Conversation
updates: - [github.com/astral-sh/ruff-pre-commit: v0.14.10 β v0.15.9](astral-sh/ruff-pre-commit@v0.14.10...v0.15.9) - [github.com/psf/black-pre-commit-mirror: 25.12.0 β 26.3.1](psf/black-pre-commit-mirror@25.12.0...26.3.1) - [github.com/python-jsonschema/check-jsonschema: 0.36.0 β 0.37.1](python-jsonschema/check-jsonschema@0.36.0...0.37.1) - [github.com/rhysd/actionlint: v1.7.10 β v1.7.12](rhysd/actionlint@v1.7.10...v1.7.12) - [github.com/pre-commit/mirrors-mypy: v1.19.1 β v1.20.0](pre-commit/mirrors-mypy@v1.19.1...v1.20.0) - [github.com/woodruffw/zizmor-pre-commit: v1.14.2 β v1.23.1](zizmorcore/zizmor-pre-commit@v1.14.2...v1.23.1) - [github.com/tox-dev/pyproject-fmt: v2.11.1 β v2.21.0](tox-dev/pyproject-fmt@v2.11.1...v2.21.0) - [github.com/abravalheri/validate-pyproject: v0.24.1 β v0.25](abravalheri/validate-pyproject@v0.24.1...v0.25) - [github.com/codespell-project/codespell: v2.4.1 β v2.4.2](codespell-project/codespell@v2.4.1...v2.4.2)
for more information, see https://pre-commit.ci
|
I've pushed a fix for this, which is to assign the command tuple to a variable first, which is how the other tests do it. Feels more like a workaround. |
|
The other one: We can either allow this, as I think the worst someone can do if they manage to get hold of this is to fake the coverage report. But better, we should be able to get rid of the token altogether now. I've pushed a commit to remove it. In another org, we just flipped this from required to not required:
@JacobCoffee Please could you select "Not required" at https://app.codecov.io/account/gh/python/org-upload-token ? |
|
Thanks! I've restarted the tests, and it's uploading okay: https://github.com/python/cherry-picker/actions/runs/24131641814/job/70428993264?pr=177 |
| git_version_cmd = "git", "--version" | ||
| version = subprocess.run(git_version_cmd, capture_output=True) |
There was a problem hiding this comment.
| git_version_cmd = "git", "--version" | |
| version = subprocess.run(git_version_cmd, capture_output=True) | |
| git_version_cmd = ("git", "--version") | |
| version = subprocess.run(git_version_cmd, capture_output=True) |
There was a problem hiding this comment.
We can commit this if you like, and we can watch the formatter remove them again :)
There was a problem hiding this comment.
Sounds like a broken formatter :). Took me a bit too long to see that git_version_cmd was in fact still a tuple and not just a string (I was apparently just glossing over the ", " as ).
| git_version_cmd = "git", "--version" | |
| version = subprocess.run(git_version_cmd, capture_output=True) | |
| git_version_cmd = ["git", "--version"] | |
| version = subprocess.run(git_version_cmd, capture_output=True) |
then. Not sure why it's changing in the first place, though...
There was a problem hiding this comment.
It's actually my memory that is broken π
Here's what was added to Black's 2026 style:
remove_parens_from_assignment_lhs: Remove unnecessary parentheses from the left-hand side of assignments while preserving magic trailing commas and intentional multiline formatting
https://github.com/psf/black/blob/main/CHANGES.md#version-2610
So it's safe to add parentheses. PR welcome! There's more than this one instance though, so do them all.
There was a problem hiding this comment.
Oh, now I see the comment about this being a hack around S607. Honestly, I'd rather just ignore S607 (or use comment suppression), at least in test_*.py.
I'm just complaining with no intent to put time into fixing it in any kind of reasonable time though, so please feel free to ignore me! :)
cherry_picker/test_cherry_picker.py
Outdated
| """\ | ||
| tmp_git_repo_dir.join(relative_config_path).write("""\ | ||
| team = "python" | ||
| repo = "core-workfolow" |
There was a problem hiding this comment.
Unrelated to this PR, but this seems to be unintentionally misspelled in several tests?
There was a problem hiding this comment.
Oh, didn't notice that! Might as well fix it. I'll slip it into this PR.


updates: