⚡️ Speed up method GoogleDriveIndexer.count_files_recursively by 29% (#593)

Author: aseembits93

CHANGELOG.md

@@ -1,3 +1,7 @@
+## 1.2.17-dev2
+
+* **Optimize `GoogleDriveIndexer.count_files_recursively`**
+
 ## 1.2.17-dev1
 
 * **Optimize `MilvusUploadStager.parse_date_string`**

unstructured_ingest/__version__.py

@@ -1 +1 @@
-__version__ = "1.2.17-dev1"  # pragma: no cover
+__version__ = "1.2.17-dev2"  # pragma: no cover

unstructured_ingest/processes/connectors/google_drive.py

@@ -189,6 +189,9 @@ def count_files_recursively(
         """
         count = 0
         stack = [folder_id]
+        # Pre-compute lower-case extension set for O(1) lookup
+        valid_exts = set(e.lower() for e in extensions) if extensions else None
+
         while stack:
             current_folder = stack.pop()
             # Always list all items under the current folder.
@@ -212,7 +215,6 @@ def count_files_recursively(
                         if extensions:
                             # Use a case-insensitive comparison for the file extension.
                             file_ext = (item.get("fileExtension") or "").lower()
-                            valid_exts = [e.lower() for e in extensions]
                             if file_ext in valid_exts:
                                 count += 1
                         else: