Update app.py

app.py

@@ -311,19 +311,27 @@ def retrieve_rec_texts(
         print(f"Error in retrieve_rec_texts: {e}")
         return []
 
-def retrieve_metadata(document_indices: List[str], metadata_path: str = 'recipes_metadata.xlsx') -> Dict[str, Dict[str, str]]:
+def retrieve_metadata(document_indices: List[int], metadata_path: str = 'recipes_metadata.xlsx') -> Dict[int, Dict[str, str]]:
     try:
-        metadata_df = pd.read_excel(metadata_path)        
+        metadata_df = pd.read_excel(metadata_path)
         required_columns = {'id', 'original_file_name', 'url'}
         if not required_columns.issubset(metadata_df.columns):
-            raise ValueError(f"Metadata file must contain the following columns: {required_columns}")       
-        metadata_mapping = metadata_df.set_index('id')[['original_file_name', 'url']].to_dict('index')       
-        result = {doc_id: metadata_mapping.get(doc_id, {}) for doc_id in document_indices}
-        return result
+            raise ValueError(f"Metadata file must contain columns: {required_columns}")
+        metadata_df['id'] = metadata_df['id'].astype(int)  # Ensure 'id' is of type int
+        filtered_metadata = metadata_df[metadata_df['id'].isin(document_indices)]
+        metadata_dict = {
+            int(row['id']): {
+                "original_file_name": row['original_file_name'],
+                "url": row['url']
+            }
+            for _, row in filtered_metadata.iterrows()
+        }
+        return metadata_dict
     except Exception as e:
         print(f"Error retrieving metadata: {e}")
         return {}
 
+
 def retrieve_metadata(document_indices: List[str], metadata_path: str = 'recipes_metadata.xlsx') -> Dict[str, Dict[str, str]]:
     try:
         metadata_df = pd.read_excel(metadata_path)        
@@ -598,10 +606,7 @@ async def recipes_endpoint(profile: MedicalProfile):
         metadata = retrieve_metadata(document_indices, metadata_path=metadata_path)
         print(f"Retrieved Metadata: {metadata}")
         response = {           
-            "metadata": [
-                {"id": doc_id, "original_file_name": metadata.get(doc_id, {}).get("original_file_name"), "url": metadata.get(doc_id, {}).get("url")}
-                for doc_id in document_indices
-            ],
+            "metadata": metadata,
         }
         return response       
     except ValueError as ve: