As the 1.2.0 release seems close, I wonder if we can merge branches of medium or large size targeting stable-1.2. In my case, it was #4976 . The review was successful, but it's one of those fixes that could have side-effects.
If we merge now, there may not be enough time to test before the 1.2.0 release. If we wait, it will go into 1.2.1, right? What would be appropriate?
In my opinion we should merge the fix and fix the side effects in 1.2.1. 馃槈
stable-1.2 in the first place. That said, if you guys are willing to take on the risk, merge it before release. 馃憤At this point I think it's worth postponing even fixes, assuming they're not fixing severe regressions. We do have a known bugs section in the release notes.
If #4980 is fixed and no worse bugs are found in the meanwhile, I think it's time to release.
If #4980 isn't fixed by June 16th (a month since the issue was opened)... release anyway?
Update: #4980 is fixed for 1.2, and no worse bugs have been found in the meantime.
We might want to wait for #4996
As we are about to release, I'm closing this thread. Feel free to reopen.