UserProxy in RoundRobinGroupChat not acting as expected.

[Wei-Cheng881221]

What happened?

I was trying to setup a simple RoundRobinGroupChat with UserProxy being apart of the Team, but ran into some issue that the UserProxy's input prompt came too early.
Use case: The RoundRobin seq is like this: [user_agent, assistant_agent, code_executor_agent], the user_agner(UserProxy) will block the output msg from code_executor_agent.

Therefore, I follow the userguide from : Providing Feedback During a Run. And found the same result from the user-guide. To make the question more easier to understand, I use the user-guide code instead my code.
For example, the reference expected output are:

---------- user ----------
Write a 4-line poem about the ocean.
---------- assistant ----------
In endless blue where whispers play,  
The ocean's waves dance night and day.  
A world of depths, both calm and wild,  
Nature's heart, forever beguiled.  
TERMINATE
---------- user_proxy ----------
APPROVE

but my results looks like this after running:

---------- user ----------
Write a 4-line poem about the ocean.
---------- assistant ----------
Enter your response: The ocean's waves crash on the shore,
A soothing sound that calms once more.
Its depths are dark, its beauty bright,
A treasure trove of wonder in sight.

TERMINATE

And after typing APPORVE, it then shows the following then end:

APPROVE
---------- user_proxy ----------
APPROVE

I am using autogen 0.4.9.1 and python 3.12.2 under Redhat 8 machine.
Also, I am using llama 3.3_70B through ollama.

Which packages was the bug in?

Python AgentChat (autogen-agentchat>=0.4.0)

AutoGen library version.

Python dev (main branch)

Other library version.

No response

Model used

llama_3.3_70B

Model provider

Ollama

Other model provider

No response

Python version

3.12

.NET version

None

Operating system

Other

[ekzhu]

@jackgerrits thoughts?

[caserzer]

same issue on ubuntu 24 python version 3.12 /3.10 , but works on macos python 3.10