Upload app.py

app.py

@@ -284,9 +284,7 @@ def build_generative_core():
     
     return GenerativeCore(model, tokenizer, gen_cfg)
 
-# ========================
 # 3) SOTA Enhanced Retrieval (EXACT SAME)
-# ========================
 
 class HybridCodeAssistant:
     """Main assistant class - EXACT SAME IMPLEMENTATION"""
@@ -475,45 +473,6 @@ Improved:"""
         
         thread.join()
 
-# 4) Gradio UI (Optimized for Hugging Face)
-
-ASSISTANT: Optional[HybridCodeAssistant] = None
-
-def initialize_assistant():
-    """Initialize assistant"""
-    global ASSISTANT
-    if ASSISTANT is None:
-        ASSISTANT = HybridCodeAssistant()
-        return "Codey Bryant 3.0 Ready! Ask your Python questions below."
-    else:
-        return "Assistant already initialized!"
-
-def chat(message: str, history):
-    """Chat function with proper history handling"""
-    if ASSISTANT is None:
-        # If not initialized, return error message
-        history.append([message, "Please click 'Initialize Assistant' first!"])
-        return history
-    
-    # Append user message to history
-    history.append([message, ""])
-    
-    try:
-        # Stream the response
-        response_text = ""
-        for token in ASSISTANT.answer_stream(message):
-            response_text += token
-            # Update the last message in history
-            history[-1] = [message, response_text]
-            yield history
-        
-    except Exception as e:
-        logger.error(f"Error generating response: {e}")
-        history[-1] = [message, f"Error: {str(e)}"]
-        yield history
-
-# 4) Gradio UI
-
 ASSISTANT: Optional[HybridCodeAssistant] = None
 
 def initialize_assistant():
@@ -551,79 +510,104 @@ def chat(message: str, history: list):
         history[-1][1] = f"Error: {str(e)}"
         yield history
 
-# 5) Main Entry Point
+# 5) Main Entry Point - SIMPLE WORKING UI
 
 if __name__ == "__main__":
     # Configure for Hugging Face Spaces
     server_name = os.environ.get("GRADIO_SERVER_NAME", "0.0.0.0")
     server_port = int(os.environ.get("GRADIO_SERVER_PORT", 7860))
     
-    # Create a simple, robust UI
+    # SIMPLE, WORKING UI
     with gr.Blocks(title="Codey Bryant 3.0") as demo:
         gr.Markdown("""
         # 🤖 Codey Bryant 3.0
-        ## SOTA RAG Python Coding Assistant
+        ## **SOTA RAG Coding Assistant**
         
-        **Features:** HyDE + Query Rewriting + Multi-Query + Answer-Space Retrieval
+        **Advanced Features:** HyDE + Query Rewriting + Multi-Query + Answer-Space Retrieval
         """)
         
         # Status display
-        status = gr.Markdown("### Status: Click 'Initialize Assistant' to start")
+        status_output = gr.Textbox(
+            label="Status",
+            value="Click 'Initialize Assistant' to start",
+            interactive=False
+        )
         
         # Initialize button
         init_btn = gr.Button("🚀 Initialize Assistant", variant="primary")
         
         # Chat interface
-        chatbot = gr.Chatbot(height=500)
+        chatbot = gr.Chatbot(label="Chat", height=500)
         
-        # Input and send
         with gr.Row():
             msg = gr.Textbox(
                 placeholder="Ask Python coding questions...",
                 label="Your Question",
+                lines=2,
                 scale=4
             )
             submit_btn = gr.Button("Send", variant="secondary", scale=1)
         
-        # Clear button
         clear_btn = gr.Button("Clear Chat")
         
         # Event handlers
         def on_init():
-            return initialize_assistant()
+            """Handle initialization and update status"""
+            status_text = ""
+            for status in initialize_assistant():
+                status_text = status
+                yield status
+            # Enable the chat interface after initialization
+            yield status_text
         
-        init_btn.click(on_init, outputs=status)
+        init_btn.click(
+            fn=on_init,
+            outputs=status_output
+        )
         
         def process_message(message, chat_history):
+            """Process a new message"""
+            if not message.strip():
+                return "", chat_history
+            
             # Add user message
             chat_history.append([message, ""])
             return "", chat_history
         
         def generate_response(message, chat_history):
-            # Generate assistant response
-            for updated_history in chat(message, chat_history):
-                yield updated_history
+            """Generate response from assistant"""
+            if not message.strip():
+                yield chat_history
+                return
+            
+            try:
+                # Get streaming response
+                for updated_history in chat(message, chat_history):
+                    yield updated_history
+            except Exception as e:
+                chat_history[-1][1] = f"Error: {str(e)}"
+                yield chat_history
         
         # Connect submit button
         submit_btn.click(
-            process_message,
-            [msg, chatbot],
-            [msg, chatbot]
+            fn=process_message,
+            inputs=[msg, chatbot],
+            outputs=[msg, chatbot]
         ).then(
-            generate_response,
-            [msg, chatbot],
-            chatbot
+            fn=generate_response,
+            inputs=[msg, chatbot],
+            outputs=chatbot
         )
         
         # Connect Enter key
         msg.submit(
-            process_message,
-            [msg, chatbot],
-            [msg, chatbot]
+            fn=process_message,
+            inputs=[msg, chatbot],
+            outputs=[msg, chatbot]
         ).then(
-            generate_response,
-            [msg, chatbot],
-            chatbot
+            fn=generate_response,
+            inputs=[msg, chatbot],
+            outputs=chatbot
         )
         
         # Clear chat
@@ -631,8 +615,11 @@ if __name__ == "__main__":
     
     # Launch the app
     logger.info(f"Starting Codey Bryant 3.0 on {server_name}:{server_port}")
+    logger.info("SOTA RAG Architecture: HyDE + Query Rewriting + Multi-Query + Answer-Space Retrieval")
+    
     demo.launch(
         server_name=server_name,
         server_port=server_port,
-        share=False
+        share=False,
+        debug=False
     )