Miloš Prchlík

> I'll rebase when #2742, and if all goes well, we can delay this for after #2557 to make sure we don't have any surprises. > > Seems neither `pyright`...

> > This sucks, and I would hate to use Any :( We still have to live with older Pint versions, and newer ones seem to be too incompatible on...

> > A lot, I would say, because mypy and pyright are not just about type checking, they also detect attributes that are not defined, for example. Most tmt users...

> As for upper bounds everywhere - We could add it but we will need to update it quite often to not block adoption of newer versions into fedora-rawhide. So...

Resurrecting. I'm working on https://github.com/teemtee/tmt/pull/2841 and Pint tends to pop its ugly head again and again, so we'll need to resolve this one sooner rather than later: do something about...

> Seems xarray could be blocked by this. There seem to be [unit tests](https://github.com/teemtee/tmt/blob/main/tests/unit/test_hardware.py) for hardware, so how about rebasing and adding this to pyright/mypy ignores? Yeah, let's give it...

Those unrelated changes are really ugly... Why would they appear here and now when this morning's pre-commit output complained about nothing to me, no idea. Your patch does not even...

> Those unrelated changes are really ugly... Why would they appear here and now when this morning's pre-commit output complained about nothing to me, no idea. Your patch does not...

My 2 cents: I for one find encoding parameters into the test extremely displeasing :( Parsing them out of there, even realizing there's anything to extract first, could be ridden...

FTR, it's caused by the `show_exception()` and `render_exception()` ignoring the colorization settings. They don't depend on any logger, and they may get called very early, and they use `click.style()`. It...