PROTOCOL_SSLv2'):
+ try_protocol_combo(ssl.PROTOCOL_SSLv3, ssl.PROTOCOL_SSLv2, False)
try_protocol_combo(ssl.PROTOCOL_SSLv3, ssl.PROTOCOL_SSLv23, False)
try_protocol_combo(ssl.PROTOCOL_SSLv3, ssl.PROTOCOL_TLSv1, False)
@@ -978,7 +982,8 @@
try_protocol_combo(ssl.PROTOCOL_TLSv1, ssl.PROTOCOL_TLSv1, True)
try_protocol_combo(ssl.PROTOCOL_TLSv1, ssl.PROTOCOL_TLSv1, True, ssl.CERT_OPTIONAL)
try_protocol_combo(ssl.PROTOCOL_TLSv1, ssl.PROTOCOL_TLSv1, True, ssl.CERT_REQUIRED)
- try_protocol_combo(ssl.PROTOCOL_TLSv1, ssl.PROTOCOL_SSLv2, False)
+ if hasattr(ssl, 'PROTOCOL_SSLv2'):
+ try_protocol_combo(ssl.PROTOCOL_TLSv1, ssl.PROTOCOL_SSLv2, False)
try_protocol_combo(ssl.PROTOCOL_TLSv1, ssl.PROTOCOL_SSLv3, False)
try_protocol_combo(ssl.PROTOCOL_TLSv1, ssl.PROTOCOL_SSLv23, False)
diff --git a/lib-python/modified-2.7/test/test_sys_settrace.py b/lib-python/modified-2.7/test/test_sys_settrace.py
--- a/lib-python/modified-2.7/test/test_sys_settrace.py
+++ b/lib-python/modified-2.7/test/test_sys_settrace.py
@@ -286,11 +286,11 @@
self.compare_events(func.func_code.co_firstlineno,
tracer.events, func.events)
- def set_and_retrieve_none(self):
+ def test_set_and_retrieve_none(self):
sys.settrace(None)
assert sys.gettrace() is None
- def set_and_retrieve_func(self):
+ def test_set_and_retrieve_func(self):
def fn(*args):
pass
diff --git a/lib-python/2.7/test/test_tarfile.py b/lib-python/modified-2.7/test/test_tarfile.py
copy from lib-python/2.7/test/test_tarfile.py
copy to lib-python/modified-2.7/test/test_tarfile.py
--- a/lib-python/2.7/test/test_tarfile.py
+++ b/lib-python/modified-2.7/test/test_tarfile.py
@@ -169,6 +169,7 @@
except tarfile.ReadError:
self.fail("tarfile.open() failed on empty archive")
self.assertListEqual(tar.getmembers(), [])
+ tar.close()
def test_null_tarfile(self):
# Test for issue6123: Allow opening empty archives.
@@ -207,16 +208,21 @@
fobj = open(self.tarname, "rb")
tar = tarfile.open(fileobj=fobj, mode=self.mode)
self.assertEqual(tar.name, os.path.abspath(fobj.name))
+ tar.close()
def test_no_name_attribute(self):
- data = open(self.tarname, "rb").read()
+ f = open(self.tarname, "rb")
+ data = f.read()
+ f.close()
fobj = StringIO.StringIO(data)
self.assertRaises(AttributeError, getattr, fobj, "name")
tar = tarfile.open(fileobj=fobj, mode=self.mode)
self.assertEqual(tar.name, None)
def test_empty_name_attribute(self):
- data = open(self.tarname, "rb").read()
+ f = open(self.tarname, "rb")
+ data = f.read()
+ f.close()
fobj = StringIO.StringIO(data)
fobj.name = ""
tar = tarfile.open(fileobj=fobj, mode=self.mode)
@@ -515,6 +521,7 @@
self.tar = tarfile.open(self.tarname, mode=self.mode, encoding="iso8859-1")
tarinfo = self.tar.getmember("pax/umlauts-�������")
self._test_member(tarinfo, size=7011, chksum=md5_regtype)
+ self.tar.close()
class LongnameTest(ReadTest):
@@ -675,6 +682,7 @@
tar = tarfile.open(tmpname, self.mode)
tarinfo = tar.gettarinfo(path)
self.assertEqual(tarinfo.size, 0)
+ tar.close()
finally:
os.rmdir(path)
@@ -692,6 +700,7 @@
tar.gettarinfo(target)
tarinfo = tar.gettarinfo(link)
self.assertEqual(tarinfo.size, 0)
+ tar.close()
finally:
os.remove(target)
os.remove(link)
@@ -704,6 +713,7 @@
tar = tarfile.open(tmpname, self.mode)
tarinfo = tar.gettarinfo(path)
self.assertEqual(tarinfo.size, 0)
+ tar.close()
finally:
os.remove(path)
@@ -722,6 +732,7 @@
tar.add(dstname)
os.chdir(cwd)
self.assertTrue(tar.getnames() == [], "added the archive to itself")
+ tar.close()
def test_exclude(self):
tempdir = os.path.join(TEMPDIR, "exclude")
@@ -742,6 +753,7 @@
tar = tarfile.open(tmpname, "r")
self.assertEqual(len(tar.getmembers()), 1)
self.assertEqual(tar.getnames()[0], "empty_dir")
+ tar.close()
finally:
shutil.rmtree(tempdir)
@@ -859,7 +871,9 @@
fobj.close()
elif self.mode.endswith("bz2"):
dec = bz2.BZ2Decompressor()
- data = open(tmpname, "rb").read()
+ f = open(tmpname, "rb")
+ data = f.read()
+ f.close()
data = dec.decompress(data)
self.assertTrue(len(dec.unused_data) == 0,
"found trailing data")
@@ -938,6 +952,7 @@
"unable to read longname member")
self.assertEqual(tarinfo.linkname, member.linkname,
"unable to read longname member")
+ tar.close()
def test_longname_1023(self):
self._test(("longnam/" * 127) + "longnam")
@@ -1030,6 +1045,7 @@
else:
n = tar.getmembers()[0].name
self.assertTrue(name == n, "PAX longname creation failed")
+ tar.close()
def test_pax_global_header(self):
pax_headers = {
@@ -1058,6 +1074,7 @@
tarfile.PAX_NUMBER_FIELDS[key](val)
except (TypeError, ValueError):
self.fail("unable to convert pax header field")
+ tar.close()
def test_pax_extended_header(self):
# The fields from the pax header have priority over the
@@ -1077,6 +1094,7 @@
self.assertEqual(t.pax_headers, pax_headers)
self.assertEqual(t.name, "foo")
self.assertEqual(t.uid, 123)
+ tar.close()
class UstarUnicodeTest(unittest.TestCase):
@@ -1120,6 +1138,7 @@
tarinfo.name = "foo"
tarinfo.uname = u"���"
self.assertRaises(UnicodeError, tar.addfile, tarinfo)
+ tar.close()
def test_unicode_argument(self):
tar = tarfile.open(tarname, "r", encoding="iso8859-1", errors="strict")
@@ -1174,6 +1193,7 @@
tar = tarfile.open(tmpname, format=self.format, encoding="ascii",
errors=handler)
self.assertEqual(tar.getnames()[0], name)
+ tar.close()
self.assertRaises(UnicodeError, tarfile.open, tmpname,
encoding="ascii", errors="strict")
@@ -1186,6 +1206,7 @@
tar = tarfile.open(tmpname, format=self.format, encoding="iso8859-1",
errors="utf-8")
self.assertEqual(tar.getnames()[0], "���/" + u"�".encode("utf8"))
+ tar.close()
class AppendTest(unittest.TestCase):
@@ -1213,6 +1234,7 @@
def _test(self, names=["bar"], fileobj=None):
tar = tarfile.open(self.tarname, fileobj=fileobj)
self.assertEqual(tar.getnames(), names)
+ tar.close()
def test_non_existing(self):
self._add_testfile()
@@ -1231,7 +1253,9 @@
def test_fileobj(self):
self._create_testtar()
- data = open(self.tarname).read()
+ f = open(self.tarname)
+ data = f.read()
+ f.close()
fobj = StringIO.StringIO(data)
self._add_testfile(fobj)
fobj.seek(0)
@@ -1257,7 +1281,9 @@
# Append mode is supposed to fail if the tarfile to append to
# does not end with a zero block.
def _test_error(self, data):
- open(self.tarname, "wb").write(data)
+ f = open(self.tarname, "wb")
+ f.write(data)
+ f.close()
self.assertRaises(tarfile.ReadError, self._add_testfile)
def test_null(self):
diff --git a/lib-python/modified-2.7/test/test_tempfile.py b/lib-python/modified-2.7/test/test_tempfile.py
--- a/lib-python/modified-2.7/test/test_tempfile.py
+++ b/lib-python/modified-2.7/test/test_tempfile.py
@@ -23,8 +23,8 @@
# TEST_FILES may need to be tweaked for systems depending on the maximum
# number of files that can be opened at one time (see ulimit -n)
-if sys.platform in ('openbsd3', 'openbsd4'):
- TEST_FILES = 48
+if sys.platform.startswith("openbsd"):
+ TEST_FILES = 64 # ulimit -n defaults to 128 for normal users
else:
TEST_FILES = 100
diff --git a/lib-python/modified-2.7/test/test_urllib2.py b/lib-python/modified-2.7/test/test_urllib2.py
--- a/lib-python/modified-2.7/test/test_urllib2.py
+++ b/lib-python/modified-2.7/test/test_urllib2.py
@@ -307,6 +307,9 @@
def getresponse(self):
return MockHTTPResponse(MockFile(), {}, 200, "OK")
+ def close(self):
+ pass
+
class MockHandler:
# useful for testing handler machinery
# see add_ordered_mock_handlers() docstring
diff --git a/lib-python/modified-2.7/urllib2.py b/lib-python/modified-2.7/urllib2.py
new file mode 100644
--- /dev/null
+++ b/lib-python/modified-2.7/urllib2.py
@@ -0,0 +1,1436 @@
+"""An extensible library for opening URLs using a variety of protocols
+
+The simplest way to use this module is to call the urlopen function,
+which accepts a string containing a URL or a Request object (described
+below). It opens the URL and returns the results as file-like
+object; the returned object has some extra methods described below.
+
+The OpenerDirector manages a collection of Handler objects that do
+all the actual work. Each Handler implements a particular protocol or
+option. The OpenerDirector is a composite object that invokes the
+Handlers needed to open the requested URL. For example, the
+HTTPHandler performs HTTP GET and POST requests and deals with
+non-error returns. The HTTPRedirectHandler automatically deals with
+HTTP 301, 302, 303 and 307 redirect errors, and the HTTPDigestAuthHandler
+deals with digest authentication.
+
+urlopen(url, data=None) -- Basic usage is the same as original
+urllib. pass the url and optionally data to post to an HTTP URL, and
+get a file-like object back. One difference is that you can also pass
+a Request instance instead of URL. Raises a URLError (subclass of
+IOError); for HTTP errors, raises an HTTPError, which can also be
+treated as a valid response.
+
+build_opener -- Function that creates a new OpenerDirector instance.
+Will install the default handlers. Accepts one or more Handlers as
+arguments, either instances or Handler classes that it will
+instantiate. If one of the argument is a subclass of the default
+handler, the argument will be installed instead of the default.
+
+install_opener -- Installs a new opener as the default opener.
+
+objects of interest:
+
+OpenerDirector -- Sets up the User Agent as the Python-urllib client and manages
+the Handler classes, while dealing with requests and responses.
+
+Request -- An object that encapsulates the state of a request. The
+state can be as simple as the URL. It can also include extra HTTP
+headers, e.g. a User-Agent.
+
+BaseHandler --
+
+exceptions:
+URLError -- A subclass of IOError, individual protocols have their own
+specific subclass.
+
+HTTPError -- Also a valid HTTP response, so you can treat an HTTP error
+as an exceptional event or valid response.
+
+internals:
+BaseHandler and parent
+_call_chain conventions
+
+Example usage:
+
+import urllib2
+
+# set up authentication info
+authinfo = urllib2.HTTPBasicAuthHandler()
+authinfo.add_password(realm='PDQ Application',
+ uri='https://mahler:8092/site-updates.py',
+ user='klem',
+ passwd='geheim$parole')
+
+proxy_support = urllib2.ProxyHandler({"http" : "http://ahad-haam:3128"})
+
+# build a new opener that adds authentication and caching FTP handlers
+opener = urllib2.build_opener(proxy_support, authinfo, urllib2.CacheFTPHandler)
+
+# install it
+urllib2.install_opener(opener)
+
+f = urllib2.urlopen('http://www.python.org/')
+
+
+"""
+
+# XXX issues:
+# If an authentication error handler that tries to perform
+# authentication for some reason but fails, how should the error be
+# signalled? The client needs to know the HTTP error code. But if
+# the handler knows that the problem was, e.g., that it didn't know
+# that hash algo that requested in the challenge, it would be good to
+# pass that information along to the client, too.
+# ftp errors aren't handled cleanly
+# check digest against correct (i.e. non-apache) implementation
+
+# Possible extensions:
+# complex proxies XXX not sure what exactly was meant by this
+# abstract factory for opener
+
+import base64
+import hashlib
+import httplib
+import mimetools
+import os
+import posixpath
+import random
+import re
+import socket
+import sys
+import time
+import urlparse
+import bisect
+
+try:
+ from cStringIO import StringIO
+except ImportError:
+ from StringIO import StringIO
+
+from urllib import (unwrap, unquote, splittype, splithost, quote,
+ addinfourl, splitport, splittag,
+ splitattr, ftpwrapper, splituser, splitpasswd, splitvalue)
+
+# support for FileHandler, proxies via environment variables
+from urllib import localhost, url2pathname, getproxies, proxy_bypass
+
+# used in User-Agent header sent
+__version__ = sys.version[:3]
+
+_opener = None
+def urlopen(url, data=None, timeout=socket._GLOBAL_DEFAULT_TIMEOUT):
+ global _opener
+ if _opener is None:
+ _opener = build_opener()
+ return _opener.open(url, data, timeout)
+
+def install_opener(opener):
+ global _opener
+ _opener = opener
+
+# do these error classes make sense?
+# make sure all of the IOError stuff is overridden. we just want to be
+# subtypes.
+
+class URLError(IOError):
+ # URLError is a sub-type of IOError, but it doesn't share any of
+ # the implementation. need to override __init__ and __str__.
+ # It sets self.args for compatibility with other EnvironmentError
+ # subclasses, but args doesn't have the typical format with errno in
+ # slot 0 and strerror in slot 1. This may be better than nothing.
+ def __init__(self, reason):
+ self.args = reason,
+ self.reason = reason
+
+ def __str__(self):
+ return '<urlopen error %s>' % self.reason
+
+class HTTPError(URLError, addinfourl):
+ """Raised when HTTP error occurs, but also acts like non-error return"""
+ __super_init = addinfourl.__init__
+
+ def __init__(self, url, code, msg, hdrs, fp):
+ self.code = code
+ self.msg = msg
+ self.hdrs = hdrs
+ self.fp = fp
+ self.filename = url
+ # The addinfourl classes depend on fp being a valid file
+ # object. In some cases, the HTTPError may not have a valid
+ # file object. If this happens, the simplest workaround is to
+ # not initialize the base classes.
+ if fp is not None:
+ self.__super_init(fp, hdrs, url, code)
+
+ def __str__(self):
+ return 'HTTP Error %s: %s' % (self.code, self.msg)
+
+# copied from cookielib.py
+_cut_port_re = re.compile(r":\d+$")
+def request_host(request):
+ """Return request-host, as defined by RFC 2965.
+
+ Variation from RFC: returned value is lowercased, for convenient
+ comparison.
+
+ """
+ url = request.get_full_url()
+ host = urlparse.urlparse(url)[1]
+ if host == "":
+ host = request.get_header("Host", "")
+
+ # remove port, if present
+ host = _cut_port_re.sub("", host, 1)
+ return host.lower()
+
+class Request:
+
+ def __init__(self, url, data=None, headers={},
+ origin_req_host=None, unverifiable=False):
+ # unwrap('<URL:type://host/path>') --> 'type://host/path'
+ self.__original = unwrap(url)
+ self.__original, fragment = splittag(self.__original)
+ self.type = None
+ # self.__r_type is what's left after doing the splittype
+ self.host = None
+ self.port = None
+ self._tunnel_host = None
+ self.data = data
+ self.headers = {}
+ for key, value in headers.items():
+ self.add_header(key, value)
+ self.unredirected_hdrs = {}
+ if origin_req_host is None:
+ origin_req_host = request_host(self)
+ self.origin_req_host = origin_req_host
+ self.unverifiable = unverifiable
+
+ def __getattr__(self, attr):
+ # XXX this is a fallback mechanism to guard against these
+ # methods getting called in a non-standard order. this may be
+ # too complicated and/or unnecessary.
+ # XXX should the __r_XXX attributes be public?
+ if attr[:12] == '_Request__r_':
+ name = attr[12:]
+ if hasattr(Request, 'get_' + name):
+ getattr(self, 'get_' + name)()
+ return getattr(self, attr)
+ raise AttributeError, attr
+
+ def get_method(self):
+ if self.has_data():
+ return "POST"
+ else:
+ return "GET"
+
+ # XXX these helper methods are lame
+
+ def add_data(self, data):
+ self.data = data
+
+ def has_data(self):
+ return self.data is not None
+
+ def get_data(self):
+ return self.data
+
+ def get_full_url(self):
+ return self.__original
+
+ def get_type(self):
+ if self.type is None:
+ self.type, self.__r_type = splittype(self.__original)
+ if self.type is None:
+ raise ValueError, "unknown url type: %s" % self.__original
+ return self.type
+
+ def get_host(self):
+ if self.host is None:
+ self.host, self.__r_host = splithost(self.__r_type)
+ if self.host:
+ self.host = unquote(self.host)
+ return self.host
+
+ def get_selector(self):
+ return self.__r_host
+
+ def set_proxy(self, host, type):
+ if self.type == 'https' and not self._tunnel_host:
+ self._tunnel_host = self.host
+ else:
+ self.type = type
+ self.__r_host = self.__original
+
+ self.host = host
+
+ def has_proxy(self):
+ return self.__r_host == self.__original
+
+ def get_origin_req_host(self):
+ return self.origin_req_host
+
+ def is_unverifiable(self):
+ return self.unverifiable
+
+ def add_header(self, key, val):
+ # useful for something like authentication
+ self.headers[key.capitalize()] = val
+
+ def add_unredirected_header(self, key, val):
+ # will not be added to a redirected request
+ self.unredirected_hdrs[key.capitalize()] = val
+
+ def has_header(self, header_name):
+ return (header_name in self.headers or
+ header_name in self.unredirected_hdrs)
+
+ def get_header(self, header_name, default=None):
+ return self.headers.get(
+ header_name,
+ self.unredirected_hdrs.get(header_name, default))
+
+ def header_items(self):
+ hdrs = self.unredirected_hdrs.copy()
+ hdrs.update(self.headers)
+ return hdrs.items()
+
+class OpenerDirector:
+ def __init__(self):
+ client_version = "Python-urllib/%s" % __version__
+ self.addheaders = [('User-agent', client_version)]
+ # manage the individual handlers
+ self.handlers = []
+ self.handle_open = {}
+ self.handle_error = {}
+ self.process_response = {}
+ self.process_request = {}
+
+ def add_handler(self, handler):
+ if not hasattr(handler, "add_parent"):
+ raise TypeError("expected BaseHandler instance, got %r" %
+ type(handler))
+
+ added = False
+ for meth in dir(handler):
+ if meth in ["redirect_request", "do_open", "proxy_open"]:
+ # oops, coincidental match
+ continue
+
+ i = meth.find("_")
+ protocol = meth[:i]
+ condition = meth[i+1:]
+
+ if condition.startswith("error"):
+ j = condition.find("_") + i + 1
+ kind = meth[j+1:]
+ try:
+ kind = int(kind)
+ except ValueError:
+ pass
+ lookup = self.handle_error.get(protocol, {})
+ self.handle_error[protocol] = lookup
+ elif condition == "open":
+ kind = protocol
+ lookup = self.handle_open
+ elif condition == "response":
+ kind = protocol
+ lookup = self.process_response
+ elif condition == "request":
+ kind = protocol
+ lookup = self.process_request
+ else:
+ continue
+
+ handlers = lookup.setdefault(kind, [])
+ if handlers:
+ bisect.insort(handlers, handler)
+ else:
+ handlers.append(handler)
+ added = True
+
+ if added:
+ # the handlers must work in an specific order, the order
+ # is specified in a Handler attribute
+ bisect.insort(self.handlers, handler)
+ handler.add_parent(self)
+
+ def close(self):
+ # Only exists for backwards compatibility.
+ pass
+
+ def _call_chain(self, chain, kind, meth_name, *args):
+ # Handlers raise an exception if no one else should try to handle
+ # the request, or return None if they can't but another handler
+ # could. Otherwise, they return the response.
+ handlers = chain.get(kind, ())
+ for handler in handlers:
+ func = getattr(handler, meth_name)
+
+ result = func(*args)
+ if result is not None:
+ return result
+
+ def open(self, fullurl, data=None, timeout=socket._GLOBAL_DEFAULT_TIMEOUT):
+ # accept a URL or a Request object
+ if isinstance(fullurl, basestring):
+ req = Request(fullurl, data)
+ else:
+ req = fullurl
+ if data is not None:
+ req.add_data(data)
+
+ req.timeout = timeout
+ protocol = req.get_type()
+
+ # pre-process request
+ meth_name = protocol+"_request"
+ for processor in self.process_request.get(protocol, []):
+ meth = getattr(processor, meth_name)
+ req = meth(req)
+
+ response = self._open(req, data)
+
+ # post-process response
+ meth_name = protocol+"_response"
+ for processor in self.process_response.get(protocol, []):
+ meth = getattr(processor, meth_name)
+ response = meth(req, response)
+
+ return response
+
+ def _open(self, req, data=None):
+ result = self._call_chain(self.handle_open, 'default',
+ 'default_open', req)
+ if result:
+ return result
+
+ protocol = req.get_type()
+ result = self._call_chain(self.handle_open, protocol, protocol +
+ '_open', req)
+ if result:
+ return result
+
+ return self._call_chain(self.handle_open, 'unknown',
+ 'unknown_open', req)
+
+ def error(self, proto, *args):
+ if proto in ('http', 'https'):
+ # XXX http[s] protocols are special-cased
+ dict = self.handle_error['http'] # https is not different than http
+ proto = args[2] # YUCK!
+ meth_name = 'http_error_%s' % proto
+ http_err = 1
+ orig_args = args
+ else:
+ dict = self.handle_error
+ meth_name = proto + '_error'
+ http_err = 0
+ args = (dict, proto, meth_name) + args
+ result = self._call_chain(*args)
+ if result:
+ return result
+
+ if http_err:
+ args = (dict, 'default', 'http_error_default') + orig_args
+ return self._call_chain(*args)
+
+# XXX probably also want an abstract factory that knows when it makes
+# sense to skip a superclass in favor of a subclass and when it might
+# make sense to include both
+
+def build_opener(*handlers):
+ """Create an opener object from a list of handlers.
+
+ The opener will use several default handlers, including support
+ for HTTP, FTP and when applicable, HTTPS.
+
+ If any of the handlers passed as arguments are subclasses of the
+ default handlers, the default handlers will not be used.
+ """
+ import types
+ def isclass(obj):
+ return isinstance(obj, (types.ClassType, type))
+
+ opener = OpenerDirector()
+ default_classes = [ProxyHandler, UnknownHandler, HTTPHandler,
+ HTTPDefaultErrorHandler, HTTPRedirectHandler,
+ FTPHandler, FileHandler, HTTPErrorProcessor]
+ if hasattr(httplib, 'HTTPS'):
+ default_classes.append(HTTPSHandler)
+ skip = set()
+ for klass in default_classes:
+ for check in handlers:
+ if isclass(check):
+ if issubclass(check, klass):
+ skip.add(klass)
+ elif isinstance(check, klass):
+ skip.add(klass)
+ for klass in skip:
+ default_classes.remove(klass)
+
+ for klass in default_classes:
+ opener.add_handler(klass())
+
+ for h in handlers:
+ if isclass(h):
+ h = h()
+ opener.add_handler(h)
+ return opener
+
+class BaseHandler:
+ handler_order = 500
+
+ def add_parent(self, parent):
+ self.parent = parent
+
+ def close(self):
+ # Only exists for backwards compatibility
+ pass
+
+ def __lt__(self, other):
+ if not hasattr(other, "handler_order"):
+ # Try to preserve the old behavior of having custom classes
+ # inserted after default ones (works only for custom user
+ # classes which are not aware of handler_order).
+ return True
+ return self.handler_order < other.handler_order
+
+
+class HTTPErrorProcessor(BaseHandler):
+ """Process HTTP error responses."""
+ handler_order = 1000 # after all other processing
+
+ def http_response(self, request, response):
+ code, msg, hdrs = response.code, response.msg, response.info()
+
+ # According to RFC 2616, "2xx" code indicates that the client's
+ # request was successfully received, understood, and accepted.
+ if not (200 <= code < 300):
+ response = self.parent.error(
+ 'http', request, response, code, msg, hdrs)
+
+ return response
+
+ https_response = http_response
+
+class HTTPDefaultErrorHandler(BaseHandler):
+ def http_error_default(self, req, fp, code, msg, hdrs):
+ raise HTTPError(req.get_full_url(), code, msg, hdrs, fp)
+
+class HTTPRedirectHandler(BaseHandler):
+ # maximum number of redirections to any single URL
+ # this is needed because of the state that cookies introduce
+ max_repeats = 4
+ # maximum total number of redirections (regardless of URL) before
+ # assuming we're in a loop
+ max_redirections = 10
+
+ def redirect_request(self, req, fp, code, msg, headers, newurl):
+ """Return a Request or None in response to a redirect.
+
+ This is called by the http_error_30x methods when a
+ redirection response is received. If a redirection should
+ take place, return a new Request to allow http_error_30x to
+ perform the redirect. Otherwise, raise HTTPError if no-one
+ else should try to handle this url. Return None if you can't
+ but another Handler might.
+ """
+ m = req.get_method()
+ if (code in (301, 302, 303, 307) and m in ("GET", "HEAD")
+ or code in (301, 302, 303) and m == "POST"):
+ # Strictly (according to RFC 2616), 301 or 302 in response
+ # to a POST MUST NOT cause a redirection without confirmation
+ # from the user (of urllib2, in this case). In practice,
+ # essentially all clients do redirect in this case, so we
+ # do the same.
+ # be conciliant with URIs containing a space
+ newurl = newurl.replace(' ', '%20')
+ newheaders = dict((k,v) for k,v in req.headers.items()
+ if k.lower() not in ("content-length", "content-type")
+ )
+ return Request(newurl,
+ headers=newheaders,
+ origin_req_host=req.get_origin_req_host(),
+ unverifiable=True)
+ else:
+ raise HTTPError(req.get_full_url(), code, msg, headers, fp)
+
+ # Implementation note: To avoid the server sending us into an
+ # infinite loop, the request object needs to track what URLs we
+ # have already seen. Do this by adding a handler-specific
+ # attribute to the Request object.
+ def http_error_302(self, req, fp, code, msg, headers):
+ # Some servers (incorrectly) return multiple Location headers
+ # (so probably same goes for URI). Use first header.
+ if 'location' in headers:
+ newurl = headers.getheaders('location')[0]
+ elif 'uri' in headers:
+ newurl = headers.getheaders('uri')[0]
+ else:
+ return
+
+ # fix a possible malformed URL
+ urlparts = urlparse.urlparse(newurl)
+ if not urlparts.path:
+ urlparts = list(urlparts)
+ urlparts[2] = "/"
+ newurl = urlparse.urlunparse(urlparts)
+
+ newurl = urlparse.urljoin(req.get_full_url(), newurl)
+
+ # XXX Probably want to forget about the state of the current
+ # request, although that might interact poorly with other
+ # handlers that also use handler-specific request attributes
+ new = self.redirect_request(req, fp, code, msg, headers, newurl)
+ if new is None:
+ return
+
+ # loop detection
+ # .redirect_dict has a key url if url was previously visited.
+ if hasattr(req, 'redirect_dict'):
+ visited = new.redirect_dict = req.redirect_dict
+ if (visited.get(newurl, 0) >= self.max_repeats or
+ len(visited) >= self.max_redirections):
+ raise HTTPError(req.get_full_url(), code,
+ self.inf_msg + msg, headers, fp)
+ else:
+ visited = new.redirect_dict = req.redirect_dict = {}
+ visited[newurl] = visited.get(newurl, 0) + 1
+
+ # Don't close the fp until we are sure that we won't use it
+ # with HTTPError.
+ fp.read()
+ fp.close()
+
+ return self.parent.open(new, timeout=req.timeout)
+
+ http_error_301 = http_error_303 = http_error_307 = http_error_302
+
+ inf_msg = "The HTTP server returned a redirect error that would " \
+ "lead to an infinite loop.\n" \
+ "The last 30x error message was:\n"
+
+
+def _parse_proxy(proxy):
+ """Return (scheme, user, password, host/port) given a URL or an authority.
+
+ If a URL is supplied, it must have an authority (host:port) component.
+ According to RFC 3986, having an authority component means the URL must
+ have two slashes after the scheme:
+
+ >>> _parse_proxy('file:/ftp.example.com/')
+ Traceback (most recent call last):
+ ValueError: proxy URL with no authority: 'file:/ftp.example.com/'
+
+ The first three items of the returned tuple may be None.
+
+ Examples of authority parsing:
+
+ >>> _parse_proxy('proxy.example.com')
+ (None, None, None, 'proxy.example.com')
+ >>> _parse_proxy('proxy.example.com:3128')
+ (None, None, None, 'proxy.example.com:3128')
+
+ The authority component may optionally include userinfo (assumed to be
+ username:password):
+
+ >>> _parse_proxy('joe:password at proxy.example.com')
+ (None, 'joe', 'password', 'proxy.example.com')
+ >>> _parse_proxy('joe:password at proxy.example.com:3128')
+ (None, 'joe', 'password', 'proxy.example.com:3128')
+
+ Same examples, but with URLs instead:
+
+ >>> _parse_proxy('http://proxy.example.com/')
+ ('http', None, None, 'proxy.example.com')
+ >>> _parse_proxy('http://proxy.example.com:3128/')
+ ('http', None, None, 'proxy.example.com:3128')
+ >>> _parse_proxy('http://joe:password@proxy.example.com/')
+ ('http', 'joe', 'password', 'proxy.example.com')
+ >>> _parse_proxy('http://joe:password@proxy.example.com:3128')
+ ('http', 'joe', 'password', 'proxy.example.com:3128')
+
+ Everything after the authority is ignored:
+
+ >>> _parse_proxy('ftp://joe:password@proxy.example.com/rubbish:3128')
+ ('ftp', 'joe', 'password', 'proxy.example.com')
+
+ Test for no trailing '/' case:
+
+ >>> _parse_proxy('http://joe:password@proxy.example.com')
+ ('http', 'joe', 'password', 'proxy.example.com')
+
+ """
+ scheme, r_scheme = splittype(proxy)
+ if not r_scheme.startswith("/"):
+ # authority
+ scheme = None
+ authority = proxy
+ else:
+ # URL
+ if not r_scheme.startswith("//"):
+ raise ValueError("proxy URL with no authority: %r" % proxy)
+ # We have an authority, so for RFC 3986-compliant URLs (by ss 3.
