43

My life lately

Comments
  • 1
    the protocol is the same 4 all...
  • 4
    @We3D I was doing the pull request, "Must be approved by at least one approver" Okay, add self as approver
  • 0
    Does bitbucket/gitlab allow this?
    AFAIK Github doesn't.
  • 1
    Yeah, i traditionally skip the charade and just push it right onto staging.

    We are all adults here and test our code locally before we push it up. Code review isn't expected to find anything major, so it happens after. If shit would really hit the fan, we could still just revert. So there isn't really that much a risk and the whole process is slightly more agile this way.
  • 2
    @Oktokolo my two cents, this also depends on the number of people working on and the connectedness of the service.

    In bigger teams with regular activity, with upstream, downstream clients, automated deployment, it just makes sense to have rigorous reviews, because a breakage just wastes everyone's time (the time taken to deploy the initial code, revert and integration tests running twice - in big codebases, this whole process can take a couple of hours), delays other teams regressions.
  • 1
    @beefdead I don't know. We're on azure
  • 0
    *laughs in github org admin*

    If I want to merge an unapproved PR, they can't stop me.
  • 1
    @beefdead gitlab has a checkbox fornthis in the project settings.
    You can actually configure not so few things about approvers and approvals there
Add Comment