Make _strptime.TimeRE().pattern() use ``\s+`` for matching whitespace instead
of ``\s*``. This prevents patterns from "stealing" bits from other patterns in
order to make a match work.
Closes bug #1730389. Will be backported.
diff --git a/Lib/_strptime.py b/Lib/_strptime.py
index 9e7823a..166cf82 100644
--- a/Lib/_strptime.py
+++ b/Lib/_strptime.py
@@ -250,7 +250,7 @@
regex_chars = re_compile(r"([\\.^$*+?\(\){}\[\]|])")
format = regex_chars.sub(r"\\\1", format)
whitespace_replacement = re_compile('\s+')
- format = whitespace_replacement.sub('\s*', format)
+ format = whitespace_replacement.sub('\s+', format)
while '%' in format:
directive_index = format.index('%')+1
processed_format = "%s%s%s" % (processed_format,
diff --git a/Lib/test/test_strptime.py b/Lib/test/test_strptime.py
index f474752..92c722a 100644
--- a/Lib/test/test_strptime.py
+++ b/Lib/test/test_strptime.py
@@ -190,6 +190,15 @@
"locale data that contains regex metacharacters is not"
" properly escaped")
+ def test_whitespace_substitution(self):
+ # When pattern contains whitespace, make sure it is taken into account
+ # so as to not allow to subpatterns to end up next to each other and
+ # "steal" characters from each other.
+ pattern = self.time_re.pattern('%j %H')
+ self.failUnless(not re.match(pattern, "180"))
+ self.failUnless(re.match(pattern, "18 0"))
+
+
class StrptimeTests(unittest.TestCase):
"""Tests for _strptime.strptime."""