Upload supabase_storage.py

supabase_storage.py

@@ -49,9 +49,9 @@ def is_supabase_available() -> bool:
 
 def _get_dataset_registry(client) -> List[str]:
     """
-    Get the list of dataset names from the registry file.
-    Returns None if there's an error reading (to prevent accidental overwrite).
-    Returns [] only if file doesn't exist yet.
+    DEPRECATED: Registry file is no longer used.
+    Datasets are now discovered by listing folders directly via list_datasets_from_supabase().
+    Keeping this function for backwards compatibility only.
     """
     try:
         storage = client.storage.from_(BUCKET_NAME)
@@ -71,7 +71,11 @@ def _get_dataset_registry(client) -> List[str]:
 
 
 def _update_dataset_registry(client, dataset_name: str, action: str = 'add'):
-    """Update the registry file with dataset names."""
+    """
+    DEPRECATED: Registry file is no longer used.
+    Datasets are now discovered by listing folders directly.
+    Keeping this function for backwards compatibility only.
+    """
     try:
         storage = client.storage.from_(BUCKET_NAME)
 
@@ -166,8 +170,7 @@ def save_dataset_to_supabase(
         )
         print(f"[Supabase] Uploaded {metadata_path}")
 
-        # 4. Update the registry file (list of all dataset names)
-        _update_dataset_registry(client, dataset_name, action='add')
+        # No registry update needed - datasets are discovered by listing folders directly
 
         print(f"[Supabase] Dataset '{dataset_name}' saved successfully")
         return True
@@ -223,7 +226,8 @@ def load_dataset_from_supabase(dataset_name: str) -> Optional[Dict[str, Any]]:
 
 def list_datasets_from_supabase() -> List[Dict[str, Any]]:
     """
-    List all datasets stored in Supabase.
+    List all datasets stored in Supabase by listing folders directly.
+    No registry file needed - just lists all folders in the bucket.
 
     Returns:
         List of dataset metadata dictionaries
@@ -236,28 +240,23 @@ def list_datasets_from_supabase() -> List[Dict[str, Any]]:
     try:
         storage = client.storage.from_(BUCKET_NAME)
 
-        # Get dataset names from registry
-        dataset_names = _get_dataset_registry(client)
-        print(f"[Supabase] Registry contains: {dataset_names}")
-
-        # If registry read failed (None), return empty to be safe
-        if dataset_names is None:
-            print("[Supabase] Could not read registry, returning empty list")
-            return []
-
-        # If registry is empty, try to find existing datasets by checking known names
-        # This handles the case where datasets were saved before registry was implemented
-        if not dataset_names:
-            print("[Supabase] Registry empty, checking for existing datasets...")
-            # Try some known/common dataset names
-            potential_names = ['testing']
-            for name in potential_names:
-                try:
-                    storage.download(f"{name}/metadata.json")
-                    dataset_names.append(name)
-                    print(f"[Supabase] Found existing dataset: {name}")
-                except Exception:
-                    pass
+        # List all items in the bucket root
+        items = storage.list()
+        print(f"[Supabase] Bucket contains {len(items)} items")
+
+        # Filter to get only folders (datasets), excluding files like _registry.json
+        # Folders have metadata.id = None or have no 'metadata' key with 'mimetype'
+        dataset_names = []
+        for item in items:
+            name = item.get('name', '')
+            # Skip hidden files and registry
+            if name.startswith('_') or name.startswith('.'):
+                continue
+            # Check if it's a folder (no metadata.mimetype means it's a folder)
+            item_metadata = item.get('metadata')
+            if item_metadata is None or item_metadata.get('mimetype') is None:
+                dataset_names.append(name)
+                print(f"[Supabase] Found dataset folder: {name}")
 
         datasets = []
         for folder_name in dataset_names:
@@ -316,8 +315,7 @@ def delete_dataset_from_supabase(dataset_name: str) -> bool:
             client.storage.from_(BUCKET_NAME).remove(file_paths)
             print(f"[Supabase] Deleted {len(file_paths)} files from '{dataset_name}'")
 
-        # Remove from registry
-        _update_dataset_registry(client, dataset_name, action='remove')
+        # No registry update needed - datasets are discovered by listing folders directly
 
         print(f"[Supabase] Dataset '{dataset_name}' deleted successfully")
         return True