Before and After Retry, and Logging

[estahn]

I'd like to log retries similar to what is done in the tenacity documentation.

https://tenacity.readthedocs.io/en/latest/#before-and-after-retry-and-logging

[hynek]

ENRICO!?!?!? You here!? How long it's been? 20 years? 😅

Anyhow, I think this is subproblem of #21 but an important addition: there should be a plug before and after a retry.

[estahn]

HYNEK, indeed it has. I was gobsmacked when I realised it was you.

In regard to the issue, a simpler resolution might be to set the logger to the built-in Python logger instead of None 🤔

stamina/src/stamina/_instrumentation.py

Line 13 in 27f283e

logger = None

[estahn]

I made it work for my case. In case anyone else needs this.

import logging

import stamina


logger = logging.getLogger("stamina")


def log_retries(
    attempt: int, idle_for: float, exc: Exception, name: str, args: tuple[object, ...], kwargs: dict[str, object]
) -> None:
    logger.warning(
        "stamina.retry_scheduled",
        extra=dict(attempt=attempt, idle_for=idle_for, exc=repr(exc), origin=name),
    )


stamina._instrumentation.INSTRUMENTS.append(log_retries)

[hynek]

This has been fixed with #35 – the snippet won't work in stamina 23.1 and later.