Merge branch 'master' into add-windows-runner

Author: mherrmann

.github/workflows/ci.yml

@@ -11,7 +11,7 @@ jobs:
     runs-on: ${{ matrix.os }}
     strategy:
       matrix:
-        python-version: ['3.12', '3.11', '3.10', '3.9', '3.8', '3.7']
+        python-version: ['3.13', '3.12', '3.11', '3.10', '3.9']
         os: [ubuntu-latest, windows-latest]
     steps:
     - uses: actions/checkout@v2

README.md

@@ -29,6 +29,16 @@ Then:
     >>> matches('/home/michael/project/__pycache__')
     True
 
+Alternatively, you can use the `parse_gitignore_str` function:
+
+    >>> from gitignore_parser import parse_gitignore_str
+    >>> matches = parse_gitignore_str(
+        '__pycache__/\n*.py[cod]', base_dir='/home/michael/project')
+    >>> matches('/home/michael/project/main.py')
+    False
+    >>> matches('/home/michael/project/main.pyc')
+    True
+
 ## Motivation
 
 I couldn't find a good library for doing the above on PyPI. There are

gitignore_parser.py

@@ -2,7 +2,7 @@
 import os
 import re
 
-from os.path import abspath, dirname
+from os.path import abspath, dirname, join
 from pathlib import Path
 import sys
 from typing import Reversible, Union
@@ -16,16 +16,22 @@ def handle_negation(file_path, rules: Reversible["IgnoreRule"]):
 def parse_gitignore(full_path, base_dir=None):
     if base_dir is None:
         base_dir = dirname(full_path)
-    rules = []
     with open(full_path) as ignore_file:
-        counter = 0
-        for line in ignore_file:
-            counter += 1
-            line = line.rstrip('\n')
-            rule = rule_from_pattern(line, base_path=Path(base_dir).resolve(),
-                                     source=(full_path, counter))
-            if rule:
-                rules.append(rule)
+        return _parse_gitignore_lines(ignore_file, full_path, base_dir)
+
+def parse_gitignore_str(gitignore_str, base_dir):
+    full_path = join(base_dir, '.gitignore')
+    lines = gitignore_str.splitlines()
+    return _parse_gitignore_lines(lines, full_path, base_dir)
+
+def _parse_gitignore_lines(lines, full_path, base_dir):
+    base_dir = _normalize_path(base_dir)
+    rules = []
+    for line_no, line in enumerate(lines, start=1):
+        rule = rule_from_pattern(
+            line.rstrip('\n'), base_path=base_dir, source=(full_path, line_no))
+        if rule:
+            rules.append(rule)
     if not any(r.negation for r in rules):
         return lambda file_path: any(r.match(file_path) for r in rules)
     else:
@@ -42,8 +48,6 @@ def rule_from_pattern(pattern, base_path=None, source=None):
     Because git allows for nested .gitignore files, a base_path value
     is required for correct behavior. The base path should be absolute.
     """
-    if base_path and base_path != Path(base_path).resolve():
-        raise ValueError('base_path must be absolute')
     # Store the exact pattern for our repr and string functions
     orig_pattern = pattern
     # Early returns follow
@@ -103,7 +107,7 @@ def rule_from_pattern(pattern, base_path=None, source=None):
         negation=negation,
         directory_only=directory_only,
         anchored=anchored,
-        base_path=_normalize_path(base_path) if base_path else None,
+        base_path=base_path if base_path else None,
         source=source
     )
 

setup.py

@@ -9,7 +9,7 @@
 description = 'A spec-compliant gitignore parser for Python 3.5+'
 setup(
     name='gitignore_parser',
-    version='0.1.9',
+    version='0.1.12',
     description=description,
     long_description=
         description + '\n\nhttps://github.com/mherrmann/gitignore_parser',
@@ -18,7 +18,7 @@
     url='https://github.com/mherrmann/gitignore_parser',
     py_modules=['gitignore_parser'],
     classifiers=[
-        'Development Status :: 3 - Alpha',
+        'Development Status :: 4 - Beta',
         'Intended Audience :: Developers',
 
         'License :: OSI Approved :: MIT License',
@@ -34,6 +34,7 @@
         'Programming Language :: Python :: 3.9',
         'Programming Language :: Python :: 3.10',
         'Programming Language :: Python :: 3.11',
+        'Programming Language :: Python :: 3.12',
 
