use gitignore file

README.rst

@@ -106,6 +106,13 @@ Exclude anything in the tests folder:
 
     lizard mySource/ -x"./tests/*"
 
+Use .gitignore file:
+
+::
+
+    lizard mySource/
+
+If there is a .gitignore file in the given path, lizard will automatically use it as an additional filter to exclude files that match the gitignore patterns. This is useful when you want to analyze only the tracked files in your git repository.
 
 Options
 ~~~~~~~

lizard.py

@@ -896,14 +896,43 @@ def md5_hash_file(full_path_name):
 def get_all_source_files(paths, exclude_patterns, lans):
     '''
     Function counts md5 hash for the given file and checks if it isn't a
-    duplicate using set of hashes for previous files '''
+    duplicate using set of hashes for previous files.
+    
+    If a .gitignore file is found in any of the given paths, it will be used
+    to filter out files that match the gitignore patterns.
+    '''
     hash_set = set()
+    gitignore_spec = None
+    base_path = None
+
+    def _load_gitignore():
+        nonlocal gitignore_spec, base_path
+        try:
+            import pathspec
+            for path in paths:
+                gitignore_path = os.path.join(path, '.gitignore')
+                if os.path.exists(gitignore_path):
+                    with open(gitignore_path, 'r') as gitignore_file:
+                        # Read lines and strip whitespace and empty lines
+                        patterns = [line.strip() for line in gitignore_file.readlines()]
+                        patterns = [p for p in patterns if p and not p.startswith('#')]
+                        gitignore_spec = pathspec.PathSpec.from_lines('gitwildmatch', patterns)
+                        base_path = path
+                    break
+        except ImportError:
+            pass
 
     def _support(reader):
         return not lans or set(lans).intersection(
             reader.language_names)
 
     def _validate_file(pathname):
+        if gitignore_spec is not None and base_path is not None:
+            rel_path = os.path.relpath(pathname, base_path)
+            # Normalize path separators for consistent matching
+            rel_path = rel_path.replace(os.sep, '/')
+            if gitignore_spec.match_file(rel_path):
+                return False
         return (
             pathname in paths or (
                 get_reader_for(pathname) and
@@ -926,6 +955,7 @@ def all_listed_files(paths):
                     for filename in files:
                         yield os.path.join(root, filename)
 
+    _load_gitignore()
     return filter(_validate_file, all_listed_files(paths))
 
 

setup.py

@@ -95,7 +95,7 @@ def run(self):
     packages=['lizard_ext', 'lizard_languages'],
     #data_files=[('lizard_ext', [])],
     py_modules=['lizard'],
-    install_requires=['pygments'],
+    install_requires=['pygments', 'pathspec'],
     entry_points={'console_scripts': ['lizard = lizard:main']},
     author='Terry Yin',
     author_email='[email protected]',

test/testFilesFilter.py

@@ -126,3 +126,34 @@ def test_fail_to_open_file_should_be_allowed(self, mock_open, mock_os_walk):
         else:
             file_names = ["./f1.cpp", "./f2.cpp"]
         self.assertEqual(file_names, list(files))
+
+    @patch.object(os.path, "exists")
+    @patch.object(os.path, "relpath")
+    @patch.object(os, "walk")
+    @patch("builtins.open", create=True)
+    def test_gitignore_filter(self, mock_open, mock_os_walk, mock_relpath, mock_exists):
+        mock_os_walk.return_value = (['.',
+                                    None,
+                                    ['temp.c', 'node_modules/file.js', 'useful.cpp']], )
+
+        def exists_side_effect(path):
+            return path.endswith('.gitignore')
+        mock_exists.side_effect = exists_side_effect
+
+        def relpath_side_effect(path, start):
+            # Return paths in a normalized format
+            if path.startswith('./'):
+                path = path[2:]
+            return path.replace(os.sep, '/')
+        mock_relpath.side_effect = relpath_side_effect
+
+        mock_file = mock_open.return_value.__enter__.return_value
+        mock_file.readlines.return_value = ["node_modules/\n", "*.c\n"]
+        mock_file.read.return_value = "node_modules/\n*.c\n"
+
+        files = get_all_source_files(["dir"], [], [])
+        if which_system() == "Windows":
+            file_names = [".\\useful.cpp"]
+        else:
+            file_names = ["./useful.cpp"]
+        self.assertEqual(file_names, list(files))