On Wed, Jul 9, 2014 at 1:08 PM, Brady Eidson <[email protected]> wrote:
> > On Jul 9, 2014, at 12:39 PM, Ryosuke Niwa <[email protected]> wrote: > > On Wed, Jul 9, 2014 at 12:35 PM, Tim Horton <[email protected]> > wrote: > >> >> On Jul 9, 2014, at 12:10 PM, Maciej Stachowiak <[email protected]> wrote: >> >> >> Could we teach webkitbot to do an appropriate notification with a waiting >> period? Either as part of rollout or add a new command to do it. >> >> >> It already does. The "waiting period" is defined by when the person who >> asked for the rollout sets the cq+ bit on the rollout patch. >> > > I don't think creating a rollout patch should be the standard method of > notifying the author/reviewer. We should be informing the author/reviewer > ahead of the time. > > > We already have an automated tool that quickly and easily notifies the > author/reviewer, and that tool also happens to create the rollout patch. > > As Tim points out, the rollout patch is never landed unless a reviewer > (usually the person who created the rollout patch) sets the cq+ bit on it. > > I don't see what negative effect the mere existence of the rollout patch > has, or why we should codify into the process that a rollout patch is *not* > created when notifying the author/reviewer. > When the bug for a rollout is created, the original bug is automatically reopened. Also, the bot doesn't provide enough information as to what's breaking because it only takes a single line of description on IRC. It's crucial that whoever reverting a patch provide a detailed explanation on what build or test failed and provide a hyper link to build.webkit.org. Otherwise the original author and the reviewer may have no idea what went wrong. - R. Niwa
_______________________________________________ webkit-dev mailing list [email protected] https://lists.webkit.org/mailman/listinfo/webkit-dev
