Looks like I made an argument in `datetime.pyi` positional-only a while back, and didn't realise the class was given as an example in `CONTRIBUTING.md`.
Also remove the python2 markers of packages that don't list Python 2
as supported in the latest version.
Don't special case version '0.1'
Co-authored-by: Akuli <akuviljanen17@gmail.com>
This is a first step towards #6095, where x.y and x.y.* are treated as
equivalent. The next step is to update existing versions to use x.y.*
where applicable and then treat x.y differently.
Co-authored-by: Akuli <akuviljanen17@gmail.com>
Remove the check in check_consistency that ensures that only other
type packages from typeshed are being depended on. Instead, add an
explanation to CONTRIBUTING that spells out the requirements. This
adds a higher, but reasonable burden on maintainers to check the
dependencies manually.
Part of #5768
Move and consolidate venv setup and running isort/black/flake8 into
separate sections and link those sections from the ToC. Also extend
those sections slightly.
Move the tests section into a separate file.
Make venv name in pre-commit match name in CONTRIBUTING.
This shuffles sections around between README.md and CONTRIBUTING.md. CONTRIBUTING now contains information pertaining to opening PRs, README all other information. I have also moved the list of maintainers to a separate file.
I have kept most information intact for now, with two main exceptions:
I removed duplicated information.
For brevity's sake, I trimmed some explanations from the section about version checks.
I have restructured the CONTRIBUTING file to follow the order of the introductory "contribution process at a glance" section. This now serves as a bit of a table of contents.
Closes: #5422
* Remove conventions enforced by black
Remove old note about optional default arguments (now part of
PEP 484 and enforced by CI)
* Recommend to use PEP 585
Cf. #4820
* Try out using collections.abc
* Reference mypy bug
This was discussed on the typing-sig mailing list and in
python/peps#1218, has met the approval of the steering council, and
was incorporated into PEP 484.
This pull request is a follow-up to https://github.com/python/mypy/issues/7214.
In short, within that mypy issue, we found it would be helpful to
determine between contextmanagers that can "swallow" exceptions vs ones
that can't. This helps prevent some false positive when using flags that
analyze control flow such as `--warn-unreachable`. To do this,
Jelle proposed assuming that only contextmanagers where the `__exit__`
returns `bool` are assumed to swallow exceptions.
This unfortunately required the following typeshed changes:
1. The typing.IO, threading.Lock, and concurrent.futures.Executor
were all modified so `__exit__` returns `Optional[None]` instead
of None -- along with all of their subclasses.
I believe these three types are meant to be subclassed, so I felt
picking the more general type was correct.
2. There were also a few concrete types (e.g. see socketserver,
subprocess, ftplib...) that I modified to return `None` -- I checked
the source code, and these all seem to return None (and don't appear
to be meant to be subclassable).
3. contextlib.suppress was changed to return bool. I also double-checked
the unittest modules and modified a subset of those contextmanagers,
leaving ones like `_AssertRaisesContext` alone.
