Skip to content

GH-79634: Accept path-like objects as pathlib glob patterns. #114017

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
barneygale merged 7 commits into from
Jan 20, 2024
Merged

GH-79634: Accept path-like objects as pathlib glob patterns. #114017

Show file tree
Hide file tree
Changes from 1 commit
Commits
Show all changes
7 commits
Select commit Hold shift + click to select a range
4e80705
GH-79634: Accept path-like objects as pathlib glob patterns.
barneygale Jan 12, 2024
b9e6b05
Merge branch 'main' into glob-pathlike-2
barneygale Jan 16, 2024
3eb53bc
Tiny speedup
barneygale Jan 19, 2024
980e4f3
Process pattern as a stack.
barneygale Jan 20, 2024
3e71a3e
Move validation into _pattern_stack
barneygale Jan 20, 2024
0c20f70
Simplify handling of '', '.' and '..' segments in patterns.
barneygale Jan 20, 2024
b500051
Fix tests
barneygale Jan 20, 2024
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
Prev Previous commit
Next Next commit
Simplify handling of '', '.' and '..' segments in patterns.
  • Loading branch information
@barneygale
barneygale committed Jan 20, 2024
commit 0c20f70a7e02c0b790409553cee58c12a5fff240
4 changes: 3 additions & 1 deletion Lib/pathlib/__init__.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -603,7 +603,7 @@ def iterdir(self):
def _scandir(self):
return os.scandir(self)

def _make_child_entry(self, entry, is_dir=False):
def _make_child_entry(self, entry):
# Transform an entry yielded from _scandir() into a path object.
path_str = entry.name if str(self) == '.' else entry.path
path = self.with_segments(path_str)
Expand All @@ -614,6 +614,8 @@ def _make_child_entry(self, entry, is_dir=False):
return path

def _make_child_relpath(self, name):
if not name:
return self
path_str = str(self)
tail = self._tail
if tail:
Expand Down
51 changes: 24 additions & 27 deletions Lib/pathlib/_abc.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -63,6 +63,12 @@ def _compile_pattern(pat, sep, case_sensitive):
return re.compile(regex, flags=flags).match


def _select_special(paths, part):
"""Yield special literal children of the given paths."""
for path in paths:
yield path._make_child_relpath(part)


def _select_children(parent_paths, dir_only, follow_symlinks, match):
"""Yield direct children of given paths, filtering by name and type."""
if follow_symlinks is None:
Expand All @@ -84,7 +90,7 @@ def _select_children(parent_paths, dir_only, follow_symlinks, match):
except OSError:
continue
if match(entry.name):
yield parent_path._make_child_entry(entry, dir_only)
yield parent_path._make_child_entry(entry)


def _select_recursive(parent_paths, dir_only, follow_symlinks):
Expand All @@ -107,7 +113,7 @@ def _select_recursive(parent_paths, dir_only, follow_symlinks):
for entry in entries:
try:
if entry.is_dir(follow_symlinks=follow_symlinks):
paths.append(path._make_child_entry(entry, dir_only))
paths.append(path._make_child_entry(entry))
continue
except OSError:
pass
Expand Down Expand Up @@ -721,10 +727,8 @@ def _scandir(self):
from contextlib import nullcontext
return nullcontext(self.iterdir())

def _make_child_entry(self, entry, is_dir=False):
def _make_child_entry(self, entry):
# Transform an entry yielded from _scandir() into a path object.
if is_dir:
return entry.joinpath('')
return entry

def _make_child_relpath(self, name):
Expand All @@ -740,39 +744,27 @@ def glob(self, pattern, *, case_sensitive=None, follow_symlinks=None):
# TODO: evaluate case-sensitivity of each directory in _select_children().
case_sensitive = _is_case_sensitive(self.pathmod)

# If symlinks are handled consistently, and the pattern does not
# contain '..' components, then we can use a 'walk-and-match' strategy
# when expanding '**' wildcards. When a '**' wildcard is encountered,
# all following pattern parts are immediately consumed and used to
# build a `re.Pattern` object. This pattern is used to filter the
# recursive walk. As a result, pattern parts following a '**' wildcard
# do not perform any filesystem access, which can be much faster!
stack = pattern._pattern_stack
filter_paths = follow_symlinks is not None and '..' not in stack
specials = ('', '.', '..')
filter_paths = False
deduplicate_paths = False
sep = self.pathmod.sep
paths = iter([self.joinpath('')] if self.is_dir() else [])
while stack:
part = stack.pop()
if part == '':
# Trailing slash.
pass
elif part == '..':
paths = (path._make_child_relpath('..') for path in paths)
if part in specials:
paths = _select_special(paths, part)
elif part == '**':
# Consume adjacent '**' components.
while stack and stack[-1] == '**':
stack.pop()

if filter_paths and stack and stack[-1] != '':
dir_only = stack[0] == ''
paths = _select_recursive(paths, dir_only, follow_symlinks)

# Filter out paths that don't match pattern.
prefix_len = len(str(self._make_child_relpath('_'))) - 1
match = _compile_pattern(str(pattern), sep, case_sensitive)
paths = (path for path in paths if match(str(path), prefix_len))
return paths
# Consume adjacent non-special components and enable post-walk
# regex filtering, provided we're treating symlinks consistently.
if follow_symlinks is not None:
while stack and stack[-1] not in specials:
filter_paths = True
stack.pop()

dir_only = bool(stack)
paths = _select_recursive(paths, dir_only, follow_symlinks)
Expand All @@ -786,6 +778,11 @@ def glob(self, pattern, *, case_sensitive=None, follow_symlinks=None):
dir_only = bool(stack)
match = _compile_pattern(part, sep, case_sensitive)
paths = _select_children(paths, dir_only, follow_symlinks, match)
if filter_paths:
# Filter out paths that don't match pattern.
prefix_len = len(str(self._make_child_relpath('_'))) - 1
match = _compile_pattern(str(pattern), sep, case_sensitive)
paths = (path for path in paths if match(str(path), prefix_len))
return paths

def rglob(self, pattern, *, case_sensitive=None, follow_symlinks=None):
Expand Down