+ # and 3.3.), path is empty or starts with '/'
+ end = r_scheme.find("/", 2)
+ if end == -1:
+ end = None
+ authority = r_scheme[2:end]
+ userinfo, hostport = splituser(authority)
+ if userinfo is not None:
+ user, password = splitpasswd(userinfo)
+ else:
+ user = password = None
+ return scheme, user, password, hostport
+
+class ProxyHandler(BaseHandler):
+ # Proxies must be in front
+ handler_order = 100
+
+ def __init__(self, proxies=None):
+ if proxies is None:
+ proxies = getproxies()
+ assert hasattr(proxies, 'has_key'), "proxies must be a mapping"
+ self.proxies = proxies
+ for type, url in proxies.items():
+ setattr(self, '%s_open' % type,
+ lambda r, proxy=url, type=type, meth=self.proxy_open: \
+ meth(r, proxy, type))
+
+ def proxy_open(self, req, proxy, type):
+ orig_type = req.get_type()
+ proxy_type, user, password, hostport = _parse_proxy(proxy)
+
+ if proxy_type is None:
+ proxy_type = orig_type
+
+ if req.host and proxy_bypass(req.host):
+ return None
+
+ if user and password:
+ user_pass = '%s:%s' % (unquote(user), unquote(password))
+ creds = base64.b64encode(user_pass).strip()
+ req.add_header('Proxy-authorization', 'Basic ' + creds)
+ hostport = unquote(hostport)
+ req.set_proxy(hostport, proxy_type)
+
+ if orig_type == proxy_type or orig_type == 'https':
+ # let other handlers take care of it
+ return None
+ else:
+ # need to start over, because the other handlers don't
+ # grok the proxy's URL type
+ # e.g. if we have a constructor arg proxies like so:
+ # {'http': 'ftp://proxy.example.com'}, we may end up turning
+ # a request for http://acme.example.com/a into one for
+ # ftp://proxy.example.com/a
+ return self.parent.open(req, timeout=req.timeout)
+
+class HTTPPasswordMgr:
+
+ def __init__(self):
+ self.passwd = {}
+
+ def add_password(self, realm, uri, user, passwd):
+ # uri could be a single URI or a sequence
+ if isinstance(uri, basestring):
+ uri = [uri]
+ if not realm in self.passwd:
+ self.passwd[realm] = {}
+ for default_port in True, False:
+ reduced_uri = tuple(
+ [self.reduce_uri(u, default_port) for u in uri])
+ self.passwd[realm][reduced_uri] = (user, passwd)
+
+ def find_user_password(self, realm, authuri):
+ domains = self.passwd.get(realm, {})
+ for default_port in True, False:
+ reduced_authuri = self.reduce_uri(authuri, default_port)
+ for uris, authinfo in domains.iteritems():
+ for uri in uris:
+ if self.is_suburi(uri, reduced_authuri):
+ return authinfo
+ return None, None
+
+ def reduce_uri(self, uri, default_port=True):
+ """Accept authority or URI and extract only the authority and path."""
+ # note HTTP URLs do not have a userinfo component
+ parts = urlparse.urlsplit(uri)
+ if parts[1]:
+ # URI
+ scheme = parts[0]
+ authority = parts[1]
+ path = parts[2] or '/'
+ else:
+ # host or host:port
+ scheme = None
+ authority = uri
+ path = '/'
+ host, port = splitport(authority)
+ if default_port and port is None and scheme is not None:
+ dport = {"http": 80,
+ "https": 443,
+ }.get(scheme)
+ if dport is not None:
+ authority = "%s:%d" % (host, dport)
+ return authority, path
+
+ def is_suburi(self, base, test):
+ """Check if test is below base in a URI tree
+
+ Both args must be URIs in reduced form.
+ """
+ if base == test:
+ return True
+ if base[0] != test[0]:
+ return False
+ common = posixpath.commonprefix((base[1], test[1]))
+ if len(common) == len(base[1]):
+ return True
+ return False
+
+
+class HTTPPasswordMgrWithDefaultRealm(HTTPPasswordMgr):
+
+ def find_user_password(self, realm, authuri):
+ user, password = HTTPPasswordMgr.find_user_password(self, realm,
+ authuri)
+ if user is not None:
+ return user, password
+ return HTTPPasswordMgr.find_user_password(self, None, authuri)
+
+
+class AbstractBasicAuthHandler:
+
+ # XXX this allows for multiple auth-schemes, but will stupidly pick
+ # the last one with a realm specified.
+
+ # allow for double- and single-quoted realm values
+ # (single quotes are a violation of the RFC, but appear in the wild)
+ rx = re.compile('(?:.*,)*[ \t]*([^ \t]+)[ \t]+'
+ 'realm=(["\'])(.*?)\\2', re.I)
+
+ # XXX could pre-emptively send auth info already accepted (RFC 2617,
+ # end of section 2, and section 1.2 immediately after "credentials"
+ # production).
+
+ def __init__(self, password_mgr=None):
+ if password_mgr is None:
+ password_mgr = HTTPPasswordMgr()
+ self.passwd = password_mgr
+ self.add_password = self.passwd.add_password
+ self.retried = 0
+
+ def reset_retry_count(self):
+ self.retried = 0
+
+ def http_error_auth_reqed(self, authreq, host, req, headers):
+ # host may be an authority (without userinfo) or a URL with an
+ # authority
+ # XXX could be multiple headers
+ authreq = headers.get(authreq, None)
+
+ if self.retried > 5:
+ # retry sending the username:password 5 times before failing.
+ raise HTTPError(req.get_full_url(), 401, "basic auth failed",
+ headers, None)
+ else:
+ self.retried += 1
+
+ if authreq:
+ mo = AbstractBasicAuthHandler.rx.search(authreq)
+ if mo:
+ scheme, quote, realm = mo.groups()
+ if scheme.lower() == 'basic':
+ response = self.retry_http_basic_auth(host, req, realm)
+ if response and response.code != 401:
+ self.retried = 0
+ return response
+
+ def retry_http_basic_auth(self, host, req, realm):
+ user, pw = self.passwd.find_user_password(realm, host)
+ if pw is not None:
+ raw = "%s:%s" % (user, pw)
+ auth = 'Basic %s' % base64.b64encode(raw).strip()
+ if req.headers.get(self.auth_header, None) == auth:
+ return None
+ req.add_unredirected_header(self.auth_header, auth)
+ return self.parent.open(req, timeout=req.timeout)
+ else:
+ return None
+
+
+class HTTPBasicAuthHandler(AbstractBasicAuthHandler, BaseHandler):
+
+ auth_header = 'Authorization'
+
+ def http_error_401(self, req, fp, code, msg, headers):
+ url = req.get_full_url()
+ response = self.http_error_auth_reqed('www-authenticate',
+ url, req, headers)
+ self.reset_retry_count()
+ return response
+
+
+class ProxyBasicAuthHandler(AbstractBasicAuthHandler, BaseHandler):
+
+ auth_header = 'Proxy-authorization'
+
+ def http_error_407(self, req, fp, code, msg, headers):
+ # http_error_auth_reqed requires that there is no userinfo component in
+ # authority. Assume there isn't one, since urllib2 does not (and
+ # should not, RFC 3986 s. 3.2.1) support requests for URLs containing
+ # userinfo.
+ authority = req.get_host()
+ response = self.http_error_auth_reqed('proxy-authenticate',
+ authority, req, headers)
+ self.reset_retry_count()
+ return response
+
+
+def randombytes(n):
+ """Return n random bytes."""
+ # Use /dev/urandom if it is available. Fall back to random module
+ # if not. It might be worthwhile to extend this function to use
+ # other platform-specific mechanisms for getting random bytes.
+ if os.path.exists("/dev/urandom"):
+ f = open("/dev/urandom")
+ s = f.read(n)
+ f.close()
+ return s
+ else:
+ L = [chr(random.randrange(0, 256)) for i in range(n)]
+ return "".join(L)
+
+class AbstractDigestAuthHandler:
+ # Digest authentication is specified in RFC 2617.
+
+ # XXX The client does not inspect the Authentication-Info header
+ # in a successful response.
+
+ # XXX It should be possible to test this implementation against
+ # a mock server that just generates a static set of challenges.
+
+ # XXX qop="auth-int" supports is shaky
+
+ def __init__(self, passwd=None):
+ if passwd is None:
+ passwd = HTTPPasswordMgr()
+ self.passwd = passwd
+ self.add_password = self.passwd.add_password
+ self.retried = 0
+ self.nonce_count = 0
+ self.last_nonce = None
+
+ def reset_retry_count(self):
+ self.retried = 0
+
+ def http_error_auth_reqed(self, auth_header, host, req, headers):
+ authreq = headers.get(auth_header, None)
+ if self.retried > 5:
+ # Don't fail endlessly - if we failed once, we'll probably
+ # fail a second time. Hm. Unless the Password Manager is
+ # prompting for the information. Crap. This isn't great
+ # but it's better than the current 'repeat until recursion
+ # depth exceeded' approach <wink>
+ raise HTTPError(req.get_full_url(), 401, "digest auth failed",
+ headers, None)
+ else:
+ self.retried += 1
+ if authreq:
+ scheme = authreq.split()[0]
+ if scheme.lower() == 'digest':
+ return self.retry_http_digest_auth(req, authreq)
+
+ def retry_http_digest_auth(self, req, auth):
+ token, challenge = auth.split(' ', 1)
+ chal = parse_keqv_list(parse_http_list(challenge))
+ auth = self.get_authorization(req, chal)
+ if auth:
+ auth_val = 'Digest %s' % auth
+ if req.headers.get(self.auth_header, None) == auth_val:
+ return None
+ req.add_unredirected_header(self.auth_header, auth_val)
+ resp = self.parent.open(req, timeout=req.timeout)
+ return resp
+
+ def get_cnonce(self, nonce):
+ # The cnonce-value is an opaque
+ # quoted string value provided by the client and used by both client
+ # and server to avoid chosen plaintext attacks, to provide mutual
+ # authentication, and to provide some message integrity protection.
+ # This isn't a fabulous effort, but it's probably Good Enough.
+ dig = hashlib.sha1("%s:%s:%s:%s" % (self.nonce_count, nonce, time.ctime(),
+ randombytes(8))).hexdigest()
+ return dig[:16]
+
+ def get_authorization(self, req, chal):
+ try:
+ realm = chal['realm']
+ nonce = chal['nonce']
+ qop = chal.get('qop')
+ algorithm = chal.get('algorithm', 'MD5')
+ # mod_digest doesn't send an opaque, even though it isn't
+ # supposed to be optional
+ opaque = chal.get('opaque', None)
+ except KeyError:
+ return None
+
+ H, KD = self.get_algorithm_impls(algorithm)
+ if H is None:
+ return None
+
+ user, pw = self.passwd.find_user_password(realm, req.get_full_url())
+ if user is None:
+ return None
+
+ # XXX not implemented yet
+ if req.has_data():
+ entdig = self.get_entity_digest(req.get_data(), chal)
+ else:
+ entdig = None
+
+ A1 = "%s:%s:%s" % (user, realm, pw)
+ A2 = "%s:%s" % (req.get_method(),
+ # XXX selector: what about proxies and full urls
+ req.get_selector())
+ if qop == 'auth':
+ if nonce == self.last_nonce:
+ self.nonce_count += 1
+ else:
+ self.nonce_count = 1
+ self.last_nonce = nonce
+
+ ncvalue = '%08x' % self.nonce_count
+ cnonce = self.get_cnonce(nonce)
+ noncebit = "%s:%s:%s:%s:%s" % (nonce, ncvalue, cnonce, qop, H(A2))
+ respdig = KD(H(A1), noncebit)
+ elif qop is None:
+ respdig = KD(H(A1), "%s:%s" % (nonce, H(A2)))
+ else:
+ # XXX handle auth-int.
+ raise URLError("qop '%s' is not supported." % qop)
+
+ # XXX should the partial digests be encoded too?
+
+ base = 'username="%s", realm="%s", nonce="%s", uri="%s", ' \
+ 'response="%s"' % (user, realm, nonce, req.get_selector(),
+ respdig)
+ if opaque:
+ base += ', opaque="%s"' % opaque
+ if entdig:
+ base += ', digest="%s"' % entdig
+ base += ', algorithm="%s"' % algorithm
+ if qop:
+ base += ', qop=auth, nc=%s, cnonce="%s"' % (ncvalue, cnonce)
+ return base
+
+ def get_algorithm_impls(self, algorithm):
+ # algorithm should be case-insensitive according to RFC2617
+ algorithm = algorithm.upper()
+ # lambdas assume digest modules are imported at the top level
+ if algorithm == 'MD5':
+ H = lambda x: hashlib.md5(x).hexdigest()
+ elif algorithm == 'SHA':
+ H = lambda x: hashlib.sha1(x).hexdigest()
+ # XXX MD5-sess
+ KD = lambda s, d: H("%s:%s" % (s, d))
+ return H, KD
+
+ def get_entity_digest(self, data, chal):
+ # XXX not implemented yet
+ return None
+
+
+class HTTPDigestAuthHandler(BaseHandler, AbstractDigestAuthHandler):
+ """An authentication protocol defined by RFC 2069
+
+ Digest authentication improves on basic authentication because it
+ does not transmit passwords in the clear.
+ """
+
+ auth_header = 'Authorization'
+ handler_order = 490 # before Basic auth
+
+ def http_error_401(self, req, fp, code, msg, headers):
+ host = urlparse.urlparse(req.get_full_url())[1]
+ retry = self.http_error_auth_reqed('www-authenticate',
+ host, req, headers)
+ self.reset_retry_count()
+ return retry
+
+
+class ProxyDigestAuthHandler(BaseHandler, AbstractDigestAuthHandler):
+
+ auth_header = 'Proxy-Authorization'
+ handler_order = 490 # before Basic auth
+
+ def http_error_407(self, req, fp, code, msg, headers):
+ host = req.get_host()
+ retry = self.http_error_auth_reqed('proxy-authenticate',
+ host, req, headers)
+ self.reset_retry_count()
+ return retry
+
+class AbstractHTTPHandler(BaseHandler):
+
+ def __init__(self, debuglevel=0):
+ self._debuglevel = debuglevel
+
+ def set_http_debuglevel(self, level):
+ self._debuglevel = level
+
+ def do_request_(self, request):
+ host = request.get_host()
+ if not host:
+ raise URLError('no host given')
+
+ if request.has_data(): # POST
+ data = request.get_data()
+ if not request.has_header('Content-type'):
+ request.add_unredirected_header(
+ 'Content-type',
+ 'application/x-www-form-urlencoded')
+ if not request.has_header('Content-length'):
+ request.add_unredirected_header(
+ 'Content-length', '%d' % len(data))
+
+ sel_host = host
+ if request.has_proxy():
+ scheme, sel = splittype(request.get_selector())
+ sel_host, sel_path = splithost(sel)
+
+ if not request.has_header('Host'):
+ request.add_unredirected_header('Host', sel_host)
+ for name, value in self.parent.addheaders:
+ name = name.capitalize()
+ if not request.has_header(name):
+ request.add_unredirected_header(name, value)
+
+ return request
+
+ def do_open(self, http_class, req):
+ """Return an addinfourl object for the request, using http_class.
+
+ http_class must implement the HTTPConnection API from httplib.
+ The addinfourl return value is a file-like object. It also
+ has methods and attributes including:
+ - info(): return a mimetools.Message object for the headers
+ - geturl(): return the original request URL
+ - code: HTTP status code
+ """
+ host = req.get_host()
+ if not host:
+ raise URLError('no host given')
+
+ h = http_class(host, timeout=req.timeout) # will parse host:port
+ h.set_debuglevel(self._debuglevel)
+
+ headers = dict(req.unredirected_hdrs)
+ headers.update(dict((k, v) for k, v in req.headers.items()
+ if k not in headers))
+
+ # We want to make an HTTP/1.1 request, but the addinfourl
+ # class isn't prepared to deal with a persistent connection.
+ # It will try to read all remaining data from the socket,
+ # which will block while the server waits for the next request.
+ # So make sure the connection gets closed after the (only)
+ # request.
+ headers["Connection"] = "close"
+ headers = dict(
+ (name.title(), val) for name, val in headers.items())
+
+ if req._tunnel_host:
+ tunnel_headers = {}
+ proxy_auth_hdr = "Proxy-Authorization"
+ if proxy_auth_hdr in headers:
+ tunnel_headers[proxy_auth_hdr] = headers[proxy_auth_hdr]
+ # Proxy-Authorization should not be sent to origin
+ # server.
+ del headers[proxy_auth_hdr]
+ h.set_tunnel(req._tunnel_host, headers=tunnel_headers)
+
+ try:
+ h.request(req.get_method(), req.get_selector(), req.data, headers)
+ try:
+ r = h.getresponse(buffering=True)
+ except TypeError: #buffering kw not supported
+ r = h.getresponse()
+ except socket.error, err: # XXX what error?
+ h.close()
+ raise URLError(err)
+
+ # Pick apart the HTTPResponse object to get the addinfourl
+ # object initialized properly.
+
+ # Wrap the HTTPResponse object in socket's file object adapter
+ # for Windows. That adapter calls recv(), so delegate recv()
+ # to read(). This weird wrapping allows the returned object to
+ # have readline() and readlines() methods.
+
+ # XXX It might be better to extract the read buffering code
+ # out of socket._fileobject() and into a base class.
+
+ r.recv = r.read
+ fp = socket._fileobject(r, close=True)
+
+ resp = addinfourl(fp, r.msg, req.get_full_url())
+ resp.code = r.status
+ resp.msg = r.reason
+ return resp
+
+
+class HTTPHandler(AbstractHTTPHandler):
+
+ def http_open(self, req):
+ return self.do_open(httplib.HTTPConnection, req)
+
+ http_request = AbstractHTTPHandler.do_request_
+
+if hasattr(httplib, 'HTTPS'):
+ class HTTPSHandler(AbstractHTTPHandler):
+
+ def https_open(self, req):
+ return self.do_open(httplib.HTTPSConnection, req)
+
+ https_request = AbstractHTTPHandler.do_request_
+
+class HTTPCookieProcessor(BaseHandler):
+ def __init__(self, cookiejar=None):
+ import cookielib
+ if cookiejar is None:
+ cookiejar = cookielib.CookieJar()
+ self.cookiejar = cookiejar
+
+ def http_request(self, request):
+ self.cookiejar.add_cookie_header(request)
+ return request
+
+ def http_response(self, request, response):
+ self.cookiejar.extract_cookies(response, request)
+ return response
+
+ https_request = http_request
+ https_response = http_response
+
+class UnknownHandler(BaseHandler):
+ def unknown_open(self, req):
+ type = req.get_type()
+ raise URLError('unknown url type: %s' % type)
+
+def parse_keqv_list(l):
+ """Parse list of key=value strings where keys are not duplicated."""
+ parsed = {}
+ for elt in l:
+ k, v = elt.split('=', 1)
+ if v[0] == '"' and v[-1] == '"':
+ v = v[1:-1]
+ parsed[k] = v
+ return parsed
+
+def parse_http_list(s):
+ """Parse lists as described by RFC 2068 Section 2.
+
+ In particular, parse comma-separated lists where the elements of
+ the list may include quoted-strings. A quoted-string could
+ contain a comma. A non-quoted string could have quotes in the
+ middle. Neither commas nor quotes count if they are escaped.
+ Only double-quotes count, not single-quotes.
+ """
+ res = []
+ part = ''
+
+ escape = quote = False
+ for cur in s:
+ if escape:
+ part += cur
+ escape = False
+ continue
+ if quote:
+ if cur == '\\':
+ escape = True
+ continue
+ elif cur == '"':
+ quote = False
+ part += cur
+ continue
+
+ if cur == ',':
+ res.append(part)
+ part = ''
+ continue
+
+ if cur == '"':
+ quote = True
+
+ part += cur
+
+ # append last part
+ if part:
+ res.append(part)
+
+ return [part.strip() for part in res]
+
+def _safe_gethostbyname(host):
+ try:
+ return socket.gethostbyname(host)
+ except socket.gaierror:
+ return None
+
+class FileHandler(BaseHandler):
+ # 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] != '/' and (req.host and
+ req.host != 'localhost'):
+ req.type = 'ftp'
+ return self.parent.open(req)
+ else:
+ return self.open_local_file(req)
+
+ # names for the localhost
+ names = None
+ def get_names(self):
+ if FileHandler.names is None:
+ try:
+ FileHandler.names = tuple(
+ socket.gethostbyname_ex('localhost')[2] +
+ socket.gethostbyname_ex(socket.gethostname())[2])
+ except socket.gaierror:
+ FileHandler.names = (socket.gethostbyname('localhost'),)
+ return FileHandler.names
+
+ # not entirely sure what the rules are here
+ def open_local_file(self, req):
+ import email.utils
+ import mimetypes
+ host = req.get_host()
+ filename = req.get_selector()
+ localfile = url2pathname(filename)
+ try:
+ stats = os.stat(localfile)
+ size = stats.st_size
+ modified = email.utils.formatdate(stats.st_mtime, usegmt=True)
+ mtype = mimetypes.guess_type(filename)[0]
+ headers = mimetools.Message(StringIO(
+ 'Content-type: %s\nContent-length: %d\nLast-modified: %s\n' %
+ (mtype or 'text/plain', size, modified)))
+ if host:
+ host, port = splitport(host)
+ if not host or \
+ (not port and _safe_gethostbyname(host) in self.get_names()):
+ if host:
+ origurl = 'file://' + host + filename
+ else:
+ origurl = 'file://' + filename
+ return addinfourl(open(localfile, 'rb'), headers, origurl)
+ except OSError, msg:
+ # urllib2 users shouldn't expect OSErrors coming from urlopen()
+ raise URLError(msg)
+ raise URLError('file not on local host')
+
+class FTPHandler(BaseHandler):
+ def ftp_open(self, req):
+ import ftplib
+ import mimetypes
+ host = req.get_host()
+ if not host:
+ raise URLError('ftp error: no host given')
+ host, port = splitport(host)
+ if port is None:
+ port = ftplib.FTP_PORT
+ else:
+ port = int(port)
+
+ # username/password handling
+ user, host = splituser(host)
+ if user:
+ user, passwd = splitpasswd(user)
+ else:
+ passwd = None
+ host = unquote(host)
+ user = user or ''
+ passwd = passwd or ''
+
+ try:
+ host = socket.gethostbyname(host)
+ except socket.error, msg:
+ raise URLError(msg)
+ path, attrs = splitattr(req.get_selector())
+ dirs = path.split('/')
+ dirs = map(unquote, dirs)
+ dirs, file = dirs[:-1], dirs[-1]
+ if dirs and not dirs[0]:
+ dirs = dirs[1:]
+ try:
+ fw = self.connect_ftp(user, passwd, host, port, dirs, req.timeout)
+ type = file and 'I' or 'D'
+ for attr in attrs:
+ attr, value = splitvalue(attr)
+ if attr.lower() == 'type' and \
+ value in ('a', 'A', 'i', 'I', 'd', 'D'):
+ type = value.upper()
+ fp, retrlen = fw.retrfile(file, type)
+ headers = ""
+ mtype = mimetypes.guess_type(req.get_full_url())[0]
+ if mtype:
+ headers += "Content-type: %s\n" % mtype
+ if retrlen is not None and retrlen >= 0:
+ headers += "Content-length: %d\n" % retrlen
+ sf = StringIO(headers)
+ headers = mimetools.Message(sf)
+ return addinfourl(fp, headers, req.get_full_url())
+ except ftplib.all_errors, msg:
+ raise URLError, ('ftp error: %s' % msg), sys.exc_info()[2]
+
+ def connect_ftp(self, user, passwd, host, port, dirs, timeout):
+ fw = ftpwrapper(user, passwd, host, port, dirs, timeout)
+## fw.ftp.set_debuglevel(1)
+ return fw
+
+class CacheFTPHandler(FTPHandler):
+ # XXX would be nice to have pluggable cache strategies
+ # XXX this stuff is definitely not thread safe
+ def __init__(self):
+ self.cache = {}
+ self.timeout = {}
+ self.soonest = 0
+ self.delay = 60
+ self.max_conns = 16
+
+ def setTimeout(self, t):
+ self.delay = t
+
+ def setMaxConns(self, m):
+ self.max_conns = m
+
+ def connect_ftp(self, user, passwd, host, port, dirs, timeout):
+ key = user, host, port, '/'.join(dirs), timeout
+ if key in self.cache:
+ self.timeout[key] = time.time() + self.delay
+ else:
+ self.cache[key] = ftpwrapper(user, passwd, host, port, dirs, timeout)
+ self.timeout[key] = time.time() + self.delay
+ self.check_cache()
+ return self.cache[key]
+
+ def check_cache(self):
+ # first check for old ones
+ t = time.time()
+ if self.soonest <= t:
+ for k, v in self.timeout.items():
+ if v < t:
+ self.cache[k].close()
+ del self.cache[k]
+ del self.timeout[k]
+ self.soonest = min(self.timeout.values())
+
+ # then check the size
+ if len(self.cache) == self.max_conns:
+ for k, v in self.timeout.items():
+ if v == self.soonest:
+ del self.cache[k]
+ del self.timeout[k]
+ break
+ self.soonest = min(self.timeout.values())
diff --git a/lib_pypy/_ctypes/basics.py b/lib_pypy/_ctypes/basics.py
--- a/lib_pypy/_ctypes/basics.py
+++ b/lib_pypy/_ctypes/basics.py
@@ -54,7 +54,8 @@
def get_ffi_argtype(self):
if self._ffiargtype:
return self._ffiargtype
- return _shape_to_ffi_type(self._ffiargshape)
+ self._ffiargtype = _shape_to_ffi_type(self._ffiargshape)
+ return self._ffiargtype
def _CData_output(self, resbuffer, base=None, index=-1):
#assert isinstance(resbuffer, _rawffi.ArrayInstance)
@@ -166,7 +167,8 @@
return tp._alignmentofinstances()
def byref(cdata):
- from ctypes import pointer
+ # "pointer" is imported at the end of this module to avoid circular
+ # imports
return pointer(cdata)
def cdata_from_address(self, address):
@@ -224,5 +226,9 @@
'Z' : _ffi.types.void_p,
'X' : _ffi.types.void_p,
'v' : _ffi.types.sshort,
+ '?' : _ffi.types.ubyte,
}
+
+# used by "byref"
+from _ctypes.pointer import pointer
diff --git a/lib_pypy/_ctypes/function.py b/lib_pypy/_ctypes/function.py
--- a/lib_pypy/_ctypes/function.py
+++ b/lib_pypy/_ctypes/function.py
@@ -78,8 +78,6 @@
_com_iid = None
_is_fastpath = False
- __restype_set = False
-
def _getargtypes(self):
return self._argtypes_
@@ -93,13 +91,15 @@
raise TypeError(
"item %d in _argtypes_ has no from_param method" % (
i + 1,))
- #
- if all([hasattr(argtype, '_ffiargshape') for argtype in argtypes]):
- fastpath_cls = make_fastpath_subclass(self.__class__)
- fastpath_cls.enable_fastpath_maybe(self)
self._argtypes_ = list(argtypes)
+ self._check_argtypes_for_fastpath()
argtypes = property(_getargtypes, _setargtypes)
+ def _check_argtypes_for_fastpath(self):
+ if all([hasattr(argtype, '_ffiargshape') for argtype in self._argtypes_]):
+ fastpath_cls = make_fastpath_subclass(self.__class__)
+ fastpath_cls.enable_fastpath_maybe(self)
+
def _getparamflags(self):
return self._paramflags
@@ -149,7 +149,6 @@
return self._restype_
def _setrestype(self, restype):
- self.__restype_set = True
self._ptr = None
if restype is int:
from ctypes import c_int
@@ -219,6 +218,7 @@
import ctypes
restype = ctypes.c_int
self._ptr = self._getfuncptr_fromaddress(self._argtypes_, restype)
+ self._check_argtypes_for_fastpath()
return
@@ -296,13 +296,12 @@
"This function takes %d argument%s (%s given)"
% (len(self._argtypes_), plural, len(args)))
- # check that arguments are convertible
- ## XXX Not as long as ctypes.cast is a callback function with
- ## py_object arguments...
- ## self._convert_args(self._argtypes_, args, {})
-
try:
- res = self.callable(*args)
+ newargs = self._convert_args_for_callback(argtypes, args)
+ except (UnicodeError, TypeError, ValueError), e:
+ raise ArgumentError(str(e))
+ try:
+ res = self.callable(*newargs)
except:
exc_info = sys.exc_info()
traceback.print_tb(exc_info[2], file=sys.stderr)
@@ -316,10 +315,6 @@
warnings.warn('C function without declared arguments called',
RuntimeWarning, stacklevel=2)
argtypes = []
-
- if not self.__restype_set:
- warnings.warn('C function without declared return type called',
- RuntimeWarning, stacklevel=2)
if self._com_index:
from ctypes import cast, c_void_p, POINTER
@@ -366,7 +361,10 @@
if self._flags_ & _rawffi.FUNCFLAG_USE_LASTERROR:
set_last_error(_rawffi.get_last_error())
#
- return self._build_result(self._restype_, result, newargs)
+ try:
+ return self._build_result(self._restype_, result, newargs)
+ finally:
+ funcptr.free_temp_buffers()
def _do_errcheck(self, result, args):
# The 'errcheck' protocol
@@ -466,6 +464,19 @@
return cobj, cobj._to_ffi_param(), type(cobj)
+ def _convert_args_for_callback(self, argtypes, args):
+ assert len(argtypes) == len(args)
+ newargs = []
+ for argtype, arg in zip(argtypes, args):
+ param = argtype.from_param(arg)
+ _type_ = getattr(argtype, '_type_', None)
+ if _type_ == 'P': # special-case for c_void_p
+ param = param._get_buffer_value()
+ elif self._is_primitive(argtype):
+ param = param.value
+ newargs.append(param)
+ return newargs
+
def _convert_args(self, argtypes, args, kwargs, marker=object()):
newargs = []
outargs = []
@@ -556,6 +567,9 @@
newargtypes.append(newargtype)
return keepalives, newargs, newargtypes, outargs
+ @staticmethod
+ def _is_primitive(argtype):
+ return argtype.__bases__[0] is _SimpleCData
def _wrap_result(self, restype, result):
"""
@@ -564,7 +578,7 @@
"""
# hack for performance: if restype is a "simple" primitive type, don't
# allocate the buffer because it's going to be thrown away immediately
- if restype.__bases__[0] is _SimpleCData and not restype._is_pointer_like():
+ if self._is_primitive(restype) and not restype._is_pointer_like():
return result
#
shape = restype._ffishape
@@ -680,7 +694,7 @@
try:
result = self._call_funcptr(funcptr, *args)
result = self._do_errcheck(result, args)
- except (TypeError, ArgumentError): # XXX, should be FFITypeError
+ except (TypeError, ArgumentError, UnicodeDecodeError):
assert self._slowpath_allowed
return CFuncPtr.__call__(self, *args)
return result
diff --git a/lib_pypy/_ctypes/primitive.py b/lib_pypy/_ctypes/primitive.py
--- a/lib_pypy/_ctypes/primitive.py
+++ b/lib_pypy/_ctypes/primitive.py
@@ -10,6 +10,8 @@
from _ctypes.builtin import ConvMode
from _ctypes.array import Array
from _ctypes.pointer import _Pointer, as_ffi_pointer
+#from _ctypes.function import CFuncPtr # this import is moved at the bottom
+ # because else it's circular
class NULL(object):
pass
@@ -86,7 +88,7 @@
return res
if isinstance(value, Array):
return value
- if isinstance(value, _Pointer):
+ if isinstance(value, (_Pointer, CFuncPtr)):
return cls.from_address(value._buffer.buffer)
if isinstance(value, (int, long)):
return cls(value)
@@ -338,3 +340,5 @@
def __nonzero__(self):
return self._buffer[0] not in (0, '\x00')
+
+from _ctypes.function import CFuncPtr
diff --git a/lib_pypy/_ctypes/structure.py b/lib_pypy/_ctypes/structure.py
--- a/lib_pypy/_ctypes/structure.py
+++ b/lib_pypy/_ctypes/structure.py
@@ -14,6 +14,15 @@
raise TypeError("Expected CData subclass, got %s" % (tp,))
if isinstance(tp, StructOrUnionMeta):
tp._make_final()
+ if len(f) == 3:
+ if (not hasattr(tp, '_type_')
+ or not isinstance(tp._type_, str)
+ or tp._type_ not in "iIhHbBlL"):
+ #XXX: are those all types?
