Update app.py

app.py

@@ -133,9 +133,9 @@ def extract_with_llm(text):
 def extract_metadata_from_query(query: str):
     """Use spaCy + LLM to extract role/location/department from user query."""
     spacy_res = extract_with_spacy(query)
-    logging.info("spaCy results ## ==>%s", spacy_res)
+    print("spaCy results ## ==>%s", spacy_res)
     llm_res = extract_with_llm(query)
-    logging.info("LLM Extraction Results ## ==>%s", llm_res)
+    print("LLM Extraction Results ## ==>%s", llm_res)
 
 
     return {
@@ -187,31 +187,31 @@ def generate_rag_based_response(user_input, retriever, k=3, max_tokens=800, temp
     # Step 1: Extract personalization metadata from query
     query_metadata = extract_metadata_from_query(user_input)
 
-    logging.info("\n======================")
-    logging.info("User Query: %s", user_input)
-    logging.info("Extracted metadata from query: %s", query_metadata)  # Investigatory log
+    print("\n======================")
+    print("User Query: %s", user_input)
+    print("Extracted metadata from query: %s", query_metadata)  # Investigatory log
 
     # 2. Retrieve top-k docs semantically
     retrieved_docs = retriever.get_relevant_documents(user_input, k=k)
-    logging.info("Retrieved %d docs before filtering", len(retrieved_docs))
+    print("Retrieved %d docs before filtering", len(retrieved_docs))
 
     # 3. Apply metadata filtering
     filtered_docs = filter_docs_by_metadata(retrieved_docs, query_metadata)
     if filtered_docs:
         selected_docs = filtered_docs
-        logging.info("✅ %d docs kept after metadata filtering", len(selected_docs))
+        print("✅ %d docs kept after metadata filtering", len(selected_docs))
     else:
         selected_docs = retrieved_docs  # fallback if no metadata match
-        logging.info("⚠️ No metadata match, falling back to semantic retrieval only")
+        print("⚠️ No metadata match, falling back to semantic retrieval only")
 
 
 
      #  Step 4: Log retrieved docs metadata
-    logging.info("✅ Retrieved %d docs", len(selected_docs))
+    print("✅ Retrieved %d docs", len(selected_docs))
     for i, d in enumerate(selected_docs, 1):
-        logging.info("\n--- Chunk %d ---", i)
-        logging.info("Text: %s...", d.page_content[:200])  # preview first 200 chars
-        logging.info("Metadata: %s", d.metadata)
+        print("\n--- Chunk %d ---", i)
+        print("Text: %s...", d.page_content[:200])  # preview first 200 chars
+        print("Metadata: %s", d.metadata)
 
 
 
@@ -358,4 +358,3 @@ gr.ChatInterface(
 ).launch()
 
 
-