Skip to content

Commit b82f076

Browse files
serhiy-storchakaserhiy-storchaka
authored
gh-124130: Increase test coverage for \b and \B in regular expressions (GH-124330)
1 parent 536bc8a commit b82f076

File tree

1 file changed

+113
-7
lines changed

1 file changed

+113
-7
lines changed

‎Lib/test/test_re.py‎

Lines changed: 113 additions & 7 deletions
Original file line numberDiff line numberDiff line change
@@ -883,31 +883,137 @@ def test_named_unicode_escapes(self):
883883
self.checkPatternError(br'\N{LESS-THAN SIGN}', r'bad escape \N', 0)
884884
self.checkPatternError(br'[\N{LESS-THAN SIGN}]', r'bad escape \N', 1)
885885

886-
def test_string_boundaries(self):
886+
def test_word_boundaries(self):
887887
# See http://bugs.python.org/issue10713
888-
self.assertEqual(re.search(r"\b(abc)\b", "abc").group(1),
889-
"abc")
888+
self.assertEqual(re.search(r"\b(abc)\b", "abc").group(1), "abc")
889+
self.assertEqual(re.search(r"\b(abc)\b", "abc", re.ASCII).group(1), "abc")
890+
self.assertEqual(re.search(br"\b(abc)\b", b"abc").group(1), b"abc")
891+
self.assertEqual(re.search(br"\b(abc)\b", b"abc", re.LOCALE).group(1), b"abc")
892+
self.assertEqual(re.search(r"\b(ьюя)\b", "ьюя").group(1), "ьюя")
893+
self.assertIsNone(re.search(r"\b(ьюя)\b", "ьюя", re.ASCII))
894+
# There's a word boundary between a word and a non-word.
895+
self.assertTrue(re.match(r".\b", "a="))
896+
self.assertTrue(re.match(r".\b", "a=", re.ASCII))
897+
self.assertTrue(re.match(br".\b", b"a="))
898+
self.assertTrue(re.match(br".\b", b"a=", re.LOCALE))
899+
self.assertTrue(re.match(r".\b", "я="))
900+
self.assertIsNone(re.match(r".\b", "я=", re.ASCII))
901+
# There's a word boundary between a non-word and a word.
902+
self.assertTrue(re.match(r".\b", "=a"))
903+
self.assertTrue(re.match(r".\b", "=a", re.ASCII))
904+
self.assertTrue(re.match(br".\b", b"=a"))
905+
self.assertTrue(re.match(br".\b", b"=a", re.LOCALE))
906+
self.assertTrue(re.match(r".\b", "=я"))
907+
self.assertIsNone(re.match(r".\b", "=я", re.ASCII))
908+
# There is no word boundary inside a word.
909+
self.assertIsNone(re.match(r".\b", "ab"))
910+
self.assertIsNone(re.match(r".\b", "ab", re.ASCII))
911+
self.assertIsNone(re.match(br".\b", b"ab"))
912+
self.assertIsNone(re.match(br".\b", b"ab", re.LOCALE))
913+
self.assertIsNone(re.match(r".\b", "юя"))
914+
self.assertIsNone(re.match(r".\b", "юя", re.ASCII))
915+
# There is no word boundary between a non-word characters.
916+
self.assertIsNone(re.match(r".\b", "=-"))
917+
self.assertIsNone(re.match(r".\b", "=-", re.ASCII))
918+
self.assertIsNone(re.match(br".\b", b"=-"))
919+
self.assertIsNone(re.match(br".\b", b"=-", re.LOCALE))
920+
# There is no non-boundary match between a word and a non-word.
921+
self.assertIsNone(re.match(r".\B", "a="))
922+
self.assertIsNone(re.match(r".\B", "a=", re.ASCII))
923+
self.assertIsNone(re.match(br".\B", b"a="))
924+
self.assertIsNone(re.match(br".\B", b"a=", re.LOCALE))
925+
self.assertIsNone(re.match(r".\B", "я="))
926+
self.assertTrue(re.match(r".\B", "я=", re.ASCII))
927+
# There is no non-boundary match between a non-word and a word.
928+
self.assertIsNone(re.match(r".\B", "=a"))
929+
self.assertIsNone(re.match(r".\B", "=a", re.ASCII))
930+
self.assertIsNone(re.match(br".\B", b"=a"))
931+
self.assertIsNone(re.match(br".\B", b"=a", re.LOCALE))
932+
self.assertIsNone(re.match(r".\B", "=я"))
933+
self.assertTrue(re.match(r".\B", "=я", re.ASCII))
934+
# There's a non-boundary match inside a word.
935+
self.assertTrue(re.match(r".\B", "ab"))
936+
self.assertTrue(re.match(r".\B", "ab", re.ASCII))
937+
self.assertTrue(re.match(br".\B", b"ab"))
938+
self.assertTrue(re.match(br".\B", b"ab", re.LOCALE))
939+
self.assertTrue(re.match(r".\B", "юя"))
940+
self.assertTrue(re.match(r".\B", "юя", re.ASCII))
941+
# There's a non-boundary match between a non-word characters.
942+
self.assertTrue(re.match(r".\B", "=-"))
943+
self.assertTrue(re.match(r".\B", "=-", re.ASCII))
944+
self.assertTrue(re.match(br".\B", b"=-"))
945+
self.assertTrue(re.match(br".\B", b"=-", re.LOCALE))
890946
# There's a word boundary at the start of a string.
891947
self.assertTrue(re.match(r"\b", "abc"))
948+
self.assertTrue(re.match(r"\b", "abc", re.ASCII))
949+