+ # we just dont get the type name
+ # in the interp levle thrown TypeError
+ # from rawffi if there are more
+ raise TypeError('bit fields not allowed for type ' + tp.__name__)
all_fields = []
for cls in reversed(inspect.getmro(superclass)):
@@ -34,34 +43,37 @@
for i, field in enumerate(all_fields):
name = field[0]
value = field[1]
+ is_bitfield = (len(field) == 3)
fields[name] = Field(name,
self._ffistruct.fieldoffset(name),
self._ffistruct.fieldsize(name),
- value, i)
+ value, i, is_bitfield)
if anonymous_fields:
resnames = []
for i, field in enumerate(all_fields):
name = field[0]
value = field[1]
+ is_bitfield = (len(field) == 3)
startpos = self._ffistruct.fieldoffset(name)
if name in anonymous_fields:
for subname in value._names:
resnames.append(subname)
- relpos = startpos + value._fieldtypes[subname].offset
- subvalue = value._fieldtypes[subname].ctype
+ subfield = getattr(value, subname)
+ relpos = startpos + subfield.offset
+ subvalue = subfield.ctype
fields[subname] = Field(subname,
relpos, subvalue._sizeofinstances(),
- subvalue, i)
+ subvalue, i, is_bitfield)
else:
resnames.append(name)
names = resnames
self._names = names
- self._fieldtypes = fields
+ self.__dict__.update(fields)
class Field(object):
- def __init__(self, name, offset, size, ctype, num):
- for k in ('name', 'offset', 'size', 'ctype', 'num'):
+ def __init__(self, name, offset, size, ctype, num, is_bitfield):
+ for k in ('name', 'offset', 'size', 'ctype', 'num', 'is_bitfield'):
self.__dict__[k] = locals()[k]
def __setattr__(self, name, value):
@@ -71,6 +83,35 @@
return "<Field '%s' offset=%d size=%d>" % (self.name, self.offset,
self.size)
+ def __get__(self, obj, cls=None):
+ if obj is None:
+ return self
+ if self.is_bitfield:
+ # bitfield member, use direct access
+ return obj._buffer.__getattr__(self.name)
+ else:
+ fieldtype = self.ctype
+ offset = self.num
+ suba = obj._subarray(fieldtype, self.name)
+ return fieldtype._CData_output(suba, obj, offset)
+
+
+ def __set__(self, obj, value):
+ fieldtype = self.ctype
+ cobj = fieldtype.from_param(value)
+ if ensure_objects(cobj) is not None:
+ key = keepalive_key(self.num)
+ store_reference(obj, key, cobj._objects)
+ arg = cobj._get_buffer_value()
+ if fieldtype._fficompositesize is not None:
+ from ctypes import memmove
+ dest = obj._buffer.fieldaddress(self.name)
+ memmove(dest, arg, fieldtype._fficompositesize)
+ else:
+ obj._buffer.__setattr__(self.name, arg)
+
+
+
# ________________________________________________________________
def _set_shape(tp, rawfields, is_union=False):
@@ -79,17 +120,12 @@
tp._ffiargshape = tp._ffishape = (tp._ffistruct, 1)
tp._fficompositesize = tp._ffistruct.size
-def struct_getattr(self, name):
- if name not in ('_fields_', '_fieldtypes'):
- if hasattr(self, '_fieldtypes') and name in self._fieldtypes:
- return self._fieldtypes[name]
- return _CDataMeta.__getattribute__(self, name)
def struct_setattr(self, name, value):
if name == '_fields_':
if self.__dict__.get('_fields_', None) is not None:
raise AttributeError("_fields_ is final")
- if self in [v for k, v in value]:
+ if self in [f[1] for f in value]:
raise AttributeError("Structure or union cannot contain itself")
names_and_fields(
self,
@@ -127,14 +163,14 @@
if '_fields_' not in self.__dict__:
self._fields_ = []
self._names = []
- self._fieldtypes = {}
_set_shape(self, [], self._is_union)
- __getattr__ = struct_getattr
__setattr__ = struct_setattr
def from_address(self, address):
instance = StructOrUnion.__new__(self)
+ if isinstance(address, _rawffi.StructureInstance):
+ address = address.buffer
instance.__dict__['_buffer'] = self._ffistruct.fromaddress(address)
return instance
@@ -200,40 +236,6 @@
A = _rawffi.Array(fieldtype._ffishape)
return A.fromaddress(address, 1)
- def __setattr__(self, name, value):
- try:
- field = self._fieldtypes[name]
- except KeyError:
- return _CData.__setattr__(self, name, value)
- fieldtype = field.ctype
- cobj = fieldtype.from_param(value)
- if ensure_objects(cobj) is not None:
- key = keepalive_key(field.num)
- store_reference(self, key, cobj._objects)
- arg = cobj._get_buffer_value()
- if fieldtype._fficompositesize is not None:
- from ctypes import memmove
- dest = self._buffer.fieldaddress(name)
- memmove(dest, arg, fieldtype._fficompositesize)
- else:
- self._buffer.__setattr__(name, arg)
-
- def __getattribute__(self, name):
- if name == '_fieldtypes':
- return _CData.__getattribute__(self, '_fieldtypes')
- try:
- field = self._fieldtypes[name]
- except KeyError:
- return _CData.__getattribute__(self, name)
- if field.size >> 16:
- # bitfield member, use direct access
- return self._buffer.__getattr__(name)
- else:
- fieldtype = field.ctype
- offset = field.num
- suba = self._subarray(fieldtype, name)
- return fieldtype._CData_output(suba, self, offset)
-
def _get_buffer_for_param(self):
return self
diff --git a/lib_pypy/_elementtree.py b/lib_pypy/_elementtree.py
new file mode 100644
--- /dev/null
+++ b/lib_pypy/_elementtree.py
@@ -0,0 +1,6 @@
+# Just use ElementTree.
+
+from xml.etree import ElementTree
+
+globals().update(ElementTree.__dict__)
+del __all__
diff --git a/lib_pypy/_functools.py b/lib_pypy/_functools.py
--- a/lib_pypy/_functools.py
+++ b/lib_pypy/_functools.py
@@ -14,10 +14,9 @@
raise TypeError("the first argument must be callable")
self.func = func
self.args = args
- self.keywords = keywords
+ self.keywords = keywords or None
def __call__(self, *fargs, **fkeywords):
- newkeywords = self.keywords.copy()
- newkeywords.update(fkeywords)
- return self.func(*(self.args + fargs), **newkeywords)
-
+ if self.keywords is not None:
+ fkeywords = dict(self.keywords, **fkeywords)
+ return self.func(*(self.args + fargs), **fkeywords)
diff --git a/lib_pypy/_pypy_interact.py b/lib_pypy/_pypy_interact.py
--- a/lib_pypy/_pypy_interact.py
+++ b/lib_pypy/_pypy_interact.py
@@ -56,6 +56,10 @@
prompt = getattr(sys, 'ps1', '>>> ')
try:
line = raw_input(prompt)
+ # Can be None if sys.stdin was redefined
+ encoding = getattr(sys.stdin, 'encoding', None)
+ if encoding and not isinstance(line, unicode):
+ line = line.decode(encoding)
except EOFError:
console.write("\n")
break
diff --git a/lib_pypy/_pypy_irc_topic.py b/lib_pypy/_pypy_irc_topic.py
--- a/lib_pypy/_pypy_irc_topic.py
+++ b/lib_pypy/_pypy_irc_topic.py
@@ -1,117 +1,6 @@
-"""qvfgbcvna naq hgbcvna punvef
-qlfgbcvna naq hgbcvna punvef
-V'z fbeel, pbhyq lbh cyrnfr abg nterr jvgu gur png nf jryy?
-V'z fbeel, pbhyq lbh cyrnfr abg nterr jvgu gur punve nf jryy?
-jr cnffrq gur RH erivrj
-cbfg RhebClguba fcevag fgnegf 12.IVV.2007, 10nz
-RhebClguba raqrq
-n Pyrna Ragrecevfrf cebqhpgvba
-npnqrzl vf n pbzcyvpngrq ebyr tnzr
-npnqrzvn vf n pbzcyvpngrq ebyr tnzr
-jbexvat pbqr vf crn fbhc
-abg lbhe snhyg, zber yvxr vg'f n zbivat gnetrg
-guvf fragrapr vf snyfr
-abguvat vf gehr
-Yncfnat Fbhpubat
-Oenpunzhgnaqn
-fbeel, V'yy grnpu gur pnpghf ubj gb fjvz yngre
-Jul fb znal znal znal znal znal ivbyvaf?
-Jul fb znal znal znal znal znal bowrpgf?
-"eha njnl naq yvir ba n snez" nccebnpu gb fbsgjner qrirybczrag
-"va snpg, lbh zvtug xabj zber nobhg gur genafyngvba gbbypunva nsgre znfgrevat eclguba guna fbzr angvir fcrnxre xabjf nobhg uvf zbgure gbathr" - kbeNkNk
-"jurer qvq nyy gur ivbyvaf tb?"
-- ClCl fgnghf oybt: uggc://zberclcl.oybtfcbg.pbz/
-uggc://kxpq.pbz/353/
-pnfhnyvgl ivbyngvbaf naq sylvat
-wrgmg abpu fpubxbynqvtre
-R09 2X @PNN:85?
-vs lbh'er gelvat gb oybj hc fghss, jub pnerf?
-vs fghss oybjf hc, lbh pner
-2008 jvyy or gur lrne bs clcl ba gur qrfxgbc
-2008 jvyy or gur lrne bs gur qrfxgbc ba #clcl
-2008 jvyy or gur lrne bs gur qrfxgbc ba #clcl, Wnahnel jvyy or gur zbagu bs gur nyc gbcf
-lrf, ohg jung'g gur frafr bs 0 < "qhena qhena"
-eclguba: flagnk naq frznagvpf bs clguba, fcrrq bs p, erfgevpgvbaf bs wnin naq pbzcvyre reebe zrffntrf nf crargenoyr nf ZHZCF
+"""eclguba: flagnk naq frznagvpf bs clguba, fcrrq bs p, erfgevpgvbaf bs wnin naq pbzcvyre reebe zrffntrf nf crargenoyr nf ZHZCF
pglcrf unf n fcva bs 1/3
' ' vf n fcnpr gbb
-2009 jvyy or gur lrne bs WVG ba gur qrfxgbc
-N ynathntr vf n qvnyrpg jvgu na nezl naq anil
-gbcvpf ner sbe gur srroyr zvaqrq
-2009 vf gur lrne bs ersyrpgvba ba gur qrfxgbc
-gur tybor vf bhe cbal, gur pbfzbf bhe erny ubefr
-jub nz V naq vs lrf, ubj znal?
-cebtenzzvat va orq vf n cresrpgyl svar npgvivgl
-zbber'f ynj vf n qeht jvgu gur jbefg pbzr qbja
-EClguba: jr hfr vg fb lbh qba'g unir gb
-Zbber'f ynj vf n qeht jvgu gur jbefg pbzr qbja. EClguba: haqrpvqrq.
-guvatf jvyy or avpr naq fghss
-qba'g cbfg yvaxf gb cngragf urer
-Abg lbhe hfhny nanylfrf.
-Gur Neg bs gur Punaary
-Clguba 300
-V fhccbfr ZO bs UGZY cre frpbaq vf abg gur hfhny fcrrq zrnfher crbcyr jbhyq rkcrpg sbe n wvg
-gur fha arire frgf ba gur ClCl rzcver
-ghegyrf ner snfgre guna lbh guvax
-cebtenzzvat vf na nrfgrguvp raqrnibhe
-P vf tbbq sbe fbzrguvat, whfg abg sbe jevgvat fbsgjner
-trezna vf tbbq sbe fbzrguvat, whfg abg sbe jevgvat fbsgjner
-trezna vf tbbq sbe artngvbaf, whfg abg sbe jevgvat fbsgjner
-# nffreg qvq abg penfu
-lbh fubhyq fgneg n cresrpg fbsgjner zbirzrag
-lbh fubhyq fgneg n cresrpg punaary gbcvp zbirzrag
-guvf vf n cresrpg punaary gbcvp
-guvf vf n frys-ersreragvny punaary gbcvp
-crrcubcr bcgvzvmngvbaf ner jung n Fhssvpvragyl Fzneg Pbzcvyre hfrf
-"crrcubcr" bcgvzvmngvbaf ner jung na bcgvzvfgvp Pbzcvyre hfrf
-pubbfr lbhe unpx
-gur 'fhcre' xrljbeq vf abg gung uhttnoyr
-wlguba cngpurf ner abg rabhtu sbe clcl
-- qb lbh xabj oreyva? - nyy bs vg? - jryy, whfg oreyva
-- ubj jvyy gur snpg gung gurl ner hfrq va bhe ercy punatr bhe gbcvpf?
-- ubj pna vg rire unir jbexrq?
-- jurer fubhyq gur unpx or fgberq?
-- Vg'f uneq gb fnl rknpgyl jung pbafgvghgrf erfrnepu va gur pbzchgre jbeyq, ohg nf n svefg nccebkvzngvba, vg'f fbsgjner gung qbrfa'g unir hfref.
-- Cebtenzzvat vf nyy nobhg xabjvat jura gb obvy gur benatr fcbatr qbaxrl npebff gur cuvyyvcvarf
-- Jul fb znal, znal, znal, znal, znal, znal qhpxyvatf?
-- ab qrgnvy vf bofpher rabhtu gb abg unir fbzr pbqr qrcraqvat ba vg.
-- jung V trarenyyl jnag vf serr fcrrqhcf
-- nyy bs ClCl vf kv-dhnyvgl
-"lbh pna nyjnlf xvyy -9 be bf._rkvg() vs lbh'er va n uheel"
-Ohernhpengf ohvyq npnqrzvp rzcverf juvpu puhea bhg zrnavatyrff fbyhgvbaf gb veeryrinag ceboyrzf.
-vg'f abg n unpx, vg'f n jbexnebhaq
-ClCl qbrfa'g unir pbcbylinevnqvp qrcraqragyl-zbabzbecurq ulcresyhknqf
-ClCl qbrfa'g punatr gur shaqnzragny culfvpf pbafgnagf
-Qnapr bs gur Fhtnecyhz Snvel
-Wnin vf whfg tbbq rabhtu gb or cenpgvpny, ohg abg tbbq rabhtu gb or hfnoyr.
-RhebClguba vf unccravat, qba'g rkcrpg nal dhvpx erfcbafr gvzrf.
-"V jbhyq yvxr gb fgnl njnl sebz ernyvgl gura"
-"gung'f jul gur 'be' vf ernyyl na 'naq' "
-jvgu nyy nccebcevngr pbagrkghnyvfngvbavat
-qba'g gevc ba gur cbjre pbeq
-vzcyrzragvat YBTB va YBTB: "ghegyrf nyy gur jnl qbja"
-gur ohooyrfbeg jbhyq or gur jebat jnl gb tb
-gur cevapvcyr bs pbafreingvba bs zrff
-gb fnir n gerr, rng n ornire
-Qre Ovore znpugf evpugvt: Antg nyyrf xnchgg.
-"Nal jbeyqivrj gung vfag jenpxrq ol frys-qbhog naq pbashfvba bire vgf bja vqragvgl vf abg n jbeyqivrj sbe zr." - Fpbgg Nnebafba
-jr oryvrir va cnapnxrf, znlor
-jr oryvrir va ghegyrf, znlor
-jr qrsvavgryl oryvrir va zrgn
-gur zngevk unf lbh
-"Yvsr vf uneq, gura lbh anc" - n png
-Vf Nezva ubzr jura gur havirefr prnfrf gb rkvfg?
-Qhrffryqbes fcevag fgnegrq
-frys.nobeeg("pnaabg ybnq negvpyrf")
-QRAGVFGEL FLZOBY YVTUG IREGVPNY NAQ JNIR
-"Gur UUH pnzchf vf n tbbq Dhnxr yriry" - Nezva
-"Gur UUH pnzchf jbhyq or n greevoyr dhnxr yriry - lbh'q arire unir n pyhr jurer lbh ner" - zvpunry
-N enqvbnpgvir png unf 18 unys-yvirf.
-<rfp> : j [fvtu] <onpxfcnpr> <onpxfcnpr> <pgey>-f
-pbybe-pbqrq oyhrf
-"Neebtnapr va pbzchgre fpvrapr vf zrnfherq va anab-Qvwxfgenf."
-ClCl arrqf n Whfg-va-Gvzr WVG
-"Lbh pna'g gvzr geniry whfg ol frggvat lbhe pybpxf jebat"
-Gjb guernqf jnyx vagb n one. Gur onexrrcre ybbxf hc naq lryyf, "url, V jnag qba'g nal pbaqvgvbaf enpr yvxr gvzr ynfg!"
Clguba 2.k rfg cerfdhr zbeg, ivir Clguba!
Clguba 2.k vf abg qrnq
Riregvzr fbzrbar nethrf jvgu "Fznyygnyx unf nyjnlf qbar K", vg vf nyjnlf n tbbq uvag gung fbzrguvat arrqf gb or punatrq snfg. - Znephf Qraxre
@@ -119,7 +8,6 @@
__kkk__ naq __ekkk__ if bcrengvba fybgf: cnegvpyr dhnaghz fhcrecbfvgvba xvaq bs sha
ClCl vf na rkpvgvat grpuabybtl gung yrgf lbh gb jevgr snfg, cbegnoyr, zhygv-cyngsbez vagrecergref jvgu yrff rssbeg
Nezva: "Cebybt vf n zrff.", PS: "Ab, vg'f irel pbby!", Nezva: "Vfa'g guvf jung V fnvq?"
-<nevtngb> tbbq, grfgf ner hfrshy fbzrgvzrf :-)
ClCl vf yvxr nofheq gurngre
jr unir ab nagv-vzcbffvoyr fgvpx gung znxrf fher gung nyy lbhe cebtenzf unyg
clcl vf n enpr orgjrra crbcyr funivat lnxf naq gur havirefr cebqhpvat zber orneqrq lnxf. Fb sne, gur havirefr vf jvaavat
@@ -136,14 +24,14 @@
ClCl 1.1.0orgn eryrnfrq: uggc://pbqrfcrnx.arg/clcl/qvfg/clcl/qbp/eryrnfr-1.1.0.ugzy
"gurer fubhyq or bar naq bayl bar boivbhf jnl gb qb vg". ClCl inevnag: "gurer pna or A unys-ohttl jnlf gb qb vg"
1.1 svany eryrnfrq: uggc://pbqrfcrnx.arg/clcl/qvfg/clcl/qbp/eryrnfr-1.1.0.ugzy
-1.1 svany eryrnfrq | <svwny> nzq64 naq ccp ner bayl ninvynoyr va ragrecevfr irefvba
+<svwny> nzq64 naq ccp ner bayl ninvynoyr va ragrecevfr irefvba
Vf gurer n clcl gvzr? - vs lbh pna srry vg (?) gura gurer vf
<nevtngb> ab, abezny jbex vf fhpu zhpu yrff gvevat guna inpngvbaf
<nevtngb> ab, abezny jbex vf fb zhpu yrff gvevat guna inpngvbaf
-SVEFG gurl vtaber lbh, gura gurl ynhtu ng lbh, gura gurl svtug lbh, gura lbh jva.
+-SVEFG gurl vtaber lbh, gura gurl ynhtu ng lbh, gura gurl svtug lbh, gura lbh jva.-
vg'f Fhaqnl, znlor
vg'f Fhaqnl, ntnva
-"3 + 3 = 8" Nagb va gur WVG gnyx
+"3 + 3 = 8" - Nagb va gur WVG gnyx
RPBBC vf unccravat
RPBBC vf svavfurq
cflpb rngf bar oenva cre vapu bs cebterff
@@ -175,10 +63,108 @@
"nu, whfg va gvzr qbphzragngvba" (__nc__)
ClCl vf abg n erny IZ: ab frtsnhyg unaqyref gb qb gur ener pnfrf
lbh pna'g unir obgu pbairavrapr naq fcrrq
-gur WVG qbrfa'g jbex ba BF/K (abi'09)
-ab fhccbeg sbe BF/K evtug abj! (abi'09)
fyvccref urvtug pna or zrnfherq va k86 ertvfgref
clcl vf n enpr orgjrra gur vaqhfgel gelvat gb ohvyq znpuvarf jvgu zber naq zber erfbheprf, naq gur clcl qrirybcref gelvat gb rng nyy bs gurz. Fb sne, gur jvaare vf fgvyy hapyrne
+"znl pbagnva ahgf naq/be lbhat cbvagref"
+vg'f nyy irel fvzcyr, yvxr gur ubyvqnlf
+unccl ClCl'f lrne 2010!
+fnzhryr fnlf gung jr ybfg n enmbe. fb jr pna'g funir lnxf
+"yrg'f abg or bofpher, hayrff jr ernyyl arrq gb"
+<nevtngb> (abg guernq-fnsr, ohg jryy, abguvat vf)
+clcl unf znal ceboyrzf, ohg rnpu bar unf znal fbyhgvbaf
+whfg nabgure vgrz (1.333...) ba bhe erny-ahzorerq gbqb yvfg
+ClCl vf Fuveg Bevtnzv erfrnepu
+<svwny> nafjrevat n dhrfgvba: "ab -- sbe ng yrnfg bar cbffvoyr vagrecergngvba bs lbhe fragrapr"
+eryrnfr 1.2 hcpbzvat
+ClCl 1.2 eryrnfrq - uggc://clcl.bet/
+AB IPF QVFPHFFVBAF
+EClguba vf n svar pnzry unve oehfu
+ClCl vf n npghnyyl n ivfhnyvmngvba cebwrpg, jr whfg ohvyq vagrecergref gb unir vagrerfgvat qngn gb ivfhnyvmr
+clcl vf yvxr fnhfntrf
+naq abj sbe fbzrguvat pbzcyrgryl qvssrerag
+n 10gu bs sberire vf 1u45
+pbeerpg pbqr qbrfag arrq nal grfgf <qhpx>
+cbfgfgehpghenyvfz rgp.
+clcl UVG trarengbe
+gur arj clcl fcbeg vf gb cnff clcl ohtf nf pclguba ohtf
+jr unir zhpu zber vagrecergref guna hfref
+ClCl 1.3 njnvgvat eryrnfr
+ClCl 1.3 eryrnfrq
+vg frrzf gb zr gung bapr lbh frggyr ba na rkrphgvba / bowrpg zbqry naq / be olgrpbqr sbezng, lbh'ir nyernql qrpvqrq jung ynathntrf (jurer gur 'f' frrzf fhcresyhbhf) fhccbeg vf tbvat gb or svefg pynff sbe
+"Nyy ceboyrzf va ClCl pna or fbyirq ol nabgure yriry bs vagrecergngvba"
+ClCl 1.3 eryrnfrq (jvaqbjf ovanevrf vapyhqrq)
+jul qvq lbh thlf unir gb znxr gur ohvygva sbeghar zber vagrerfgvat guna npghny jbex? v whfg pngpurq zlfrys erfgnegvat clcl 20 gvzrf
+"jr hfrq gb unir n zrff jvgu na bofpher vagresnpr, abj jr unir zrff urer naq bofpher vagresnpr gurer. cebterff" crqebavf ba n clcl fcevag
+"phcf bs pbssrr ner yvxr nanybtvrf va gung V'z znxvat bar evtug abj"
+"vg'f nyjnlf hc gb hf, va n jnl be gur bgure"
+ClCl vf infg, naq pbagnvaf zhygvghqrf
+qravny vf eneryl n tbbq qrohttvat grpuavdhr
+"Yrg'f tb." - "Jr pna'g" - "Jul abg?" - "Jr'er jnvgvat sbe n Genafyngvba." - (qrfcnvevatyl) "Nu!"
+'gung'f qrsvavgryl n pnfr bs "hu????"'
+va gurbel gurer vf gur Ybbc, va cenpgvpr gurer ner oevqtrf
+gur uneqqevir - pbafgnag qngn cvytevzntr
+ClCl vf n gbby gb xrrc bgurejvfr qnatrebhf zvaqf fnsryl bpphcvrq.
+jr ner n trareny senzrjbex ohvyg ba pbafvfgrag nccyvpngvba bs nqubp-arff
+gur jnl gb nibvq n jbexnebhaq vf gb vagebqhpr n fgebatre jbexnebhaq fbzrjurer ryfr
+pnyyvat gur genafyngvba gbby punva n 'fpevcg' vf xvaq bs bssrafvir
+ehaavat clcl-p genafyngr.cl vf n ovg yvxr jngpuvat n guevyyre zbivr, vg pbhyq pbafhzr nyy gur zrzbel ng nal gvzr
+ehaavat clcl-p genafyngr.cl vf n ovg yvxr jngpuvat n guevyyre zbivr, vg pbhyq qvr ng nal gvzr orpnhfr bs gur 32-ovg 4TO yvzvg bs ENZ
+Qh jvefg rora tranh qnf reervpura, jbena xrvare tynhog
+vs fjvgmreynaq jrer jurer terrpr vf (ba vfynaqf) jbhyq gurl nyy or pbaarpgrq ol oevqtrf?
+genafyngvat clcl jvgu pclguba vf fbbbbbb fybj
+ClCl 1.4 eryrnfrq!
+Jr ner abg urebrf, whfg irel cngvrag.
+QBAR zrnaf vg'f qbar
+jul gurer vf ab "ClCl 1.4 eryrnfrq" va gbcvp nal zber?
+fabj! fabj!
+svanyyl, zrephevny zvtengvba vf unccravat!
+Gur zvtengvba gb zrephevny vf pbzcyrgrq! uggc://ovgohpxrg.bet/clcl/clcl
+fabj! fabj! (gre)
+unccl arj lrne
+naq anaanaw gb lbh nf jryy
+Frrvat nf gur ynjf bs culfvpf ner ntnvafg lbh, lbh unir gb pnershyyl pbafvqre lbhe fpbcr fb gung lbhe tbnyf ner ernfbanoyr.
+nf hfhny va clcl, gur fbyhgvba nccrnef pbzcyrgryl qvfcebcbegvbangr gb gur ceboyrz naq vafgrnq jr'yy tb sbe n pbzcyrgryl qvssrerag fvzcyre nccebnpu gb gur bevtvany ceboyrz
+fabj, fabj!
+va clcl lbh ner nyjnlf ng gur jebat yriry, va bar jnl be gur bgure
+jryy, vg'f jebat ohg abg fb "irel jebat" nf vg ybbxrq
+<svwny> V ybir clcl
+ynmvarff vzcngvrapr naq uhoevf
+fabj, fabj
+EClguba: guvatf lbh jbhyqa'g qb va Clguba, naq pna'g qb va P.
+vg vf gur rkcrpgrq orunivbe, rkprcg jura lbh qba'g rkcrpg vg
+erqrsvavat lryybj frrzf yvxr n orggre vqrn
+"gung'f ubjrire whfg ratvarrevat" (svwny)
+"[vg] whfg fubjf ntnva gung eclguba vf bofpher" (psobym)
+"naljnl, clguba vf n infg ynathntr" (svwny)
+bhg-bs-yvr-thneqf
+"gurer ner qnlf ba juvpu lbh ybbx nebhaq naq abguvat fubhyq unir rire jbexrq" (svwny)
+clcl vf n orggre xvaq bs sbbyvfuarff - ynp
+ehaavat grfgf vf rffragvny sbe qrirybcvat clcl -- hu? qvq V oernx gur grfg? (svwny)
+V'ir tbg guvf sybbe jnk gung'f nyfb n TERNG qrffreg gbccvat!!
+rknexha: "gur cneg gung V gubhtug jnf tbvat gb or uneq jnf gevivny, fb abj V whfg unir guvf cneg gung V qvqa'g rira guvax bs gung vf uneq"
+V fhccbfr jr pna yvir jvgu gur bofphevgl, nf ybat nf gurer vf n pbzzrag znxvat vg yvtugre
+V nz n ovt oryvrire va ernfbaf. ohg gur nccnerag xvaq ner zl snibevgr.
+clcl: trg n WVG sbe serr (jryy gur svefg qnl lbh jba'g znantr naq vg jvyy or irel sehfgengvat)
+<Nyrk_Tnlabe> thgjbegu: bu, jr fubhyq znxr gur WVG zntvpnyyl orggre, jvgu qrpbengbef naq fghss
+vg'f n pbzcyrgr unpx, ohg n irel zvavzny bar (nevtngb)
+svefg gurl ynhtu ng lbh, gura gurl vtaber lbh, gura gurl svtug lbh, gura lbh jva
+ClCl vf snzvyl sevraqyl
+jr yvxr pbzcynvagf
+gbqnl jr'er snfgre guna lrfgreqnl (hfhnyyl)
+ClCl naq PClguba: gurl ner zbegny rarzvrf vagrag ba xvyyvat rnpu bgure
+nethnoyl, rirelguvat vf n avpur
+clcl unf ynlref yvxr bavbaf: crryvat gurz onpx jvyy znxr lbh pel
+EClguba zntvpnyyl znxrf lbh evpu naq snzbhf (fnlf fb ba gur gva)
+Vf evtbobg nebhaq jura gur havirefr prnfrf gb rkvfg?
+ClCl vf gbb pbby sbe dhrelfgevatf.
+< nevtngb> gura jung bpphef? < svwny> tbbq fghss V oryvrir
+ClCl 1.6 eryrnfrq!
+<psobym> jurer ner gur grfgf?
+uggc://gjvgcvp.pbz/52nr8s
+N enaqbz dhbgr
+Nyy rkprcgoybpxf frrz fnar.
