Update app.py

app.py

@@ -226,11 +226,7 @@ def rag_index():
 @app.route("/rag/query", methods=["POST"])
 def rag_query():
     """
-    Process user input with an in-depth approach:
-    - Dynamically extract brands, product types, and price limits from the query.
-    - Perform precise DB filtering based on extracted parameters.
-    - Construct a structured prompt for Gemini using the filtered results.
-    - Parse Gemini's response and update the conversation history.
+    Process user input with an in-depth approach.
     """
     if "rag_chat" not in session:
         session["rag_chat"] = []
@@ -239,23 +235,29 @@ def rag_query():
     if not user_input:
         return redirect(url_for("rag_index"))
 
+    # Add user query to chat history
     session["rag_chat"].append(("user", user_input))
 
+    # Extract and process the query
     brand_keyword, product_type, price_val = extract_query_parameters(user_input)
-
     matched_items = filter_database(brand_keyword, product_type, price_val)
-
     db_context = build_db_context(matched_items, brand_keyword, product_type, price_val)
-
     conversation_text = construct_prompt(session["rag_chat"], db_context)
-
+    
+    # Get response from Gemini API
     gemini_response = gemini_generate_content(
         api_key=GEMINI_API_KEY,
         conversation_text=conversation_text
     )
 
+    # Add assistant's response to chat history
     session["rag_chat"].append(("assistant", gemini_response))
-    return redirect(url_for("rag_index"))
+    
+    # Save session to persist the chat history
+    session.modified = True
+
+    # Render the chat page with updated history
+    return render_template("rag.html", chat_history=session["rag_chat"])
 
 def extract_query_parameters(user_query):
     """