[Python-checkins] cpython (2.7): #13899: \A, \Z, and \B now correctly match the A, Z, and B literals when used

[ezio.melotti]

http://hg.python.org/cpython/rev/2bc04449fd8c
changeset:   81387:2bc04449fd8c
branch:      2.7
parent:      81378:2a417ad8bfbf
user:        Ezio Melotti <ezio.melotti at gmail.com>
date:        Fri Jan 11 08:32:01 2013 +0200
summary:
  #13899: \A, \Z, and \B now correctly match the A, Z, and B literals when used inside character classes (e.g. [A]).  Patch by Matthew Barnett.

files:
  Lib/sre_parse.py    |  2 +-
  Lib/test/test_re.py |  6 ++++++
  Misc/ACKS           |  1 +
  Misc/NEWS           |  3 +++
  4 files changed, 11 insertions(+), 1 deletions(-)


diff --git a/Lib/sre_parse.py b/Lib/sre_parse.py
--- a/Lib/sre_parse.py
+++ b/Lib/sre_parse.py
@@ -228,7 +228,7 @@
     if code:
         return code
     code = CATEGORIES.get(escape)
-    if code:
+    if code and code[0] == IN:
         return code
     try:
         c = escape[1:2]
diff --git a/Lib/test/test_re.py b/Lib/test/test_re.py
--- a/Lib/test/test_re.py
+++ b/Lib/test/test_re.py
@@ -821,6 +821,12 @@
         # Test behaviour when not given a string or pattern as parameter
         self.assertRaises(TypeError, re.compile, 0)
 
+    def test_bug_13899(self):
+        # Issue #13899: re pattern r"[\A]" should work like "A" but matches
+        # nothing. Ditto B and Z.
+        self.assertEqual(re.findall(r'[\A\B\b\C\Z]', 'AB\bCZ'),
+                         ['A', 'B', '\b', 'C', 'Z'])
+
     @precisionbigmemtest(size=_2G, memuse=1)
     def test_large_search(self, size):
         # Issue #10182: indices were 32-bit-truncated.
diff --git a/Misc/ACKS b/Misc/ACKS
--- a/Misc/ACKS
+++ b/Misc/ACKS
@@ -58,6 +58,7 @@
 Anton Barkovsky
 Nick Barnes
 Quentin Barnes
+Matthew Barnett
 Richard Barran
 Cesar Eduardo Barros
 Des Barry
diff --git a/Misc/NEWS b/Misc/NEWS
--- a/Misc/NEWS
+++ b/Misc/NEWS
@@ -186,6 +186,9 @@
 Library
 -------
 
+- Issue #13899: \A, \Z, and \B now correctly match the A, Z, and B literals
+  when used inside character classes (e.g. '[\A]').  Patch by Matthew Barnett.
+
 - Issue #15109: Fix regression in sqlite3's iterdump method where it would
   die with an encoding error if the database contained string values
   containing non-ASCII.  (Regression was introduced by fix for 9750).

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