+N cvax tyvggrel ebgngvat ynzoqn
+"vg'f yvxryl grzcbenel hagvy sberire" nevtb
"""
def some_topic():
diff --git a/lib_pypy/_sqlite3.py b/lib_pypy/_sqlite3.py
--- a/lib_pypy/_sqlite3.py
+++ b/lib_pypy/_sqlite3.py
@@ -24,6 +24,7 @@
from ctypes import c_void_p, c_int, c_double, c_int64, c_char_p, cdll
from ctypes import POINTER, byref, string_at, CFUNCTYPE, cast
from ctypes import sizeof, c_ssize_t
+from collections import OrderedDict
import datetime
import sys
import time
@@ -274,6 +275,28 @@
def unicode_text_factory(x):
return unicode(x, 'utf-8')
+
+class StatementCache(object):
+ def __init__(self, connection, maxcount):
+ self.connection = connection
+ self.maxcount = maxcount
+ self.cache = OrderedDict()
+
+ def get(self, sql, cursor, row_factory):
+ try:
+ stat = self.cache[sql]
+ except KeyError:
+ stat = Statement(self.connection, sql)
+ self.cache[sql] = stat
+ if len(self.cache) > self.maxcount:
+ self.cache.popitem(0)
+ #
+ if stat.in_use:
+ stat = Statement(self.connection, sql)
+ stat.set_row_factory(row_factory)
+ return stat
+
+
class Connection(object):
def __init__(self, database, timeout=5.0, detect_types=0, isolation_level="",
check_same_thread=True, factory=None, cached_statements=100):
@@ -291,6 +314,7 @@
self.row_factory = None
self._isolation_level = isolation_level
self.detect_types = detect_types
+ self.statement_cache = StatementCache(self, cached_statements)
self.cursors = []
@@ -399,7 +423,7 @@
cur = Cursor(self)
if not isinstance(sql, (str, unicode)):
raise Warning("SQL is of wrong type. Must be string or unicode.")
- statement = Statement(cur, sql, self.row_factory)
+ statement = self.statement_cache.get(sql, cur, self.row_factory)
return statement
def _get_isolation_level(self):
@@ -681,6 +705,8 @@
from sqlite3.dump import _iterdump
return _iterdump(self)
+DML, DQL, DDL = range(3)
+
class Cursor(object):
def __init__(self, con):
if not isinstance(con, Connection):
@@ -708,12 +734,12 @@
if type(sql) is unicode:
sql = sql.encode("utf-8")
self._check_closed()
- self.statement = Statement(self, sql, self.row_factory)
+ self.statement = self.connection.statement_cache.get(sql, self, self.row_factory)
if self.connection._isolation_level is not None:
- if self.statement.kind == "DDL":
+ if self.statement.kind == DDL:
self.connection.commit()
- elif self.statement.kind == "DML":
+ elif self.statement.kind == DML:
self.connection._begin()
self.statement.set_params(params)
@@ -724,19 +750,18 @@
self.statement.reset()
raise self.connection._get_exception(ret)
- if self.statement.kind == "DQL":
- if ret == SQLITE_ROW:
- self.statement._build_row_cast_map()
- self.statement._readahead()
- else:
- self.statement.item = None
- self.statement.exhausted = True
+ if self.statement.kind == DQL and ret == SQLITE_ROW:
+ self.statement._build_row_cast_map()
+ self.statement._readahead(self)
+ else:
+ self.statement.item = None
+ self.statement.exhausted = True
- if self.statement.kind in ("DML", "DDL"):
+ if self.statement.kind == DML or self.statement.kind == DDL:
self.statement.reset()
self.rowcount = -1
- if self.statement.kind == "DML":
+ if self.statement.kind == DML:
self.rowcount = sqlite.sqlite3_changes(self.connection.db)
return self
@@ -747,8 +772,9 @@
if type(sql) is unicode:
sql = sql.encode("utf-8")
self._check_closed()
- self.statement = Statement(self, sql, self.row_factory)
- if self.statement.kind == "DML":
+ self.statement = self.connection.statement_cache.get(sql, self, self.row_factory)
+
+ if self.statement.kind == DML:
self.connection._begin()
else:
raise ProgrammingError, "executemany is only for DML statements"
@@ -800,7 +826,7 @@
return self
def __iter__(self):
- return self.statement
+ return iter(self.fetchone, None)
def _check_reset(self):
if self.reset:
@@ -817,7 +843,7 @@
return None
try:
- return self.statement.next()
+ return self.statement.next(self)
except StopIteration:
return None
@@ -831,7 +857,7 @@
if size is None:
size = self.arraysize
lst = []
- for row in self.statement:
+ for row in self:
lst.append(row)
if len(lst) == size:
break
@@ -842,7 +868,7 @@
self._check_reset()
if self.statement is None:
return []
- return list(self.statement)
+ return list(self)
def _getdescription(self):
if self._description is None:
@@ -872,39 +898,47 @@
lastrowid = property(_getlastrowid)
class Statement(object):
- def __init__(self, cur, sql, row_factory):
+ def __init__(self, connection, sql):
self.statement = None
if not isinstance(sql, str):
raise ValueError, "sql must be a string"
- self.con = cur.connection
- self.cur = weakref.ref(cur)
+ self.con = connection
self.sql = sql # DEBUG ONLY
- self.row_factory = row_factory
first_word = self._statement_kind = sql.lstrip().split(" ")[0].upper()
if first_word in ("INSERT", "UPDATE", "DELETE", "REPLACE"):
- self.kind = "DML"
+ self.kind = DML
elif first_word in ("SELECT", "PRAGMA"):
- self.kind = "DQL"
+ self.kind = DQL
else:
- self.kind = "DDL"
+ self.kind = DDL
self.exhausted = False
+ self.in_use = False
+ #
+ # set by set_row_factory
+ self.row_factory = None
self.statement = c_void_p()
next_char = c_char_p()
- ret = sqlite.sqlite3_prepare_v2(self.con.db, sql, -1, byref(self.statement), byref(next_char))
+ sql_char = c_char_p(sql)
+ ret = sqlite.sqlite3_prepare_v2(self.con.db, sql_char, -1, byref(self.statement), byref(next_char))
if ret == SQLITE_OK and self.statement.value is None:
# an empty statement, we work around that, as it's the least trouble
ret = sqlite.sqlite3_prepare_v2(self.con.db, "select 42", -1, byref(self.statement), byref(next_char))
- self.kind = "DQL"
+ self.kind = DQL
if ret != SQLITE_OK:
raise self.con._get_exception(ret)
self.con._remember_statement(self)
if _check_remaining_sql(next_char.value):
- raise Warning, "One and only one statement required"
+ raise Warning, "One and only one statement required: %r" % (
+ next_char.value,)
+ # sql_char should remain alive until here
self._build_row_cast_map()
+ def set_row_factory(self, row_factory):
+ self.row_factory = row_factory
+
def _build_row_cast_map(self):
self.row_cast_map = []
for i in xrange(sqlite.sqlite3_column_count(self.statement)):
@@ -974,6 +1008,7 @@
ret = sqlite.sqlite3_reset(self.statement)
if ret != SQLITE_OK:
raise self.con._get_exception(ret)
+ self.mark_dirty()
if params is None:
if sqlite.sqlite3_bind_parameter_count(self.statement) != 0:
@@ -1004,10 +1039,7 @@
raise ProgrammingError("missing parameter '%s'" %param)
self.set_param(idx, param)
- def __iter__(self):
- return self
-
- def next(self):
+ def next(self, cursor):
self.con._check_closed()
self.con._check_thread()
if self.exhausted:
@@ -1023,10 +1055,10 @@
sqlite.sqlite3_reset(self.statement)
raise exc
- self._readahead()
+ self._readahead(cursor)
return item
- def _readahead(self):
+ def _readahead(self, cursor):
self.column_count = sqlite.sqlite3_column_count(self.statement)
row = []
for i in xrange(self.column_count):
@@ -1061,23 +1093,30 @@
row = tuple(row)
if self.row_factory is not None:
- row = self.row_factory(self.cur(), row)
+ row = self.row_factory(cursor, row)
self.item = row
def reset(self):
self.row_cast_map = None
- return sqlite.sqlite3_reset(self.statement)
+ ret = sqlite.sqlite3_reset(self.statement)
+ self.in_use = False
+ self.exhausted = False
+ return ret
def finalize(self):
sqlite.sqlite3_finalize(self.statement)
self.statement = None
+ self.in_use = False
+
+ def mark_dirty(self):
+ self.in_use = True
def __del__(self):
sqlite.sqlite3_finalize(self.statement)
self.statement = None
def _get_description(self):
- if self.kind == "DML":
+ if self.kind == DML:
return None
desc = []
for i in xrange(sqlite.sqlite3_column_count(self.statement)):
diff --git a/lib_pypy/_subprocess.py b/lib_pypy/_subprocess.py
--- a/lib_pypy/_subprocess.py
+++ b/lib_pypy/_subprocess.py
@@ -35,7 +35,7 @@
_DuplicateHandle.restype = ctypes.c_int
_WaitForSingleObject = _kernel32.WaitForSingleObject
-_WaitForSingleObject.argtypes = [ctypes.c_int, ctypes.c_int]
+_WaitForSingleObject.argtypes = [ctypes.c_int, ctypes.c_uint]
_WaitForSingleObject.restype = ctypes.c_int
_GetExitCodeProcess = _kernel32.GetExitCodeProcess
diff --git a/lib_pypy/distributed/test/test_distributed.py b/lib_pypy/distributed/test/test_distributed.py
--- a/lib_pypy/distributed/test/test_distributed.py
+++ b/lib_pypy/distributed/test/test_distributed.py
@@ -9,7 +9,7 @@
class AppTestDistributed(object):
def setup_class(cls):
cls.space = gettestobjspace(**{"objspace.std.withtproxy": True,
- "usemodules":("_stackless",)})
+ "usemodules":("_continuation",)})
def test_init(self):
import distributed
@@ -91,10 +91,8 @@
class AppTestDistributedTasklets(object):
spaceconfig = {"objspace.std.withtproxy": True,
- "objspace.usemodules._stackless": True}
+ "objspace.usemodules._continuation": True}
def setup_class(cls):
- #cls.space = gettestobjspace(**{"objspace.std.withtproxy": True,
- # "usemodules":("_stackless",)})
cls.w_test_env = cls.space.appexec([], """():
from distributed import test_env
return test_env
diff --git a/lib_pypy/distributed/test/test_greensock.py b/lib_pypy/distributed/test/test_greensock.py
--- a/lib_pypy/distributed/test/test_greensock.py
+++ b/lib_pypy/distributed/test/test_greensock.py
@@ -10,7 +10,7 @@
if not option.runappdirect:
py.test.skip("Cannot run this on top of py.py because of PopenGateway")
cls.space = gettestobjspace(**{"objspace.std.withtproxy": True,
- "usemodules":("_stackless",)})
+ "usemodules":("_continuation",)})
cls.w_remote_side_code = cls.space.appexec([], """():
import sys
sys.path.insert(0, '%s')
diff --git a/lib_pypy/distributed/test/test_socklayer.py b/lib_pypy/distributed/test/test_socklayer.py
--- a/lib_pypy/distributed/test/test_socklayer.py
+++ b/lib_pypy/distributed/test/test_socklayer.py
@@ -9,7 +9,8 @@
class AppTestSocklayer:
def setup_class(cls):
cls.space = gettestobjspace(**{"objspace.std.withtproxy": True,
- "usemodules":("_stackless","_socket", "select")})
+ "usemodules":("_continuation",
+ "_socket", "select")})
def test_socklayer(self):
class X(object):
diff --git a/lib_pypy/greenlet.py b/lib_pypy/greenlet.py
--- a/lib_pypy/greenlet.py
+++ b/lib_pypy/greenlet.py
@@ -1,1 +1,144 @@
-from _stackless import greenlet
+import _continuation, sys
+
+
+# ____________________________________________________________
+# Exceptions
+
+class GreenletExit(Exception):
+ """This special exception does not propagate to the parent greenlet; it
+can be used to kill a single greenlet."""
+
+error = _continuation.error
+
+# ____________________________________________________________
+# Helper function
+
+def getcurrent():
+ "Returns the current greenlet (i.e. the one which called this function)."
+ try:
+ return _tls.current
+ except AttributeError:
+ # first call in this thread: current == main
+ _green_create_main()
+ return _tls.current
+
+# ____________________________________________________________
+# The 'greenlet' class
+
+_continulet = _continuation.continulet
+
+class greenlet(_continulet):
+ getcurrent = staticmethod(getcurrent)
+ error = error
+ GreenletExit = GreenletExit
+ __main = False
+ __started = False
+
+ def __new__(cls, *args, **kwds):
+ self = _continulet.__new__(cls)
+ self.parent = getcurrent()
+ return self
+
+ def __init__(self, run=None, parent=None):
+ if run is not None:
+ self.run = run
+ if parent is not None:
+ self.parent = parent
+
+ def switch(self, *args):
+ "Switch execution to this greenlet, optionally passing the values "
+ "given as argument(s). Returns the value passed when switching back."
+ return self.__switch('switch', args)
+
+ def throw(self, typ=GreenletExit, val=None, tb=None):
+ "raise exception in greenlet, return value passed when switching back"
+ return self.__switch('throw', typ, val, tb)
+
+ def __switch(target, methodname, *args):
+ current = getcurrent()
+ #
+ while not target:
+ if not target.__started:
+ if methodname == 'switch':
+ greenlet_func = _greenlet_start
+ else:
+ greenlet_func = _greenlet_throw
+ _continulet.__init__(target, greenlet_func, *args)
+ methodname = 'switch'
+ args = ()
+ target.__started = True
+ break
+ # already done, go to the parent instead
+ # (NB. infinite loop possible, but unlikely, unless you mess
+ # up the 'parent' explicitly. Good enough, because a Ctrl-C
+ # will show that the program is caught in this loop here.)
+ target = target.parent
+ #
+ try:
+ unbound_method = getattr(_continulet, methodname)
+ args = unbound_method(current, *args, to=target)
+ except GreenletExit, e:
+ args = (e,)
+ finally:
+ _tls.current = current
+ #
+ if len(args) == 1:
+ return args[0]
+ else:
+ return args
+
+ def __nonzero__(self):
+ return self.__main or _continulet.is_pending(self)
+
+ @property
+ def dead(self):
+ return self.__started and not self
+
+ @property
+ def gr_frame(self):
+ # xxx this doesn't work when called on either the current or
+ # the main greenlet of another thread
+ if self is getcurrent():
+ return None
+ if self.__main:
+ self = getcurrent()
+ f = _continulet.__reduce__(self)[2][0]
+ if not f:
+ return None
+ return f.f_back.f_back.f_back # go past start(), __switch(), switch()
+
+# ____________________________________________________________
+# Internal stuff
+
+try:
+ from thread import _local
+except ImportError:
+ class _local(object): # assume no threads
+ pass
+
+_tls = _local()
+
+def _green_create_main():
+ # create the main greenlet for this thread
+ _tls.current = None
+ gmain = greenlet.__new__(greenlet)
+ gmain._greenlet__main = True
+ gmain._greenlet__started = True
+ assert gmain.parent is None
+ _tls.main = gmain
+ _tls.current = gmain
+
+def _greenlet_start(greenlet, args):
+ _tls.current = greenlet
+ try:
+ res = greenlet.run(*args)
+ finally:
+ _continuation.permute(greenlet, greenlet.parent)
+ return (res,)
+
+def _greenlet_throw(greenlet, exc, value, tb):
+ _tls.current = greenlet
+ try:
+ raise exc, value, tb
+ finally:
+ _continuation.permute(greenlet, greenlet.parent)
diff --git a/lib_pypy/pypy_test/test_coroutine.py b/lib_pypy/pypy_test/test_coroutine.py
--- a/lib_pypy/pypy_test/test_coroutine.py
+++ b/lib_pypy/pypy_test/test_coroutine.py
@@ -2,7 +2,7 @@
from py.test import skip, raises
try:
- from lib_pypy.stackless import coroutine, CoroutineExit
+ from stackless import coroutine, CoroutineExit
except ImportError, e:
skip('cannot import stackless: %s' % (e,))
@@ -20,10 +20,6 @@
assert not co.is_zombie
def test_is_zombie_del_without_frame(self):
- try:
- import _stackless # are we on pypy with a stackless build?
- except ImportError:
- skip("only works on pypy-c-stackless")
import gc
res = []
class MyCoroutine(coroutine):
@@ -45,10 +41,6 @@
assert res[0], "is_zombie was False in __del__"
def test_is_zombie_del_with_frame(self):
- try:
- import _stackless # are we on pypy with a stackless build?
- except ImportError:
- skip("only works on pypy-c-stackless")
import gc
res = []
class MyCoroutine(coroutine):
diff --git a/lib_pypy/pypy_test/test_stackless_pickling.py b/lib_pypy/pypy_test/test_stackless_pickling.py
--- a/lib_pypy/pypy_test/test_stackless_pickling.py
+++ b/lib_pypy/pypy_test/test_stackless_pickling.py
@@ -1,7 +1,3 @@
-"""
-this test should probably not run from CPython or py.py.
-I'm not entirely sure, how to do that.
-"""
from __future__ import absolute_import
from py.test import skip
try:
@@ -16,11 +12,15 @@
class Test_StacklessPickling:
+ def test_pickle_main_coroutine(self):
+ import stackless, pickle
+ s = pickle.dumps(stackless.coroutine.getcurrent())
+ print s
+ c = pickle.loads(s)
+ assert c is stackless.coroutine.getcurrent()
+
def test_basic_tasklet_pickling(self):
- try:
- import stackless
- except ImportError:
- skip("can't load stackless and don't know why!!!")
+ import stackless
from stackless import run, schedule, tasklet
import pickle
diff --git a/lib_pypy/pyrepl/completing_reader.py b/lib_pypy/pyrepl/completing_reader.py
--- a/lib_pypy/pyrepl/completing_reader.py
+++ b/lib_pypy/pyrepl/completing_reader.py
@@ -229,7 +229,8 @@
def after_command(self, cmd):
super(CompletingReader, self).after_command(cmd)
- if not isinstance(cmd, complete) and not isinstance(cmd, self_insert):
+ if not isinstance(cmd, self.commands['complete']) \
+ and not isinstance(cmd, self.commands['self_insert']):
self.cmpltn_reset()
def calc_screen(self):
diff --git a/lib_pypy/pyrepl/reader.py b/lib_pypy/pyrepl/reader.py
--- a/lib_pypy/pyrepl/reader.py
+++ b/lib_pypy/pyrepl/reader.py
@@ -401,13 +401,19 @@
return "(arg: %s) "%self.arg
if "\n" in self.buffer:
if lineno == 0:
- return self._ps2
+ res = self.ps2
elif lineno == self.buffer.count("\n"):
- return self._ps4
+ res = self.ps4
else:
- return self._ps3
+ res = self.ps3
else:
- return self._ps1
+ res = self.ps1
+ # Lazily call str() on self.psN, and cache the results using as key
+ # the object on which str() was called. This ensures that even if the
+ # same object is used e.g. for ps1 and ps2, str() is called only once.
+ if res not in self._pscache:
+ self._pscache[res] = str(res)
+ return self._pscache[res]
def push_input_trans(self, itrans):
self.input_trans_stack.append(self.input_trans)
@@ -473,8 +479,7 @@
self.pos = 0
self.dirty = 1
self.last_command = None
- self._ps1, self._ps2, self._ps3, self._ps4 = \
- map(str, [self.ps1, self.ps2, self.ps3, self.ps4])
+ self._pscache = {}
except:
self.restore()
raise
@@ -571,7 +576,7 @@
self.console.push_char(char)
self.handle1(0)
- def readline(self):
+ def readline(self, returns_unicode=False):
"""Read a line. The implementation of this method also shows
how to drive Reader if you want more control over the event
loop."""
@@ -580,6 +585,8 @@
self.refresh()
while not self.finished:
self.handle1()
+ if returns_unicode:
+ return self.get_unicode()
return self.get_buffer()
finally:
self.restore()
diff --git a/lib_pypy/pyrepl/readline.py b/lib_pypy/pyrepl/readline.py
--- a/lib_pypy/pyrepl/readline.py
+++ b/lib_pypy/pyrepl/readline.py
@@ -33,7 +33,7 @@
from pyrepl.unix_console import UnixConsole, _error
-ENCODING = 'latin1' # XXX hard-coded
+ENCODING = sys.getfilesystemencoding() or 'latin1' # XXX review
__all__ = ['add_history',
'clear_history',
@@ -198,7 +198,7 @@
reader.ps1 = prompt
return reader.readline()
- def multiline_input(self, more_lines, ps1, ps2):
+ def multiline_input(self, more_lines, ps1, ps2, returns_unicode=False):
"""Read an input on possibly multiple lines, asking for more
lines as long as 'more_lines(unicodetext)' returns an object whose
boolean value is true.
@@ -209,7 +209,7 @@
reader.more_lines = more_lines
reader.ps1 = reader.ps2 = ps1
reader.ps3 = reader.ps4 = ps2
- return reader.readline()
+ return reader.readline(returns_unicode=returns_unicode)
finally:
reader.more_lines = saved
@@ -231,7 +231,11 @@
return ''.join(chars)
def _histline(self, line):
- return unicode(line.rstrip('\n'), ENCODING)
+ line = line.rstrip('\n')
+ try:
+ return unicode(line, ENCODING)
+ except UnicodeDecodeError: # bah, silently fall back...
+ return unicode(line, 'utf-8')
def get_history_length(self):
return self.saved_history_length
@@ -268,7 +272,10 @@
f = open(os.path.expanduser(filename), 'w')
for entry in history:
if isinstance(entry, unicode):
- entry = entry.encode(ENCODING)
+ try:
+ entry = entry.encode(ENCODING)
+ except UnicodeEncodeError: # bah, silently fall back...
+ entry = entry.encode('utf-8')
entry = entry.replace('\n', '\r\n') # multiline history support
f.write(entry + '\n')
f.close()
@@ -395,9 +402,21 @@
_wrapper.f_in = f_in
_wrapper.f_out = f_out
- if hasattr(sys, '__raw_input__'): # PyPy
- _old_raw_input = sys.__raw_input__
+ if '__pypy__' in sys.builtin_module_names: # PyPy
+
+ def _old_raw_input(prompt=''):
+ # sys.__raw_input__() is only called when stdin and stdout are
+ # as expected and are ttys. If it is the case, then get_reader()
+ # should not really fail in _wrapper.raw_input(). If it still
+ # does, then we will just cancel the redirection and call again
+ # the built-in raw_input().
+ try:
+ del sys.__raw_input__
+ except AttributeError:
+ pass
+ return raw_input(prompt)
sys.__raw_input__ = _wrapper.raw_input
+
else:
# this is not really what readline.c does. Better than nothing I guess
import __builtin__
diff --git a/lib_pypy/pyrepl/simple_interact.py b/lib_pypy/pyrepl/simple_interact.py
--- a/lib_pypy/pyrepl/simple_interact.py
+++ b/lib_pypy/pyrepl/simple_interact.py
@@ -54,7 +54,8 @@
ps1 = getattr(sys, 'ps1', '>>> ')
ps2 = getattr(sys, 'ps2', '... ')
try:
- statement = multiline_input(more_lines, ps1, ps2)
+ statement = multiline_input(more_lines, ps1, ps2,
+ returns_unicode=True)
except EOFError:
break
more = console.push(statement)
diff --git a/lib_pypy/pyrepl/unix_console.py b/lib_pypy/pyrepl/unix_console.py
--- a/lib_pypy/pyrepl/unix_console.py
+++ b/lib_pypy/pyrepl/unix_console.py
@@ -384,15 +384,19 @@
self.__maybe_write_code(self._smkx)
- self.old_sigwinch = signal.signal(
- signal.SIGWINCH, self.__sigwinch)
+ try:
+ self.old_sigwinch = signal.signal(
+ signal.SIGWINCH, self.__sigwinch)
+ except ValueError:
+ pass
def restore(self):
self.__maybe_write_code(self._rmkx)
self.flushoutput()
tcsetattr(self.input_fd, termios.TCSADRAIN, self.__svtermstate)
- signal.signal(signal.SIGWINCH, self.old_sigwinch)
+ if hasattr(self, 'old_sigwinch'):
+ signal.signal(signal.SIGWINCH, self.old_sigwinch)
def __sigwinch(self, signum, frame):
self.height, self.width = self.getheightwidth()
diff --git a/lib_pypy/resource.py b/lib_pypy/resource.py
--- a/lib_pypy/resource.py
+++ b/lib_pypy/resource.py
@@ -7,7 +7,7 @@
from ctypes_support import standard_c_lib as libc
from ctypes_support import get_errno
-from ctypes import Structure, c_int, c_long, byref, sizeof, POINTER
+from ctypes import Structure, c_int, c_long, byref, POINTER
from errno import EINVAL, EPERM
import _structseq
@@ -165,7 +165,6 @@
@builtinify
def getpagesize():
- pagesize = 0
if _getpagesize:
return _getpagesize()
else:
diff --git a/lib_pypy/stackless.py b/lib_pypy/stackless.py
--- a/lib_pypy/stackless.py
+++ b/lib_pypy/stackless.py
@@ -4,121 +4,110 @@
Please refer to their documentation.
"""
-DEBUG = True
-def dprint(*args):
- for arg in args:
- print arg,
- print
+import _continuation
-import traceback
-import sys
+class TaskletExit(Exception):
+ pass
+
+CoroutineExit = TaskletExit
+
+
+def _coroutine_getcurrent():
+ "Returns the current coroutine (i.e. the one which called this function)."
+ try:
+ return _tls.current_coroutine
+ except AttributeError:
+ # first call in this thread: current == main
+ return _coroutine_getmain()
+
+def _coroutine_getmain():
+ try:
+ return _tls.main_coroutine
+ except AttributeError:
+ # create the main coroutine for this thread
+ continulet = _continuation.continulet
+ main = coroutine()
+ main._frame = continulet.__new__(continulet)
+ main._is_started = -1
+ _tls.current_coroutine = _tls.main_coroutine = main
+ return _tls.main_coroutine
+
+
+class coroutine(object):
+ _is_started = 0 # 0=no, 1=yes, -1=main
+
+ def __init__(self):
+ self._frame = None
+
+ def bind(self, func, *argl, **argd):
+ """coro.bind(f, *argl, **argd) -> None.
+ binds function f to coro. f will be called with
+ arguments *argl, **argd
+ """
+ if self.is_alive:
+ raise ValueError("cannot bind a bound coroutine")
+ def run(c):
+ _tls.current_coroutine = self
+ self._is_started = 1
+ return func(*argl, **argd)
+ self._is_started = 0
+ self._frame = _continuation.continulet(run)
+
+ def switch(self):
+ """coro.switch() -> returnvalue
+ switches to coroutine coro. If the bound function
+ f finishes, the returnvalue is that of f, otherwise
+ None is returned
+ """
+ current = _coroutine_getcurrent()
+ try:
+ current._frame.switch(to=self._frame)
+ finally:
+ _tls.current_coroutine = current
+
+ def kill(self):
+ """coro.kill() : kill coroutine coro"""
+ current = _coroutine_getcurrent()
+ try:
+ current._frame.throw(CoroutineExit, to=self._frame)
+ finally:
+ _tls.current_coroutine = current
+
+ @property
+ def is_alive(self):
+ return self._is_started < 0 or (
+ self._frame is not None and self._frame.is_pending())
+
+ @property
+ def is_zombie(self):
+ return self._is_started > 0 and not self._frame.is_pending()
+
+ getcurrent = staticmethod(_coroutine_getcurrent)
+
+ def __reduce__(self):
+ if self._is_started < 0:
+ return _coroutine_getmain, ()
+ else:
+ return type(self), (), self.__dict__
+
+
try:
- # If _stackless can be imported then TaskletExit and CoroutineExit are
- # automatically added to the builtins.
- from _stackless import coroutine, greenlet
-except ImportError: # we are running from CPython
- from greenlet import greenlet, GreenletExit
- TaskletExit = CoroutineExit = GreenletExit
- del GreenletExit
- try:
- from functools import partial
- except ImportError: # we are not running python 2.5
- class partial(object):
- # just enough of 'partial' to be usefull
- def __init__(self, func, *argl, **argd):
- self.func = func
- self.argl = argl
- self.argd = argd
+ from thread import _local
+except ImportError:
+ class _local(object): # assume no threads
+ pass
- def __call__(self):
- return self.func(*self.argl, **self.argd)
+_tls = _local()
- class GWrap(greenlet):
- """This is just a wrapper around greenlets to allow
- to stick additional attributes to a greenlet.
- To be more concrete, we need a backreference to
- the coroutine object"""
- class MWrap(object):
- def __init__(self,something):
- self.something = something
+# ____________________________________________________________
- def __getattr__(self, attr):
- return getattr(self.something, attr)
-
- class coroutine(object):
- "we can't have greenlet as a base, because greenlets can't be rebound"
-
- def __init__(self):
- self._frame = None
- self.is_zombie = False
-
- def __getattr__(self, attr):
- return getattr(self._frame, attr)
-
- def __del__(self):
- self.is_zombie = True
- del self._frame
- self._frame = None
-
- def bind(self, func, *argl, **argd):
- """coro.bind(f, *argl, **argd) -> None.
- binds function f to coro. f will be called with
- arguments *argl, **argd
- """
- if self._frame is None or self._frame.dead:
- self._frame = frame = GWrap()
- frame.coro = self
- if hasattr(self._frame, 'run') and self._frame.run:
- raise ValueError("cannot bind a bound coroutine")
- self._frame.run = partial(func, *argl, **argd)
-
- def switch(self):
- """coro.switch() -> returnvalue
- switches to coroutine coro. If the bound function
- f finishes, the returnvalue is that of f, otherwise
- None is returned
- """
- try:
- return greenlet.switch(self._frame)
- except TypeError, exp: # self._frame is the main coroutine
- return greenlet.switch(self._frame.something)
-
- def kill(self):
- """coro.kill() : kill coroutine coro"""
- self._frame.throw()
-
- def _is_alive(self):
- if self._frame is None:
- return False
- return not self._frame.dead
- is_alive = property(_is_alive)
- del _is_alive
-
- def getcurrent():
- """coroutine.getcurrent() -> the currently running coroutine"""
- try:
- return greenlet.getcurrent().coro
- except AttributeError:
- return _maincoro
- getcurrent = staticmethod(getcurrent)
-
- def __reduce__(self):
- raise TypeError, 'pickling is not possible based upon greenlets'
-
- _maincoro = coroutine()
- maingreenlet = greenlet.getcurrent()
- _maincoro._frame = frame = MWrap(maingreenlet)
- frame.coro = _maincoro
- del frame
- del maingreenlet
from collections import deque
import operator
-__all__ = 'run getcurrent getmain schedule tasklet channel coroutine \
- greenlet'.split()
+__all__ = 'run getcurrent getmain schedule tasklet channel coroutine'.split()
_global_task_id = 0
_squeue = None
@@ -131,7 +120,8 @@
def _scheduler_remove(value):
try:
del _squeue[operator.indexOf(_squeue, value)]
- except ValueError:pass
+ except ValueError:
+ pass
def _scheduler_append(value, normal=True):
if normal:
@@ -157,10 +147,7 @@
_last_task = next
assert not next.blocked
if next is not current:
- try:
- next.switch()
- except CoroutineExit:
- raise TaskletExit
+ next.switch()
return current
def set_schedule_callback(callback):
@@ -184,34 +171,6 @@
raise self.type, self.value, self.traceback
#
-# helpers for pickling
-#
-
-_stackless_primitive_registry = {}
-
-def register_stackless_primitive(thang, retval_expr='None'):
- import types
- func = thang
- if isinstance(thang, types.MethodType):
- func = thang.im_func
- code = func.func_code
- _stackless_primitive_registry[code] = retval_expr
- # It is not too nice to attach info via the code object, but
- # I can't think of a better solution without a real transform.
