Improve performance by skipping unnecessary normalisation (#3751)

This speeds up black by about 40% when the cache is full

diff --git a/CHANGES.md b/CHANGES.md
index 15027afbf0b24f3f33f96460d35f08bbcf2c7de2..93d8ee1921ac3e8c79a3c4e9cf8fc8c8bd1e8ab1 100644 (file)
--- a/CHANGES.md
+++ b/CHANGES.md
@@ -55,6 +55,7 @@
 
 <!-- Changes that improve Black's performance. -->
 
+- Speed up _Black_ significantly when the cache is full (#3751)
 - Avoid importing `IPython` in a case where we wouldn't need it (#3748)
 
 ### Output

diff --git a/src/black/files.py b/src/black/files.py
index 4e2209e557d2cc9c26d60db5d471c79e80a8f7da..ef6895ee3af5341c6e32b9ff0521426dd9107c59 100644 (file)
--- a/src/black/files.py
+++ b/src/black/files.py
@@ -276,15 +276,24 @@ def normalize_path_maybe_ignore(
     return root_relative_path
 
 
-def path_is_ignored(
-    path: Path, gitignore_dict: Dict[Path, PathSpec], report: Report
+def _path_is_ignored(
+    root_relative_path: str,
+    root: Path,
+    gitignore_dict: Dict[Path, PathSpec],
+    report: Report,
 ) -> bool:
+    path = root / root_relative_path
+    # Note that this logic is sensitive to the ordering of gitignore_dict. Callers must
+    # ensure that gitignore_dict is ordered from least specific to most specific.
     for gitignore_path, pattern in gitignore_dict.items():
-        relative_path = normalize_path_maybe_ignore(path, gitignore_path, report)
-        if relative_path is None:
+        try:
+            relative_path = path.relative_to(gitignore_path).as_posix()
+        except ValueError:
             break
         if pattern.match_file(relative_path):
-            report.path_ignored(path, "matches a .gitignore file content")
+            report.path_ignored(
+                path.relative_to(root), "matches a .gitignore file content"
+            )
             return True
     return False
 
@@ -326,7 +335,9 @@ def gen_python_files(
             continue
 
         # First ignore files matching .gitignore, if passed
-        if gitignore_dict and path_is_ignored(child, gitignore_dict, report):
+        if gitignore_dict and _path_is_ignored(
+            normalized_path, root, gitignore_dict, report
+        ):
             continue
 
         # Then ignore with `--exclude` `--extend-exclude` and `--force-exclude` options.

diff --git a/tests/test_black.py b/tests/test_black.py
index dd21d0a7ae6a00d7614e3c4a995ce96d9f7c44c0..3b3ab721c5f21d1671057e07fa7c0a1f43de18ac 100644 (file)
--- a/tests/test_black.py
+++ b/tests/test_black.py
@@ -508,6 +508,8 @@ class BlackTestCase(BlackBaseTestCase):
             "pathlib.Path.cwd", return_value=working_directory
         ), patch("pathlib.Path.is_dir", side_effect=mock_n_calls([True])):
             ctx = FakeContext()
+            # Note that the root folder (project_root) isn't the folder
+            # named "root" (aka working_directory)
             ctx.obj["root"] = project_root
             report = MagicMock(verbose=True)
             black.get_sources(
@@ -527,7 +529,7 @@ class BlackTestCase(BlackBaseTestCase):
             for _, mock_args, _ in report.path_ignored.mock_calls
         ), "A symbolic link was reported."
         report.path_ignored.assert_called_once_with(
-            Path("child", "b.py"), "matches a .gitignore file content"
+            Path("root", "child", "b.py"), "matches a .gitignore file content"
         )
 
     def test_report_verbose(self) -> None: