[Python-checkins] r78729 - in python/branches/py3k: Lib/sre_parse.py Lib/test/test_re.py Misc/NEWS

[ezio.melotti]

Author: ezio.melotti
Date: Sat Mar  6 16:24:08 2010
New Revision: 78729

Log:
#6509: fix re.sub to work properly when the pattern, the string, and the replacement were all bytes. Patch by Antoine Pitrou.

Modified:
   python/branches/py3k/Lib/sre_parse.py
   python/branches/py3k/Lib/test/test_re.py
   python/branches/py3k/Misc/NEWS

Modified: python/branches/py3k/Lib/sre_parse.py
==============================================================================
--- python/branches/py3k/Lib/sre_parse.py	(original)
+++ python/branches/py3k/Lib/sre_parse.py	Sat Mar  6 16:24:08 2010
@@ -786,12 +786,18 @@
     groups = []
     groupsappend = groups.append
     literals = [None] * len(p)
+    if isinstance(source, str):
+        encode = lambda x: x
+    else:
+        # The tokenizer implicitly decodes bytes objects as latin-1, we must
+        # therefore re-encode the final representation.
+        encode = lambda x: x.encode('latin1')
     for c, s in p:
         if c is MARK:
             groupsappend((i, s))
             # literal[i] is already None
         else:
-            literals[i] = s
+            literals[i] = encode(s)
         i = i + 1
     return groups, literals
 

Modified: python/branches/py3k/Lib/test/test_re.py
==============================================================================
--- python/branches/py3k/Lib/test/test_re.py	(original)
+++ python/branches/py3k/Lib/test/test_re.py	Sat Mar  6 16:24:08 2010
@@ -717,6 +717,24 @@
         self.assertRaises(ValueError, re.compile, '(?a)\w', re.UNICODE)
         self.assertRaises(ValueError, re.compile, '(?au)\w')
 
+    def test_bug_6509(self):
+        # Replacement strings of both types must parse properly.
+        # all strings
+        pat = re.compile('a(\w)')
+        self.assertEqual(pat.sub('b\\1', 'ac'), 'bc')
+        pat = re.compile('a(.)')
+        self.assertEqual(pat.sub('b\\1', 'a\u1234'), 'b\u1234')
+        pat = re.compile('..')
+        self.assertEqual(pat.sub(lambda m: 'str', 'a5'), 'str')
+
+        # all bytes
+        pat = re.compile(b'a(\w)')
+        self.assertEqual(pat.sub(b'b\\1', b'ac'), b'bc')
+        pat = re.compile(b'a(.)')
+        self.assertEqual(pat.sub(b'b\\1', b'a\xCD'), b'b\xCD')
+        pat = re.compile(b'..')
+        self.assertEqual(pat.sub(lambda m: b'bytes', b'a5'), b'bytes')
+
     def test_dealloc(self):
         # issue 3299: check for segfault in debug build
         import _sre

Modified: python/branches/py3k/Misc/NEWS
==============================================================================
--- python/branches/py3k/Misc/NEWS	(original)
+++ python/branches/py3k/Misc/NEWS	Sat Mar  6 16:24:08 2010
@@ -268,6 +268,9 @@
 Library
 -------
 
+- Issue #6509: fix re.sub to work properly when the pattern, the string, and
+  the replacement were all bytes. Patch by Antoine Pitrou.
+
 - The sqlite3 module was updated to pysqlite 2.6.0. This fixes several obscure
   bugs and allows loading SQLite extensions from shared libraries.