refactor: Create chunk_metadata dictionary to store doc_id key-value pairs and preserve original metadata of each data chunk

ldw129 · ldw129 · commit b65f548ec9d7 · 2025-06-24T17:19:11.000+08:00
diff --git a/embedder_service/routers/embed.py b/embedder_service/routers/embed.py
@@ -96,8 +96,7 @@ async def data_chunking(request:DataRequest, chunker) -> List[Dict[str, Any]]:
             'end_char': chunk_end,
             'page_number': None,
             'chunk_index': len(chunk_data),
-            # 'metadata': chunk_metadata # {"doc_id": request.doc_id}
-            'metadata': {"doc_id": request.doc_id}
+            'metadata': chunk_metadata
             })
 
             current_pos = chunk_end
