[3.12] GH-106330: Fix matching of empty path in pathlib.PurePath.match() (GH-106331) (GH-106372)

We match paths using the `_lines` attribute, which is derived from the
path's string representation. The bug arises because an empty path's string
representation is `'.'` (not `''`), which is matched by the `'*'` wildcard.
(cherry picked from commit b4efdf8cda)
This commit is contained in:
Barney Gale 2023-07-04 21:51:36 +01:00 committed by GitHub
parent 930df7b07e
commit d5ed72b696
No known key found for this signature in database
GPG key ID: 4AEE18F83AFDEB23
3 changed files with 25 additions and 8 deletions

View file

@ -127,12 +127,19 @@ def _compile_pattern_lines(pattern_lines, case_sensitive):
# Match the start of the path, or just after a path separator # Match the start of the path, or just after a path separator
parts = ['^'] parts = ['^']
for part in pattern_lines.splitlines(keepends=True): for part in pattern_lines.splitlines(keepends=True):
# We slice off the common prefix and suffix added by translate() to if part == '*\n':
# ensure that re.DOTALL is not set, and the end of the string not part = r'.+\n'
# matched, respectively. With DOTALL not set, '*' wildcards will not elif part == '*':
# match path separators, because the '.' characters in the pattern part = r'.+'
# will not match newlines. else:
parts.append(fnmatch.translate(part)[_FNMATCH_SLICE]) # Any other component: pass to fnmatch.translate(). We slice off
# the common prefix and suffix added by translate() to ensure that
# re.DOTALL is not set, and the end of the string not matched,
# respectively. With DOTALL not set, '*' wildcards will not match
# path separators, because the '.' characters in the pattern will
# not match newlines.
part = fnmatch.translate(part)[_FNMATCH_SLICE]
parts.append(part)
# Match the end of the path, always. # Match the end of the path, always.
parts.append(r'\Z') parts.append(r'\Z')
flags = re.MULTILINE flags = re.MULTILINE
@ -501,8 +508,12 @@ class PurePath(object):
try: try:
return self._lines_cached return self._lines_cached
except AttributeError: except AttributeError:
path_str = str(self)
if path_str == '.':
self._lines_cached = ''
else:
trans = _SWAP_SEP_AND_NEWLINE[self._flavour.sep] trans = _SWAP_SEP_AND_NEWLINE[self._flavour.sep]
self._lines_cached = str(self).translate(trans) self._lines_cached = path_str.translate(trans)
return self._lines_cached return self._lines_cached
def __eq__(self, other): def __eq__(self, other):

View file

@ -317,6 +317,10 @@ class _BasePurePathTest(object):
self.assertTrue(P('A.py').match('a.PY', case_sensitive=False)) self.assertTrue(P('A.py').match('a.PY', case_sensitive=False))
self.assertFalse(P('c:/a/B.Py').match('C:/A/*.pY', case_sensitive=True)) self.assertFalse(P('c:/a/B.Py').match('C:/A/*.pY', case_sensitive=True))
self.assertTrue(P('/a/b/c.py').match('/A/*/*.Py', case_sensitive=False)) self.assertTrue(P('/a/b/c.py').match('/A/*/*.Py', case_sensitive=False))
# Matching against empty path
self.assertFalse(P().match('*'))
self.assertTrue(P().match('**'))
self.assertFalse(P().match('**/*'))
def test_ordering_common(self): def test_ordering_common(self):
# Ordering is tuple-alike. # Ordering is tuple-alike.

View file

@ -0,0 +1,2 @@
Fix incorrect matching of empty paths in :meth:`pathlib.PurePath.match`.
This bug was introduced in Python 3.12.0 beta 1.