The patch-available label was added by me, through the HBaseQA account. I'm currently testing whether adding labels can cause an email notification by default and it seems to work.
So I think we could change the behavior to "link label", then we could get the notification then. What do you guys think? Let's have a try? Thanks. Nick Dimiduk <ndimi...@apache.org> 于2024年1月25日周四 17:31写道： > > I reverted HBASE-28325 for now. > > On Thu, Jan 25, 2024 at 10:12 AM Nick Dimiduk <ndimi...@apache.org> wrote: > > > On Thu, Jan 25, 2024 at 3:51 AM 张铎(Duo Zhang) <palomino...@gmail.com> > > wrote: > > > >> Was wondering if we could do something by our own, like jenkins job, > >> use github api to scan all the PRs and change the status related jira > >> issues to patch available through jenkins api. > >> > >> Not sure how difficult to do this as I'm not a jenkins expert... > >> > > > > Yes, we could probably hack something up. But pursuing this route, a > > GitHub Action would be nicer as it can work from api callbacks, which I > > don't think is available to us from Jenkins -- all the GitHub interaction > > from Apache's Jenkins that I've seen is via polling. > > > > Bryan Beaudreault <bbeaudrea...@apache.org> 于2024年1月25日周四 02:00写道： > >> > >> > > >> > My vote would be to disable it, mostly because of the noise from > >> pre-commit > >> > hooks that get mirrored over. That said I appreciate what you were > >> trying > >> > to do here, and would definitely be in support of your original > >> intention > >> > if that were possible to have without the pre-commit noise. > >> > > >> > On Wed, Jan 24, 2024 at 12:55 PM Nick Dimiduk <ndimi...@apache.org> > >> wrote: > >> > > >> > > Hey team, > >> > > > >> > > I executed the below change in a unilateral manner that was not > >> > > appropriate. This was a mistake on my part. However, I thought the > >> change I > >> > > pushed would be less disruptive than it turned out to be. > >> > > > >> > > My intention was to fill a gap in workflow that has long bothered me. > >> I > >> > > find it annoying that I subscribe to a Jira issue as a watcher but > >> then > >> > > don’t find out when a PR is opened for that issue. I inquired with > >> infra > >> > > about this, and they pointed me toward this feature of the .asf.yaml. > >> Only > >> > > after I enabled it did I realize the mistake. With the current > >> setting, > >> > > every comment on a PR associated with a Jira is repeated into that > >> Jira. I > >> > > believe that this is the behavior that Hadoop uses, which I > >> personally find > >> > > to be not useful. > >> > > > >> > > So, do folks here have a preference about this notification setting? > >> Should > >> > > I continue pursuit of a more limited version of this comment feature > >> that I > >> > > initially desired? Or do folks here prefer the more verbose version > >> that is > >> > > enabled now? > >> > > > >> > > Thanks for your time, > >> > > Nick > >> > > > >> > > On Tue, 23 Jan 2024 at 14:27, Nick Dimiduk <ndimi...@apache.org> > >> wrote: > >> > > > >> > > > FYI, this was me via [0], [1]. It's not actually what I want... > >> > > > > >> > > > [0]: https://issues.apache.org/jira/browse/HBASE-28325 > >> > > > [1]: https://issues.apache.org/jira/browse/INFRA-25382 > >> > > > > >> > > > On Tue, Jan 23, 2024 at 2:21 PM GitBox <git...@apache.org> wrote: > >> > > > > >> > > >> > >> > > >> The following notification schemes have been changed on hbase by > >> > > ndimiduk: > >> > > >> > >> > > >> - updating scheme jira_options: 'link' -> 'link comment' > >> > > >> > >> > > >> > >> > > >> With regards, > >> > > >> ASF Infra. > >> > > >> > >> > > >> > >> > > >> > >> > > > >> > >
