[Python-3000-checkins] r55224 - in python/branches/py3k-struni/Lib: encodings/punycode.py test/test_codecs.py

walter.doerwald python-3000-checkins at python.org
Thu May 10 14:36:30 CEST 2007


Author: walter.doerwald
Date: Thu May 10 14:36:25 2007
New Revision: 55224

Modified:
   python/branches/py3k-struni/Lib/encodings/punycode.py
   python/branches/py3k-struni/Lib/test/test_codecs.py
Log:
Fix punycode codec and tests.


Modified: python/branches/py3k-struni/Lib/encodings/punycode.py
==============================================================================
--- python/branches/py3k-struni/Lib/encodings/punycode.py	(original)
+++ python/branches/py3k-struni/Lib/encodings/punycode.py	Thu May 10 14:36:25 2007
@@ -10,15 +10,15 @@
 
 def segregate(str):
     """3.1 Basic code point segregation"""
-    base = []
-    extended = {}
+    base = b""
+    extended = set()
     for c in str:
         if ord(c) < 128:
-            base.append(c)
+            base.append(ord(c))
         else:
-            extended[c] = 1
-    extended = sorted(extended.keys())
-    return "".join(base).encode("ascii"),extended
+            extended.add(c)
+    extended = sorted(extended)
+    return (base, extended)
 
 def selective_len(str, max):
     """Return the length of str, considering only characters below max."""
@@ -75,10 +75,10 @@
     if res > 26: return 26
     return res
 
-digits = "abcdefghijklmnopqrstuvwxyz0123456789"
+digits = b"abcdefghijklmnopqrstuvwxyz0123456789"
 def generate_generalized_integer(N, bias):
     """3.3 Generalized variable-length integers"""
-    result = []
+    result = b""
     j = 0
     while 1:
         t = T(j, bias)
@@ -107,21 +107,20 @@
 def generate_integers(baselen, deltas):
     """3.4 Bias adaptation"""
     # Punycode parameters: initial bias = 72, damp = 700, skew = 38
-    result = []
+    result = b""
     bias = 72
     for points, delta in enumerate(deltas):
         s = generate_generalized_integer(delta, bias)
         result.extend(s)
         bias = adapt(delta, points==0, baselen+points+1)
-    return "".join(result)
+    return result
 
 def punycode_encode(text):
     base, extended = segregate(text)
-    base = base.encode("ascii")
     deltas = insertion_unsort(text, extended)
     extended = generate_integers(len(base), deltas)
     if base:
-        return base + "-" + extended
+        return base + b"-" + extended
     return extended
 
 ##################### Decoding #####################################
@@ -182,15 +181,13 @@
     return base
 
 def punycode_decode(text, errors):
-    pos = text.rfind("-")
+    pos = text.rfind(b"-")
     if pos == -1:
         base = ""
-        extended = text
+        extended = str(text, "ascii").upper()
     else:
-        base = text[:pos]
-        extended = text[pos+1:]
-    base = str(base, "ascii", errors)
-    extended = extended.upper()
+        base = str(text[:pos], "ascii", errors)
+        extended = str(text[pos+1:], "ascii").upper()
     return insertion_sort(base, extended, errors)
 
