Abhijit Menon-Sen wrote:
> At 2014-07-02 12:52:51 -0700, mail@joeconway.com wrote:
> >
> > Doesn't mean that to me but feel free to change it to Waiting on
> > Author if you prefer :-)
> >
> > Is there any official explanation as to what those states mean
> > documented anywhere?
>
> I don't know if there's an official definition, but my understanding of
> "returned with feedback" was always pretty much in agreement with what
> Fujii wrote. If the author is expected to post a revised patch soon, it
> gets marked "waiting on author", and "returned with feedback" means it
> will take longer, probably in the next CF.
I think the main thing with "returned with feedback" is the patch is not
supposed to be handled any further in the current commitfest. Normally
if a new version is submitted, it's opened as a new entry in a future
commitfest. So it's one of the three final states for a patch, the
other two being "committed" and "rejected". The other status values,
"needs review", "waiting on author", and "ready for committer" are
transient and can change to any other state.
So I disagree with you here:
> But I also treat these labels as a matter of convenience, and definitely
> not some mark of shame where the author should feel upset that the patch
> was put in one state or the other. As far as I'm concerned, patches can
> be switched from "returned with feedback" to "needs review" to "ready
> for committer" at the drop of a hat if updates are made in time.
A patch that is Returned with Feedback is *not* supposed to go back to
"needs review" or any of the other states. If we expect that the author
is going to update the patch, the right state to use is "Waiting on
author".
In any case, no state is a mark of shame on the author. We are not
supposed to judge people here.
--
Álvaro Herrera http://www.2ndQuadrant.com/
PostgreSQL Development, 24x7 Support, Training & Services