self.assertTrue(re.match(br"\b", b"abc"))
950+
self.assertTrue(re.match(br"\b", b"abc", re.LOCALE))
951+
self.assertTrue(re.match(r"\b", "ьюя"))
952+
self.assertIsNone(re.match(r"\b", "ьюя", re.ASCII))
953+
# There's a word boundary at the end of a string.
954+
self.assertTrue(re.fullmatch(r".+\b", "abc"))
955+
self.assertTrue(re.fullmatch(r".+\b", "abc", re.ASCII))
956+
self.assertTrue(re.fullmatch(br".+\b", b"abc"))
957+
self.assertTrue(re.fullmatch(br".+\b", b"abc", re.LOCALE))
958+
self.assertTrue(re.fullmatch(r".+\b", "ьюя"))
959+
self.assertIsNone(re.search(r"\b", "ьюя", re.ASCII))
892960
# A non-empty string includes a non-boundary zero-length match.
893-
self.assertTrue(re.search(r"\B", "abc"))
961+
self.assertEqual(re.search(r"\B", "abc").span(), (1, 1))
962+
self.assertEqual(re.search(r"\B", "abc", re.ASCII).span(), (1, 1))
963+
self.assertEqual(re.search(br"\B", b"abc").span(), (1, 1))
964+
self.assertEqual(re.search(br"\B", b"abc", re.LOCALE).span(), (1, 1))
965+
self.assertEqual(re.search(r"\B", "ьюя").span(), (1, 1))
966+
self.assertEqual(re.search(r"\B", "ьюя", re.ASCII).span(), (0, 0))
894967
# There is no non-boundary match at the start of a string.
895-
self.assertFalse(re.match(r"\B", "abc"))
968+
self.assertIsNone(re.match(r"\B", "abc"))
969+
self.assertIsNone(re.match(r"\B", "abc", re.ASCII))
970+
self.assertIsNone(re.match(br"\B", b"abc"))
971+
self.assertIsNone(re.match(br"\B", b"abc", re.LOCALE))
972+
self.assertIsNone(re.match(r"\B", "ьюя"))
973+
self.assertTrue(re.match(r"\B", "ьюя", re.ASCII))
974+
# There is no non-boundary match at the end of a string.
975+
self.assertIsNone(re.fullmatch(r".+\B", "abc"))
976+
self.assertIsNone(re.fullmatch(r".+\B", "abc", re.ASCII))
977+
self.assertIsNone(re.fullmatch(br".+\B", b"abc"))
978+
self.assertIsNone(re.fullmatch(br".+\B", b"abc", re.LOCALE))
979+
self.assertIsNone(re.fullmatch(r".+\B", "ьюя"))
980+
self.assertTrue(re.fullmatch(r".+\B", "ьюя", re.ASCII))
896981
# However, an empty string contains no word boundaries, and also no
897982
# non-boundaries.
898-
self.assertIsNone(re.search(r"\B", ""))
983+
self.assertIsNone(re.search(r"\b", ""))
984+
self.assertIsNone(re.search(r"\b", "", re.ASCII))
985+
self.assertIsNone(re.search(br"\b", b""))
986+
self.assertIsNone(re.search(br"\b", b"", re.LOCALE))
899987
# This one is questionable and different from the perlre behaviour,
900988
# but describes current behavior.
901-
self.assertIsNone(re.search(r"\b", ""))
989+
self.assertIsNone(re.search(r"\B", ""))
990+
self.assertIsNone(re.search(r"\B", "", re.ASCII))
991+
self.assertIsNone(re.search(br"\B", b""))
992+
self.assertIsNone(re.search(br"\B", b"", re.LOCALE))
902993
# A single word-character string has two boundaries, but no
903994
# non-boundary gaps.
904995
self.assertEqual(len(re.findall(r"\b", "a")), 2)
996+
self.assertEqual(len(re.findall(r"\b", "a", re.ASCII)), 2)
997+
self.assertEqual(len(re.findall(br"\b", b"a")), 2)
998+
self.assertEqual(len(re.findall(br"\b", b"a", re.LOCALE)), 2)
905999
self.assertEqual(len(re.findall(r"\B", "a")), 0)
1000+
self.assertEqual(len(re.findall(r"\B", "a", re.ASCII)), 0)
1001+
self.assertEqual(len(re.findall(br"\B", b"a")), 0)
1002+
self.assertEqual(len(re.findall(br"\B", b"a", re.LOCALE)), 0)
9061003
# If there are no words, there are no boundaries
9071004
self.assertEqual(len(re.findall(r"\b", " ")), 0)
1005+
self.assertEqual(len(re.findall(r"\b", " ", re.ASCII)), 0)
1006+
self.assertEqual(len(re.findall(br"\b", b" ")), 0)
1007+
self.assertEqual(len(re.findall(br"\b", b" ", re.LOCALE)), 0)
9081008
self.assertEqual(len(re.findall(r"\b", " ")), 0)
1009+
self.assertEqual(len(re.findall(r"\b", " ", re.ASCII)), 0)
1010+
self.assertEqual(len(re.findall(br"\b", b" ")), 0)
1011+
self.assertEqual(len(re.findall(br"\b", b" ", re.LOCALE)), 0)
9091012
# Can match around the whitespace.
9101013
self.assertEqual(len(re.findall(r"\B", " ")), 2)
1014+
self.assertEqual(len(re.findall(r"\B", " ", re.ASCII)), 2)
1015+
self.assertEqual(len(re.findall(br"\B", b" ")), 2)
1016+
self.assertEqual(len(re.findall(br"\B", b" ", re.LOCALE)), 2)
9111017

9121018
def test_bigcharset(self):
9131019
self.assertEqual(re.match("([\u2222\u2223])",

0 commit comments

Comments
 (0)