 ### Codec APIs

Modified: python/branches/py3k-struni/Lib/test/test_codecs.py
==============================================================================
--- python/branches/py3k-struni/Lib/test/test_codecs.py	(original)
+++ python/branches/py3k-struni/Lib/test/test_codecs.py	Thu May 10 14:36:25 2007
@@ -505,48 +505,48 @@
     # A Arabic (Egyptian):
     ("\u0644\u064A\u0647\u0645\u0627\u0628\u062A\u0643\u0644"
      "\u0645\u0648\u0634\u0639\u0631\u0628\u064A\u061F",
-     "egbpdaj6bu4bxfgehfvwxn"),
+     b"egbpdaj6bu4bxfgehfvwxn"),
     # B Chinese (simplified):
     ("\u4ED6\u4EEC\u4E3A\u4EC0\u4E48\u4E0D\u8BF4\u4E2D\u6587",
-     "ihqwcrb4cv8a8dqg056pqjye"),
+     b"ihqwcrb4cv8a8dqg056pqjye"),
     # C Chinese (traditional):
     ("\u4ED6\u5011\u7232\u4EC0\u9EBD\u4E0D\u8AAA\u4E2D\u6587",
-     "ihqwctvzc91f659drss3x8bo0yb"),
+     b"ihqwctvzc91f659drss3x8bo0yb"),
     # D Czech: Pro<ccaron>prost<ecaron>nemluv<iacute><ccaron>esky
     ("\u0050\u0072\u006F\u010D\u0070\u0072\u006F\u0073\u0074"
      "\u011B\u006E\u0065\u006D\u006C\u0075\u0076\u00ED\u010D"
      "\u0065\u0073\u006B\u0079",
-     "Proprostnemluvesky-uyb24dma41a"),
+     b"Proprostnemluvesky-uyb24dma41a"),
     # E Hebrew:
     ("\u05DC\u05DE\u05D4\u05D4\u05DD\u05E4\u05E9\u05D5\u05D8"
      "\u05DC\u05D0\u05DE\u05D3\u05D1\u05E8\u05D9\u05DD\u05E2"
      "\u05D1\u05E8\u05D9\u05EA",
-     "4dbcagdahymbxekheh6e0a7fei0b"),
+     b"4dbcagdahymbxekheh6e0a7fei0b"),
     # F Hindi (Devanagari):
     ("\u092F\u0939\u0932\u094B\u0917\u0939\u093F\u0928\u094D"
-    "\u0926\u0940\u0915\u094D\u092F\u094B\u0902\u0928\u0939"
-    "\u0940\u0902\u092C\u094B\u0932\u0938\u0915\u0924\u0947"
-    "\u0939\u0948\u0902",
-    "i1baa7eci9glrd9b2ae1bj0hfcgg6iyaf8o0a1dig0cd"),
+     "\u0926\u0940\u0915\u094D\u092F\u094B\u0902\u0928\u0939"
+     "\u0940\u0902\u092C\u094B\u0932\u0938\u0915\u0924\u0947"
+     "\u0939\u0948\u0902",
+     b"i1baa7eci9glrd9b2ae1bj0hfcgg6iyaf8o0a1dig0cd"),
 
     #(G) Japanese (kanji and hiragana):
     ("\u306A\u305C\u307F\u3093\u306A\u65E5\u672C\u8A9E\u3092"
-    "\u8A71\u3057\u3066\u304F\u308C\u306A\u3044\u306E\u304B",
-     "n8jok5ay5dzabd5bym9f0cm5685rrjetr6pdxa"),
+     "\u8A71\u3057\u3066\u304F\u308C\u306A\u3044\u306E\u304B",
+     b"n8jok5ay5dzabd5bym9f0cm5685rrjetr6pdxa"),
 
     # (H) Korean (Hangul syllables):
     ("\uC138\uACC4\uC758\uBAA8\uB4E0\uC0AC\uB78C\uB4E4\uC774"
      "\uD55C\uAD6D\uC5B4\uB97C\uC774\uD574\uD55C\uB2E4\uBA74"
      "\uC5BC\uB9C8\uB098\uC88B\uC744\uAE4C",
-     "989aomsvi5e83db1d2a355cv1e0vak1dwrv93d5xbh15a0dt30a5j"
-     "psd879ccm6fea98c"),
+     b"989aomsvi5e83db1d2a355cv1e0vak1dwrv93d5xbh15a0dt30a5j"
+     b"psd879ccm6fea98c"),
 
     # (I) Russian (Cyrillic):
     ("\u043F\u043E\u0447\u0435\u043C\u0443\u0436\u0435\u043E"
      "\u043D\u0438\u043D\u0435\u0433\u043E\u0432\u043E\u0440"
      "\u044F\u0442\u043F\u043E\u0440\u0443\u0441\u0441\u043A"
      "\u0438",
-     "b1abfaaepdrnnbgefbaDotcwatmq2g4l"),
+     b"b1abfaaepdrnnbgefbaDotcwatmq2g4l"),
 
     # (J) Spanish: Porqu<eacute>nopuedensimplementehablarenEspa<ntilde>ol
     ("\u0050\u006F\u0072\u0071\u0075\u00E9\u006E\u006F\u0070"
@@ -554,7 +554,7 @@
      "\u006C\u0065\u006D\u0065\u006E\u0074\u0065\u0068\u0061"
      "\u0062\u006C\u0061\u0072\u0065\u006E\u0045\u0073\u0070"
      "\u0061\u00F1\u006F\u006C",
-     "PorqunopuedensimplementehablarenEspaol-fmd56a"),
+     b"PorqunopuedensimplementehablarenEspaol-fmd56a"),
 
