dwsmith1983 commented on PR #7377: URL: https://github.com/apache/datafusion-comet/pull/7377#issuecomment-6552305599
> I'm wondering why PRs cap imposed by GH and ASF is not enough? The cap only counts pull requests in this repo, so an account with 500 open elsewhere still gets three here. I replayed the check from this PR, read-only, against the 211 PRs open today. It would close PRs from four accounts. Three of them have between 261 and 3,175 open PRs across 29 to 93 repos, and hold eight open PRs here between them. So these accounts do show up here. Before it starts closing PRs, I'd change three things: - **Dry run by default.** Right now it closes PRs unless someone sets `HIGH_VOLUME_DRY_RUN`. I'd make dry run the default in the workflow and switch to closing once a couple of weeks of logs look right. - **Draft instead of close when the author has merged work here.** The fourth account looks like a person: 91 open PRs across 23 repos, 75 merged across GitHub, and a draft here from June. Two of the other three also have PRs merged here, and one of their open PRs is already approved. For an author with a merged PR here, converting to draft with the label keeps it out of the review queue, keeps its history, and needs no committer to reopen it. - **Check write access directly.** Through the API, several committers come back as `CONTRIBUTOR`, I think because their apache org membership is private. If the event payload reports the same, the check runs for them too. Nobody is near the thresholds today, but asking `repos/$GH_REPO/collaborators/$login/permission` would make the skip independent of membership visibility. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