-
-def rewrite_stackless_primitive(coro_state, alive, tempval):
- flags, frame, thunk, parent = coro_state
- while frame is not None:
- retval_expr = _stackless_primitive_registry.get(frame.f_code)
- if retval_expr:
- # this tasklet needs to stop pickling here and return its value.
- tempval = eval(retval_expr, globals(), frame.f_locals)
- coro_state = flags, frame, thunk, parent
- break
- frame = frame.f_back
- return coro_state, alive, tempval
-
-#
#
class channel(object):
@@ -363,8 +322,6 @@
"""
return self._channel_action(None, -1)
- register_stackless_primitive(receive, retval_expr='receiver.tempval')
-
def send_exception(self, exp_type, msg):
self.send(bomb(exp_type, exp_type(msg)))
@@ -381,9 +338,8 @@
the runnables list.
"""
return self._channel_action(msg, 1)
-
- register_stackless_primitive(send)
-
+
+
class tasklet(coroutine):
"""
A tasklet object represents a tiny task in a Python thread.
@@ -455,6 +411,7 @@
def _func():
try:
try:
+ coroutine.switch(back)
func(*argl, **argd)
except TaskletExit:
pass
@@ -464,6 +421,8 @@
self.func = None
coroutine.bind(self, _func)
+ back = _coroutine_getcurrent()
+ coroutine.switch(self)
self.alive = True
_scheduler_append(self)
return self
@@ -486,39 +445,6 @@
raise RuntimeError, "The current tasklet cannot be removed."
# not sure if I will revive this " Use t=tasklet().capture()"
_scheduler_remove(self)
-
- def __reduce__(self):
- one, two, coro_state = coroutine.__reduce__(self)
- assert one is coroutine
- assert two == ()
- # we want to get rid of the parent thing.
- # for now, we just drop it
- a, frame, c, d = coro_state
-
- # Removing all frames related to stackless.py.
- # They point to stuff we don't want to be pickled.
-
- pickleframe = frame
- while frame is not None:
- if frame.f_code == schedule.func_code:
- # Removing everything including and after the
- # call to stackless.schedule()
- pickleframe = frame.f_back
- break
- frame = frame.f_back
- if d:
- assert isinstance(d, coroutine)
- coro_state = a, pickleframe, c, None
- coro_state, alive, tempval = rewrite_stackless_primitive(coro_state, self.alive, self.tempval)
- inst_dict = self.__dict__.copy()
- inst_dict.pop('tempval', None)
- return self.__class__, (), (coro_state, alive, tempval, inst_dict)
-
- def __setstate__(self, (coro_state, alive, tempval, inst_dict)):
- coroutine.__setstate__(self, coro_state)
- self.__dict__.update(inst_dict)
- self.alive = alive
- self.tempval = tempval
def getmain():
"""
@@ -607,30 +533,7 @@
global _last_task
_global_task_id = 0
_main_tasklet = coroutine.getcurrent()
- try:
- _main_tasklet.__class__ = tasklet
- except TypeError: # we are running pypy-c
- class TaskletProxy(object):
- """TaskletProxy is needed to give the _main_coroutine tasklet behaviour"""
- def __init__(self, coro):
- self._coro = coro
-
- def __getattr__(self,attr):
- return getattr(self._coro,attr)
-
- def __str__(self):
- return '<tasklet %s a:%s>' % (self._task_id, self.is_alive)
-
- def __reduce__(self):
- return getmain, ()
-
- __repr__ = __str__
-
-
- global _main_coroutine
- _main_coroutine = _main_tasklet
- _main_tasklet = TaskletProxy(_main_tasklet)
- assert _main_tasklet.is_alive and not _main_tasklet.is_zombie
+ _main_tasklet.__class__ = tasklet # XXX HAAAAAAAAAAAAAAAAAAAAACK
_last_task = _main_tasklet
tasklet._init.im_func(_main_tasklet, label='main')
_squeue = deque()
diff --git a/py/_code/source.py b/py/_code/source.py
--- a/py/_code/source.py
+++ b/py/_code/source.py
@@ -139,7 +139,7 @@
trysource = self[start:end]
if trysource.isparseable():
return start, end
- return start, end
+ return start, len(self)
def getblockend(self, lineno):
# XXX
diff --git a/pypy/annotation/annrpython.py b/pypy/annotation/annrpython.py
--- a/pypy/annotation/annrpython.py
+++ b/pypy/annotation/annrpython.py
@@ -149,7 +149,7 @@
desc = olddesc.bind_self(classdef)
args = self.bookkeeper.build_args("simple_call", args_s[:])
desc.consider_call_site(self.bookkeeper, desc.getcallfamily(), [desc],
- args, annmodel.s_ImpossibleValue)
+ args, annmodel.s_ImpossibleValue, None)
result = []
def schedule(graph, inputcells):
result.append((graph, inputcells))
diff --git a/pypy/annotation/bookkeeper.py b/pypy/annotation/bookkeeper.py
--- a/pypy/annotation/bookkeeper.py
+++ b/pypy/annotation/bookkeeper.py
@@ -209,8 +209,8 @@
self.consider_call_site(call_op)
for pbc, args_s in self.emulated_pbc_calls.itervalues():
- self.consider_call_site_for_pbc(pbc, 'simple_call',
- args_s, s_ImpossibleValue)
+ self.consider_call_site_for_pbc(pbc, 'simple_call',
+ args_s, s_ImpossibleValue, None)
self.emulated_pbc_calls = {}
finally:
self.leave()
@@ -257,18 +257,18 @@
args_s = [lltype_to_annotation(adtmeth.ll_ptrtype)] + args_s
if isinstance(s_callable, SomePBC):
s_result = binding(call_op.result, s_ImpossibleValue)
- self.consider_call_site_for_pbc(s_callable,
- call_op.opname,
- args_s, s_result)
+ self.consider_call_site_for_pbc(s_callable, call_op.opname, args_s,
+ s_result, call_op)
- def consider_call_site_for_pbc(self, s_callable, opname, args_s, s_result):
+ def consider_call_site_for_pbc(self, s_callable, opname, args_s, s_result,
+ call_op):
descs = list(s_callable.descriptions)
if not descs:
return
family = descs[0].getcallfamily()
args = self.build_args(opname, args_s)
s_callable.getKind().consider_call_site(self, family, descs, args,
- s_result)
+ s_result, call_op)
def getuniqueclassdef(self, cls):
"""Get the ClassDef associated with the given user cls.
@@ -656,6 +656,7 @@
whence = None
else:
whence = emulated # callback case
+ op = None
s_previous_result = s_ImpossibleValue
def schedule(graph, inputcells):
@@ -663,7 +664,7 @@
results = []
for desc in descs:
- results.append(desc.pycall(schedule, args, s_previous_result))
+ results.append(desc.pycall(schedule, args, s_previous_result, op))
s_result = unionof(*results)
return s_result
diff --git a/pypy/annotation/builtin.py b/pypy/annotation/builtin.py
--- a/pypy/annotation/builtin.py
+++ b/pypy/annotation/builtin.py
@@ -308,9 +308,6 @@
clsdef = clsdef.commonbase(cdef)
return SomeInstance(clsdef)
-def robjmodel_we_are_translated():
- return immutablevalue(True)
-
def robjmodel_r_dict(s_eqfn, s_hashfn, s_force_non_null=None):
if s_force_non_null is None:
force_non_null = False
@@ -376,8 +373,6 @@
BUILTIN_ANALYZERS[pypy.rlib.rarithmetic.intmask] = rarith_intmask
BUILTIN_ANALYZERS[pypy.rlib.objectmodel.instantiate] = robjmodel_instantiate
-BUILTIN_ANALYZERS[pypy.rlib.objectmodel.we_are_translated] = (
- robjmodel_we_are_translated)
BUILTIN_ANALYZERS[pypy.rlib.objectmodel.r_dict] = robjmodel_r_dict
BUILTIN_ANALYZERS[pypy.rlib.objectmodel.hlinvoke] = robjmodel_hlinvoke
BUILTIN_ANALYZERS[pypy.rlib.objectmodel.keepalive_until_here] = robjmodel_keepalive_until_here
@@ -416,7 +411,8 @@
from pypy.annotation.model import SomePtr
from pypy.rpython.lltypesystem import lltype
-def malloc(s_T, s_n=None, s_flavor=None, s_zero=None, s_track_allocation=None):
+def malloc(s_T, s_n=None, s_flavor=None, s_zero=None, s_track_allocation=None,
+ s_add_memory_pressure=None):
assert (s_n is None or s_n.knowntype == int
or issubclass(s_n.knowntype, pypy.rlib.rarithmetic.base_int))
assert s_T.is_constant()
@@ -432,6 +428,8 @@
else:
assert s_flavor.is_constant()
assert s_track_allocation is None or s_track_allocation.is_constant()
+ assert (s_add_memory_pressure is None or
+ s_add_memory_pressure.is_constant())
# not sure how to call malloc() for the example 'p' in the
# presence of s_extraargs
r = SomePtr(lltype.Ptr(s_T.const))
diff --git a/pypy/annotation/classdef.py b/pypy/annotation/classdef.py
--- a/pypy/annotation/classdef.py
+++ b/pypy/annotation/classdef.py
@@ -276,8 +276,8 @@
# create the Attribute and do the generalization asked for
newattr = Attribute(attr, self.bookkeeper)
if s_value:
- if newattr.name == 'intval' and getattr(s_value, 'unsigned', False):
- import pdb; pdb.set_trace()
+ #if newattr.name == 'intval' and getattr(s_value, 'unsigned', False):
+ # import pdb; pdb.set_trace()
newattr.s_value = s_value
# keep all subattributes' values
diff --git a/pypy/annotation/description.py b/pypy/annotation/description.py
--- a/pypy/annotation/description.py
+++ b/pypy/annotation/description.py
@@ -255,7 +255,11 @@
raise TypeError, "signature mismatch: %s" % e.getmsg(self.name)
return inputcells
- def specialize(self, inputcells):
+ def specialize(self, inputcells, op=None):
+ if (op is None and
+ getattr(self.bookkeeper, "position_key", None) is not None):
+ _, block, i = self.bookkeeper.position_key
+ op = block.operations[i]
if self.specializer is None:
# get the specializer based on the tag of the 'pyobj'
# (if any), according to the current policy
@@ -269,11 +273,14 @@
enforceargs = Sig(*enforceargs)
self.pyobj._annenforceargs_ = enforceargs
enforceargs(self, inputcells) # can modify inputcells in-place
- return self.specializer(self, inputcells)
+ if getattr(self.pyobj, '_annspecialcase_', '').endswith("call_location"):
+ return self.specializer(self, inputcells, op)
+ else:
+ return self.specializer(self, inputcells)
- def pycall(self, schedule, args, s_previous_result):
+ def pycall(self, schedule, args, s_previous_result, op=None):
inputcells = self.parse_arguments(args)
- result = self.specialize(inputcells)
+ result = self.specialize(inputcells, op)
if isinstance(result, FunctionGraph):
graph = result # common case
# if that graph has a different signature, we need to re-parse
@@ -296,17 +303,17 @@
None, # selfclassdef
name)
- def consider_call_site(bookkeeper, family, descs, args, s_result):
+ def consider_call_site(bookkeeper, family, descs, args, s_result, op):
shape = rawshape(args)
- row = FunctionDesc.row_to_consider(descs, args)
+ row = FunctionDesc.row_to_consider(descs, args, op)
family.calltable_add_row(shape, row)
consider_call_site = staticmethod(consider_call_site)
- def variant_for_call_site(bookkeeper, family, descs, args):
+ def variant_for_call_site(bookkeeper, family, descs, args, op):
shape = rawshape(args)
bookkeeper.enter(None)
try:
- row = FunctionDesc.row_to_consider(descs, args)
+ row = FunctionDesc.row_to_consider(descs, args, op)
finally:
bookkeeper.leave()
index = family.calltable_lookup_row(shape, row)
@@ -316,7 +323,7 @@
def rowkey(self):
return self
- def row_to_consider(descs, args):
+ def row_to_consider(descs, args, op):
# see comments in CallFamily
from pypy.annotation.model import s_ImpossibleValue
row = {}
@@ -324,7 +331,7 @@
def enlist(graph, ignore):
row[desc.rowkey()] = graph
return s_ImpossibleValue # meaningless
- desc.pycall(enlist, args, s_ImpossibleValue)
+ desc.pycall(enlist, args, s_ImpossibleValue, op)
return row
row_to_consider = staticmethod(row_to_consider)
@@ -399,9 +406,7 @@
if b1 is object:
continue
if b1.__dict__.get('_mixin_', False):
- assert b1.__bases__ == () or b1.__bases__ == (object,), (
- "mixin class %r should have no base" % (b1,))
- self.add_sources_for_class(b1, mixin=True)
+ self.add_mixin(b1)
else:
assert base is object, ("multiple inheritance only supported "
"with _mixin_: %r" % (cls,))
@@ -469,6 +474,15 @@
return
self.classdict[name] = Constant(value)
+ def add_mixin(self, base):
+ for subbase in base.__bases__:
+ if subbase is object:
+ continue
+ assert subbase.__dict__.get("_mixin_", False), ("Mixin class %r has non"
+ "mixin base class %r" % (base, subbase))
+ self.add_mixin(subbase)
+ self.add_sources_for_class(base, mixin=True)
+
def add_sources_for_class(self, cls, mixin=False):
for name, value in cls.__dict__.items():
self.add_source_attribute(name, value, mixin)
@@ -514,7 +528,7 @@
"specialization" % (self.name,))
return self.getclassdef(None)
- def pycall(self, schedule, args, s_previous_result):
+ def pycall(self, schedule, args, s_previous_result, op=None):
from pypy.annotation.model import SomeInstance, SomeImpossibleValue
if self.specialize:
if self.specialize == 'specialize:ctr_location':
@@ -657,7 +671,7 @@
cdesc = cdesc.basedesc
return s_result # common case
- def consider_call_site(bookkeeper, family, descs, args, s_result):
+ def consider_call_site(bookkeeper, family, descs, args, s_result, op):
from pypy.annotation.model import SomeInstance, SomePBC, s_None
if len(descs) == 1:
# call to a single class, look at the result annotation
@@ -702,7 +716,7 @@
initdescs[0].mergecallfamilies(*initdescs[1:])
initfamily = initdescs[0].getcallfamily()
MethodDesc.consider_call_site(bookkeeper, initfamily, initdescs,
- args, s_None)
+ args, s_None, op)
consider_call_site = staticmethod(consider_call_site)
def getallbases(self):
@@ -775,13 +789,13 @@
def getuniquegraph(self):
return self.funcdesc.getuniquegraph()
- def pycall(self, schedule, args, s_previous_result):
+ def pycall(self, schedule, args, s_previous_result, op=None):
from pypy.annotation.model import SomeInstance
if self.selfclassdef is None:
raise Exception("calling %r" % (self,))
s_instance = SomeInstance(self.selfclassdef, flags = self.flags)
args = args.prepend(s_instance)
- return self.funcdesc.pycall(schedule, args, s_previous_result)
+ return self.funcdesc.pycall(schedule, args, s_previous_result, op)
def bind_under(self, classdef, name):
self.bookkeeper.warning("rebinding an already bound %r" % (self,))
@@ -794,10 +808,10 @@
self.name,
flags)
- def consider_call_site(bookkeeper, family, descs, args, s_result):
+ def consider_call_site(bookkeeper, family, descs, args, s_result, op):
shape = rawshape(args, nextra=1) # account for the extra 'self'
funcdescs = [methoddesc.funcdesc for methoddesc in descs]
- row = FunctionDesc.row_to_consider(descs, args)
+ row = FunctionDesc.row_to_consider(descs, args, op)
family.calltable_add_row(shape, row)
consider_call_site = staticmethod(consider_call_site)
@@ -949,16 +963,16 @@
return '<MethodOfFrozenDesc %r of %r>' % (self.funcdesc,
self.frozendesc)
- def pycall(self, schedule, args, s_previous_result):
+ def pycall(self, schedule, args, s_previous_result, op=None):
from pypy.annotation.model import SomePBC
s_self = SomePBC([self.frozendesc])
args = args.prepend(s_self)
- return self.funcdesc.pycall(schedule, args, s_previous_result)
+ return self.funcdesc.pycall(schedule, args, s_previous_result, op)
- def consider_call_site(bookkeeper, family, descs, args, s_result):
+ def consider_call_site(bookkeeper, family, descs, args, s_result, op):
shape = rawshape(args, nextra=1) # account for the extra 'self'
funcdescs = [mofdesc.funcdesc for mofdesc in descs]
- row = FunctionDesc.row_to_consider(descs, args)
+ row = FunctionDesc.row_to_consider(descs, args, op)
family.calltable_add_row(shape, row)
consider_call_site = staticmethod(consider_call_site)
diff --git a/pypy/annotation/policy.py b/pypy/annotation/policy.py
--- a/pypy/annotation/policy.py
+++ b/pypy/annotation/policy.py
@@ -1,7 +1,7 @@
# base annotation policy for specialization
from pypy.annotation.specialize import default_specialize as default
-from pypy.annotation.specialize import specialize_argvalue, specialize_argtype, specialize_arglistitemtype
-from pypy.annotation.specialize import memo
+from pypy.annotation.specialize import specialize_argvalue, specialize_argtype, specialize_arglistitemtype, specialize_arg_or_var
+from pypy.annotation.specialize import memo, specialize_call_location
# for some reason, model must be imported first,
# or we create a cycle.
from pypy.annotation import model as annmodel
@@ -73,8 +73,10 @@
default_specialize = staticmethod(default)
specialize__memo = staticmethod(memo)
specialize__arg = staticmethod(specialize_argvalue) # specialize:arg(N)
+ specialize__arg_or_var = staticmethod(specialize_arg_or_var)
specialize__argtype = staticmethod(specialize_argtype) # specialize:argtype(N)
specialize__arglistitemtype = staticmethod(specialize_arglistitemtype)
+ specialize__call_location = staticmethod(specialize_call_location)
def specialize__ll(pol, *args):
from pypy.rpython.annlowlevel import LowLevelAnnotatorPolicy
diff --git a/pypy/annotation/specialize.py b/pypy/annotation/specialize.py
--- a/pypy/annotation/specialize.py
+++ b/pypy/annotation/specialize.py
@@ -353,6 +353,16 @@
key = tuple(key)
return maybe_star_args(funcdesc, key, args_s)
+def specialize_arg_or_var(funcdesc, args_s, *argindices):
+ for argno in argindices:
+ if not args_s[argno].is_constant():
+ break
+ else:
+ # all constant
+ return specialize_argvalue(funcdesc, args_s, *argindices)
+ # some not constant
+ return maybe_star_args(funcdesc, None, args_s)
+
def specialize_argtype(funcdesc, args_s, *argindices):
key = tuple([args_s[i].knowntype for i in argindices])
for cls in key:
@@ -370,3 +380,7 @@
else:
key = s.listdef.listitem.s_value.knowntype
return maybe_star_args(funcdesc, key, args_s)
+
+def specialize_call_location(funcdesc, args_s, op):
+ assert op is not None
+ return maybe_star_args(funcdesc, op, args_s)
diff --git a/pypy/annotation/test/test_annrpython.py b/pypy/annotation/test/test_annrpython.py
--- a/pypy/annotation/test/test_annrpython.py
+++ b/pypy/annotation/test/test_annrpython.py
@@ -1099,8 +1099,8 @@
allocdesc = a.bookkeeper.getdesc(alloc)
s_C1 = a.bookkeeper.immutablevalue(C1)
s_C2 = a.bookkeeper.immutablevalue(C2)
- graph1 = allocdesc.specialize([s_C1])
- graph2 = allocdesc.specialize([s_C2])
+ graph1 = allocdesc.specialize([s_C1], None)
+ graph2 = allocdesc.specialize([s_C2], None)
assert a.binding(graph1.getreturnvar()).classdef == C1df
assert a.binding(graph2.getreturnvar()).classdef == C2df
assert graph1 in a.translator.graphs
@@ -1135,8 +1135,8 @@
allocdesc = a.bookkeeper.getdesc(alloc)
s_C1 = a.bookkeeper.immutablevalue(C1)
s_C2 = a.bookkeeper.immutablevalue(C2)
- graph1 = allocdesc.specialize([s_C1, s_C2])
- graph2 = allocdesc.specialize([s_C2, s_C2])
+ graph1 = allocdesc.specialize([s_C1, s_C2], None)
+ graph2 = allocdesc.specialize([s_C2, s_C2], None)
assert a.binding(graph1.getreturnvar()).classdef == C1df
assert a.binding(graph2.getreturnvar()).classdef == C2df
assert graph1 in a.translator.graphs
@@ -1194,6 +1194,33 @@
assert len(executedesc._cache[(0, 'star', 2)].startblock.inputargs) == 4
assert len(executedesc._cache[(1, 'star', 3)].startblock.inputargs) == 5
+ def test_specialize_arg_or_var(self):
+ def f(a):
+ return 1
+ f._annspecialcase_ = 'specialize:arg_or_var(0)'
+
+ def fn(a):
+ return f(3) + f(a)
+
+ a = self.RPythonAnnotator()
+ a.build_types(fn, [int])
+ executedesc = a.bookkeeper.getdesc(f)
+ assert sorted(executedesc._cache.keys()) == [None, (3,)]
+ # we got two different special
+
+ def test_specialize_call_location(self):
+ def g(a):
+ return a
+ g._annspecialcase_ = "specialize:call_location"
+ def f(x):
+ return g(x)
+ f._annspecialcase_ = "specialize:argtype(0)"
+ def h(y):
+ w = f(y)
+ return int(f(str(y))) + w
+ a = self.RPythonAnnotator()
+ assert a.build_types(h, [int]) == annmodel.SomeInteger()
+
def test_assert_list_doesnt_lose_info(self):
class T(object):
pass
@@ -3177,6 +3204,8 @@
s = a.build_types(f, [])
assert isinstance(s, annmodel.SomeList)
assert not s.listdef.listitem.resized
+ assert not s.listdef.listitem.immutable
+ assert s.listdef.listitem.mutated
def test_delslice(self):
def f():
diff --git a/pypy/annotation/unaryop.py b/pypy/annotation/unaryop.py
--- a/pypy/annotation/unaryop.py
+++ b/pypy/annotation/unaryop.py
@@ -352,6 +352,7 @@
check_negative_slice(s_start, s_stop)
if not isinstance(s_iterable, SomeList):
raise Exception("list[start:stop] = x: x must be a list")
+ lst.listdef.mutate()
lst.listdef.agree(s_iterable.listdef)
# note that setslice is not allowed to resize a list in RPython
diff --git a/pypy/config/config.py b/pypy/config/config.py
--- a/pypy/config/config.py
+++ b/pypy/config/config.py
@@ -81,6 +81,12 @@
(self.__class__, name))
return self._cfgimpl_values[name]
+ def __dir__(self):
+ from_type = dir(type(self))
+ from_dict = list(self.__dict__)
+ extras = list(self._cfgimpl_values)
+ return sorted(set(extras + from_type + from_dict))
+
def __delattr__(self, name):
# XXX if you use delattr you are responsible for all bad things
# happening
diff --git a/pypy/config/makerestdoc.py b/pypy/config/makerestdoc.py
--- a/pypy/config/makerestdoc.py
+++ b/pypy/config/makerestdoc.py
@@ -134,7 +134,7 @@
for child in self._children:
subpath = fullpath + "." + child._name
toctree.append(subpath)
- content.add(Directive("toctree", *toctree, maxdepth=4))
+ content.add(Directive("toctree", *toctree, **{'maxdepth': 4}))
content.join(
ListItem(Strong("name:"), self._name),
ListItem(Strong("description:"), self.doc))
diff --git a/pypy/config/pypyoption.py b/pypy/config/pypyoption.py
--- a/pypy/config/pypyoption.py
+++ b/pypy/config/pypyoption.py
@@ -27,13 +27,14 @@
# --allworkingmodules
working_modules = default_modules.copy()
working_modules.update(dict.fromkeys(
- ["_socket", "unicodedata", "mmap", "fcntl", "_locale",
+ ["_socket", "unicodedata", "mmap", "fcntl", "_locale", "pwd",
"rctime" , "select", "zipimport", "_lsprof",
"crypt", "signal", "_rawffi", "termios", "zlib", "bz2",
"struct", "_hashlib", "_md5", "_sha", "_minimal_curses", "cStringIO",
"thread", "itertools", "pyexpat", "_ssl", "cpyext", "array",
"_bisect", "binascii", "_multiprocessing", '_warnings',
- "_collections", "_multibytecodec", "micronumpy", "_ffi"]
+ "_collections", "_multibytecodec", "micronumpy", "_ffi",
+ "_continuation"]
))
translation_modules = default_modules.copy()
@@ -57,6 +58,7 @@
# unix only modules
del working_modules["crypt"]
del working_modules["fcntl"]
+ del working_modules["pwd"]
del working_modules["termios"]
del working_modules["_minimal_curses"]
@@ -70,6 +72,7 @@
del working_modules['fcntl'] # LOCK_NB not defined
del working_modules["_minimal_curses"]
del working_modules["termios"]
+ del working_modules["_multiprocessing"] # depends on rctime
@@ -89,7 +92,7 @@
module_import_dependencies = {
# no _rawffi if importing pypy.rlib.clibffi raises ImportError
- # or CompilationError
+ # or CompilationError or py.test.skip.Exception
"_rawffi" : ["pypy.rlib.clibffi"],
"_ffi" : ["pypy.rlib.clibffi"],
@@ -99,6 +102,7 @@
"_ssl" : ["pypy.module._ssl.interp_ssl"],
"_hashlib" : ["pypy.module._ssl.interp_ssl"],
"_minimal_curses": ["pypy.module._minimal_curses.fficurses"],
+ "_continuation": ["pypy.rlib.rstacklet"],
}
def get_module_validator(modname):
@@ -109,7 +113,7 @@
try:
for name in modlist:
__import__(name)
- except (ImportError, CompilationError), e:
+ except (ImportError, CompilationError, py.test.skip.Exception), e:
errcls = e.__class__.__name__
config.add_warning(
"The module %r is disabled\n" % (modname,) +
@@ -124,7 +128,7 @@
pypy_optiondescription = OptionDescription("objspace", "Object Space Options", [
ChoiceOption("name", "Object Space name",
- ["std", "flow", "thunk", "dump", "taint"],
+ ["std", "flow", "thunk", "dump"],
"std",
cmdline='--objspace -o'),
@@ -327,6 +331,9 @@
BoolOption("mutable_builtintypes",
"Allow the changing of builtin types", default=False,
requires=[("objspace.std.builtinshortcut", True)]),
+ BoolOption("withidentitydict",
+ "track types that override __hash__, __eq__ or __cmp__ and use a special dict strategy for those which do not",
+ default=True),
]),
])
diff --git a/pypy/config/support.py b/pypy/config/support.py
--- a/pypy/config/support.py
+++ b/pypy/config/support.py
@@ -9,7 +9,7 @@
return 1 # don't override MAKEFLAGS. This will call 'make' without any '-j' option
if sys.platform == 'darwin':
return darwin_get_cpu_count()
- elif sys.platform != 'linux2':
+ elif not sys.platform.startswith('linux'):
return 1 # implement me
try:
if isinstance(filename_or_file, str):
diff --git a/pypy/config/test/test_config.py b/pypy/config/test/test_config.py
--- a/pypy/config/test/test_config.py
+++ b/pypy/config/test/test_config.py
@@ -1,5 +1,5 @@
from pypy.config.config import *
-import py
+import py, sys
def make_description():
gcoption = ChoiceOption('name', 'GC name', ['ref', 'framework'], 'ref')
@@ -63,6 +63,22 @@
py.test.raises(ConfigError, 'config.gc.name = "ref"')
config.gc.name = "framework"
+def test___dir__():
+ descr = make_description()
+ config = Config(descr, bool=False)
+ attrs = dir(config)
+ assert '__repr__' in attrs # from the type
+ assert '_cfgimpl_values' in attrs # from self
+ if sys.version_info >= (2, 6):
+ assert 'gc' in attrs # custom attribute
+ assert 'objspace' in attrs # custom attribute
+ #
+ attrs = dir(config.gc)
+ if sys.version_info >= (2, 6):
+ assert 'name' in attrs
+ assert 'dummy' in attrs
+ assert 'float' in attrs
+
def test_arbitrary_option():
descr = OptionDescription("top", "", [
ArbitraryOption("a", "no help", default=None)
@@ -265,11 +281,11 @@
def test_underscore_in_option_name():
descr = OptionDescription("opt", "", [
- BoolOption("_stackless", "", default=False),
+ BoolOption("_foobar", "", default=False),
])
config = Config(descr)
parser = to_optparse(config)
- assert parser.has_option("--_stackless")
+ assert parser.has_option("--_foobar")
def test_none():
dummy1 = BoolOption('dummy1', 'doc dummy', default=False, cmdline=None)
diff --git a/pypy/config/test/test_support.py b/pypy/config/test/test_support.py
--- a/pypy/config/test/test_support.py
+++ b/pypy/config/test/test_support.py
@@ -40,7 +40,7 @@
return self._value
def test_cpuinfo_linux():
- if sys.platform != 'linux2':
+ if not sys.platform.startswith('linux'):
py.test.skip("linux only")
saved = os.environ
try:
diff --git a/pypy/config/translationoption.py b/pypy/config/translationoption.py
--- a/pypy/config/translationoption.py
+++ b/pypy/config/translationoption.py
@@ -13,6 +13,10 @@
DEFL_LOW_INLINE_THRESHOLD = DEFL_INLINE_THRESHOLD / 2.0
DEFL_GC = "minimark"
+if sys.platform.startswith("linux"):
+ DEFL_ROOTFINDER_WITHJIT = "asmgcc"
+else:
+ DEFL_ROOTFINDER_WITHJIT = "shadowstack"
IS_64_BITS = sys.maxint > 2147483647
@@ -24,10 +28,9 @@
translation_optiondescription = OptionDescription(
"translation", "Translation Options", [
- BoolOption("stackless", "enable stackless features during compilation",
- default=False, cmdline="--stackless",
- requires=[("translation.type_system", "lltype"),
- ("translation.gcremovetypeptr", False)]), # XXX?