     # (K) Vietnamese:
     #  T<adotbelow>isaoh<odotbelow>kh<ocirc>ngth<ecirchookabove>ch\
@@ -563,45 +563,45 @@
      "\u0068\u00F4\u006E\u0067\u0074\u0068\u1EC3\u0063\u0068"
      "\u1EC9\u006E\u00F3\u0069\u0074\u0069\u1EBF\u006E\u0067"
      "\u0056\u0069\u1EC7\u0074",
-     "TisaohkhngthchnitingVit-kjcr8268qyxafd2f1b9g"),
+     b"TisaohkhngthchnitingVit-kjcr8268qyxafd2f1b9g"),
 
     #(L) 3<nen>B<gumi><kinpachi><sensei>
     ("\u0033\u5E74\u0042\u7D44\u91D1\u516B\u5148\u751F",
-     "3B-ww4c5e180e575a65lsy2b"),
+     b"3B-ww4c5e180e575a65lsy2b"),
 
     # (M) <amuro><namie>-with-SUPER-MONKEYS
     ("\u5B89\u5BA4\u5948\u7F8E\u6075\u002D\u0077\u0069\u0074"
      "\u0068\u002D\u0053\u0055\u0050\u0045\u0052\u002D\u004D"
      "\u004F\u004E\u004B\u0045\u0059\u0053",
-     "-with-SUPER-MONKEYS-pc58ag80a8qai00g7n9n"),
+     b"-with-SUPER-MONKEYS-pc58ag80a8qai00g7n9n"),
 
     # (N) Hello-Another-Way-<sorezore><no><basho>
     ("\u0048\u0065\u006C\u006C\u006F\u002D\u0041\u006E\u006F"
      "\u0074\u0068\u0065\u0072\u002D\u0057\u0061\u0079\u002D"
      "\u305D\u308C\u305E\u308C\u306E\u5834\u6240",
-     "Hello-Another-Way--fc4qua05auwb3674vfr0b"),
+     b"Hello-Another-Way--fc4qua05auwb3674vfr0b"),
 
     # (O) <hitotsu><yane><no><shita>2
     ("\u3072\u3068\u3064\u5C4B\u6839\u306E\u4E0B\u0032",
-     "2-u9tlzr9756bt3uc0v"),
+     b"2-u9tlzr9756bt3uc0v"),
 
     # (P) Maji<de>Koi<suru>5<byou><mae>
     ("\u004D\u0061\u006A\u0069\u3067\u004B\u006F\u0069\u3059"
      "\u308B\u0035\u79D2\u524D",
-     "MajiKoi5-783gue6qz075azm5e"),
+     b"MajiKoi5-783gue6qz075azm5e"),
 
      # (Q) <pafii>de<runba>
     ("\u30D1\u30D5\u30A3\u30FC\u0064\u0065\u30EB\u30F3\u30D0",
-     "de-jg4avhby1noc0d"),
+     b"de-jg4avhby1noc0d"),
 
     # (R) <sono><supiido><de>
     ("\u305D\u306E\u30B9\u30D4\u30FC\u30C9\u3067",
-     "d9juau41awczczp"),
+     b"d9juau41awczczp"),
 
     # (S) -> $1.00 <-
     ("\u002D\u003E\u0020\u0024\u0031\u002E\u0030\u0030\u0020"
      "\u003C\u002D",
-     "-> $1.00 <--")
+     b"-> $1.00 <--")
     ]
 
 for i in punycode_testcases:
@@ -616,7 +616,10 @@
             # code produces only lower case. Converting just puny to
             # lower is also insufficient, since some of the input characters
             # are upper case.
-            self.assertEquals(uni.encode("punycode").lower(), puny.lower())
+            self.assertEquals(
+                str(uni.encode("punycode"), "ascii").lower(),
+                str(puny, "ascii").lower()
+            )
 
     def test_decode(self):
         for uni, puny in punycode_testcases:


More information about the Python-3000-checkins mailing list