Skip to content

DOC: Remove hint on PRs from origin/main #28638

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 1 commit into from
Aug 1, 2024
Merged

Conversation

timhoffm
Copy link
Member

@timhoffm timhoffm commented Aug 1, 2024

As poposed in #28575 (review), we don't need this. If developers have followed above instructions, they won't end up having changes on origin/main. But if they do, it's not really a problem on our side. We're still able to handle that. - As this PR shows, which I've intentionally made from my origin/main 😄 ping @story645

For an inexperienced user, it is simpler to continue with a PR from main rather than roll everything back and create a new PR.

The only problem exists on the user side in that they can't handle multiple concurrent changes when they only use main. But that's on them. And if they are inexperienced, they likely won't do that anyway.

@github-actions github-actions bot added the Documentation: devdocs files in doc/devel label Aug 1, 2024
@timhoffm timhoffm added this to the v3.10.0 milestone Aug 1, 2024
@story645
Copy link
Member

story645 commented Aug 1, 2024

This didn't fail our pre-commit?
ETA: am slightly concerned that our "don't push to main" pre-commit didn't fail :/

@tacaswell
Copy link
Member

you can either not installs it or explicitly skip it when committing (e.g. I'm currently getting spurious mypy failures via precommit locally (which I assume is mypy version issue due to virtualenv trying to be about 10% too clever and failing at its job) so I do every commit twice (once to let precommit fail and once skipping it to actually commit).

@timhoffm
Copy link
Member Author

timhoffm commented Aug 1, 2024

This didn't fail our pre-commit? ETA: am slightly concerned that our "don't push to main" pre-commit didn't fail :/

It failed the local pre-commit. I had to deactivate it to be able to commit to main 😄. Putting it the other way round: Precommit makes the case of the hint even less likely. One more argument to remove it 😆

@story645
Copy link
Member

story645 commented Aug 1, 2024

Putting it the other way round: Precommit makes the case of the hint even less likely. One more argument to remove it 😆

So I try and make sure anything we do as a pre-commit we somewhat document in our workflow, which may be how this came in in the first place 😅

But also I think we use hints way too liberally and Juanita's suggestion of moving this down to "open a pull request" as an optional type of thing makes sense & can be removed from here for now.

Which the approval is self merge or move the sentence then self merge. 🤷‍♀️

@ksunden
Copy link
Member

ksunden commented Aug 1, 2024

The one way it can roll onto slightly more problematic for us as maintainers is that PRs from main do not allow us to push to a contributor's branch (because default branch protections are in place). In my experience this is true even if github says "Maintainers are allowed to edit this pull request" because that is a separate layer of protection.

This isn't that much of a problem, but can mean a slight slow down for simple fixes we might otherwise just make and merge (e.g. typos etc). Just means we on occasion may have to wait for an extra round of the contributor responding, merge and open a followup, or open a PR from our own fork that branches from their main.

It is relatively rare that we want to do all of that at once, but can happen. However, overall, I kind of doubt this particular note significantly decreases the prevalence of doing this.

@timhoffm timhoffm merged commit f4f40ba into matplotlib:main Aug 1, 2024
25 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

Successfully merging this pull request may close these issues.

4 participants