Open vojvodics opened 8 months ago
Hi, I'm still looking at this issue, but just from a quick look, you can also use fail_on_error: false
in add-reviewers
, and then you probably don't need the reject
filter on the action itself.
@vim-zz thanks. Would gitstream still request a review from others? E.g. I have 3 people in the array, one of which is a codeowner - would the other 2 get a request for a review or would it skip everything?
It shouls still assign the other 2
Describe the bug
I'm unsure if this is a bug with our config, or in gitstream. We have a rule to require multiple approves if the PR is > than some estimated time. And also a rule that automatically assigns a review for a person based on the files changed and who created a PR:
// reject is added because gitstream would fail if it required a review from a person that created a PR and also codeowners who github adds automatically.
The issue doesn't occur on every PR, but sometimes if the PR is >= 10min, gitstream bot re-requests a review from the last person that approved the PR:
This causes the approve to never be accepted and an infinite loop.
To Reproduce
Provided above, not sure if I can create a reproduction as it requires a bit more complex setup than usual
Expected behavior
Approve shouldn't be re-requested