Handling py_canoe in a multi-thread environment

[AlvaroSC7]

I am trying to implement a tool that executes some commands in CANoe, using for that the py_canoe library. If I implement all the CANoe calls in the main thread everything works fine, but since some of them take a lot of time (highest being 4 mins) I want to execute it a parallel thread and keep the main one only for GUI updates.

When I do this, I can see some strange cases where the CANoe execution stops after opening the instance, depending on the machine I execute the tool on, being a sporadic issue. As a hint, I'm using pySimpleGUI for the UI and I use the built-in thread implementation, but I have already tried to code it myself and the result is the same.

I don't see whether this library is thread safe or not in the documentation, so I'm assuming it's not, and that's why I am having such problems.

Has anyone seen anything similar? Is anyone executing this library in a GUI or multi-thread application? If so, how?

This is an example of my code:

Pd_conf.MainWindow.perform_long_operation(func=lambda: execute_smoketests(choice), end_key="--PRECONDITION_CHECK_DONE--")
    event, values = Pd_conf.MainWindow.read(timeout=0)
    event = None
    print(f"TARGET   : Execute the following smoke tests: {choice}")
    print(f"Duration : 0:00:00")  # noqa: F541
    timeDelta = 0
    compileStart = datetime.now()
    while event != "--PRECONDITION_CHECK_DONE--":
        event, values = Pd_conf.MainWindow.read(timeout=1000)
        if (datetime.now() - compileStart).seconds > timeDelta:
            # Get time delta since start of CANoe operation
            timeDelta = (datetime.now() - compileStart).seconds
            # Update print output if one second has passed
            print(f"Duration : {timedelta(seconds=timeDelta)}", replace=f"Duration : {timedelta(seconds=timeDelta-1)}")