You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Was going to try out the new formatting support on the nodejs.dev content an ran into (likely unrelated) issues.
I was thinking it might make sense to add some integration tests on a few of the main repos that use the config to see if there are any diffs/failures with running remark . -o.
Ones that pop into my head would be:
nodejs/node
nodejs/nodejs.org
nodejs/nodejs.dev
nodejs/TSC
The text was updated successfully, but these errors were encountered:
There will be a lot of diffs with nodejs/node but I'm working through them. Hopefully no bugs left. I think TSC is likely to be the easiest one to get where we need it. The problem is that there's nothing stopping new meeting notes from having diffs. It's perfectly possible to write valid no-lint-error markdown that is different from what the formatter would output. I suppose we could try to add the formatter to some kind of commit-queue pipeline but that's more work than I want to do at the moment.
Those four repos are a good list, though. If we can get them all into shape, that will be a good sign that the formatter doesn't cause more problems than it solves.
I suppose we could try to add the formatter to some kind of commit-queue pipeline but that's more work than I want to do at the moment.
We should also keep in mind that sometimes auto-formatter are doing the wring thing, I think it'd be a bit dangerous to land something that is not also human reviewed.
Another way of doing this would be to have an action that runs every time the main branch is being pushed, run the formattter, and opens a PR if there's a diff. I can try to work on that if that sounds worth it to you.
Was going to try out the new formatting support on the nodejs.dev content an ran into (likely unrelated) issues.
I was thinking it might make sense to add some integration tests on a few of the main repos that use the config to see if there are any diffs/failures with running
remark . -o
.Ones that pop into my head would be:
The text was updated successfully, but these errors were encountered: