Applied patch #725106, by Greg Chapman, fixing capturing groups

within repeats of alternatives. The only change to the original
patch was to convert the tests to the new test_re.py file.

This patch fixes cases like:

>>> re.match('((a)|b)*', 'abc').groups()
('b', '')

Which is wrong (it's impossible to match the empty string),
and incompatible with other regex systems, like the following
examples show:

% perl -e '"abc" =~ /^((a)|b)*/; print "$1 $2\n";'
b a

% echo "abc" | sed -r -e "s/^((a)|b)*/\1 \2|/"
b a|c

1 files changed, 19 insertions, 0 deletions

diff --git a/Lib/test/test_re.py b/Lib/test/test_re.py
index 2430790..7ba9a1b 100644
--- a/Lib/test/test_re.py
+++ b/Lib/test/test_re.py
@@ -276,6 +276,25 @@ class ReTests(unittest.TestCase):
             self.assertEqual(re.match(r'((.%s):)?z'%op, 'a:z').groups(),
                              ('a:', 'a'))
 
+    def test_bug_725106(self):
+        # capturing groups in alternatives in repeats
+        self.assertEqual(re.match('^((a)|b)*', 'abc').groups(),
+                         ('b', 'a'))
+        self.assertEqual(re.match('^(([ab])|c)*', 'abc').groups(),
+                         ('c', 'b'))
+        self.assertEqual(re.match('^((d)|[ab])*', 'abc').groups(),
+                         ('b', None))
+        self.assertEqual(re.match('^((a)c|[ab])*', 'abc').groups(),
+                         ('b', None))
+        self.assertEqual(re.match('^((a)|b)*?c', 'abc').groups(),
+                         ('b', 'a'))
+        self.assertEqual(re.match('^(([ab])|c)*?d', 'abcd').groups(),
+                         ('c', 'b'))
+        self.assertEqual(re.match('^((d)|[ab])*?c', 'abc').groups(),
+                         ('b', None))
+        self.assertEqual(re.match('^((a)c|[ab])*?c', 'abc').groups(),
+                         ('b', None))
+
     def test_finditer(self):
         iter = re.finditer(r":+", "a:b::c:::d")
         self.assertEqual([item.group(0) for item in iter],