mirrors/cpython
1
0
Fork 0
mirror of https://github.com/python/cpython.git synced 2025-08-30 21:48:47 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions 7

gh-100061: Proper fix of the bug in the matching of possessive quantifiers (GH-102612)

Browse source 
Restore the global Input Stream pointer after trying to match a sub-pattern.

Co-authored-by: Ma Lin <animalize@users.noreply.github.com>
This commit is contained in:
SKO 2023-08-16 16:43:45 +09:00 • committed by GitHub
parent a86df298df
commit abd9cc52d9
No known key found for this signature in database
GPG key ID: 4AEE18F83AFDEB23
4 changed files with 17 additions and 10 deletions
Download patch file Download diff file Expand all files Collapse all files

7
Lib/re/_compiler.py
View file

@ -100,13 +100,6 @@ def _compile(code, pattern, flags):
emit(ANY_ALL) emit(ANY_ALL)
else: else:
emit(ANY) emit(ANY)
elif op is POSSESSIVE_REPEAT:
# gh-106052: Possessive quantifiers do not work when the
# subpattern contains backtracking, i.e. "(?:ab?c)*+".
# Implement it as equivalent greedy qualifier in atomic group.
p = [(MAX_REPEAT, av)]
p = [(ATOMIC_GROUP, p)]
_compile(code, p, flags)
elif op in REPEATING_CODES: elif op in REPEATING_CODES:
if _simple(av[2]): if _simple(av[2]):
emit(REPEATING_CODES[op][2]) emit(REPEATING_CODES[op][2])

14
Lib/test/test_re.py
View file

@ -2342,7 +2342,17 @@ class ReTests(unittest.TestCase):
self.assertTrue(re.fullmatch(r'(?s:(?>.*?\.).*)\Z', "a.txt")) # reproducer self.assertTrue(re.fullmatch(r'(?s:(?>.*?\.).*)\Z', "a.txt")) # reproducer
self.assertTrue(re.fullmatch(r'(?s:(?=(?P<g0>.*?\.))(?P=g0).*)\Z', "a.txt")) self.assertTrue(re.fullmatch(r'(?s:(?=(?P<g0>.*?\.))(?P=g0).*)\Z', "a.txt"))
def test_bug_gh106052(self): def test_bug_gh100061(self):
# gh-100061
self.assertEqual(re.match('(?>(?:.(?!D))+)', 'ABCDE').span(), (0, 2))
self.assertEqual(re.match('(?:.(?!D))++', 'ABCDE').span(), (0, 2))
self.assertEqual(re.match('(?>(?:.(?!D))*)', 'ABCDE').span(), (0, 2))
self.assertEqual(re.match('(?:.(?!D))*+', 'ABCDE').span(), (0, 2))
self.assertEqual(re.match('(?>(?:.(?!D))?)', 'CDE').span(), (0, 0))
self.assertEqual(re.match('(?:.(?!D))?+', 'CDE').span(), (0, 0))
self.assertEqual(re.match('(?>(?:.(?!D)){1,3})', 'ABCDE').span(), (0, 2))
self.assertEqual(re.match('(?:.(?!D)){1,3}+', 'ABCDE').span(), (0, 2))
# gh-106052
self.assertEqual(re.match("(?>(?:ab?c)+)", "aca").span(), (0, 2)) self.assertEqual(re.match("(?>(?:ab?c)+)", "aca").span(), (0, 2))
self.assertEqual(re.match("(?:ab?c)++", "aca").span(), (0, 2)) self.assertEqual(re.match("(?:ab?c)++", "aca").span(), (0, 2))
self.assertEqual(re.match("(?>(?:ab?c)*)", "aca").span(), (0, 2)) self.assertEqual(re.match("(?>(?:ab?c)*)", "aca").span(), (0, 2))
@ -2451,7 +2461,6 @@ ATOMIC_GROUP
17: SUCCESS 17: SUCCESS
''') ''')
@unittest.expectedFailure # gh-106052
def test_possesive_repeat_one(self): def test_possesive_repeat_one(self):
self.assertEqual(get_debug_out(r'a?+'), '''\ self.assertEqual(get_debug_out(r'a?+'), '''\
POSSESSIVE_REPEAT 0 1 POSSESSIVE_REPEAT 0 1
@ -2464,7 +2473,6 @@ POSSESSIVE_REPEAT 0 1
12: SUCCESS 12: SUCCESS
''') ''')
@unittest.expectedFailure # gh-106052
def test_possesive_repeat(self): def test_possesive_repeat(self):
self.assertEqual(get_debug_out(r'(?:ab)?+'), '''\ self.assertEqual(get_debug_out(r'(?:ab)?+'), '''\
POSSESSIVE_REPEAT 0 1 POSSESSIVE_REPEAT 0 1

2
Misc/NEWS.d/next/Library/2023-03-14-01-19-57.gh-issue-100061.CiXJYn.rst Normal file
View file

@ -0,0 +1,2 @@
Fix a bug that causes wrong matches for regular expressions with possessive
qualifier.

4
Modules/_sre/sre_lib.h
View file

@ -1336,6 +1336,10 @@ dispatch:
MARK_POP(ctx->lastmark); MARK_POP(ctx->lastmark);
LASTMARK_RESTORE(); LASTMARK_RESTORE();
/* Restore the global Input Stream pointer
since it can change after jumps. */
state->ptr = ptr;
/* We have sufficient matches, so exit loop. */ /* We have sufficient matches, so exit loop. */
break; break;
} }