Update hf_utils.py

hf_utils.py

@@ -17,29 +17,28 @@ def download_filtered_space_files(space_id: str, local_dir: str = "repo_files",
 
     print(f"Downloading Space '{space_id}' and filtering for: {', '.join(file_extensions)}")
 
-    # Download the full snapshot to a temp directory
-    repo_path = snapshot_download(repo_id=space_id, repo_type="space")
-
     # Clear out local_dir if it already exists
     if os.path.exists(local_dir):
         shutil.rmtree(local_dir)
 
-    os.makedirs(local_dir, exist_ok=True)
+    # Convert file extensions to allow_patterns format (e.g., ['.py', '.md'] -> ['*.py', '*.md'])
+    allow_patterns = [f"*{ext}" for ext in file_extensions]
+    
+    # Download directly to local_dir with filtering during download
+    repo_path = snapshot_download(
+        repo_id=space_id,
+        repo_type="space",
+        local_dir=local_dir,
+        allow_patterns=allow_patterns
+    )
+    
+    # Count downloaded files for feedback
     copied_files = 0
-
-    # Walk through the snapshot and copy only files with desired extensions
-    for root, _, files in os.walk(repo_path):
+    for root, _, files in os.walk(local_dir):
         for file in files:
             if any(file.endswith(ext) for ext in file_extensions):
-                src_file = os.path.join(root, file)
-                rel_path = os.path.relpath(src_file, repo_path)
-                dest_file = os.path.join(local_dir, rel_path)
-                os.makedirs(os.path.dirname(dest_file), exist_ok=True)
-                
-                # Debug: Show exactly which file is being downloaded
-                print(f"DEBUG: Downloading file: {rel_path}")
-                
-                shutil.copy2(src_file, dest_file)
+                rel_path = os.path.relpath(os.path.join(root, file), local_dir)
+                print(f"DEBUG: Downloaded file: {rel_path}")
                 copied_files += 1
 
     print(f"Downloaded {copied_files} filtered file(s) to: {local_dir}")