         'Topic :: Software Development :: Libraries',
         'Topic :: Software Development :: Libraries :: Python Modules'

tests.py

@@ -2,25 +2,35 @@
 from pathlib import Path
 from tempfile import TemporaryDirectory
 
-from gitignore_parser import parse_gitignore
+from gitignore_parser import parse_gitignore, parse_gitignore_str
 
 from unittest import TestCase, main, SkipTest
 
 
 class Test(TestCase):
     def test_simple(self):
-        matches = _parse_gitignore_string(
+        matches = parse_gitignore_str(
             '__pycache__/\n'
             '*.py[cod]',
-            fake_base_dir='/home/michael'
+            base_dir='/home/michael'
         )
         self.assertFalse(matches('/home/michael/main.py'))
         self.assertTrue(matches('/home/michael/main.pyc'))
         self.assertTrue(matches('/home/michael/dir/main.pyc'))
         self.assertTrue(matches('/home/michael/__pycache__'))
 
+    def test_simple_parse_file(self):
+        with patch('builtins.open', mock_open(read_data=
+                                              '__pycache__/\n'
+                                              '*.py[cod]')):
+            matches = parse_gitignore('/home/michael/.gitignore')
+            self.assertFalse(matches('/home/michael/main.py'))
+            self.assertTrue(matches('/home/michael/main.pyc'))
+            self.assertTrue(matches('/home/michael/dir/main.pyc'))
+            self.assertTrue(matches('/home/michael/__pycache__'))
+
     def test_incomplete_filename(self):
-        matches = _parse_gitignore_string('o.py', fake_base_dir='/home/michael')
+        matches = parse_gitignore_str('o.py', base_dir='/home/michael')
         self.assertTrue(matches('/home/michael/o.py'))
         self.assertFalse(matches('/home/michael/foo.py'))
         self.assertFalse(matches('/home/michael/o.pyc'))
@@ -29,9 +39,9 @@ def test_incomplete_filename(self):
         self.assertFalse(matches('/home/michael/dir/o.pyc'))
 
     def test_wildcard(self):
-        matches = _parse_gitignore_string(
+        matches = parse_gitignore_str(
             'hello.*',
-            fake_base_dir='/home/michael'
+            base_dir='/home/michael'
         )
         self.assertTrue(matches('/home/michael/hello.txt'))
         self.assertTrue(matches('/home/michael/hello.foobar/'))
@@ -41,22 +51,22 @@ def test_wildcard(self):
         self.assertFalse(matches('/home/michael/helloX'))
 
     def test_anchored_wildcard(self):
-        matches = _parse_gitignore_string(
+        matches = parse_gitignore_str(
             '/hello.*',
-            fake_base_dir='/home/michael'
+            base_dir='/home/michael'
         )
         self.assertTrue(matches('/home/michael/hello.txt'))
         self.assertTrue(matches('/home/michael/hello.c'))
         self.assertFalse(matches('/home/michael/a/hello.java'))
 
     def test_trailingspaces(self):
-        matches = _parse_gitignore_string(
+        matches = parse_gitignore_str(
             'ignoretrailingspace \n'
             'notignoredspace\\ \n'
             'partiallyignoredspace\\  \n'
             'partiallyignoredspace2 \\  \n'
             'notignoredmultiplespace\\ \\ \\ ',
-            fake_base_dir='/home/michael'
+            base_dir='/home/michael'
         )
         self.assertTrue(matches('/home/michael/ignoretrailingspace'))
         self.assertFalse(matches('/home/michael/ignoretrailingspace '))
@@ -73,59 +83,66 @@ def test_trailingspaces(self):
         self.assertFalse(matches('/home/michael/notignoredmultiplespace'))
 
     def test_comment(self):
-        matches = _parse_gitignore_string(
+        matches = parse_gitignore_str(
                         'somematch\n'
                         '#realcomment\n'
                         'othermatch\n'
                         '\\#imnocomment',
-            fake_base_dir='/home/michael'
+            base_dir='/home/michael'
         )
         self.assertTrue(matches('/home/michael/somematch'))
         self.assertFalse(matches('/home/michael/#realcomment'))
         self.assertTrue(matches('/home/michael/othermatch'))
         self.assertTrue(matches('/home/michael/#imnocomment'))
 
