Fix a false negative for unused-import

[mbyrnepr2]

Fix a false negative for unused-import when linting multiple modules & one module uses an import in a type annotation & the same import is unused in a subsequent module.

Type of Changes

	Type
✓	🐛 Bug fix
	✨ New feature
	🔨 Refactoring
	📜 Docs

Description

Closes #4150

[coveralls]

Pull Request Test Coverage Report for Build 3351949749

• 1 of 1 (100.0%) changed or added relevant line in 1 file are covered.
• No unchanged relevant lines lost coverage.
• Overall coverage increased (+0.0003%) to 95.364%

---

Totals
Change from base Build 3348231752:	0.0003%
Covered Lines:	17178
Relevant Lines:	18013

---

💛 - Coveralls

[Pierre-Sassoulas]

Great fix !

[Pierre-Sassoulas]

The fix itself is really simple 😄

[Pierre-Sassoulas]

Not sure about what to do regarding the release we're aiming. It's a bug but at the same time it was a false negative. Maybe 2.16.0 so we do not raise new messages in a patch. If there's an oversight on our side we don't want cascading patches backport ?

[mbyrnepr2]

Not sure about what to do regarding the release we're aiming. It's a bug but at the same time it was a false negative. Maybe 2.16.0 so we do not raise new messages in a patch. If there's an oversight on our side we don't want cascading patches backport ?

Agreed. Happy to wait until 2.16.0.

[mbyrnepr2]

About the primer comment.

I notice that the output for Django, Pandas & Sentry is similar to some other open MRs; it doesn't look like it is caused by this MR.
The black output looks relevant to this MR but, at the same time, it is saying that a new message is emitted while I also see that message emitted on main.

[Pierre-Sassoulas]

In black I think os is used here : https://github.com/psf/black/blob/f16333e78ba77ab29ab0b75e10891e6c65c6da7e/src/blib2to3/pgen2/pgen.py#L25. Feel like a false positive.

[mbyrnepr2]

In black I think os is used here : https://github.com/psf/black/blob/f16333e78ba77ab29ab0b75e10891e6c65c6da7e/src/blib2to3/pgen2/pgen.py#L25. Feel like a false positive.

I agree with you & I'm happy to dig into that one; but this false positive isn't created by the current MR.

The sample from the black primer will emit an unused-import on main also:

import os
from typing import Union
Path = Union[str, "os.PathLike[str]"]

[Pierre-Sassoulas]

I agree with you & I'm happy to dig into that one; but this false positive isn't created by the current MR.

Ha, something is wrong with the primer then, sorry.

[Pierre-Sassoulas]

@mbyrnepr2 can you merge or rebase on main so we launch the new python 3.11 tests please ?

[github-actions]

🤖 Effect of this PR on checked open source code: 🤖

Effect on black:
The following messages are now emitted:

1. unused-import:
   Unused import os
   https://github.com/psf/black/blob/4bb6e4f64ab3820ab9fae6716cd59479d34b7edf/src/blib2to3/pgen2/pgen.py#L22

Effect on pytest:
The following messages are now emitted:

1. inconsistent-return-statements:
   Either all return statements in a function should return an expression, or none of them should.
   https://github.com/pytest-dev/pytest/blob/646a46e5f4b1f1ae5a06dcbc91fcdebfc235a28a/src/_pytest/pytester.py#L301
2. no-name-in-module:
   No name 'NOSE_SUPPORT_METHOD' in module '_pytest.deprecated'
   https://github.com/pytest-dev/pytest/blob/646a46e5f4b1f1ae5a06dcbc91fcdebfc235a28a/src/_pytest/python.py#L62
3. no-name-in-module:
   No name 'PytestReturnNotNoneWarning' in module '_pytest.warning_types'
   https://github.com/pytest-dev/pytest/blob/646a46e5f4b1f1ae5a06dcbc91fcdebfc235a28a/src/_pytest/python.py#L81
4. inconsistent-return-statements:
   Either all return statements in a function should return an expression, or none of them should.
   https://github.com/pytest-dev/pytest/blob/646a46e5f4b1f1ae5a06dcbc91fcdebfc235a28a/src/_pytest/python.py#L1085
5. inconsistent-return-statements:
   Either all return statements in a function should return an expression, or none of them should.
   https://github.com/pytest-dev/pytest/blob/646a46e5f4b1f1ae5a06dcbc91fcdebfc235a28a/src/_pytest/fixtures.py#L130
6. inconsistent-return-statements:
   Either all return statements in a function should return an expression, or none of them should.
   https://github.com/pytest-dev/pytest/blob/646a46e5f4b1f1ae5a06dcbc91fcdebfc235a28a/src/_pytest/python_api.py#L798
7. no-name-in-module:
   No name 'NOSE_SUPPORT' in module '_pytest.deprecated'
   https://github.com/pytest-dev/pytest/blob/646a46e5f4b1f1ae5a06dcbc91fcdebfc235a28a/src/_pytest/nose.py#L5
8. no-name-in-module:
   No name 'warn_explicit_for' in module '_pytest.warning_types'
   https://github.com/pytest-dev/pytest/blob/646a46e5f4b1f1ae5a06dcbc91fcdebfc235a28a/src/_pytest/config/__init__.py#L62
9. no-member:
   Module '_pytest.deprecated' has no 'HOOK_LEGACY_MARKING' member
   https://github.com/pytest-dev/pytest/blob/646a46e5f4b1f1ae5a06dcbc91fcdebfc235a28a/src/_pytest/config/__init__.py#L369

This comment was generated for commit 295d0ce

[DanielNoord]

@mbyrnepr2 The message in black seems like a false positive?

[mbyrnepr2]

It does seem that way, I agree @DanielNoord but we can also see the same happening on the main branch (before it was merged). Perhaps an issue with the primer?

[DanielNoord]

Probably! Went through my notifications of last week and noticed this message, but pytest is also broken in that comment so could very well be a fluke.