feat: add ignore_hidden parameter to extract_from_dir

CHANGES.rst

@@ -1,6 +1,15 @@
 Babel Changelog
 ===============
 
+Version 2.19.0 (Unreleased)
+----------------------------
+
+Features
+~~~~~~~~
+
+* Messages: Add ``ignore_hidden`` parameter to ``extract_from_dir()`` to skip
+  files and directories whose names start with a dot by @2025sl93041-gkb in :gh:`471`
+
 Version 2.18.0
 --------------
 

babel/messages/extract.py

@@ -160,6 +160,7 @@ def extract_from_dir(
     callback: Callable[[str, str, dict[str, Any]], object] | None = None,
     strip_comment_tags: bool = False,
     directory_filter: Callable[[str], bool] | None = None,
+    ignore_hidden: bool = False,
 ) -> Generator[_FileExtractionResult, None, None]:
     """Extract messages from any source files found in the given directory.
 
@@ -228,6 +229,8 @@ def extract_from_dir(
     :param directory_filter: a callback to determine whether a directory should
                              be recursed into. Receives the full directory path;
                              should return True if the directory is valid.
+    :param ignore_hidden: if True, skip files and directories whose names start
+                          with a dot (e.g. ``.git``, ``.hidden.py``).
     :see: `pathmatch`
     """
     if dirname is None:
@@ -246,11 +249,15 @@ def extract_from_dir(
 
     for root, dirnames, filenames in os.walk(dirname):
         dirnames[:] = [
-            subdir for subdir in dirnames if directory_filter(os.path.join(root, subdir))
+            subdir for subdir in dirnames
+            if directory_filter(os.path.join(root, subdir))
+            and (not ignore_hidden or not subdir.startswith('.'))
         ]
         dirnames.sort()
         filenames.sort()
         for filename in filenames:
+            if ignore_hidden and filename.startswith('.'):
+                continue
             filepath = os.path.join(root, filename).replace(os.sep, '/')
 
             yield from check_and_call_extract_file(

tests/messages/test_extract.py

@@ -174,3 +174,35 @@ def test_issue_1195_2():
         'NOTE: This should still be considered, even if',
         'the text is far away',
     ]
+
+
+def test_extract_from_dir_ignore_hidden(tmp_path):
+    (tmp_path / "visible.py").write_text("_('visible')\n")
+    (tmp_path / ".hidden.py").write_text("_('hidden_file')\n")
+    hidden_dir = tmp_path / ".hiddendir"
+    hidden_dir.mkdir()
+    (hidden_dir / "inside.py").write_text("_('inside_hidden_dir')\n")
+
+    # ignore_hidden=False (default): hidden file extracted; hidden dir skipped
+    # by default directory_filter which already ignores dot dirs
+    results_default = list(extract.extract_from_dir(tmp_path))
+    msgids_default = [r[2] for r in results_default]
+    assert 'visible' in msgids_default
+    assert 'hidden_file' in msgids_default
+
+    # ignore_hidden=True: hidden files AND hidden dirs skipped
+    results_no_hidden = list(extract.extract_from_dir(tmp_path, ignore_hidden=True))
+    msgids_no_hidden = [r[2] for r in results_no_hidden]
+    assert 'visible' in msgids_no_hidden
+    assert 'hidden_file' not in msgids_no_hidden
+
+    # ignore_hidden=True also overrides a permissive directory_filter
+    results_override = list(extract.extract_from_dir(
+        tmp_path,
+        directory_filter=lambda p: True,
+        ignore_hidden=True,
+    ))
+    msgids_override = [r[2] for r in results_override]
+    assert 'visible' in msgids_override
+    assert 'hidden_file' not in msgids_override
+    assert 'inside_hidden_dir' not in msgids_override