Update app.py

app.py

@@ -6,35 +6,32 @@ from huggingface_hub import InferenceClient
 model_id = "Qwen/Qwen2.5-Coder-32B-Instruct"
 
 def respond(message, history, system_message, temperature, request: gr.Request):
+    # --- 1. Authentication Logic ---
     token = None
-    
-    # 1. Safely try to get the user's token (works on HF Spaces if logged in)
+    # Safely access token (handles both Local run and Spaces)
     if request:
-        # We use getattr() to avoid the "AttributeError" if running locally
         token = getattr(request, "token", None)
     
-    # 2. If no user token, fall back to the Space's secret token (HF_TOKEN)
     if token is None:
         token = os.getenv("HF_TOKEN")
 
-    # 3. If still no token, stop.
     if token is None:
-        yield "Error: No authentication token found. Please add 'HF_TOKEN' to Space Secrets or run locally with an env variable."
+        yield "Error: No authentication token found. Please add 'HF_TOKEN' to Space Secrets."
         return
 
-    # Initialize client
+    # --- 2. Setup Client ---
     client = InferenceClient(model_id, token=token)
     
-    # Build the message format
+    # --- 3. Build Messages (handling history correctly) ---
     messages = [{"role": "system", "content": system_message}]
     
-    # 'history' comes as tuples [(user, bot), (user, bot)...]
-    for user_msg, bot_msg in history:
-        messages.append({"role": "user", "content": user_msg})
-        messages.append({"role": "assistant", "content": bot_msg})
+    # We use type="messages" in ChatInterface, so history is already a list of dicts
+    for msg in history:
+        messages.append(msg)
     
     messages.append({"role": "user", "content": message})
 
+    # --- 4. Generate Response ---
     try:
         stream = client.chat_completion(
             messages, 
@@ -43,15 +40,22 @@ def respond(message, history, system_message, temperature, request: gr.Request):
             temperature=temperature,
             top_p=0.9
         )
+        
         response_text = ""
         for chunk in stream:
+            # FIX: Check if choices exist before accessing index [0]
+            if not chunk.choices:
+                continue
+            
             content = chunk.choices[0].delta.content
             if content:
                 response_text += content
                 yield response_text
+
     except Exception as e:
         yield f"Error: {str(e)}"
 
+# --- 5. Build UI ---
 with gr.Blocks(fill_height=True) as demo:
     with gr.Sidebar():
         gr.Markdown("# AI Coding Assistant")
@@ -60,6 +64,8 @@ with gr.Blocks(fill_height=True) as demo:
     
     gr.ChatInterface(
         respond,
+        # 'type="messages"' fixes the deprecation warning and makes parsing easier
+        type="messages", 
         additional_inputs=[
             gr.Textbox(value="You are a helpful assistant.", label="System Instruction", lines=2),
             gr.Slider(minimum=0.1, maximum=2.0, value=0.7, step=0.1, label="Temperature")