Can't diagnose=False used in sys.stdout?

[qq745639151]

I tried the example， but I set the diagnose=False

logger.add("out.log", backtrace=True, diagnose=False)
def func(a, b):
    return a / b

def nested(c):
    try:
        func(5, c)
    except ZeroDivisionError:
        logger.exception("What?!")

nested(0)

but the output is the same as in the case of diagnose=True.

[qq745639151]

This is the output when diagnose=True

2024-06-15 19:12:27.677 | ERROR    | __main__:nested:15 - What?!
Traceback (most recent call last):

  File "F:\WorkSpace\PycharmProjects\fastapi_learning\async_log.py", line 17, in <module>
    nested(0)
    └ <function nested at 0x000001D782E3D090>

> File "F:\WorkSpace\PycharmProjects\fastapi_learning\async_log.py", line 13, in nested
    func(5, c)
    │       └ 0
    └ <function func at 0x000001D780773E20>

  File "F:\WorkSpace\PycharmProjects\fastapi_learning\async_log.py", line 9, in func
    return a / b
           │   └ 0
           └ 5

ZeroDivisionError: division by zero

and This is the output when diagnose=False

2024-06-15 19:15:01.122 | ERROR    | __main__:nested:15 - What?!
Traceback (most recent call last):

  File "F:\WorkSpace\PycharmProjects\fastapi_learning\async_log.py", line 17, in <module>
    nested(0)
    └ <function nested at 0x000002B83785D090>

> File "F:\WorkSpace\PycharmProjects\fastapi_learning\async_log.py", line 13, in nested
    func(5, c)
    │       └ 0
    └ <function func at 0x000002B834E03E20>

  File "F:\WorkSpace\PycharmProjects\fastapi_learning\async_log.py", line 9, in func
    return a / b
           │   └ 0
           └ 5

ZeroDivisionError: division by zero

[Delgan]

The diagnose formatting should work regardless of the logging output.

Are you sure you did remove the default handler? Maybe call logger.remove() first. Also, be aware that if you run multiple FastAPI workers, your main file will be executed multiple times, which can disrupt the configuration of your logger.

[qq745639151]

Thanks for your answer. I have tried and solved the question. Thanks again