Question: this only applies for *approvals*, not for *refusals*, right?; I mean, the *merge pull request* button will remain blocked if there were some changes requested by reviewers and the author of the PR adds new commits (either addressing those requested changes or not)?. If the answer to the above is "yes", then +1.
On Wed, Oct 30, 2019 at 1:44 AM Nabarun Nag <n...@apache.org> wrote: > +1 > > On Tue, Oct 29, 2019 at 6:21 PM Darrel Schneider <dschnei...@pivotal.io> > wrote: > > > +1 > > > > On Tue, Oct 29, 2019 at 6:08 PM Owen Nichols <onich...@pivotal.io> > wrote: > > > > > +1 …this has already bitten me a few times > > > > > > > On Oct 29, 2019, at 6:01 PM, Dan Smith <dsm...@pivotal.io> wrote: > > > > > > > > Hi all, > > > > > > > > It seems we've configured our branch protection rules such that > > pushing a > > > > change to a PR that has been approved invalidates the previous > > approval. > > > > > > > > I think we should turn this off - it looks like it's an optional > > feature. > > > > We should trust people to rerequest reviews if needed. Right now this > > is > > > > adding busywork for people to reapprove minor changes (Fixing merge > > > > conflicts, spotless, etc.) > > > > > > > > If you all agree I'll ask infra to uncheck "Dismiss stale pull > request > > > > approvals when new commits are pushed." in our branch protection > rules. > > > > > > > > -Dan > > > > > > > > > -- Ju@N
