Maximilian Linhoff
Maximilian Linhoff
How do we want to handle configured values that are invalid? I would be in favor of raising an error instead of continuing to the next possible fallback in case...
@rdimaio I just saw https://github.com/rucio/rucio/issues/6846 and the commits referenced therein, so you also started to refactor this part...
@rdimaio In our tests for the different authentication methods in our setup, I also needed to make sure the rucio token is cleaned up before running a specific test, otherwise...
Will have a look again, sorry!
I will now look into adapting the tests to address @rdimaio observations.
The tests mentioned above by @rdimaio have been changed since the comments were made.
Here is a plot of the improvements, generated using the code below  ```python from pathlib import Path import astropy.units as u from astropy.units import Quantity import numpy as np...
@neutrinoceros How can I run the same benchmarks as you did in #16222?
Could you add it? I don't have the permissions to modify labels
The test failure *looks* unrelated at first glance... any idea? https://github.com/astropy/astropy/actions/runs/8798691509/job/24146244116?pr=16326