Merge branch 'main' of https://huggingface.co/spaces/Kalpokoch/ChatbotDemo

app/policy_vector_db.py

@@ -10,6 +10,7 @@ import logging
 logger = logging.getLogger("app")
 
 class PolicyVectorDB:
+    # ... (the class code is correct and does not need changes) ...
     def __init__(self, persist_directory: str, top_k_default: int = 5, relevance_threshold: float = 0.5):
         self.persist_directory = persist_directory
         self.client = chromadb.PersistentClient(path=persist_directory, settings=Settings(allow_reset=True))
@@ -42,7 +43,7 @@ class PolicyVectorDB:
         except Exception as e:
             logger.warning(f"Could not retrieve existing IDs from ChromaDB: {e}. Assuming no existing IDs for now.")
 
-        new_chunks = [chunk for chunk in chunks if chunk.get('id') and chunk['id'] not in existing_ids]
+        new_chunks = [chunk for chunk in chunks if chunk.get('id') and str(chunk.get('id')) not in existing_ids]
 
         if not new_chunks:
             logger.info("No new chunks to add.")
@@ -53,7 +54,7 @@ class PolicyVectorDB:
         for i in range(0, len(new_chunks), batch_size):
             batch = new_chunks[i:i + batch_size]
             texts = [chunk['text'] for chunk in batch]
-            ids = [chunk['id'] for chunk in batch]
+            ids = [str(chunk['id']) for chunk in batch]
             metadatas = [self._flatten_metadata(chunk['metadata']) if chunk.get('metadata') else {} for chunk in batch]
             embeddings = self.embedding_model.encode(texts, show_progress_bar=False).tolist()
             collection.add(ids=ids, embeddings=embeddings, documents=texts, metadatas=metadatas)
@@ -89,6 +90,7 @@ def ensure_db_populated(db_instance: PolicyVectorDB, chunks_file_path: str):
                 logger.error(f"Chunks file not found at {chunks_file_path}. Cannot populate DB.")
                 return False
             
+            # --- REVERTED: This is the correct method for a standard .json file ---
             with open(chunks_file_path, 'r', encoding='utf-8') as f:
                 chunks_to_add = json.load(f)