gh-140797: Forbid capturing groups in re.Scanner lexicon patterns (GH-140944)

This commit is contained in:
Abhishek Tiwari 2025-11-04 16:24:28 +05:30 committed by GitHub
parent a84181c31b
commit fa9c3eefd4
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
3 changed files with 24 additions and 1 deletions

View file

@ -1639,6 +1639,24 @@ def s_int(scanner, token): return int(token)
(['sum', 'op=', 3, 'op*', 'foo', 'op+', 312.5,
'op+', 'bar'], ''))
def test_bug_gh140797(self):
# gh140797: Capturing groups are not allowed in re.Scanner
msg = r"Cannot use capturing groups in re\.Scanner"
# Capturing group throws an error
with self.assertRaisesRegex(ValueError, msg):
Scanner([("(a)b", None)])
# Named Group
with self.assertRaisesRegex(ValueError, msg):
Scanner([("(?P<name>a)", None)])
# Non-capturing groups should pass normally
s = Scanner([("(?:a)b", lambda scanner, token: token)])
result, rem = s.scan("ab")
self.assertEqual(result,['ab'])
self.assertEqual(rem,'')
def test_bug_448951(self):
# bug 448951 (similar to 429357, but with single char match)
# (Also test greedy matches.)