[Webkit-unassigned] [Bug 135104] Commit Queue clears the review flag, making it hard to tell whether a patch was reviewed and by whom

[bugzilla-daemon at webkit.org]

https://bugs.webkit.org/show_bug.cgi?id=135104





--- Comment #2 from Alexey Proskuryakov <ap at webkit.org>  2014-07-21 09:31:24 PST ---
When landing manually, I never remove the review+ flag. Never observed any problems caused by this.

Rollouts and having multiple patches in one bug already confuse the tools, so I don't think that these make a very strong argument. I'd like to respond to these in more detail though:

1. Rollout.

I think that when the bug is re-opened due to rollout, it's even more important to have review history intact. It's essentially the same as if a problem was discovered manually, and the patch was never landed.

2. r+ with comments.

Not sure what the concern here is. Uploading a new patch obsoletes the old one.

3. Multiple patches.

I think that the only issue scripts would have with that is that the strange logic to not close the bug with more than one r+ed patch would be broken. This seems acceptable to me, assuming that commit queue will just close the bugs regardless. If someone really needs multiple fixes in a bug, that's an exceptional situation, and the burden is on them to re-open the bug, or to use appropriate arguments with webkit-patch.

-- 
Configure bugmail: https://bugs.webkit.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.