.github/labeler.yml: automatically add backport label for PRs touching ci/#374921
Conversation
wolfgangwalther
left a comment
There was a problem hiding this comment.
I like the idea. There is already a diff again, between master and release-24.11, for some of the workflows.
|
I have the impression that this CI will remove labels that does not match the yml file (example: #375210) 🤔 For example if you add the backport label to a random PR that only touches |
Right, very good catch. We use |
|
If an additional action on each PR is needed then i'm in favor of just not doing it and closing this PR. Barely any PRs touch the CI setup, and the current action usage is already massive since eval was moved from ofborg. |
|
I'm not saying we need a new job - it could be a second step in the same job. So almost no overhead. But the action itself needs to run twice. We do forget to backport those changes to CI regularly, so I think this is a really good idea here. |
I have plans to save some of that by moving many of the simple checks (editorconfig, nixfmt, nixpkgs-vet, ... all those) into a single job. Imho that would both improve the output and save resources. Just didn't get to it, yet. |
b2a0ec5 to
022a850
Compare
|
pushed the change, how may i test this on my branch? |
Since this uses
(it might also work if you just create a PR from another branch into this branch, without the master step. The PR needs to be in your fork. But I didn't test it that way, yet) |
022a850 to
15b3e2d
Compare
|
The ci runs on my own fork seem to get skipped for some reason... |
|
tested working on my fork |
|
Let's try this - we can always revert, if it causes problems. |
| # This file is used by .github/workflows/labels.yml | ||
| # This version uses `sync-labels: false`, meaning that a non-match will NOT remove the label | ||
|
|
||
| "backport: release-24.11": |
There was a problem hiding this comment.
| "backport: release-24.11": | |
| "backport release-24.11": |
😂
|
I think this is now labeling the "backport: release-24.11" label to every PR now, I guess |
|
🙈 Luckily it's the wrong label, so easy to fix that. Will open a PR soon. |
I assume it makes sense to keep master and stable in sync since these folders are also used in downstream tools such as nixpkgs-review
Things done
nix.conf? (See Nix manual)sandbox = relaxedsandbox = truenix-shell -p nixpkgs-review --run "nixpkgs-review rev HEAD". Note: all changes have to be committed, also see nixpkgs-review usage./result/bin/)Add a 👍 reaction to pull requests you find important.