+ BoolOption("continuation", "enable single-shot continuations",
+ default=False, cmdline="--continuation",
+ requires=[("translation.type_system", "lltype")]),
ChoiceOption("type_system", "Type system to use when RTyping",
["lltype", "ootype"], cmdline=None, default="lltype",
requires={
@@ -66,7 +69,8 @@
"statistics": [("translation.gctransformer", "framework")],
"generation": [("translation.gctransformer", "framework")],
"hybrid": [("translation.gctransformer", "framework")],
- "boehm": [("translation.gctransformer", "boehm")],
+ "boehm": [("translation.gctransformer", "boehm"),
+ ("translation.continuation", False)], # breaks
"markcompact": [("translation.gctransformer", "framework")],
"minimark": [("translation.gctransformer", "framework")],
},
@@ -109,7 +113,7 @@
BoolOption("jit", "generate a JIT",
default=False,
suggests=[("translation.gc", DEFL_GC),
- ("translation.gcrootfinder", "asmgcc"),
+ ("translation.gcrootfinder", DEFL_ROOTFINDER_WITHJIT),
("translation.list_comprehension_operations", True)]),
ChoiceOption("jit_backend", "choose the backend for the JIT",
["auto", "x86", "x86-without-sse2", "llvm"],
@@ -140,7 +144,10 @@
["annotate", "rtype", "backendopt", "database", "source",
"pyjitpl"],
default=None, cmdline="--fork-before"),
-
+ BoolOption("dont_write_c_files",
+ "Make the C backend write everyting to /dev/null. " +
+ "Useful for benchmarking, so you don't actually involve the disk",
+ default=False, cmdline="--dont-write-c-files"),
ArbitraryOption("instrumentctl", "internal",
default=None),
StrOption("output", "Output file name", cmdline="--output"),
@@ -382,8 +389,6 @@
config.translation.suggest(withsmallfuncsets=5)
elif word == 'jit':
config.translation.suggest(jit=True)
- if config.translation.stackless:
- raise NotImplementedError("JIT conflicts with stackless for now")
elif word == 'removetypeptr':
config.translation.suggest(gcremovetypeptr=True)
else:
diff --git a/pypy/doc/__pypy__-module.rst b/pypy/doc/__pypy__-module.rst
--- a/pypy/doc/__pypy__-module.rst
+++ b/pypy/doc/__pypy__-module.rst
@@ -37,29 +37,6 @@
.. _`thunk object space docs`: objspace-proxies.html#thunk
.. _`interface section of the thunk object space docs`: objspace-proxies.html#thunk-interface
-.. broken:
-
- Taint Object Space Functionality
- ================================
-
- When the taint object space is used (choose with :config:`objspace.name`),
- the following names are put into ``__pypy__``:
-
- - ``taint``
- - ``is_tainted``
- - ``untaint``
- - ``taint_atomic``
- - ``_taint_debug``
- - ``_taint_look``
- - ``TaintError``
-
- Those are all described in the `interface section of the taint object space
- docs`_.
-
- For more detailed explanations and examples see the `taint object space docs`_.
-
- .. _`taint object space docs`: objspace-proxies.html#taint
- .. _`interface section of the taint object space docs`: objspace-proxies.html#taint-interface
Transparent Proxy Functionality
===============================
diff --git a/pypy/doc/_ref.txt b/pypy/doc/_ref.txt
--- a/pypy/doc/_ref.txt
+++ b/pypy/doc/_ref.txt
@@ -1,11 +1,10 @@
.. _`ctypes_configure/doc/sample.py`: https://bitbucket.org/pypy/pypy/src/default/ctypes_configure/doc/sample.py
.. _`demo/`: https://bitbucket.org/pypy/pypy/src/default/demo/
-.. _`demo/pickle_coroutine.py`: https://bitbucket.org/pypy/pypy/src/default/demo/pickle_coroutine.py
.. _`lib-python/`: https://bitbucket.org/pypy/pypy/src/default/lib-python/
.. _`lib-python/2.7/dis.py`: https://bitbucket.org/pypy/pypy/src/default/lib-python/2.7/dis.py
.. _`lib_pypy/`: https://bitbucket.org/pypy/pypy/src/default/lib_pypy/
+.. _`lib_pypy/greenlet.py`: https://bitbucket.org/pypy/pypy/src/default/lib_pypy/greenlet.py
.. _`lib_pypy/pypy_test/`: https://bitbucket.org/pypy/pypy/src/default/lib_pypy/pypy_test/
-.. _`lib_pypy/stackless.py`: https://bitbucket.org/pypy/pypy/src/default/lib_pypy/stackless.py
.. _`lib_pypy/tputil.py`: https://bitbucket.org/pypy/pypy/src/default/lib_pypy/tputil.py
.. _`pypy/annotation`:
.. _`pypy/annotation/`: https://bitbucket.org/pypy/pypy/src/default/pypy/annotation/
@@ -55,7 +54,6 @@
.. _`pypy/module`:
.. _`pypy/module/`: https://bitbucket.org/pypy/pypy/src/default/pypy/module/
.. _`pypy/module/__builtin__/__init__.py`: https://bitbucket.org/pypy/pypy/src/default/pypy/module/__builtin__/__init__.py
-.. _`pypy/module/_stackless/test/test_composable_coroutine.py`: https://bitbucket.org/pypy/pypy/src/default/pypy/module/_stackless/test/test_composable_coroutine.py
.. _`pypy/objspace`:
.. _`pypy/objspace/`: https://bitbucket.org/pypy/pypy/src/default/pypy/objspace/
.. _`pypy/objspace/dump.py`: https://bitbucket.org/pypy/pypy/src/default/pypy/objspace/dump.py
@@ -117,6 +115,7 @@
.. _`pypy/translator/`: https://bitbucket.org/pypy/pypy/src/default/pypy/translator/
.. _`pypy/translator/backendopt/`: https://bitbucket.org/pypy/pypy/src/default/pypy/translator/backendopt/
.. _`pypy/translator/c/`: https://bitbucket.org/pypy/pypy/src/default/pypy/translator/c/
+.. _`pypy/translator/c/src/stacklet/`: https://bitbucket.org/pypy/pypy/src/default/pypy/translator/c/src/stacklet/
.. _`pypy/translator/cli/`: https://bitbucket.org/pypy/pypy/src/default/pypy/translator/cli/
.. _`pypy/translator/goal/`: https://bitbucket.org/pypy/pypy/src/default/pypy/translator/goal/
.. _`pypy/translator/jvm/`: https://bitbucket.org/pypy/pypy/src/default/pypy/translator/jvm/
diff --git a/pypy/doc/architecture.rst b/pypy/doc/architecture.rst
--- a/pypy/doc/architecture.rst
+++ b/pypy/doc/architecture.rst
@@ -153,7 +153,7 @@
* Optionally, `various transformations`_ can then be applied which, for
example, perform optimizations such as inlining, add capabilities
- such as stackless_-style concurrency, or insert code for the
+ such as stackless-style concurrency (deprecated), or insert code for the
`garbage collector`_.
* Then, the graphs are converted to source code for the target platform
@@ -255,7 +255,6 @@
.. _Python: http://docs.python.org/reference/
.. _Psyco: http://psyco.sourceforge.net
-.. _stackless: stackless.html
.. _`generate Just-In-Time Compilers`: jit/index.html
.. _`JIT Generation in PyPy`: jit/index.html
.. _`implement your own interpreter`: http://morepypy.blogspot.com/2011/04/tutorial-writing-interpreter-with-pypy.html
diff --git a/pypy/doc/coding-guide.rst b/pypy/doc/coding-guide.rst
--- a/pypy/doc/coding-guide.rst
+++ b/pypy/doc/coding-guide.rst
@@ -929,6 +929,19 @@
located in the ``py/bin/`` directory. For switches to
modify test execution pass the ``-h`` option.
+Coverage reports
+----------------
+
+In order to get coverage reports the `pytest-cov`_ plugin is included.
+it adds some extra requirements ( coverage_ and `cov-core`_ )
+and can once they are installed coverage testing can be invoked via::
+
+ python test_all.py --cov file_or_direcory_to_cover file_or_directory
+
+.. _`pytest-cov`: http://pypi.python.org/pypi/pytest-cov
+.. _`coverage`: http://pypi.python.org/pypi/coverage
+.. _`cov-core`: http://pypi.python.org/pypi/cov-core
+
Test conventions
----------------
diff --git a/pypy/doc/conf.py b/pypy/doc/conf.py
--- a/pypy/doc/conf.py
+++ b/pypy/doc/conf.py
@@ -45,9 +45,9 @@
# built documents.
#
# The short X.Y version.
-version = '1.5'
+version = '1.6'
# The full version, including alpha/beta/rc tags.
-release = '1.5'
+release = '1.6'
# The language for content autogenerated by Sphinx. Refer to documentation
# for a list of supported languages.
diff --git a/pypy/doc/config/objspace.name.txt b/pypy/doc/config/objspace.name.txt
--- a/pypy/doc/config/objspace.name.txt
+++ b/pypy/doc/config/objspace.name.txt
@@ -4,7 +4,6 @@
for normal usage):
* thunk_: The thunk object space adds lazy evaluation to PyPy.
- * taint_: The taint object space adds soft security features.
* dump_: Using this object spaces results in the dumpimp of all operations
to a log.
@@ -12,5 +11,4 @@
.. _`Object Space Proxies`: ../objspace-proxies.html
.. _`Standard Object Space`: ../objspace.html#standard-object-space
.. _thunk: ../objspace-proxies.html#thunk
-.. _taint: ../objspace-proxies.html#taint
.. _dump: ../objspace-proxies.html#dump
diff --git a/pypy/doc/config/objspace.std.withidentitydict.txt b/pypy/doc/config/objspace.std.withidentitydict.txt
new file mode 100644
--- /dev/null
+++ b/pypy/doc/config/objspace.std.withidentitydict.txt
@@ -0,0 +1,21 @@
+=============================
+objspace.std.withidentitydict
+=============================
+
+* **name:** withidentitydict
+
+* **description:** enable a dictionary strategy for "by identity" comparisons
+
+* **command-line:** --objspace-std-withidentitydict
+
+* **command-line for negation:** --no-objspace-std-withidentitydict
+
+* **option type:** boolean option
+
+* **default:** True
+
+
+Enable a dictionary strategy specialized for instances of classes which
+compares "by identity", which is the default unless you override ``__hash__``,
+``__eq__`` or ``__cmp__``. This strategy will be used only with new-style
+classes.
diff --git a/pypy/doc/config/objspace.usemodules._stackless.txt b/pypy/doc/config/objspace.usemodules._continuation.txt
rename from pypy/doc/config/objspace.usemodules._stackless.txt
rename to pypy/doc/config/objspace.usemodules._continuation.txt
--- a/pypy/doc/config/objspace.usemodules._stackless.txt
+++ b/pypy/doc/config/objspace.usemodules._continuation.txt
@@ -1,6 +1,4 @@
-Use the '_stackless' module.
+Use the '_continuation' module.
-Exposes the `stackless` primitives, and also implies a stackless build.
-See also :config:`translation.stackless`.
-
-.. _`stackless`: ../stackless.html
+Exposes the `continulet` app-level primitives.
+See also :config:`translation.continuation`.
diff --git a/pypy/doc/config/objspace.usemodules.pwd.txt b/pypy/doc/config/objspace.usemodules.pwd.txt
new file mode 100644
--- /dev/null
+++ b/pypy/doc/config/objspace.usemodules.pwd.txt
@@ -0,0 +1,2 @@
+Use the 'pwd' module.
+This module is expected to be fully working.
diff --git a/pypy/doc/config/translation.stackless.txt b/pypy/doc/config/translation.continuation.txt
rename from pypy/doc/config/translation.stackless.txt
rename to pypy/doc/config/translation.continuation.txt
--- a/pypy/doc/config/translation.stackless.txt
+++ b/pypy/doc/config/translation.continuation.txt
@@ -1,5 +1,2 @@
-Run the `stackless transform`_ on each generated graph, which enables the use
-of coroutines at RPython level and the "stackless" module when translating
-PyPy.
-
-.. _`stackless transform`: ../stackless.html
+Enable the use of a stackless-like primitive called "stacklet".
+In PyPy, this is exposed at app-level by the "_continuation" module.
diff --git a/pypy/doc/config/translation.dont_write_c_files.txt b/pypy/doc/config/translation.dont_write_c_files.txt
new file mode 100644
--- /dev/null
+++ b/pypy/doc/config/translation.dont_write_c_files.txt
@@ -0,0 +1,4 @@
+write the generated C files to ``/dev/null`` instead of to the disk. Useful if
+you want to use translate.py as a benchmark and don't want to access the disk.
+
+.. _`translation documentation`: ../translation.html
diff --git a/pypy/doc/config/translation.gc.txt b/pypy/doc/config/translation.gc.txt
--- a/pypy/doc/config/translation.gc.txt
+++ b/pypy/doc/config/translation.gc.txt
@@ -1,4 +1,6 @@
-Choose the Garbage Collector used by the translated program:
+Choose the Garbage Collector used by the translated program.
+The good performing collectors are "hybrid" and "minimark".
+The default is "minimark".
- "ref": reference counting. Takes very long to translate and the result is
slow.
@@ -11,3 +13,12 @@
older generation.
- "boehm": use the Boehm conservative GC.
+
+ - "hybrid": a hybrid collector of "generation" together with a
+ mark-n-sweep old space
+
+ - "markcompact": a slow, but memory-efficient collector,
+ influenced e.g. by Smalltalk systems.
+
+ - "minimark": a generational mark-n-sweep collector with good
+ performance. Includes page marking for large arrays.
diff --git a/pypy/doc/contributor.rst b/pypy/doc/contributor.rst
--- a/pypy/doc/contributor.rst
+++ b/pypy/doc/contributor.rst
@@ -9,22 +9,22 @@
Armin Rigo
Maciej Fijalkowski
Carl Friedrich Bolz
+ Antonio Cuni
Amaury Forgeot d'Arc
- Antonio Cuni
Samuele Pedroni
Michael Hudson
Holger Krekel
+ Benjamin Peterson
Christian Tismer
- Benjamin Peterson
+ Hakan Ardo
+ Alex Gaynor
Eric van Riet Paap
- Anders Chrigström
- Håkan Ardö
+ Anders Chrigstrom
+ David Schneider
Richard Emslie
Dan Villiom Podlaski Christiansen
Alexander Schremmer
- Alex Gaynor
- David Schneider
- Aurelién Campeas
+ Aurelien Campeas
Anders Lehmann
Camillo Bruni
Niklaus Haldimann
@@ -35,16 +35,17 @@
Bartosz Skowron
Jakub Gustak
Guido Wesdorp
+ Daniel Roberts
Adrien Di Mascio
Laura Creighton
Ludovic Aubry
Niko Matsakis
- Daniel Roberts
Jason Creighton
- Jacob Hallén
+ Jacob Hallen
Alex Martelli
Anders Hammarquist
Jan de Mooij
+ Wim Lavrijsen
Stephan Diehl
Michael Foord
Stefan Schwarzer
@@ -55,9 +56,13 @@
Alexandre Fayolle
Marius Gedminas
Simon Burton
+ Justin Peel
Jean-Paul Calderone
John Witulski
+ Lukas Diekmann
+ holger krekel
Wim Lavrijsen
+ Dario Bertini
Andreas Stührk
Jean-Philippe St. Pierre
Guido van Rossum
@@ -69,15 +74,16 @@
Georg Brandl
Gerald Klix
Wanja Saatkamp
+ Ronny Pfannschmidt
Boris Feigin
Oscar Nierstrasz
- Dario Bertini
David Malcolm
Eugene Oden
Henry Mason
+ Sven Hager
Lukas Renggli
+ Ilya Osadchiy
Guenter Jantzen
- Ronny Pfannschmidt
Bert Freudenberg
Amit Regmi
Ben Young
@@ -94,8 +100,8 @@
Jared Grubb
Karl Bartel
Gabriel Lavoie
+ Victor Stinner
Brian Dorsey
- Victor Stinner
Stuart Williams
Toby Watson
Antoine Pitrou
@@ -106,19 +112,23 @@
Jonathan David Riehl
Elmo Mäntynen
Anders Qvist
- Beatrice Düring
+ Beatrice During
Alexander Sedov
+ Timo Paulssen
+ Corbin Simpson
Vincent Legoll
+ Romain Guillebert
Alan McIntyre
- Romain Guillebert
Alex Perry
Jens-Uwe Mager
+ Simon Cross
Dan Stromberg
- Lukas Diekmann
+ Guillebert Romain
Carl Meyer
Pieter Zieschang
Alejandro J. Cura
Sylvain Thenault
+ Christoph Gerum
Travis Francis Athougies
Henrik Vendelbo
Lutz Paelike
@@ -129,6 +139,7 @@
Miguel de Val Borro
Ignas Mikalajunas
Artur Lisiecki
+ Philip Jenvey
Joshua Gilbert
Godefroid Chappelle
Yusei Tahara
@@ -137,24 +148,29 @@
Gustavo Niemeyer
William Leslie
Akira Li
- Kristján Valur Jónsson
+ Kristjan Valur Jonsson
Bobby Impollonia
+ Michael Hudson-Doyle
Andrew Thompson
Anders Sigfridsson
+ Floris Bruynooghe
Jacek Generowicz
Dan Colish
- Sven Hager
Zooko Wilcox-O Hearn
+ Dan Villiom Podlaski Christiansen
Anders Hammarquist
+ Chris Lambacher
Dinu Gherman
Dan Colish
+ Brett Cannon
Daniel Neuhäuser
Michael Chermside
Konrad Delong
Anna Ravencroft
Greg Price
Armin Ronacher
+ Christian Muirhead
Jim Baker
- Philip Jenvey
Rodrigo Araújo
+ Romain Guillebert
diff --git a/pypy/doc/cpython_differences.rst b/pypy/doc/cpython_differences.rst
--- a/pypy/doc/cpython_differences.rst
+++ b/pypy/doc/cpython_differences.rst
@@ -24,6 +24,7 @@
_bisect
_codecs
_collections
+ `_continuation`_
`_ffi`_
_hashlib
_io
@@ -84,9 +85,12 @@
_winreg
- Extra module with Stackless_ only:
-
- _stackless
+ Note that only some of these modules are built-in in a typical
+ CPython installation, and the rest is from non built-in extension
+ modules. This means that e.g. ``import parser`` will, on CPython,
+ find a local file ``parser.py``, while ``import sys`` will not find a
+ local file ``sys.py``. In PyPy the difference does not exist: all
+ these modules are built-in.
* Supported by being rewritten in pure Python (possibly using ``ctypes``):
see the `lib_pypy/`_ directory. Examples of modules that we
@@ -101,11 +105,11 @@
.. the nonstandard modules are listed below...
.. _`__pypy__`: __pypy__-module.html
+.. _`_continuation`: stackless.html
.. _`_ffi`: ctypes-implementation.html
.. _`_rawffi`: ctypes-implementation.html
.. _`_minimal_curses`: config/objspace.usemodules._minimal_curses.html
.. _`cpyext`: http://morepypy.blogspot.com/2010/04/using-cpython-extension-modules-with.html
-.. _Stackless: stackless.html
Differences related to garbage collection strategies
@@ -211,6 +215,38 @@
>>>> print d1['a']
42
+Mutating classes of objects which are already used as dictionary keys
+---------------------------------------------------------------------
+
+Consider the following snippet of code::
+
+ class X(object):
+ pass
+
+ def __evil_eq__(self, other):
+ print 'hello world'
+ return False
+
+ def evil(y):
+ d = {x(): 1}
+ X.__eq__ = __evil_eq__
+ d[y] # might trigger a call to __eq__?
+
+In CPython, __evil_eq__ **might** be called, although there is no way to write
+a test which reliably calls it. It happens if ``y is not x`` and ``hash(y) ==
+hash(x)``, where ``hash(x)`` is computed when ``x`` is inserted into the
+dictionary. If **by chance** the condition is satisfied, then ``__evil_eq__``
+is called.
+
+PyPy uses a special strategy to optimize dictionaries whose keys are instances
+of user-defined classes which do not override the default ``__hash__``,
+``__eq__`` and ``__cmp__``: when using this strategy, ``__eq__`` and
+``__cmp__`` are never called, but instead the lookup is done by identity, so
+in the case above it is guaranteed that ``__eq__`` won't be called.
+
+Note that in all other cases (e.g., if you have a custom ``__hash__`` and
+``__eq__`` in ``y``) the behavior is exactly the same as CPython.
+
Ignored exceptions
-----------------------
@@ -248,7 +284,14 @@
never a dictionary as it sometimes is in CPython. Assigning to
``__builtins__`` has no effect.
-* object identity of immutable keys in dictionaries is not necessarily preserved.
- Never compare immutable objects with ``is``.
+* Do not compare immutable objects with ``is``. For example on CPython
+ it is true that ``x is 0`` works, i.e. does the same as ``type(x) is
+ int and x == 0``, but it is so by accident. If you do instead
+ ``x is 1000``, then it stops working, because 1000 is too large and
+ doesn't come from the internal cache. In PyPy it fails to work in
+ both cases, because we have no need for a cache at all.
+
+* Also, object identity of immutable keys in dictionaries is not necessarily
+ preserved.
.. include:: _ref.txt
diff --git a/pypy/doc/extending.rst b/pypy/doc/extending.rst
--- a/pypy/doc/extending.rst
+++ b/pypy/doc/extending.rst
@@ -19,12 +19,12 @@
section
* Write them in pure python and use direct libffi low-level bindings, See
- \_rawffi_ module description.
+ \_ffi_ module description.
* Write them in RPython as mixedmodule_, using *rffi* as bindings.
.. _ctypes: #CTypes
-.. _\_rawffi: #LibFFI
+.. _\_ffi: #LibFFI
.. _mixedmodule: #Mixed Modules
CTypes
@@ -42,41 +42,50 @@
platform-dependent details (compiling small snippets of C code and running
them), so it'll benefit not pypy-related ctypes-based modules as well.
+ctypes call are optimized by the JIT and the resulting machine code contains a
+direct call to the target C function. However, due to the very dynamic nature
+of ctypes, some overhead over a bare C call is still present, in particular to
+check/convert the types of the parameters. Moreover, even if most calls are
+optimized, some cannot and thus need to follow the slow path, not optimized by
+the JIT.
+
.. _`ctypes-configure`: ctypes-implementation.html#ctypes-configure
+.. _`CPython ctypes`: http://docs.python.org/library/ctypes.html
Pros
----
-Stable, CPython-compatible API
+Stable, CPython-compatible API. Most calls are fast, optimized by JIT.
Cons
----
-Only pure-python code (slow), problems with platform-dependency (although
-we partially solve those). PyPy implementation is now very slow.
+Problems with platform-dependency (although we partially solve
+those). Although the JIT optimizes ctypes calls, some overhead is still
+present. The slow-path is very slow.
-_`CPython ctypes`: http://python.net/crew/theller/ctypes/
LibFFI
======
Mostly in order to be able to write a ctypes module, we developed a very
-low-level libffi bindings. (libffi is a C-level library for dynamic calling,
+low-level libffi bindings called ``_ffi``. (libffi is a C-level library for dynamic calling,
which is used by CPython ctypes). This library provides stable and usable API,
although it's API is a very low-level one. It does not contain any
-magic.
+magic. It is also optimized by the JIT, but has much less overhead than ctypes.
Pros
----
-Works. Combines disadvantages of using ctypes with disadvantages of
-using mixed modules. Probably more suitable for a delicate code
-where ctypes magic goes in a way.
+It Works. Probably more suitable for a delicate code where ctypes magic goes
+in a way. All calls are optimized by the JIT, there is no slow path as in
+ctypes.
Cons
----
-Slow. CPython-incompatible API, very rough and low-level
+It combines disadvantages of using ctypes with disadvantages of using mixed
+modules. CPython-incompatible API, very rough and low-level.
Mixed Modules
=============
@@ -87,15 +96,15 @@
* a mixed module needs to be written in RPython, which is far more
complicated than Python (XXX link)
-* due to lack of separate compilation (as of April 2008), each
+* due to lack of separate compilation (as of July 2011), each
compilation-check requires to recompile whole PyPy python interpreter,
which takes 0.5-1h. We plan to solve this at some point in near future.
* although rpython is a garbage-collected language, the border between
C and RPython needs to be managed by hand (each object that goes into the
- C level must be explicitly freed) XXX we try to solve this
+ C level must be explicitly freed).
-Some document is available `here`_
+Some documentation is available `here`_
.. _`here`: rffi.html
diff --git a/pypy/doc/faq.rst b/pypy/doc/faq.rst
--- a/pypy/doc/faq.rst
+++ b/pypy/doc/faq.rst
@@ -315,6 +315,28 @@
.. _`Andrew Brown's tutorial`: http://morepypy.blogspot.com/2011/04/tutorial-writing-interpreter-with-pypy.html
+---------------------------------------------------------
+Can RPython modules for PyPy be translated independently?
+---------------------------------------------------------
+
+No, you have to rebuild the entire interpreter. This means two things:
+
+* It is imperative to use test-driven development. You have to test
+ exhaustively your module in pure Python, before even attempting to
+ translate it. Once you translate it, you should have only a few typing
+ issues left to fix, but otherwise the result should work out of the box.
+
+* Second, and perhaps most important: do you have a really good reason
+ for writing the module in RPython in the first place? Nowadays you
+ should really look at alternatives, like writing it in pure Python,
+ using ctypes if it needs to call C code. Other alternatives are being
+ developed too (as of summer 2011), like a Cython binding.
+
+In this context it is not that important to be able to translate
+RPython modules independently of translating the complete interpreter.
+(It could be done given enough efforts, but it's a really serious
+undertaking. Consider it as quite unlikely for now.)
+
----------------------------------------------------------
Why does PyPy draw a Mandelbrot fractal while translating?
----------------------------------------------------------
diff --git a/pypy/doc/garbage_collection.rst b/pypy/doc/garbage_collection.rst
--- a/pypy/doc/garbage_collection.rst
+++ b/pypy/doc/garbage_collection.rst
@@ -147,7 +147,7 @@
You can read more about them at the start of
`pypy/rpython/memory/gc/minimark.py`_.
-In more details:
+In more detail:
- The small newly malloced objects are allocated in the nursery (case 1).
All objects living in the nursery are "young".
diff --git a/pypy/doc/getting-started-python.rst b/pypy/doc/getting-started-python.rst
--- a/pypy/doc/getting-started-python.rst
+++ b/pypy/doc/getting-started-python.rst
@@ -32,7 +32,10 @@
.. _`windows document`: windows.html
You can translate the whole of PyPy's Python interpreter to low level C code,
-or `CLI code`_.
+or `CLI code`_. If you intend to build using gcc, check to make sure that
+the version you have is not 4.2 or you will run into `this bug`_.
+
+.. _`this bug`: https://bugs.launchpad.net/ubuntu/+source/gcc-4.2/+bug/187391
1. First `download a pre-built PyPy`_ for your architecture which you will
use to translate your Python interpreter. It is, of course, possible to
@@ -64,7 +67,6 @@
* ``libssl-dev`` (for the optional ``_ssl`` module)
* ``libgc-dev`` (for the Boehm garbage collector: only needed when translating with `--opt=0, 1` or `size`)
* ``python-sphinx`` (for the optional documentation build. You need version 1.0.7 or later)
- * ``python-greenlet`` (for the optional stackless support in interpreted mode/testing)
3. Translation is time-consuming -- 45 minutes on a very fast machine --
@@ -102,7 +104,7 @@
$ ./pypy-c
Python 2.7.0 (61ef2a11b56a, Mar 02 2011, 03:00:11)
- [PyPy 1.5.0-alpha0 with GCC 4.4.3] on linux2
+ [PyPy 1.6.0 with GCC 4.4.3] on linux2
Type "help", "copyright", "credits" or "license" for more information.
And now for something completely different: ``this sentence is false''
>>>> 46 - 4
@@ -117,19 +119,8 @@
Installation_ below.
The ``translate.py`` script takes a very large number of options controlling
-what to translate and how. See ``translate.py -h``. Some of the more
-interesting options (but for now incompatible with the JIT) are:
-
- * ``--stackless``: this produces a pypy-c that includes features
- inspired by `Stackless Python <http://www.stackless.com>`__.
-
- * ``--gc=boehm|ref|marknsweep|semispace|generation|hybrid|minimark``:
- choose between using
- the `Boehm-Demers-Weiser garbage collector`_, our reference
- counting implementation or one of own collector implementations
- (the default depends on the optimization level but is usually
- ``minimark``).
-
+what to translate and how. See ``translate.py -h``. The default options
+should be suitable for mostly everybody by now.
Find a more detailed description of the various options in our `configuration
sections`_.
@@ -162,7 +153,7 @@
$ ./pypy-cli
Python 2.7.0 (61ef2a11b56a, Mar 02 2011, 03:00:11)
- [PyPy 1.5.0-alpha0] on linux2
+ [PyPy 1.6.0] on linux2
Type "help", "copyright", "credits" or "license" for more information.
And now for something completely different: ``distopian and utopian chairs''
>>>>
@@ -199,7 +190,7 @@
$ ./pypy-jvm
Python 2.7.0 (61ef2a11b56a, Mar 02 2011, 03:00:11)
- [PyPy 1.5.0-alpha0] on linux2
+ [PyPy 1.6.0] on linux2
Type "help", "copyright", "credits" or "license" for more information.
And now for something completely different: ``# assert did not crash''
>>>>
@@ -238,7 +229,7 @@
the ``bin/pypy`` executable.
To install PyPy system wide on unix-like systems, it is recommended to put the
-whole hierarchy alone (e.g. in ``/opt/pypy1.5``) and put a symlink to the
+whole hierarchy alone (e.g. in ``/opt/pypy1.6``) and put a symlink to the
``pypy`` executable into ``/usr/bin`` or ``/usr/local/bin``
If the executable fails to find suitable libraries, it will report
diff --git a/pypy/doc/getting-started.rst b/pypy/doc/getting-started.rst
--- a/pypy/doc/getting-started.rst
+++ b/pypy/doc/getting-started.rst
@@ -53,11 +53,11 @@
PyPy is ready to be executed as soon as you unpack the tarball or the zip
file, with no need to install it in any specific location::
- $ tar xf pypy-1.5-linux.tar.bz2
+ $ tar xf pypy-1.6-linux.tar.bz2
- $ ./pypy-1.5-linux/bin/pypy
+ $ ./pypy-1.6/bin/pypy
Python 2.7.1 (?, Apr 27 2011, 12:44:21)
- [PyPy 1.5.0-alpha0 with GCC 4.4.3] on linux2
+ [PyPy 1.6.0 with GCC 4.4.3] on linux2
Type "help", "copyright", "credits" or "license" for more information.
And now for something completely different: ``implementing LOGO in LOGO:
"turtles all the way down"''
@@ -73,16 +73,16 @@
$ curl -O http://python-distribute.org/distribute_setup.py
- $ curl -O https://github.com/pypa/pip/raw/master/contrib/get-pip.py
+ $ curl -O https://raw.github.com/pypa/pip/master/contrib/get-pip.py
- $ ./pypy-1.5-linux/bin/pypy distribute_setup.py
+ $ ./pypy-1.6/bin/pypy distribute_setup.py
- $ ./pypy-1.5-linux/bin/pypy get-pip.py
+ $ ./pypy-1.6/bin/pypy get-pip.py
- $ ./pypy-1.5-linux/bin/pip install pygments # for example
+ $ ./pypy-1.6/bin/pip install pygments # for example
-3rd party libraries will be installed in ``pypy-1.5-linux/site-packages``, and
-the scripts in ``pypy-1.5-linux/bin``.
+3rd party libraries will be installed in ``pypy-1.6/site-packages``, and
+the scripts in ``pypy-1.6/bin``.
