I think we don't need such a complex rule at all. More reviewers on it, more confidence it will give the author. I think there's a chance for reviewers more than one looking into one pull request because this pull request may look interesting to them. I don't worry too much since eventually they will reach a consensus and the pull request get merged by one of the reviewers. What I suggest is to make the rule simple, no pull request can be merged unless it's reviewed one reviewer at least.
Thanks, -Ian. On Fri, Mar 8, 2019 at 10:37 AM Ian Luo <[email protected]> wrote: > I think we don't need such a complex rule at all. More reviewers on it, > more confidence it will give the author. I think there's a chance for > reviewers more than one looking into one pull request because this pull > request may look interesting > > On Thu, Mar 7, 2019 at 5:04 PM Huxing Zhang <[email protected]> wrote: > >> Hi, >> >> When I am looking at the pull request, I found a pull request[1] got >> approved by 2 of our reviewers(committers), but still not getting >> merged. >> >> I am thinking why it is like this. Should we set up community rules >> for thing like this? >> For example, if a pull request has got at least N approval from >> committers, it can be merged, where N can be discussed. The more >> approval it need, the longer process it will take. >> >> For large size pull requests, the reviewer can request another one to >> help on it. >> >> I would suggest to keep it small, N=1. Even the reviewer fails to >> identify the issues, it can be fixed by sending another pull request. >> >> How do you think? >> >> [1] https://github.com/apache/incubator-dubbo/pull/3536 >> >> -- >> Best Regards！ >> Huxing >> >
