Update app.py

app.py

@@ -247,7 +247,6 @@ tools = [
 web_agent = AgentWorkflow.from_tools_or_functions(
     tools, 
     llm=llm,
-    streaming=True,
     system_prompt="""You are a helpful assistant with access to specialized tools for retrieving information about weather, and news.
     AVAILABLE TOOLS:
     1. current_weather - Get current weather conditions for a location
@@ -323,30 +322,34 @@ async def gradio_query(user_input, chat_history=None):
     history.append({"role": "user", "content": user_input})
     
     # Add initial assistant message
-    history.append({"role": "assistant", "content": ""})
+    history.append({"role": "assistant", "content": "Thinking..."})
     yield history, history
     
-    # Get streaming response generator
-    response_gen = await run_query(user_input)
+    # Run the query (non-streaming at the agent level)
+    result = await run_query(user_input)
+    text = result.response if isinstance(result.response, str) else str(result.response)
     
-    # Accumulated response
-    full_response = ""
+    # Simulate streaming by yielding partial responses
+    # This is a simple approach - we'll show the response word by word
+    words = text.split()
+    partial_response = ""
     
-    # Process the streaming response
-    async for delta in response_gen:
-        if hasattr(delta, 'delta'):
-            # For streaming text chunks
-            full_response += delta.delta
-        elif hasattr(delta, 'response') and delta.response:
-            # For final response or tool outputs
-            if isinstance(delta.response, str):
-                full_response = delta.response
-            else:
-                full_response = str(delta.response)
+    for i, word in enumerate(words):
+        # Add the word to the partial response
+        partial_response += word + " "
         
-        # Update the last message with accumulated text
-        history[-1]["content"] = full_response
+        # Update the history with the partial response
+        history[-1]["content"] = partial_response
+        
+        # Yield the updated history
         yield history, history
+        
+        # Add a small delay to simulate typing (can be adjusted)
+        await asyncio.sleep(0.05)
+    
+    # Final update with complete response
+    history[-1]["content"] = text
+    yield history, history
 
 # Build and launch Gradio app
 grb = gr.Blocks()
@@ -363,7 +366,7 @@ with grb:
     chatbot = gr.Chatbot(type="messages")
     txt = gr.Textbox(placeholder="Ask me anything...", show_label=False)
     
-    # Modify the submit handler to work with streaming
+    # Set up event handlers for streaming
     txt.submit(
         gradio_query, 
         inputs=[txt, chatbot], 
@@ -375,7 +378,8 @@ with grb:
     )
     
     # Also update the button click handler
-    gr.Button("Send").click(
+    send_btn = gr.Button("Send")
+    send_btn.click(
         gradio_query, 
         [txt, chatbot], 
         [chatbot, chatbot]