Petr Viktorin

Let me close this. It could be solved by adding `libpython3d.so`, but that doesn't seem useful. Or it could be solved by removing `libpython3.so`, butsomeone might be depending on it....

FWIW, I think the source should be a human-writable format, and we should *generate* a JSON for the machines to consume (like https://peps.python.org/api/peps.json). And we should tell people to expect...

Já ych si je rád přečetl, a už tu je pár knížek co není přímo o Pythonu, takže klidně zařaď :)

I've made the *encoding* argument mandatory for `_text` functions when multiple path names are given. > Wouldn't this be a bit late for that? Yes, sorry. Previously I couldn't commit...

@jaraco, do you think allowing path separators is OK? I'm happy to delegate the decision to you. My opinion, for what it's worth: it *is* OK. I don't think we...

> Since Traversable.join_path accepts (*descendants), it should be readily supportable. Oh, that's great! I'll [add it to the docs](https://github.com/python/cpython/issues/117074). The semantics are clear then: I'll delegate to joinpath :) >...

I've updated the PR to allow path separators. A PR for `importlib_resources` is here: https://github.com/python/importlib_resources/pull/303

Thanks for the discussion everyone, and sorry I couldn't make everyone happy. I merged the PR. If you disagree with that, please preferably continue the discussion here: https://discuss.python.org/t/11386/47

Apparently the tests fail on big-endian machines; will fix

This is due to the following in `parser.py`: ```python @dataclass class Feature: scenarios: OrderedDict[str, ScenarioTemplate] [...] @dataclass class ScenarioTemplate: feature: Feature [...] ``` That is, a `Feature` references all its...