Paul Ganssle

OK, I actually figured out how to fix this manually (though I think it is still a problem in general): on the device there is a `metadata.calibre` file in the...

I think that the scope of `rrule` mostly operates on naive datetimes because of the nature of datetime arithmetic in Python. It might be expensive to fix this in the...

One thing to note is that I believe a full implementation of `RSCALE` would be a *lot* of work, so we'll limit this to supporting only the Gregorian calendar for...

Note to anyone who wants to work on the `skip` portion of this, the work is >80% done by @yoch in #522. If someone wants to do the last part,...

I usually just click the link on the CI: Is there some advantage other than "easier to find"?

Weird, but nothing we can or should do about this, so I am going to close it. Probably a good cautionary tale to add to the many reasons to use...

I think that `ZoneInfo('__init__.py')` is also a problem, but a slightly different one. That comes from the fact that in order to support `importlib.resources`, each of the zoneinfo subdirectories needs...

By the way, it might be easiest to start with a PR against backports.zoneinfo, because I have a lot more linting, coverage and format checks set up there: https://github.com/pganssle/zoneinfo

@labrys I'm not sure if they compare equal, but equality is not super important in `tzinfo` objects, since the [semantics of `datetime` use `is` rather than `==`](https://blog.ganssle.io/articles/2018/02/a-curious-case-datetimes.html). I will be...

@justanr Definitely willing to accept this sort of thing, though I'd have to look at the details. One overall point, which is a hard circle to square, is that your...