I thought about that, create a fake branch for fake commits... I just didn't like it, that's why i didn't suggest it.
On Artemis we usually get a good response on PRs... there was just one exception that the guy never came back and I used the empty commit here: https://github.com/apache/activemq-artemis/commit/224770168bdf1b255eff2fac2464d08fa674b3cc The good thing is that we have it registered (why the commit was not merged). but if you have too many PRs, it would get annoying. On Mon, Oct 24, 2016 at 4:28 PM, John D. Ament <johndam...@apache.org> wrote: > You can also push up the commits to a separate branch, which closes the PR, > then delete that branch (note you have to eventually push a real commit on > top so there will be a short delay until someone commits to git in a > non-delete fashion). > > John > > On Mon, Oct 24, 2016 at 4:25 PM Clebert Suconic <clebert.suco...@gmail.com> > wrote: > >> On Mon, Oct 24, 2016 at 10:39 AM, Christopher Shannon >> <christopher.l.shan...@gmail.com> wrote: >> > I don't really like the idea of empty commits laying around. It would be >> > nice if infra could give us permission somehow to actually close these >> > commits on github. >> >> +1 >> >> although I can see why Infra is doing that. >> >> there's a fork update through the apache bot (the github fork is in >> sync with the apache git), and if someone pushes the "merge" button on >> github by accident, a lot of stuff would probably be broken? >> >> There's no way (afaik) to only permit closing a PR. You would get all >> or nothing. >> >> >> But if we could have at least one person on our group with that power. >> (Bruce maybe?) we would speed up the process IMO. >> -- Clebert Suconic
