[Python-Dev] Tracker reviews workflow and flags

[R. David Murray]

On Fri, 19 Mar 2010 23:20:52 -0400, Alexander Belopolsky <alexander.belopolsky at gmail.com> wrote:
> I will not go into details here beyond referring to
> http://bugs.python.org/issue8154, but if you follow the link, you'll
> see that there was not a consensus on how the issue should be
> addressed and even whether or not it was a bug.  Nevertheless the
> patch was committed both to the trunk and to 2.6 without any answer to
> my concerns and without  even an rNNN link to the committed revision.
> 
> I think it would be nice if committers would not cut the discussion
> short without at least a note explaining their decision.

I think the reason this happened in this bug is that it *appeared*
as though the same problem had been encountered and already fixed in
python3, and all that was needed was a backport.  We were groping our
way toward deciding that that wasn't actually the case when Matthias
committed the patch.  Possibly my fault, since I said that I agreed that
it should be fixed.  As I've now noted on the tracker (and in the Ubuntu
tracker just for good measure), I don't think that commit was
appropriate, at least for 2.6, and probably not for 2.7 either.

I also hope that Matthias will put in commit numbers in the future,
since it is very helpful.  Even better, I hope that we can automate this
after the switch to Mercurial (but someone will need to write the code
to do it, of course...)

--David