Installing using virtualenv
---------------------------
diff --git a/pypy/doc/how-to-release.rst b/pypy/doc/how-to-release.rst
--- a/pypy/doc/how-to-release.rst
+++ b/pypy/doc/how-to-release.rst
@@ -21,8 +21,8 @@
Release Steps
----------------
-* at code freeze make a release branch under
- http://codepeak.net/svn/pypy/release/x.y(.z). IMPORTANT: bump the
+* at code freeze make a release branch using release-x.x.x in mercurial
+ IMPORTANT: bump the
pypy version number in module/sys/version.py and in
module/cpyext/include/patchlevel.h, notice that the branch
will capture the revision number of this change for the release;
@@ -42,18 +42,11 @@
JIT: windows, linux, os/x
no JIT: windows, linux, os/x
sandbox: linux, os/x
- stackless: windows, linux, os/x
* write release announcement pypy/doc/release-x.y(.z).txt
the release announcement should contain a direct link to the download page
* update pypy.org (under extradoc/pypy.org), rebuild and commit
-* update http://codespeak.net/pypy/trunk:
- code0> + chmod -R yourname:users /www/codespeak.net/htdocs/pypy/trunk
- local> cd ..../pypy/doc && py.test
- local> cd ..../pypy
- local> rsync -az doc codespeak.net:/www/codespeak.net/htdocs/pypy/trunk/pypy/
-
* post announcement on morepypy.blogspot.com
* send announcements to pypy-dev, python-list,
python-announce, python-dev ...
diff --git a/pypy/doc/index-of-release-notes.rst b/pypy/doc/index-of-release-notes.rst
--- a/pypy/doc/index-of-release-notes.rst
+++ b/pypy/doc/index-of-release-notes.rst
@@ -16,3 +16,4 @@
release-1.4.0beta.rst
release-1.4.1.rst
release-1.5.0.rst
+ release-1.6.0.rst
diff --git a/pypy/doc/index.rst b/pypy/doc/index.rst
--- a/pypy/doc/index.rst
+++ b/pypy/doc/index.rst
@@ -15,14 +15,12 @@
* `FAQ`_: some frequently asked questions.
-* `Release 1.5`_: the latest official release
+* `Release 1.6`_: the latest official release
* `PyPy Blog`_: news and status info about PyPy
* `Papers`_: Academic papers, talks, and related projects
-* `Videos`_: Videos of PyPy talks and presentations
-
* `speed.pypy.org`_: Daily benchmarks of how fast PyPy is
* `potential project ideas`_: In case you want to get your feet wet...
@@ -35,7 +33,7 @@
* `Differences between PyPy and CPython`_
* `What PyPy can do for your objects`_
- * `Stackless and coroutines`_
+ * `Continulets and greenlets`_
* `JIT Generation in PyPy`_
* `Sandboxing Python code`_
@@ -77,7 +75,7 @@
.. _`Getting Started`: getting-started.html
.. _`Papers`: extradoc.html
.. _`Videos`: video-index.html
-.. _`Release 1.5`: http://pypy.org/download.html
+.. _`Release 1.6`: http://pypy.org/download.html
.. _`speed.pypy.org`: http://speed.pypy.org
.. _`RPython toolchain`: translation.html
.. _`potential project ideas`: project-ideas.html
@@ -122,9 +120,9 @@
Windows, on top of .NET, and on top of Java.
To dig into PyPy it is recommended to try out the current
Mercurial default branch, which is always working or mostly working,
-instead of the latest release, which is `1.5`__.
+instead of the latest release, which is `1.6`__.
-.. __: release-1.5.0.html
+.. __: release-1.6.0.html
PyPy is mainly developed on Linux and Mac OS X. Windows is supported,
but platform-specific bugs tend to take longer before we notice and fix
@@ -292,8 +290,6 @@
`pypy/translator/jvm/`_ the Java backend
-`pypy/translator/stackless/`_ the `Stackless Transform`_
-
`pypy/translator/tool/`_ helper tools for translation, including the Pygame
`graph viewer`_
@@ -313,12 +309,11 @@
.. _`object space`: objspace.html
.. _FlowObjSpace: objspace.html#the-flow-object-space
.. _`trace object space`: objspace.html#the-trace-object-space
-.. _`taint object space`: objspace-proxies.html#taint
.. _`thunk object space`: objspace-proxies.html#thunk
.. _`transparent proxies`: objspace-proxies.html#tproxy
.. _`Differences between PyPy and CPython`: cpython_differences.html
.. _`What PyPy can do for your objects`: objspace-proxies.html
-.. _`Stackless and coroutines`: stackless.html
+.. _`Continulets and greenlets`: stackless.html
.. _StdObjSpace: objspace.html#the-standard-object-space
.. _`abstract interpretation`: http://en.wikipedia.org/wiki/Abstract_interpretation
.. _`rpython`: coding-guide.html#rpython
@@ -337,7 +332,6 @@
.. _`low-level type system`: rtyper.html#low-level-type
.. _`object-oriented type system`: rtyper.html#oo-type
.. _`garbage collector`: garbage_collection.html
-.. _`Stackless Transform`: translation.html#the-stackless-transform
.. _`main PyPy-translation scripts`: getting-started-python.html#translating-the-pypy-python-interpreter
.. _`.NET`: http://www.microsoft.com/net/
.. _Mono: http://www.mono-project.com/
diff --git a/pypy/doc/jit/pyjitpl5.rst b/pypy/doc/jit/pyjitpl5.rst
--- a/pypy/doc/jit/pyjitpl5.rst
+++ b/pypy/doc/jit/pyjitpl5.rst
@@ -103,7 +103,7 @@
The meta-interpreter starts interpreting the JIT bytecode. Each operation is
executed and then recorded in a list of operations, called the trace.
-Operations can have a list of boxes that operate on, arguments. Some operations
+Operations can have a list of boxes they operate on, arguments. Some operations
(like GETFIELD and GETARRAYITEM) also have special objects that describe how
their arguments are laid out in memory. All possible operations generated by
tracing are listed in metainterp/resoperation.py. When a (interpreter-level)
diff --git a/pypy/doc/objspace-proxies.rst b/pypy/doc/objspace-proxies.rst
--- a/pypy/doc/objspace-proxies.rst
+++ b/pypy/doc/objspace-proxies.rst
@@ -129,297 +129,6 @@
function behaves lazily: all calls to it return a thunk object.
-.. broken right now:
-
- .. _taint:
-
- The Taint Object Space
- ======================
-
- Motivation
- ----------
-
- The Taint Object Space provides a form of security: "tainted objects",
- inspired by various sources, see [D12.1]_ for a more detailed discussion.
-
- The basic idea of this kind of security is not to protect against
- malicious code but to help with handling and boxing sensitive data.
- It covers two kinds of sensitive data: secret data which should not leak,
- and untrusted data coming from an external source and that must be
- validated before it is used.
-
- The idea is that, considering a large application that handles these
- kinds of sensitive data, there are typically only a small number of
- places that need to explicitly manipulate that sensitive data; all the
- other places merely pass it around, or do entirely unrelated things.
-
- Nevertheless, if a large application needs to be reviewed for security,
- it must be entirely carefully checked, because it is possible that a
- bug at some apparently unrelated place could lead to a leak of sensitive
- information in a way that an external attacker could exploit. For
- example, if any part of the application provides web services, an
- attacker might be able to issue unexpected requests with a regular web
- browser and deduce secret information from the details of the answers he
- gets. Another example is the common CGI attack where an attacker sends
- malformed inputs and causes the CGI script to do unintended things.
-
- An approach like that of the Taint Object Space allows the small parts
- of the program that manipulate sensitive data to be explicitly marked.
- The effect of this is that although these small parts still need a
- careful security review, the rest of the application no longer does,
- because even a bug would be unable to leak the information.
-
- We have implemented a simple two-level model: objects are either
- regular (untainted), or sensitive (tainted). Objects are marked as
- sensitive if they are secret or untrusted, and only declassified at
- carefully-checked positions (e.g. where the secret data is needed, or
- after the untrusted data has been fully validated).
-
- It would be simple to extend the code for more fine-grained scales of
- secrecy. For example it is typical in the literature to consider
- user-specified lattices of secrecy levels, corresponding to multiple
- "owners" that cannot access data belonging to another "owner" unless
- explicitly authorized to do so.
-
- Tainting and untainting
- -----------------------
-
- Start a py.py with the Taint Object Space and try the following example::
-
- $ py.py -o taint
- >>>> from __pypy__ import taint
- >>>> x = taint(6)
-
- # x is hidden from now on. We can pass it around and
- # even operate on it, but not inspect it. Taintness
- # is propagated to operation results.
-
- >>>> x
- TaintError
-
- >>>> if x > 5: y = 2 # see below
- TaintError
-
- >>>> y = x + 5 # ok
- >>>> lst = [x, y]
- >>>> z = lst.pop()
- >>>> t = type(z) # type() works too, tainted answer
- >>>> t
- TaintError
- >>>> u = t is int # even 'is' works
- >>>> u
- TaintError
-
- Notice that using a tainted boolean like ``x > 5`` in an ``if``
- statement is forbidden. This is because knowing which path is followed
- would give away a hint about ``x``; in the example above, if the
- statement ``if x > 5: y = 2`` was allowed to run, we would know
- something about the value of ``x`` by looking at the (untainted) value
- in the variable ``y``.
-
- Of course, there is a way to inspect tainted objects. The basic way is
- to explicitly "declassify" it with the ``untaint()`` function. In an
- application, the places that use ``untaint()`` are the places that need
- careful security review. To avoid unexpected objects showing up, the
- ``untaint()`` function must be called with the exact type of the object
- to declassify. It will raise ``TaintError`` if the type doesn't match::
-
- >>>> from __pypy__ import taint
- >>>> untaint(int, x)
- 6
- >>>> untaint(int, z)
- 11
- >>>> untaint(bool, x > 5)
- True
- >>>> untaint(int, x > 5)
- TaintError
-
-
- Taint Bombs
- -----------
-
- In this area, a common problem is what to do about failing operations.
- If an operation raises an exception when manipulating a tainted object,
- then the very presence of the exception can leak information about the
- tainted object itself. Consider::
-
- >>>> 5 / (x-6)
-
- By checking if this raises ``ZeroDivisionError`` or not, we would know
- if ``x`` was equal to 6 or not. The solution to this problem in the
- Taint Object Space is to introduce *Taint Bombs*. They are a kind of
- tainted object that doesn't contain a real object, but a pending
- exception. Taint Bombs are indistinguishable from normal tainted
- objects to unprivileged code. See::
-
- >>>> x = taint(6)
- >>>> i = 5 / (x-6) # no exception here
- >>>> j = i + 1 # nor here
- >>>> k = j + 5 # nor here
- >>>> untaint(int, k)
- TaintError
-
- In the above example, all of ``i``, ``j`` and ``k`` contain a Taint
- Bomb. Trying to untaint it raises an exception - a generic
- ``TaintError``. What we win is that the exception gives little away,
- and most importantly it occurs at the point where ``untaint()`` is
- called, not where the operation failed. This means that all calls to
- ``untaint()`` - but not the rest of the code - must be carefully
- reviewed for what occurs if they receive a Taint Bomb; they might catch
- the ``TaintError`` and give the user a generic message that something
- went wrong, if we are reasonably careful that the message or even its
- presence doesn't give information away. This might be a
- problem by itself, but there is no satisfying general solution here:
- it must be considered on a case-by-case basis. Again, what the
- Taint Object Space approach achieves is not solving these problems, but
- localizing them to well-defined small parts of the application - namely,
- around calls to ``untaint()``.
-
- The ``TaintError`` exception deliberately does not include any
- useful error messages, because they might give information away.
- Of course, this makes debugging quite a bit harder; a difficult
- problem to solve properly. So far we have implemented a way to peek in a Taint
- Box or Bomb, ``__pypy__._taint_look(x)``, and a "debug mode" that
- prints the exception as soon as a Bomb is created - both write
- information to the low-level stderr of the application, where we hope
- that it is unlikely to be seen by anyone but the application
- developer.
-
-
- Taint Atomic functions
- ----------------------
-
- Occasionally, a more complicated computation must be performed on a
- tainted object. This requires first untainting the object, performing the
- computations, and then carefully tainting the result again (including
- hiding all exceptions into Bombs).
-
- There is a built-in decorator that does this for you::
-
- >>>> @__pypy__.taint_atomic
- >>>> def myop(x, y):
- .... while x > 0:
- .... x -= y
- .... return x
- ....
- >>>> myop(42, 10)
- -8
- >>>> z = myop(taint(42), 10)
- >>>> z
- TaintError
- >>>> untaint(int, z)
- -8
-
- The decorator makes a whole function behave like a built-in operation.
- If no tainted argument is passed in, the function behaves normally. But
- if any of the arguments is tainted, it is automatically untainted - so
- the function body always sees untainted arguments - and the eventual
- result is tainted again (possibly in a Taint Bomb).
-
- It is important for the function marked as ``taint_atomic`` to have no
- visible side effects, as these could cause information leakage.
- This is currently not enforced, which means that all ``taint_atomic``
- functions have to be carefully reviewed for security (but not the
- callers of ``taint_atomic`` functions).
-
- A possible future extension would be to forbid side-effects on
- non-tainted objects from all ``taint_atomic`` functions.
-
- An example of usage: given a tainted object ``passwords_db`` that
- references a database of passwords, we can write a function
- that checks if a password is valid as follows::
-
- @taint_atomic
- def validate(passwords_db, username, password):
- assert type(passwords_db) is PasswordDatabase
- assert type(username) is str
- assert type(password) is str
- ...load username entry from passwords_db...
- return expected_password == password
-
- It returns a tainted boolean answer, or a Taint Bomb if something
- went wrong. A caller can do::
-
- ok = validate(passwords_db, 'john', '1234')
- ok = untaint(bool, ok)
-
- This can give three outcomes: ``True``, ``False``, or a ``TaintError``
- exception (with no information on it) if anything went wrong. If even
- this is considered giving too much information away, the ``False`` case
- can be made indistinguishable from the ``TaintError`` case (simply by
- raising an exception in ``validate()`` if the password is wrong).
-
- In the above example, the security results achieved are the following:
- as long as ``validate()`` does not leak information, no other part of
- the code can obtain more information about a passwords database than a
- Yes/No answer to a precise query.
-
- A possible extension of the ``taint_atomic`` decorator would be to check
- the argument types, as ``untaint()`` does, for the same reason: to
- prevent bugs where a function like ``validate()`` above is accidentally
- called with the wrong kind of tainted object, which would make it
- misbehave. For now, all ``taint_atomic`` functions should be
- conservative and carefully check all assumptions on their input
- arguments.
-
-
- .. _`taint-interface`:
-
- Interface
- ---------
-
- .. _`like a built-in operation`:
-
- The basic rule of the Tainted Object Space is that it introduces two new
- kinds of objects, Tainted Boxes and Tainted Bombs (which are not types
- in the Python sense). Each box internally contains a regular object;
- each bomb internally contains an exception object. An operation
- involving Tainted Boxes is performed on the objects contained in the
- boxes, and gives a Tainted Box or a Tainted Bomb as a result (such an
- operation does not let an exception be raised). An operation called
- with a Tainted Bomb argument immediately returns the same Tainted Bomb.
-
- In a PyPy running with (or translated with) the Taint Object Space,
- the ``__pypy__`` module exposes the following interface:
-
- * ``taint(obj)``
-
- Return a new Tainted Box wrapping ``obj``. Return ``obj`` itself
- if it is already tainted (a Box or a Bomb).
-
- * ``is_tainted(obj)``
-
- Check if ``obj`` is tainted (a Box or a Bomb).
-
- * ``untaint(type, obj)``
-
- Untaints ``obj`` if it is tainted. Raise ``TaintError`` if the type
- of the untainted object is not exactly ``type``, or if ``obj`` is a
- Bomb.
-
- * ``taint_atomic(func)``
-
- Return a wrapper function around the callable ``func``. The wrapper
- behaves `like a built-in operation`_ with respect to untainting the
- arguments, tainting the result, and returning a Bomb.
-
- * ``TaintError``
-
- Exception. On purpose, it provides no attribute or error message.
-
- * ``_taint_debug(level)``
-
- Set the debugging level to ``level`` (0=off). At level 1 or above,
- all Taint Bombs print a diagnostic message to stderr when they are
- created.
-
- * ``_taint_look(obj)``
-
- For debugging purposes: prints (to stderr) the type and address of
- the object in a Tainted Box, or prints the exception if ``obj`` is
- a Taint Bomb.
-
-
.. _dump:
The Dump Object Space
diff --git a/pypy/doc/project-ideas.rst b/pypy/doc/project-ideas.rst
--- a/pypy/doc/project-ideas.rst
+++ b/pypy/doc/project-ideas.rst
@@ -17,6 +17,12 @@
projects, or anything else in PyPy, pop up on IRC or write to us on the
`mailing list`_.
+Make big integers faster
+-------------------------
+
+PyPy's implementation of the Python ``long`` type is slower than CPython's.
+Find out why and optimize them.
+
Numpy improvements
------------------
@@ -48,17 +54,23 @@
.. image:: image/jitviewer.png
-We would like to add one level to this hierarchy, by showing the generated
-machine code for each jit operation. The necessary information is already in
-the log file produced by the JIT, so it is "only" a matter of teaching the
-jitviewer to display it. Ideally, the machine code should be hidden by
-default and viewable on request.
-
The jitviewer is a web application based on flask and jinja2 (and jQuery on
the client): if you have great web developing skills and want to help PyPy,
this is an ideal task to get started, because it does not require any deep
knowledge of the internals.
+Optimized Unicode Representation
+--------------------------------
+
+CPython 3.3 will use an `optimized unicode representation`_ which switches between
+different ways to represent a unicode string, depending on whether the string
+fits into ASCII, has only two-byte characters or needs four-byte characters.
+
+The actual details would be rather differen in PyPy, but we would like to have
+the same optimization implemented.
+
+.. _`optimized unicode representation`: http://www.python.org/dev/peps/pep-0393/
+
Translation Toolchain
---------------------
diff --git a/pypy/doc/release-1.6.0.rst b/pypy/doc/release-1.6.0.rst
new file mode 100644
--- /dev/null
+++ b/pypy/doc/release-1.6.0.rst
@@ -0,0 +1,95 @@
+========================
+PyPy 1.6 - kickass panda
+========================
+
+We're pleased to announce the 1.6 release of PyPy. This release brings a lot
+of bugfixes and performance improvements over 1.5, and improves support for
+Windows 32bit and OS X 64bit. This version fully implements Python 2.7.1 and
+has beta level support for loading CPython C extensions. You can download it
+here:
+
+ http://pypy.org/download.html
+
+What is PyPy?
+=============
+
+PyPy is a very compliant Python interpreter, almost a drop-in replacement for
+CPython 2.7.1. It's fast (`pypy 1.5 and cpython 2.6.2`_ performance comparison)
+due to its integrated tracing JIT compiler.
+
+This release supports x86 machines running Linux 32/64 or Mac OS X. Windows 32
+is beta (it roughly works but a lot of small issues have not been fixed so
+far). Windows 64 is not yet supported.
+
+The main topics of this release are speed and stability: on average on
+our benchmark suite, PyPy 1.6 is between **20% and 30%** faster than PyPy 1.5,
+which was already much faster than CPython on our set of benchmarks.
+
+The speed improvements have been made possible by optimizing many of the
+layers which compose PyPy. In particular, we improved: the Garbage Collector,
+the JIT warmup time, the optimizations performed by the JIT, the quality of
+the generated machine code and the implementation of our Python interpreter.
+
+.. _`pypy 1.5 and cpython 2.6.2`: http://speed.pypy.org
+
+
+Highlights
+==========
+
+* Numerous performance improvements, overall giving considerable speedups:
+
+ - better GC behavior when dealing with very large objects and arrays
+
+ - **fast ctypes:** now calls to ctypes functions are seen and optimized
+ by the JIT, and they are up to 60 times faster than PyPy 1.5 and 10 times
+ faster than CPython
+
+ - improved generators(1): simple generators now are inlined into the caller
+ loop, making performance up to 3.5 times faster than PyPy 1.5.
+
+ - improved generators(2): thanks to other optimizations, even generators
+ that are not inlined are between 10% and 20% faster than PyPy 1.5.
+
+ - faster warmup time for the JIT
+
+ - JIT support for single floats (e.g., for ``array('f')``)
+
+ - optimized dictionaries: the internal representation of dictionaries is now
+ dynamically selected depending on the type of stored objects, resulting in
+ faster code and smaller memory footprint. For example, dictionaries whose
+ keys are all strings, or all integers. Other dictionaries are also smaller
+ due to bugfixes.
+
+* JitViewer: this is the first official release which includes the JitViewer,
+ a web-based tool which helps you to see which parts of your Python code have
+ been compiled by the JIT, down until the assembler. The `jitviewer`_ 0.1 has
+ already been release and works well with PyPy 1.6.
+
+* The CPython extension module API has been improved and now supports many
+ more extensions. For information on which one are supported, please refer to
+ our `compatibility wiki`_.
+
+* Multibyte encoding support: this was of of the last areas in which we were
+ still behind CPython, but now we fully support them.
+
+* Preliminary support for NumPy: this release includes a preview of a very
+ fast NumPy module integrated with the PyPy JIT. Unfortunately, this does
+ not mean that you can expect to take an existing NumPy program and run it on
+ PyPy, because the module is still unfinished and supports only some of the
+ numpy API. However, barring some details, what works should be
+ blazingly fast :-)
+
+* Bugfixes: since the 1.5 release we fixed 53 bugs in our `bug tracker`_, not
+ counting the numerous bugs that were found and reported through other
+ channels than the bug tracker.
+
+Cheers,
+
+Hakan Ardo, Carl Friedrich Bolz, Laura Creighton, Antonio Cuni,
+Maciej Fijalkowski, Amaury Forgeot d'Arc, Alex Gaynor,
+Armin Rigo and the PyPy team
+
+.. _`jitviewer`: http://morepypy.blogspot.com/2011/08/visualization-of-jitted-code.html
+.. _`bug tracker`: https://bugs.pypy.org
+.. _`compatibility wiki`: https://bitbucket.org/pypy/compatibility/wiki/Home
+
diff --git a/pypy/doc/rlib.rst b/pypy/doc/rlib.rst
--- a/pypy/doc/rlib.rst
+++ b/pypy/doc/rlib.rst
@@ -134,69 +134,6 @@
a hierarchy of Address classes, in a typical static-OO-programming style.
-``rstack``
-==========
-
-The `pypy/rlib/rstack.py`_ module allows an RPython program to control its own execution stack.
-This is only useful if the program is translated using stackless. An old
-description of the exposed functions is below.
-
-We introduce an RPython type ``frame_stack_top`` and a built-in function
-``yield_current_frame_to_caller()`` that work as follows (see example below):
-
-* The built-in function ``yield_current_frame_to_caller()`` causes the current
- function's state to be captured in a new ``frame_stack_top`` object that is
- returned to the parent. Only one frame, the current one, is captured this
- way. The current frame is suspended and the caller continues to run. Note
- that the caller is only resumed once: when
- ``yield_current_frame_to_caller()`` is called. See below.
-
-* A ``frame_stack_top`` object can be jumped to by calling its ``switch()``
- method with no argument.
-
-* ``yield_current_frame_to_caller()`` and ``switch()`` themselves return a new
- ``frame_stack_top`` object: the freshly captured state of the caller of the
- source ``switch()`` that was just executed, or None in the case described
- below.
-
-* the function that called ``yield_current_frame_to_caller()`` also has a
- normal return statement, like all functions. This statement must return
- another ``frame_stack_top`` object. The latter is *not* returned to the
- original caller; there is no way to return several times to the caller.
- Instead, it designates the place to which the execution must jump, as if by
- a ``switch()``. The place to which we jump this way will see a None as the
- source frame stack top.
-
-* every frame stack top must be resumed once and only once. Not resuming
- it at all causes a leak. Resuming it several times causes a crash.
-
-* a function that called ``yield_current_frame_to_caller()`` should not raise.
- It would have no implicit parent frame to propagate the exception to. That
- would be a crashingly bad idea.
-
-The following example would print the numbers from 1 to 7 in order::
-
- def g():
- print 2
- frametop_before_5 = yield_current_frame_to_caller()
- print 4
- frametop_before_7 = frametop_before_5.switch()
- print 6
- return frametop_before_7
-
- def f():
- print 1
- frametop_before_4 = g()
- print 3
- frametop_before_6 = frametop_before_4.switch()
- print 5
- frametop_after_return = frametop_before_6.switch()
- print 7
- assert frametop_after_return is None
-
- f()
-
-
``streamio``
============
diff --git a/pypy/doc/stackless.rst b/pypy/doc/stackless.rst
--- a/pypy/doc/stackless.rst
+++ b/pypy/doc/stackless.rst
@@ -8,446 +8,312 @@
================
PyPy can expose to its user language features similar to the ones
-present in `Stackless Python`_: **no recursion depth limit**, and the
-ability to write code in a **massively concurrent style**. It actually
-exposes three different paradigms to choose from:
+present in `Stackless Python`_: the ability to write code in a
+**massively concurrent style**. (It does not (any more) offer the
+ability to run with no `recursion depth limit`_, but the same effect
+can be achieved indirectly.)
-* `Tasklets and channels`_;
+This feature is based on a custom primitive called a continulet_.
+Continulets can be directly used by application code, or it is possible
+to write (entirely at app-level) more user-friendly interfaces.
-* Greenlets_;
+Currently PyPy implements greenlets_ on top of continulets. It would be
+easy to implement tasklets and channels as well, emulating the model
+of `Stackless Python`_.
-* Plain coroutines_.
+Continulets are extremely light-weight, which means that PyPy should be
+able to handle programs containing large amounts of them. However, due
+to an implementation restriction, a PyPy compiled with
+``--gcrootfinder=shadowstack`` consumes at least one page of physical
+memory (4KB) per live continulet, and half a megabyte of virtual memory
+on 32-bit or a complete megabyte on 64-bit. Moreover, the feature is
+only available (so far) on x86 and x86-64 CPUs; for other CPUs you need
+to add a short page of custom assembler to
+`pypy/translator/c/src/stacklet/`_.
-All of them are extremely light-weight, which means that PyPy should be
-able to handle programs containing large amounts of coroutines, tasklets
-and greenlets.
+Theory
+======
-Requirements
-++++++++++++++++
+The fundamental idea is that, at any point in time, the program happens
+to run one stack of frames (or one per thread, in case of
+multi-threading). To see the stack, start at the top frame and follow
+the chain of ``f_back`` until you reach the bottom frame. From the
+point of view of one of these frames, it has a ``f_back`` pointing to
+another frame (unless it is the bottom frame), and it is itself being
+pointed to by another frame (unless it is the top frame).
-If you are running py.py on top of CPython, then you need to enable
-the _stackless module by running it as follows::
+The theory behind continulets is to literally take the previous sentence
+as definition of "an O.K. situation". The trick is that there are
+O.K. situations that are more complex than just one stack: you will
+always have one stack, but you can also have in addition one or more
+detached *cycles* of frames, such that by following the ``f_back`` chain
+you run in a circle. But note that these cycles are indeed completely
+detached: the top frame (the currently running one) is always the one
+which is not the ``f_back`` of anybody else, and it is always the top of
+a stack that ends with the bottom frame, never a part of these extra
+cycles.
- py.py --withmod-_stackless
+How do you create such cycles? The fundamental operation to do so is to
+take two frames and *permute* their ``f_back`` --- i.e. exchange them.
+You can permute any two ``f_back`` without breaking the rule of "an O.K.
+situation". Say for example that ``f`` is some frame halfway down the
+stack, and you permute its ``f_back`` with the ``f_back`` of the top
+frame. Then you have removed from the normal stack all intermediate
+frames, and turned them into one stand-alone cycle. By doing the same
+permutation again you restore the original situation.
-This is implemented internally using greenlets, so it only works on a
-platform where `greenlets`_ are supported. A few features do
-not work this way, though, and really require a translated
-``pypy-c``.
+In practice, in PyPy, you cannot change the ``f_back`` of an abitrary
+frame, but only of frames stored in ``continulets``.
-To obtain a translated version of ``pypy-c`` that includes Stackless
-support, run translate.py as follows::
-
- cd pypy/translator/goal
- python translate.py --stackless
+Continulets are internally implemented using stacklets_. Stacklets are a
+bit more primitive (they are really one-shot continuations), but that
+idea only works in C, not in Python. The basic idea of continulets is
+to have at any point in time a complete valid stack; this is important
+e.g. to correctly propagate exceptions (and it seems to give meaningful
+tracebacks too).
Application level interface
=============================
-A stackless PyPy contains a module called ``stackless``. The interface
-exposed by this module have not been refined much, so it should be
-considered in-flux (as of 2007).
-So far, PyPy does not provide support for ``stackless`` in a threaded
-environment. This limitation is not fundamental, as previous experience
-has shown, so supporting this would probably be reasonably easy.
+.. _continulet:
-An interesting point is that the same ``stackless`` module can provide
-a number of different concurrency paradigms at the same time. From a
-theoretical point of view, none of above-mentioned existing three
-paradigms considered on its own is new: two of them are from previous
-Python work, and the third one is a variant of the classical coroutine.
-The new part is that the PyPy implementation manages to provide all of
-them and let the user implement more. Moreover - and this might be an
-important theoretical contribution of this work - we manage to provide
-these concurrency concepts in a "composable" way. In other words, it
-is possible to naturally mix in a single application multiple
-concurrency paradigms, and multiple unrelated usages of the same
-paradigm. This is discussed in the Composability_ section below.
+Continulets
++++++++++++
+A translated PyPy contains by default a module called ``_continuation``
+exporting the type ``continulet``. A ``continulet`` object from this
+module is a container that stores a "one-shot continuation". It plays
+the role of an extra frame you can insert in the stack, and whose
+``f_back`` can be changed.
-Infinite recursion
-++++++++++++++++++
+To make a continulet object, call ``continulet()`` with a callable and
+optional extra arguments.
-Any stackless PyPy executable natively supports recursion that is only
-limited by the available memory. As in normal Python, though, there is
-an initial recursion limit (which is 5000 in all pypy-c's, and 1000 in
-CPython). It can be changed with ``sys.setrecursionlimit()``. With a
-stackless PyPy, any value is acceptable - use ``sys.maxint`` for
-unlimited.
+Later, the first time you ``switch()`` to the continulet, the callable
+is invoked with the same continulet object as the extra first argument.
+At that point, the one-shot continuation stored in the continulet points
+to the caller of ``switch()``. In other words you have a perfectly
+normal-looking stack of frames. But when ``switch()`` is called again,
+this stored one-shot continuation is exchanged with the current one; it
+means that the caller of ``switch()`` is suspended with its continuation
+stored in the container, and the old continuation from the continulet
+object is resumed.
-In some cases, you can write Python code that causes interpreter-level
-infinite recursion -- i.e. infinite recursion without going via
-application-level function calls. It is possible to limit that too,
-with ``_stackless.set_stack_depth_limit()``, or to unlimit it completely
-by setting it to ``sys.maxint``.
+The most primitive API is actually 'permute()', which just permutes the
+one-shot continuation stored in two (or more) continulets.
+In more details:
-Coroutines
-++++++++++
+* ``continulet(callable, *args, **kwds)``: make a new continulet.
+ Like a generator, this only creates it; the ``callable`` is only
+ actually called the first time it is switched to. It will be
+ called as follows::
-A Coroutine is similar to a very small thread, with no preemptive scheduling.
-Within a family of coroutines, the flow of execution is explicitly
-transferred from one to another by the programmer. When execution is
-transferred to a coroutine, it begins to execute some Python code. When
-it transfers execution away from itself it is temporarily suspended, and
-when execution returns to it it resumes its execution from the
-point where it was suspended. Conceptually, only one coroutine is
-actively running at any given time (but see Composability_ below).
+ callable(cont, *args, **kwds)
-The ``stackless.coroutine`` class is instantiated with no argument.
-It provides the following methods and attributes:
+ where ``cont`` is the same continulet object.
-* ``stackless.coroutine.getcurrent()``
+ Note that it is actually ``cont.__init__()`` that binds
+ the continulet. It is also possible to create a not-bound-yet
+ continulet by calling explicitly ``continulet.__new__()``, and
+ only bind it later by calling explicitly ``cont.__init__()``.
- Static method returning the currently running coroutine. There is a
- so-called "main" coroutine object that represents the "outer"
- execution context, where your main program started and where it runs
- as long as it does not switch to another coroutine.
+* ``cont.switch(value=None, to=None)``: start the continulet if
+ it was not started yet. Otherwise, store the current continuation
+ in ``cont``, and activate the target continuation, which is the
+ one that was previously stored in ``cont``. Note that the target
+ continuation was itself previously suspended by another call to
+ ``switch()``; this older ``switch()`` will now appear to return.
+ The ``value`` argument is any object that is carried to the target
+ and returned by the target's ``switch()``.
-* ``coro.bind(callable, *args, **kwds)``
+ If ``to`` is given, it must be another continulet object. In
+ that case, performs a "double switch": it switches as described
+ above to ``cont``, and then immediately switches again to ``to``.
+ This is different from switching directly to ``to``: the current
+ continuation gets stored in ``cont``, the old continuation from
+ ``cont`` gets stored in ``to``, and only then we resume the
+ execution from the old continuation out of ``to``.
- Bind the coroutine so that it will execute ``callable(*args,
- **kwds)``. The call is not performed immediately, but only the
- first time we call the ``coro.switch()`` method. A coroutine must
- be bound before it is switched to. When the coroutine finishes
- (because the call to the callable returns), the coroutine exits and
- implicitly switches back to another coroutine (its "parent"); after
- this point, it is possible to bind it again and switch to it again.
- (Which coroutine is the parent of which is not documented, as it is
- likely to change when the interface is refined.)
+* ``cont.throw(type, value=None, tb=None, to=None)``: similar to
+ ``switch()``, except that immediately after the switch is done, raise
+ the given exception in the target.
-* ``coro.switch()``
+* ``cont.is_pending()``: return True if the continulet is pending.
+ This is False when it is not initialized (because we called
+ ``__new__`` and not ``__init__``) or when it is finished (because
+ the ``callable()`` returned). When it is False, the continulet
+ object is empty and cannot be ``switch()``-ed to.
- Suspend the current (caller) coroutine, and resume execution in the
- target coroutine ``coro``.
+* ``permute(*continulets)``: a global function that permutes the
+ continuations stored in the given continulets arguments. Mostly
+ theoretical. In practice, using ``cont.switch()`` is easier and
+ more efficient than using ``permute()``; the latter does not on
+ its own change the currently running frame.
-* ``coro.kill()``
- Kill ``coro`` by sending a CoroutineExit exception and switching
- execution immediately to it. This exception can be caught in the
- coroutine itself and can be raised from any call to ``coro.switch()``.
- This exception isn't propagated to the parent coroutine.
+Genlets
++++++++
-* ``coro.throw(type, value)``
+The ``_continuation`` module also exposes the ``generator`` decorator::
- Insert an exception in ``coro`` an resume switches execution
- immediately to it. In the coroutine itself, this exception
- will come from any call to ``coro.switch()`` and can be caught. If the
- exception isn't caught, it will be propagated to the parent coroutine.
+ @generator
+ def f(cont, a, b):
+ cont.switch(a + b)
+ cont.switch(a + b + 1)
-When a coroutine is garbage-collected, it gets the ``.kill()`` method sent to
-it. This happens at the point the next ``.switch`` method is called, so the
-target coroutine of this call will be executed only after the ``.kill`` has
-finished.
+ for i in f(10, 20):
+ print i
-Example
-~~~~~~~
+This example prints 30 and 31. The only advantage over using regular
+generators is that the generator itself is not limited to ``yield``
+statements that must all occur syntactically in the same function.
+Instead, we can pass around ``cont``, e.g. to nested sub-functions, and
+call ``cont.switch(x)`` from there.
-Here is a classical producer/consumer example: an algorithm computes a
-sequence of values, while another consumes them. For our purposes we
-assume that the producer can generate several values at once, and the
-consumer can process up to 3 values in a batch - it can also process
-batches with fewer than 3 values without waiting for the producer (which
-would be messy to express with a classical Python generator). ::
+The ``generator`` decorator can also be applied to methods::
- def producer(lst):
- while True:
- ...compute some more values...
- lst.extend(new_values)
- coro_consumer.switch()
-
- def consumer(lst):
- while True:
- # First ask the producer for more values if needed
- while len(lst) == 0:
- coro_producer.switch()
- # Process the available values in a batch, but at most 3
- batch = lst[:3]
- del lst[:3]
- ...process batch...
-
- # Initialize two coroutines with a shared list as argument
- exchangelst = []
- coro_producer = coroutine()
- coro_producer.bind(producer, exchangelst)
- coro_consumer = coroutine()
- coro_consumer.bind(consumer, exchangelst)
-
- # Start running the consumer coroutine
- coro_consumer.switch()
-
-
-Tasklets and channels
-+++++++++++++++++++++
-
-The ``stackless`` module also provides an interface that is roughly
-compatible with the interface of the ``stackless`` module in `Stackless
-Python`_: it contains ``stackless.tasklet`` and ``stackless.channel``
-classes. Tasklets are also similar to microthreads, but (like coroutines)
-they don't actually run in parallel with other microthreads; instead,
-they synchronize and exchange data with each other over Channels, and
-these exchanges determine which Tasklet runs next.
-
-For usage reference, see the documentation on the `Stackless Python`_
-website.
-
-Note that Tasklets and Channels are implemented at application-level in
-`lib_pypy/stackless.py`_ on top of coroutines_. You can refer to this
-module for more details and API documentation.
-
-The stackless.py code tries to resemble the stackless C code as much
-as possible. This makes the code somewhat unpythonic.
-
-Bird's eye view of tasklets and channels
-~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
-
-Tasklets are a bit like threads: they encapsulate a function in such a way that
-they can be suspended/restarted any time. Unlike threads, they won't
-run concurrently, but must be cooperative. When using stackless
-features, it is vitally important that no action is performed that blocks
-everything else. In particular, blocking input/output should be centralized
-to a single tasklet.
-
-Communication between tasklets is done via channels.
-There are three ways for a tasklet to give up control:
-
-1. call ``stackless.schedule()``
-2. send something over a channel
-3. receive something from a channel
-
-A (live) tasklet can either be running, waiting to get scheduled, or be
-blocked by a channel.
-
-Scheduling is done in strictly round-robin manner. A blocked tasklet
-is removed from the scheduling queue and will be reinserted when it
-becomes unblocked.
-
-Example
-~~~~~~~
-
-Here is a many-producers many-consumers example, where any consumer can
-process the result of any producer. For this situation we set up a
-single channel where all producer send, and on which all consumers
-wait::
-
- def producer(chan):
- while True:
- chan.send(...next value...)
-
- def consumer(chan):
- while True:
- x = chan.receive()
- ...do something with x...
-
- # Set up the N producer and M consumer tasklets
- common_channel = stackless.channel()
- for i in range(N):
- stackless.tasklet(producer, common_channel)()
- for i in range(M):
- stackless.tasklet(consumer, common_channel)()
-
- # Run it all
- stackless.run()
-
-Each item sent over the channel is received by one of the waiting
-consumers; which one is not specified. The producers block until their
-item is consumed: the channel is not a queue, but rather a meeting point
-which causes tasklets to block until both a consumer and a producer are
-ready. In practice, the reason for having several consumers receiving
-on a single channel is that some of the consumers can be busy in other
-ways part of the time. For example, each consumer might receive a
-database request, process it, and send the result to a further channel
-before it asks for the next request. In this situation, further
-requests can still be received by other consumers.
+ class X:
+ @generator
+ def f(self, cont, a, b):
+ ...
Greenlets
+++++++++
-A Greenlet is a kind of primitive Tasklet with a lower-level interface
-and with exact control over the execution order. Greenlets are similar
-to Coroutines, with a slightly different interface: greenlets put more
-emphasis on a tree structure. The various greenlets of a program form a
-precise tree, which fully determines their order of execution.
+Greenlets are implemented on top of continulets in `lib_pypy/greenlet.py`_.
+See the official `documentation of the greenlets`_.
-For usage reference, see the `documentation of the greenlets`_.
-The PyPy interface is identical. You should use ``greenlet.greenlet``
-instead of ``stackless.greenlet`` directly, because the greenlet library
-can give you the latter when you ask for the former on top of PyPy.
+Note that unlike the CPython greenlets, this version does not suffer
+from GC issues: if the program "forgets" an unfinished greenlet, it will
+always be collected at the next garbage collection.
-PyPy's greenlets do not suffer from the cyclic GC limitation that the
-CPython greenlets have: greenlets referencing each other via local
-variables tend to leak on top of CPython (where it is mostly impossible
-to do the right thing). It works correctly on top of PyPy.
+Unimplemented features
+++++++++++++++++++++++
-Coroutine Pickling
-++++++++++++++++++
+The following features (present in some past Stackless version of PyPy)
+are for the time being not supported any more:
-Coroutines and tasklets can be pickled and unpickled, i.e. serialized to
-a string of bytes for the purpose of storage or transmission. This
-allows "live" coroutines or tasklets to be made persistent, moved to
-other machines, or cloned in any way. The standard ``pickle`` module
-works with coroutines and tasklets (at least in a translated ``pypy-c``;
-unpickling live coroutines or tasklets cannot be easily implemented on
-top of CPython).
+* Tasklets and channels (currently ``stackless.py`` seems to import,
+ but you have tasklets on top of coroutines on top of greenlets on
+ top of continulets on top of stacklets, and it's probably not too
+ hard to cut two of these levels by adapting ``stackless.py`` to
+ use directly continulets)
-To be able to achieve this result, we have to consider many objects that
-are not normally pickleable in CPython. Here again, the `Stackless
-Python`_ implementation has paved the way, and we follow the same
-general design decisions: simple internal objects like bound method
-objects and various kinds of iterators are supported; frame objects can
-be fully pickled and unpickled
-(by serializing a reference to the bytecode they are
-running in addition to all the local variables). References to globals
-and modules are pickled by name, similarly to references to functions
-and classes in the traditional CPython ``pickle``.
+* Coroutines (could be rewritten at app-level)
-The "magic" part of this process is the implementation of the unpickling
-of a chain of frames. The Python interpreter of PyPy uses
-interpreter-level recursion to represent application-level calls. The
-reason for this is that it tremendously simplifies the implementation of
-the interpreter itself. Indeed, in Python, almost any operation can
-potentially result in a non-tail-recursive call to another Python
-function. This makes writing a non-recursive interpreter extremely
-tedious; instead, we rely on lower-level transformations during the
-translation process to control this recursion. This is the `Stackless
-Transform`_, which is at the heart of PyPy's support for stackless-style
-concurrency.
+* Pickling and unpickling continulets (*)
-At any point in time, a chain of Python-level frames corresponds to a
-chain of interpreter-level frames (e.g. C frames in pypy-c), where each
-single Python-level frame corresponds to one or a few interpreter-level
-frames - depending on the length of the interpreter-level call chain
-from one bytecode evaluation loop to the next (recursively invoked) one.
+* Continuing execution of a continulet in a different thread (*)
-This means that it is not sufficient to simply create a chain of Python
-frame objects in the heap of a process before we can resume execution of
-these newly built frames. We must recreate a corresponding chain of
-interpreter-level frames. To this end, we have inserted a few *named
-resume points* (see 3.2.4, in `D07.1 Massive Parallelism and Translation Aspects`_) in the Python interpreter of PyPy. This is the
-motivation for implementing the interpreter-level primitives
-``resume_state_create()`` and ``resume_state_invoke()``, the powerful
-interface that allows an RPython program to artificially rebuild a chain
-of calls in a reflective way, completely from scratch, and jump to it.
+* Automatic unlimited stack (must be emulated__ so far)
-.. _`D07.1 Massive Parallelism and Translation Aspects`: http://codespeak.net/pypy/extradoc/eu-report/D07.1_Massive_Parallelism_and_Translation_Aspects-2007-02-28.pdf
+* Support for other CPUs than x86 and x86-64
-Example
-~~~~~~~
+.. __: `recursion depth limit`_
-(See `demo/pickle_coroutine.py`_ for the complete source of this demo.)
+(*) Pickling, as well as changing threads, could be implemented by using
+a "soft" stack switching mode again. We would get either "hard" or
+"soft" switches, similarly to Stackless Python 3rd version: you get a
+"hard" switch (like now) when the C stack contains non-trivial C frames
+to save, and a "soft" switch (like previously) when it contains only
+simple calls from Python to Python. Soft-switched continulets would
+also consume a bit less RAM, and the switch might be a bit faster too
+(unsure about that; what is the Stackless Python experience?).
-Consider a program which contains a part performing a long-running
-computation::
- def ackermann(x, y):
- if x == 0:
- return y + 1
- if y == 0:
- return ackermann(x - 1, 1)
- return ackermann(x - 1, ackermann(x, y - 1))
+Recursion depth limit
++++++++++++++++++++++
-By using pickling, we can save the state of the computation while it is
-running, for the purpose of restoring it later and continuing the
-computation at another time or on a different machine. However,
-pickling does not produce a whole-program dump: it can only pickle
-individual coroutines. This means that the computation should be
-started in its own coroutine::
+You can use continulets to emulate the infinite recursion depth present
+in Stackless Python and in stackless-enabled older versions of PyPy.
- # Make a coroutine that will run 'ackermann(3, 8)'
- coro = coroutine()
- coro.bind(ackermann, 3, 8)
+The trick is to start a continulet "early", i.e. when the recursion
+depth is very low, and switch to it "later", i.e. when the recursion
+depth is high. Example::
- # Now start running the coroutine
- result = coro.switch()
+ from _continuation import continulet
-The coroutine itself must switch back to the main program when it needs
-to be interrupted (we can only pickle suspended coroutines). Due to
-current limitations this requires an explicit check in the
-``ackermann()`` function::
+ def invoke(_, callable, arg):
+ return callable(arg)
- def ackermann(x, y):
- if interrupt_flag: # test a global flag
- main.switch() # and switch back to 'main' if it is set
- if x == 0:
- return y + 1
- if y == 0:
- return ackermann(x - 1, 1)
- return ackermann(x - 1, ackermann(x, y - 1))
+ def bootstrap(c):
+ # this loop runs forever, at a very low recursion depth
+ callable, arg = c.switch()
+ while True:
+ # start a new continulet from here, and switch to
+ # it using an "exchange", i.e. a switch with to=.
+ to = continulet(invoke, callable, arg)
+ callable, arg = c.switch(to=to)
-The global ``interrupt_flag`` would be set for example by a timeout, or
-by a signal handler reacting to Ctrl-C, etc. It causes the coroutine to
-transfer control back to the main program. The execution comes back
-just after the line ``coro.switch()``, where we can pickle the coroutine
-if necessary::
+ c = continulet(bootstrap)
+ c.switch()
- if not coro.is_alive:
- print "finished; the result is:", result
- else:
- # save the state of the suspended coroutine
- f = open('demo.pickle', 'w')
- pickle.dump(coro, f)
- f.close()
-The process can then stop. At any later time, or on another machine,
-we can reload the file and restart the coroutine with::
+ def recursive(n):
+ if n == 0:
+ return ("ok", n)
+ if n % 200 == 0:
+ prev = c.switch((recursive, n - 1))
+ else:
+ prev = recursive(n - 1)
+ return (prev[0], prev[1] + 1)
- f = open('demo.pickle', 'r')
- coro = pickle.load(f)
- f.close()
- result = coro.switch()
+ print recursive(999999) # prints ('ok', 999999)
-Limitations
-~~~~~~~~~~~
+Note that if you press Ctrl-C while running this example, the traceback
+will be built with *all* recursive() calls so far, even if this is more
+than the number that can possibly fit in the C stack. These frames are
+"overlapping" each other in the sense of the C stack; more precisely,
+they are copied out of and into the C stack as needed.
-Coroutine pickling is subject to some limitations. First of all, it is
-not a whole-program "memory dump". It means that only the "local" state
-of a coroutine is saved. The local state is defined to include the
-chain of calls and the local variables, but not for example the value of
-any global variable.
+(The example above also makes use of the following general "guideline"
+to help newcomers write continulets: in ``bootstrap(c)``, only call
+methods on ``c``, not on another continulet object. That's why we wrote
+``c.switch(to=to)`` and not ``to.switch()``, which would mess up the
+state. This is however just a guideline; in general we would recommend
+to use other interfaces like genlets and greenlets.)
-As in normal Python, the pickle will not include any function object's
-code, any class definition, etc., but only references to functions and
-classes. Unlike normal Python, the pickle contains frames. A pickled
-frame stores a bytecode index, representing the current execution
-position. This means that the user program cannot be modified *at all*
-between pickling and unpickling!
-On the other hand, the pickled data is fairly independent from the
-platform and from the PyPy version.
+Stacklets
++++++++++
-Pickling/unpickling fails if the coroutine is suspended in a state that
-involves Python frames which were *indirectly* called. To define this
-more precisely, a Python function can issue a regular function or method
-call to invoke another Python function - this is a *direct* call and can
-be pickled and unpickled. But there are many ways to invoke a Python
-function indirectly. For example, most operators can invoke a special
-method ``__xyz__()`` on a class, various built-in functions can call
-back Python functions, signals can invoke signal handlers, and so on.
-These cases are not supported yet.
+Continulets are internally implemented using stacklets, which is the
+generic RPython-level building block for "one-shot continuations". For
+more information about them please see the documentation in the C source
+at `pypy/translator/c/src/stacklet/stacklet.h`_.
+The module ``pypy.rlib.rstacklet`` is a thin wrapper around the above
+functions. The key point is that new() and switch() always return a
+fresh stacklet handle (or an empty one), and switch() additionally
+consumes one. It makes no sense to have code in which the returned
+handle is ignored, or used more than once. Note that ``stacklet.c`` is
+written assuming that the user knows that, and so no additional checking
+occurs; this can easily lead to obscure crashes if you don't use a
+wrapper like PyPy's '_continuation' module.
-Composability
-+++++++++++++
+
+Theory of composability
++++++++++++++++++++++++
Although the concept of coroutines is far from new, they have not been
generally integrated into mainstream languages, or only in limited form
(like generators in Python and iterators in C#). We can argue that a
possible reason for that is that they do not scale well when a program's
complexity increases: they look attractive in small examples, but the
-models that require explicit switching, by naming the target coroutine,
-do not compose naturally. This means that a program that uses
-coroutines for two unrelated purposes may run into conflicts caused by
-unexpected interactions.
+models that require explicit switching, for example by naming the target
+coroutine, do not compose naturally. This means that a program that
+uses coroutines for two unrelated purposes may run into conflicts caused
+by unexpected interactions.
To illustrate the problem, consider the following example (simplified
-code; see the full source in
-`pypy/module/_stackless/test/test_composable_coroutine.py`_). First, a
-simple usage of coroutine::
+code using a theorical ``coroutine`` class). First, a simple usage of
+coroutine::
main_coro = coroutine.getcurrent() # the main (outer) coroutine
data = []
@@ -530,74 +396,35 @@
main coroutine, which confuses the ``generator_iterator.next()`` method
(it gets resumed, but not as a result of a call to ``Yield()``).
-As part of trying to combine multiple different paradigms into a single
-application-level module, we have built a way to solve this problem.
-The idea is to avoid the notion of a single, global "main" coroutine (or
-a single main greenlet, or a single main tasklet). Instead, each
-conceptually separated user of one of these concurrency interfaces can
-create its own "view" on what the main coroutine/greenlet/tasklet is,
-which other coroutine/greenlet/tasklets there are, and which of these is
-the currently running one. Each "view" is orthogonal to the others. In
-particular, each view has one (and exactly one) "current"
-coroutine/greenlet/tasklet at any point in time. When the user switches
-to a coroutine/greenlet/tasklet, it implicitly means that he wants to
-switch away from the current coroutine/greenlet/tasklet *that belongs to
-the same view as the target*.
+Thus the notion of coroutine is *not composable*. By opposition, the
+primitive notion of continulets is composable: if you build two
+different interfaces on top of it, or have a program that uses twice the
+same interface in two parts, then assuming that both parts independently
+work, the composition of the two parts still works.
-The precise application-level interface has not been fixed yet; so far,
-"views" in the above sense are objects of the type
-``stackless.usercostate``. The above two examples can be rewritten in
-the following way::
+A full proof of that claim would require careful definitions, but let us
+just claim that this fact is true because of the following observation:
+the API of continulets is such that, when doing a ``switch()``, it
+requires the program to have some continulet to explicitly operate on.
+It shuffles the current continuation with the continuation stored in
+that continulet, but has no effect outside. So if a part of a program
+has a continulet object, and does not expose it as a global, then the
+rest of the program cannot accidentally influence the continuation
+stored in that continulet object.
- producer_view = stackless.usercostate() # a local view
- main_coro = producer_view.getcurrent() # the main (outer) coroutine
- ...
- producer_coro = producer_view.newcoroutine()
- ...
-
-and::
-
- generators_view = stackless.usercostate()
-
- def generator(f):
- def wrappedfunc(*args, **kwds):
- g = generators_view.newcoroutine(generator_iterator)
- ...
-
- ...generators_view.getcurrent()...
-
-Then the composition ``grab_values()`` works as expected, because the
-two views are independent. The coroutine captured as ``self.caller`` in
-the ``generator_iterator.next()`` method is the main coroutine of the
-``generators_view``. It is no longer the same object as the main
-coroutine of the ``producer_view``, so when ``data_producer()`` issues
-the following command::
-
- main_coro.switch()
-
-the control flow cannot accidentally jump back to
-``generator_iterator.next()``. In other words, from the point of view
-of ``producer_view``, the function ``grab_next_value()`` always runs in
-its main coroutine ``main_coro`` and the function ``data_producer`` in
-its coroutine ``producer_coro``. This is the case independently of
-which ``generators_view``-based coroutine is the current one when
-``grab_next_value()`` is called.
-
-Only code that has explicit access to the ``producer_view`` or its
-coroutine objects can perform switches that are relevant for the
-generator code. If the view object and the coroutine objects that share
-this view are all properly encapsulated inside the generator logic, no
-external code can accidentally temper with the expected control flow any
-longer.
-
-In conclusion: we will probably change the app-level interface of PyPy's
-stackless module in the future to not expose coroutines and greenlets at
-all, but only views. They are not much more difficult to use, and they
-scale automatically to larger programs.
+In other words, if we regard the continulet object as being essentially
+a modifiable ``f_back``, then it is just a link between the frame of
+``callable()`` and the parent frame --- and it cannot be arbitrarily
+changed by unrelated code, as long as they don't explicitly manipulate
+the continulet object. Typically, both the frame of ``callable()``
+(commonly a local function) and its parent frame (which is the frame
+that switched to it) belong to the same class or module; so from that
+point of view the continulet is a purely local link between two local
+frames. It doesn't make sense to have a concept that allows this link
+to be manipulated from outside.
.. _`Stackless Python`: http://www.stackless.com
.. _`documentation of the greenlets`: http://packages.python.org/greenlet/
-.. _`Stackless Transform`: translation.html#the-stackless-transform
.. include:: _ref.txt
diff --git a/pypy/doc/translation.rst b/pypy/doc/translation.rst
--- a/pypy/doc/translation.rst
+++ b/pypy/doc/translation.rst
@@ -552,14 +552,15 @@
The stackless transform converts functions into a form that knows how
to save the execution point and active variables into a heap structure
-and resume execution at that point. This is used to implement
+and resume execution at that point. This was used to implement
coroutines as an RPython-level feature, which in turn are used to
-implement `coroutines, greenlets and tasklets`_ as an application
+implement coroutines, greenlets and tasklets as an application
level feature for the Standard Interpreter.
-Enable the stackless transformation with :config:`translation.stackless`.
+The stackless transformation has been deprecated and is no longer
+available in trunk. It has been replaced with continulets_.
-.. _`coroutines, greenlets and tasklets`: stackless.html
+.. _continulets: stackless.html
.. _`preparing the graphs for source generation`:
diff --git a/pypy/doc/windows.rst b/pypy/doc/windows.rst
--- a/pypy/doc/windows.rst
+++ b/pypy/doc/windows.rst
@@ -32,6 +32,24 @@
modules that relies on third-party libraries. See below how to get
and build them.
+Preping Windows for the Large Build
+-----------------------------------
+
+Normally 32bit programs are limited to 2GB of memory on Windows. It is
+possible to raise this limit, to 3GB on Windows 32bit, and almost 4GB
+on Windows 64bit.
+
+On Windows 32bit, it is necessary to modify the system: follow
+http://usa.autodesk.com/adsk/servlet/ps/dl/item?siteID=123112&id=9583842&linkID=9240617
+to enable the "3GB" feature, and reboot. This step is not necessary on
+Windows 64bit.
+
+Then you need to execute::
+
+ editbin /largeaddressaware pypy.exe
+
+on the pypy.exe file you compiled.
+
Installing external packages
----------------------------
diff --git a/pypy/interpreter/argument.py b/pypy/interpreter/argument.py
--- a/pypy/interpreter/argument.py
+++ b/pypy/interpreter/argument.py
@@ -125,6 +125,7 @@
### Manipulation ###
+ @jit.look_inside_iff(lambda self: not self._dont_jit)
def unpack(self): # slowish
"Return a ([w1,w2...], {'kw':w3...}) pair."
kwds_w = {}
@@ -245,6 +246,8 @@
### Parsing for function calls ###
+ # XXX: this should be @jit.look_inside_iff, but we need key word arguments,
+ # and it doesn't support them for now.
def _match_signature(self, w_firstarg, scope_w, signature, defaults_w=None,
blindargs=0):
"""Parse args and kwargs according to the signature of a code object,
diff --git a/pypy/interpreter/astcompiler/ast.py b/pypy/interpreter/astcompiler/ast.py
--- a/pypy/interpreter/astcompiler/ast.py
+++ b/pypy/interpreter/astcompiler/ast.py
@@ -2,7 +2,7 @@
from pypy.interpreter.baseobjspace import Wrappable
from pypy.interpreter import typedef
from pypy.interpreter.gateway import interp2app
-from pypy.interpreter.error import OperationError
+from pypy.interpreter.error import OperationError, operationerrfmt
from pypy.rlib.unroll import unrolling_iterable
from pypy.tool.pairtype import extendabletype
from pypy.tool.sourcetools import func_with_new_name
@@ -2541,8 +2541,9 @@
class ASTVisitor(object):
def visit_sequence(self, seq):
- for node in seq:
- node.walkabout(self)
+ if seq is not None:
+ for node in seq:
+ node.walkabout(self)
def default_visitor(self, node):
raise NodeVisitorNotImplemented
@@ -2673,46 +2674,36 @@
class GenericASTVisitor(ASTVisitor):
def visit_Module(self, node):
- if node.body:
- self.visit_sequence(node.body)
+ self.visit_sequence(node.body)
def visit_Interactive(self, node):
- if node.body:
- self.visit_sequence(node.body)
+ self.visit_sequence(node.body)
def visit_Expression(self, node):
node.body.walkabout(self)
def visit_Suite(self, node):
- if node.body:
- self.visit_sequence(node.body)
+ self.visit_sequence(node.body)
def visit_FunctionDef(self, node):
node.args.walkabout(self)
- if node.body:
- self.visit_sequence(node.body)
- if node.decorator_list:
- self.visit_sequence(node.decorator_list)
+ self.visit_sequence(node.body)
+ self.visit_sequence(node.decorator_list)
def visit_ClassDef(self, node):
- if node.bases:
- self.visit_sequence(node.bases)
- if node.body:
- self.visit_sequence(node.body)
- if node.decorator_list:
- self.visit_sequence(node.decorator_list)
+ self.visit_sequence(node.bases)
+ self.visit_sequence(node.body)
+ self.visit_sequence(node.decorator_list)
def visit_Return(self, node):
if node.value:
node.value.walkabout(self)
def visit_Delete(self, node):
- if node.targets:
- self.visit_sequence(node.targets)
+ self.visit_sequence(node.targets)
def visit_Assign(self, node):
- if node.targets:
- self.visit_sequence(node.targets)
+ self.visit_sequence(node.targets)
node.value.walkabout(self)
def visit_AugAssign(self, node):
@@ -2722,37 +2713,29 @@
def visit_Print(self, node):
if node.dest:
node.dest.walkabout(self)
- if node.values:
- self.visit_sequence(node.values)
+ self.visit_sequence(node.values)
def visit_For(self, node):
node.target.walkabout(self)
node.iter.walkabout(self)
- if node.body:
- self.visit_sequence(node.body)
- if node.orelse:
- self.visit_sequence(node.orelse)
+ self.visit_sequence(node.body)
+ self.visit_sequence(node.orelse)
def visit_While(self, node):
node.test.walkabout(self)
- if node.body:
- self.visit_sequence(node.body)
- if node.orelse:
- self.visit_sequence(node.orelse)
+ self.visit_sequence(node.body)
+ self.visit_sequence(node.orelse)
def visit_If(self, node):
node.test.walkabout(self)
- if node.body:
- self.visit_sequence(node.body)
- if node.orelse:
- self.visit_sequence(node.orelse)
+ self.visit_sequence(node.body)
+ self.visit_sequence(node.orelse)
def visit_With(self, node):
node.context_expr.walkabout(self)
if node.optional_vars:
node.optional_vars.walkabout(self)
- if node.body:
- self.visit_sequence(node.body)
+ self.visit_sequence(node.body)
def visit_Raise(self, node):
if node.type:
@@ -2763,18 +2746,13 @@
node.tback.walkabout(self)
def visit_TryExcept(self, node):
- if node.body:
- self.visit_sequence(node.body)
- if node.handlers:
- self.visit_sequence(node.handlers)
- if node.orelse:
- self.visit_sequence(node.orelse)
+ self.visit_sequence(node.body)
+ self.visit_sequence(node.handlers)
+ self.visit_sequence(node.orelse)
def visit_TryFinally(self, node):
- if node.body:
- self.visit_sequence(node.body)
- if node.finalbody:
- self.visit_sequence(node.finalbody)
+ self.visit_sequence(node.body)
+ self.visit_sequence(node.finalbody)
def visit_Assert(self, node):
node.test.walkabout(self)
@@ -2782,12 +2760,10 @@
node.msg.walkabout(self)
def visit_Import(self, node):
- if node.names:
- self.visit_sequence(node.names)
+ self.visit_sequence(node.names)
def visit_ImportFrom(self, node):
- if node.names:
- self.visit_sequence(node.names)
+ self.visit_sequence(node.names)
def visit_Exec(self, node):
node.body.walkabout(self)
@@ -2812,8 +2788,7 @@
pass
def visit_BoolOp(self, node):
- if node.values:
- self.visit_sequence(node.values)
+ self.visit_sequence(node.values)
def visit_BinOp(self, node):
node.left.walkabout(self)
@@ -2832,35 +2807,28 @@
node.orelse.walkabout(self)
def visit_Dict(self, node):
- if node.keys:
- self.visit_sequence(node.keys)
- if node.values:
- self.visit_sequence(node.values)
+ self.visit_sequence(node.keys)
+ self.visit_sequence(node.values)
def visit_Set(self, node):
More information about the pypy-commit
mailing list