[FastAPI Example] State Persistence Not Working as Documented in Team Chat Implementation

[Edison-A-N]

[FastAPI Example] State Persistence Not Working as Documented in Team Chat Implementation

Issue Description

There is a discrepancy between the documentation and actual implementation regarding state persistence in the team chat application. The README states that "the agents and team use the load_state and save_state methods to load and save their state from and to files on each turn." However, in the actual implementation (app_team.py), the state saving mechanism doesn't work as documented.

Current Implementation Issue

In app_team.py, the state saving logic is placed after the for message in stream loop:

async for message in stream:
    if isinstance(message, TaskResult):
        continue
    await websocket.send_json(message.model_dump())
    if not isinstance(message, UserInputRequestedEvent):
        history.append(message.model_dump())

# Save team state to file.
async with aiofiles.open(state_path, "w") as file:
    state = await team.save_state()
    await file.write(json.dumps(state))

The problem is that the _user_input function blocks the execution flow when waiting for user input, causing the code to never reach the state saving logic. This means that while the application can continuously receive user input, it never gets to save the team state and chat history as promised in the documentation.

Expected Behavior

The state should be saved after each turn as documented in the README. This would require either:

1. Moving the state saving logic to a more appropriate location in the code flow
2. Implementing a different mechanism to ensure state persistence between turns
3. Updating the documentation to accurately reflect the current implementation

Additional Context

This issue affects the reliability of the application's state persistence feature, which is important for maintaining conversation context across server restarts.

Documentation Link

https://github.com/microsoft/autogen/blob/main/python/samples/agentchat_fastapi/README.md