From: Serhiy Storchaka <storchaka@gmail.com>
Date: Thu, 24 Oct 2013 19:02:42 +0000 (+0300)
Subject: Issue #19327: Fixed the working of regular expressions with too big charset.
X-Git-Tag: v2.7.6rc1~10
X-Git-Url: https://granicus.if.org/sourcecode?a=commitdiff_plain;h=22fb0dec30af4168c81744782a8bcc2453ac8055;p=python

Issue #19327: Fixed the working of regular expressions with too big charset.
---

diff --git a/Lib/sre_compile.py b/Lib/sre_compile.py
index 97c1663b0b..bd4070549a 100644
--- a/Lib/sre_compile.py
+++ b/Lib/sre_compile.py
@@ -343,7 +343,7 @@ def _optimize_unicode(charset, fixup):
     else:
         code = 'I'
     # Convert block indices to byte array of 256 bytes
-    mapping = array.array('b', mapping).tostring()
+    mapping = array.array('B', mapping).tostring()
     # Convert byte array to word array
     mapping = array.array(code, mapping)
     assert mapping.itemsize == _sre.CODESIZE
diff --git a/Lib/test/test_re.py b/Lib/test/test_re.py
index 8b277cfc02..d879bac5f6 100644
--- a/Lib/test/test_re.py
+++ b/Lib/test/test_re.py
@@ -427,6 +427,8 @@ class ReTests(unittest.TestCase):
                                   u"\u2222").group(1), u"\u2222")
         self.assertEqual(re.match(u"([\u2222\u2223])",
                                   u"\u2222", re.UNICODE).group(1), u"\u2222")
+        r = u'[%s]' % u''.join(map(unichr, range(256, 2**16, 255)))
+        self.assertEqual(re.match(r, u"\uff01", re.UNICODE).group(), u"\uff01")
 
     def test_big_codesize(self):
         # Issue #1160
diff --git a/Misc/NEWS b/Misc/NEWS
index 4e8dab8fc2..5dc2870042 100644
--- a/Misc/NEWS
+++ b/Misc/NEWS
@@ -40,6 +40,8 @@ Core and Builtins
 Library
 -------
 
+- Issue #19327: Fixed the working of regular expressions with too big charset.
+
 - Issue #19350: Increasing the test coverage of macurl2path. Patch by Colin
   Williams.