#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.

This commit is contained in:
Ezio Melotti 2013-01-11 08:32:01 +02:00
parent 26ed234052
commit fe8e6e7414
4 changed files with 11 additions and 1 deletions

View File

@ -235,7 +235,7 @@ def _class_escape(source, escape):
if code:
return code
code = CATEGORIES.get(escape)
if code:
if code and code[0] == IN:
return code
try:
c = escape[1:2]

View File

@ -857,6 +857,12 @@ class ReTests(unittest.TestCase):
# 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'])
@bigmemtest(size=_2G, memuse=character_size)
def test_large_search(self, size):
# Issue #10182: indices were 32-bit-truncated.

View File

@ -65,6 +65,7 @@ Chris Barker
Anton Barkovsky
Nick Barnes
Quentin Barnes
Matthew Barnett
Richard Barran
Cesar Eduardo Barros
Des Barry

View File

@ -199,6 +199,9 @@ Core and Builtins
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 #15545: Fix regression in sqlite3's iterdump method where it was
failing if the connection used a row factory (such as sqlite3.Row) that
produced unsortable objects. (Regression was introduced by fix for 9750).