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
Full disclaimer: I dont know if this is the way to go. Just because it worked for my workflow doesnt mean it should work with this one. Maybe it isnt even related but i thought i put it in here anyway.
i have a question about the use of the workflow. Are the following usecases valid for this workflow?
1. PR1 is dependent on PR2 (PR2 has to be merged/closed first in order to get PR1 merged). This one is pretty straight forward.
2. PR1 is dependent on issue1 (issue1 has to be closed first in order to get PR1 merged). This can be helpful when someone already created PR1 but u want issue1 to get addressed first but nobody has created a PR for it yet.
3. issue1 is dependent on PR1 (PR1 has to be merged/closed first in order to close issue1). This can be helpful when somebody forgets to link PR1 as dependency when they create PR2 for issue1
4. issue1 is dependent on issue2 (issue2 has to be closed first in order to close issue1). This can be helpful when there are no PR's yet that addresses one of the issues.
Getting current issue...
Issue is closed. Unblocking other issues...
Unblocking issues blocked by #39...
Initializing labels...
Updating 40
Re-running most recent action on testpr
Oh... I see why this is failing. To signal to GitHub that the PR is ready to be merged, it needs to re-run the check that failed and let it succeed. When an issue/PR is closed, it goes through all issues/PRs that were blocked by it and, if it's a PR, re-runs the failed action.
The problem here is how it finds the action. Currently, the code searches for all checks that have run on a given branch, with the logic being that one of them has to be blocking-issues. However, the branch in question (testpr) has no workflows because it's a fork with actions disabled.
So I need to rework the logic around re-running failed checks. Maybe we don't even need to re-run it and can just start a new check.
Full disclaimer: I dont know if this is the way to go. Just because it worked for my workflow doesnt mean it should work with this one. Maybe it isnt even related but i thought i put it in here anyway.
At the moment i can get your workflow to trigger when i create a PR in the main repo from a fork but the test result is exacly the same as #4 (comment) > Levi-Lesches/blocking-issues-test#40.
Edit: See #19 (comment)
The text was updated successfully, but these errors were encountered: