stubtest now checks this for overloads as well (previously it didn't,
due to some combination of laziness, that this used to short circuit and
prevent signature checking, and the fact that these are all not super
helpful)
Co-authored-by: hauntsaninja <>
Update pyright CI test to use the latest version of pyright (1.1.118). Enable all but one of pyright's strictest checks. Add a "# type: ignore" to `__new__` method in `weakref.KeyRef` because it uses a non-standard name for the `cls` parameter, which is flagged as an error by pyright.
Co-authored-by: Eric Traut <erictr@microsoft.com>
Run the mypy test suite using Ubuntu 18.04. Ubuntu 20.04 doesn't have a package for virtualenv on Python 2.
Also, curses.color_pair() now has a named argument.
A security fix added a "separator" argument to several URL parsing
functions and method in point releases:
* 3.6.13
* 3.7.10
* 3.8.8
* 3.9.2
Until all these versions are available on the GitHub Actions runners,
we need to whitelist the functions in the stubtests.
For now, the test requires changes that have been merged into the pytype master branch but not yet released. I'll update requirements-tests-py3.txt again once I've cut a new release (hopefully later this week).
* Fix pytype_test.
* Re-enable pytype_test.
* Remove continue-on-error from pytype_test.
* Temporarily install pytype from its master branch, as the code to
support the new directory structure hasn't been released yet.
* Set TYPESHED_HOME before using pytype.pytd.Typeshed().
* Add _collections_abc module
* Rearrange collections.abc and collections to re-export from _collections_abc
Co-authored-by: Sebastian Rittau <srittau@rittau.biz>
I noticed that the pytype parse test was getting quite slow again. This
is a somewhat hacky change that caches pytype's internal pyi loader to
avoid parsing files multiple times. On my machine, the test goes from
taking ~6m to complete to ~30s.
* Add script to generate protoc stubs using mypy-protobuf generated stubs
* Use generate_proto_stubs to generate stubs for protobuf 3.14.0
* Skip _pb2.pyi from flake8,black,isort,pytype
[tests/pytype_test.py](./tests/pytype_test.py#L176) is still using
Python `3.6`:
> python_version="2.7" if version == 2 else "3.6",
The test fails to run if `python3.6` is unavailable (for example on
Debian 10 Buster, which has 3.7).
According to (https://devguide.python.org/#status-of-python-branches
`3.6` is still maintained until end of 2021.
`pytype` itself [claims](https://github.com/google/pytype#requirements)
to run on `2.7`, `3.5`-`3.8`.
Change the script to use the version of Python from the invoking
interpreter.
The previous typing meant `threading.ExceptHookArgs` could not be used
to type a value.
The new typing follows what cpython does in the happy path (`_thread`
exists rather than the pure-python fallback being used).
Fixes#4767.
* typeshed: update stubtest version
Includes changes from https://github.com/python/mypy/pull/9680
I've already fixed all the true positives on typeshed.
* attempt to fix windows' _WarnFunction
Co-authored-by: hauntsaninja <>
Only a subset of mypy's test suite should be relevant to typeshed:
https://github.com/python/mypy/pull/9638
This should make things faster.
(Also the -n12 argument to pytest is weird; note mypy's pytest.ini
automatically specificies -nauto which is what we should want)
Resolves https://github.com/python/typeshed/issues/4333
Co-authored-by: hauntsaninja <>
In #4696 I bumped stubtest to a commit that hasn't been released. We
should have these stubtest versions match. In
https://github.com/python/mypy/pull/9426 I changed the output of
stubtest here; it's good to get this change in before I forget and this
action silently stops working.
Co-authored-by: hauntsaninja <>
