Learn from PR builds too #89
Merged
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Mark-Simulacrum
approved these changes
May 13, 2025
| @@ -171,11 +171,11 @@ impl Worker { | |||
| if !outcome.is_passed() { | |||
| self.report_failed(build_id, build.as_ref())?; | |||
| } | |||
| if build.pr_number().is_none() && build.branch_name() == "auto" { | |||
| if build.pr_number().is_some() || build.branch_name() == "auto" { | |||
There was a problem hiding this comment.
We'll need to update this for new-bors as well (cc @Kobzol) and we might want to include try jobs I guess?
There was a problem hiding this comment.
Try jobs are a subset of auto jobs, so not sure if it adds a lot of additional value. For new bors the branch name should be probably automation/bors/auto, but we haven't decided it yet. Created #90 to track this.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
After talking with some folks at RustWeek, I discovered that the RLA output for PR builds is suboptimal, as we are not learning from PR build logs. Originally we were only learning from auto builds to avoid the risk of a PR build poisoning RLA, but that risk is less than the downside of having bad error logs. This PR thus enables learning from PR build logs.