     def test_ignore_directory(self):
-        matches = _parse_gitignore_string('.venv/', fake_base_dir='/home/michael')
+        matches = \
+            parse_gitignore_str('.venv/', base_dir='/home/michael')
         self.assertTrue(matches('/home/michael/.venv'))
         self.assertTrue(matches('/home/michael/.venv/folder'))
         self.assertTrue(matches('/home/michael/.venv/file.txt'))
         self.assertFalse(matches('/home/michael/.venv_other_folder'))
         self.assertFalse(matches('/home/michael/.venv_no_folder.py'))
 
     def test_ignore_directory_asterisk(self):
-        matches = _parse_gitignore_string('.venv/*', fake_base_dir='/home/michael')
+        matches = \
+            parse_gitignore_str('.venv/*', base_dir='/home/michael')
         self.assertFalse(matches('/home/michael/.venv'))
         self.assertTrue(matches('/home/michael/.venv/folder'))
         self.assertTrue(matches('/home/michael/.venv/file.txt'))
 
     def test_negation(self):
-        matches = _parse_gitignore_string(
+        matches = parse_gitignore_str(
             '''
 *.ignore
 !keep.ignore
             ''',
-            fake_base_dir='/home/michael'
+            base_dir='/home/michael'
         )
         self.assertTrue(matches('/home/michael/trash.ignore'))
         self.assertFalse(matches('/home/michael/keep.ignore'))
         self.assertTrue(matches('/home/michael/waste.ignore'))
 
     def test_literal_exclamation_mark(self):
-        matches = _parse_gitignore_string('\\!ignore_me!', fake_base_dir='/home/michael')
+        matches = parse_gitignore_str(
+            '\\!ignore_me!', base_dir='/home/michael'
+        )
         self.assertTrue(matches('/home/michael/!ignore_me!'))
         self.assertFalse(matches('/home/michael/ignore_me!'))
         self.assertFalse(matches('/home/michael/ignore_me'))
 
     def test_double_asterisks(self):
-        matches = _parse_gitignore_string('foo/**/Bar', fake_base_dir='/home/michael')
+        matches = parse_gitignore_str(
+            'foo/**/Bar', base_dir='/home/michael'
+        )
         self.assertTrue(matches('/home/michael/foo/hello/Bar'))
         self.assertTrue(matches('/home/michael/foo/world/Bar'))
         self.assertTrue(matches('/home/michael/foo/Bar'))
         self.assertFalse(matches('/home/michael/foo/BarBar'))
 
     def test_double_asterisk_without_slashes_handled_like_single_asterisk(self):
-        matches = _parse_gitignore_string('a/b**c/d', fake_base_dir='/home/michael')
+        matches = \
+            parse_gitignore_str('a/b**c/d', base_dir='/home/michael')
         self.assertTrue(matches('/home/michael/a/bc/d'))
         self.assertTrue(matches('/home/michael/a/bXc/d'))
         self.assertTrue(matches('/home/michael/a/bbc/d'))
@@ -136,42 +153,50 @@ def test_double_asterisk_without_slashes_handled_like_single_asterisk(self):
         self.assertFalse(matches('/home/michael/a/bb/XX/cc/d'))
 
     def test_more_asterisks_handled_like_single_asterisk(self):
-        matches = _parse_gitignore_string('***a/b', fake_base_dir='/home/michael')
+        matches = \
+            parse_gitignore_str('***a/b', base_dir='/home/michael')
         self.assertTrue(matches('/home/michael/XYZa/b'))
         self.assertFalse(matches('/home/michael/foo/a/b'))
-        matches = _parse_gitignore_string('a/b***', fake_base_dir='/home/michael')
+        matches = \
+            parse_gitignore_str('a/b***', base_dir='/home/michael')
         self.assertTrue(matches('/home/michael/a/bXYZ'))
         self.assertFalse(matches('/home/michael/a/b/foo'))
 
     def test_directory_only_negation(self):
-        matches = _parse_gitignore_string('''
+        matches = parse_gitignore_str('''
 data/**
 !data/**/
 !.gitkeep
 !data/01_raw/*
             ''',
-            fake_base_dir='/home/michael'
+            base_dir='/home/michael'
         )
         self.assertFalse(matches('/home/michael/data/01_raw/'))
         self.assertFalse(matches('/home/michael/data/01_raw/.gitkeep'))
         self.assertFalse(matches('/home/michael/data/01_raw/raw_file.csv'))
         self.assertFalse(matches('/home/michael/data/02_processed/'))
         self.assertFalse(matches('/home/michael/data/02_processed/.gitkeep'))
-        self.assertTrue(matches('/home/michael/data/02_processed/processed_file.csv'))
+        self.assertTrue(
+            matches('/home/michael/data/02_processed/processed_file.csv')
+        )
 
     def test_single_asterisk(self):
-        matches = _parse_gitignore_string('*', fake_base_dir='/home/michael')
+        matches = parse_gitignore_str('*', base_dir='/home/michael')
         self.assertTrue(matches('/home/michael/file.txt'))
         self.assertTrue(matches('/home/michael/directory'))
         self.assertTrue(matches('/home/michael/directory-trailing/'))
 
     def test_supports_path_type_argument(self):
-        matches = _parse_gitignore_string('file1\n!file2', fake_base_dir='/home/michael')
+        matches = parse_gitignore_str(
+            'file1\n!file2', base_dir='/home/michael'
+        )
         self.assertTrue(matches(Path('/home/michael/file1')))
         self.assertFalse(matches(Path('/home/michael/file2')))
 
     def test_slash_in_range_does_not_match_dirs(self):
-        matches = _parse_gitignore_string('abc[X-Z/]def', fake_base_dir='/home/michael')
+        matches = parse_gitignore_str(
+            'abc[X-Z/]def', base_dir='/home/michael'
+        )
         self.assertFalse(matches('/home/michael/abcdef'))
         self.assertTrue(matches('/home/michael/abcXdef'))
         self.assertTrue(matches('/home/michael/abcYdef'))
@@ -180,37 +205,43 @@ def test_slash_in_range_does_not_match_dirs(self):
         self.assertFalse(matches('/home/michael/abcXYZdef'))
 
     def test_symlink_to_another_directory(self):
-        """Test the behavior of a symlink to another directory.
-
-        The issue https://github.com/mherrmann/gitignore_parser/issues/29 describes how
-        a symlink to another directory caused an exception to be raised during matching.
-
-        This test ensures that the issue is now fixed.
-        """
-        with TemporaryDirectory() as project_dir, TemporaryDirectory() as another_dir:
+        with TemporaryDirectory() as project_dir:
+            project_dir = Path(project_dir).resolve()
+            with TemporaryDirectory() as another_dir:
+                another_dir = Path(another_dir).resolve()
+                matches = parse_gitignore_str('link', base_dir=project_dir)
+
+                # Create a symlink to another directory.
+                link = project_dir / 'link'
+                target = another_dir / 'target'
+                try:
+                    link.symlink_to(target)
+                except OSError:
+                    raise SkipTest(
+                        "Current user does not have permissions to perform symlink."
+                    )
+
+                # Check the intended behavior according to
+                # https://git-scm.com/docs/gitignore#_notes:
+                # Symbolic links are not followed and are matched as if they
+                # were regular files.
+                self.assertTrue(matches(link))
+
+    def test_symlink_to_symlink_directory(self):
+        with TemporaryDirectory() as project_dir:
             project_dir = Path(project_dir).resolve()
-            another_dir = Path(another_dir).resolve()
-            matches = _parse_gitignore_string('link', fake_base_dir=project_dir)
-
-            # Create a symlink to another directory.
-            link = project_dir / 'link'
-            target = another_dir / 'target'
-            
-            try:
-                link.symlink_to(target)
-            except OSError:
-                e = "Current user does not have permissions to perform symlink."
-                raise SkipTest(e)
-            # Check the intended behavior according to
-            # https://git-scm.com/docs/gitignore#_notes:
-            # Symbolic links are not followed and are matched as if they were regular
-            # files.
-            self.assertTrue(matches(link))
-
-def _parse_gitignore_string(data: str, fake_base_dir: str = None):
-    with patch('builtins.open', mock_open(read_data=data)):
-        success = parse_gitignore(f'{fake_base_dir}/.gitignore', fake_base_dir)
-        return success
+            with TemporaryDirectory() as link_dir:
+                link_dir = link_dir.resolve()
+                link = link_dir / 'link'
+                try:
+                    link.symlink_to(project_dir)
+                except OSError:
+                    raise SkipTest(
+                        "Current user does not have permissions to perform symlink."
+                    )
+                file = Path(link, 'file.txt')
+                matches = parse_gitignore_str('file.txt', base_dir=str(link_dir))
+                self.assertTrue(matches(file))
 
 if __name__ == '__main__':
     main()