bluss

I have the following regression as well. I'm not sure if it's the same bug or not, so I'll put it here for now. Expectation: We repeat a metadata in...

@saecki What do you think about the grid header? I guess it makes sense that it is a repeated cell rather than just one header that's broken across several pages...

My mental model of this was wrong (I mean, see this bug report) so I don't know, but I had the idea that "inside" and before/after in document order couldn't...

Thanks for that very astute observation. I know I had experimented with rounding, but in a very naive way: just in the box.baseline. It's clear now why that didn't help:...

Thanks for the help! For me, the issue is resolved with that, I will use rounding from now on and keep a closer eye on exact float values than before..

uuids and non-sequential u128 make sense for map-type graphs (based on HashMap or similar) and not the index based graphs. (Graph and StableGraph are for example index based).

What is the more specific harm here? It would make it easier to understand what to change if it was more specific about what's wrong. IMO it's only good that...

The config `tool.hatch.build.targets.sdist.packages` looks out of place to me, and without it, I think it's doing the right thing? """ The packages option is semantically equivalent to only-include (which takes...

It's possible that https://docs.pytest.org/en/stable/explanation/goodpractices.html could help here, they have a section for your case: *Tests as part of application code*

Something that's close to the current style (including where clauses) and a line width of 90, 100 (whatever is close to the current style). I think rustfmt is getting to...