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
Make checks workflow label-driven via run-type output (#8524)
* Make checks workflow label-driven via run-type output
Replace the fragile 'removal' detection (which inferred removals from a
'Bad data []' parse-error sentinel) with an explicit, label-driven
run-type output on the preflight job:
- remove-repositories -> removal
- New default repository -> new-repository
- renamed-repositories -> renamed
- no known label -> unknown
Fail the preflight if more than one of these labels is set. Only
new-repository PRs run the preflight detection steps and the downstream
jobs (except editable). Add the labeled trigger so applying a label
re-runs the checks.
* Apply suggestion from @ludeeus
* Rename run-type output to run_type in checks workflow
* Simplify run type detection to a plain if/elif chain
* Remove paths guard from checks workflow trigger
---------
Co-authored-by: Claude <noreply@anthropic.com>
0 commit comments