Skip to content

ci(weekly-lints): pass the build status to the Zulip report - #881

Merged
chenson2018 merged 2 commits into
leanprover:mainfrom
jessealama:ci/weekly-lints-success
Sep 8, 2026
Merged

ci(weekly-lints): pass the build status to the Zulip report#881
chenson2018 merged 2 commits into
leanprover:mainfrom
jessealama:ci/weekly-lints-success

Conversation

@jessealama

Copy link
Copy Markdown
Contributor

The report script marks a run with a green checkmark only when the caller sets SUCCESS=true. This workflow never did: the switch was added to the script in mathlib-ci in March 2026, and the nightly workflows there were updated to set it, but the weekly workflows in Mathlib and CSLib were not. The result is that every weekly report has been posted with a red X, even when the build was clean and had no linter messages.

The solution is to just capture the exit status of lake build and pass it through, as in the corresponding Mathlib PR.

The report script marks a run as succeeded only when the caller sets
SUCCESS=true, and this workflow never did, so every weekly report has
been posted as failed, even for a clean build with no linter messages.
Capture the exit status of `lake build` (the step already runs with
`pipefail`) and pass it through, matching how the nightly workflows in
mathlib-ci call the same script.
Unspecified `run:` steps use `bash -e {0}` without `pipefail`, so the status
of `lake build | tee` was tee's and `build_status` stayed 0 even on a failed
build. Turn on `pipefail` at the top of the step.
@chenson2018

Copy link
Copy Markdown
Collaborator

@jessealama I see you pushed a new commit. Let me know if you'd like me to run this job from a branch before merging.

@jessealama

Copy link
Copy Markdown
Contributor Author

@jessealama I see you pushed a new commit. Let me know if you'd like me to run this job from a branch before merging.

Yeah, please do! I noticed that there was a mismatch in this action and others, in that we don't specify shell: bash explicitly. I looked at the GH docs and realized I assumed the pipefail option was set, but that's only true if you specify shell: bash.

@chenson2018 chenson2018 left a comment

Copy link
Copy Markdown
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

@jessealama Sorry for the delay, I'm on spotty hotel wifi at the moment. The run that I started seems to have worked as expected. Is this good to merge?

@jessealama

Copy link
Copy Markdown
Contributor Author

@jessealama Sorry for the delay, I'm on spotty hotel wifi at the moment. The run that I started seems to have worked as expected. Is this good to merge?

No problem! I'm also a bit delayed, looking around at CI scripts and seeing what else might be lurking here. Yes, I believe this is good to merge.

@chenson2018
chenson2018 added this pull request to the merge queue Sep 8, 2026
Merged via the queue into leanprover:main with commit 24cb101 Sep 8, 2026
3 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants