Re: pgsql: New files for MERGE

Simon Riggs <simon(at)2ndquadrant(dot)com> writes:
> On 6 April 2018 at 17:22, Bruce Momjian <bruce(at)momjian(dot)us> wrote:
>> My point was that people didn't ask you to work harder on fixing the
>> patch, but in reverting it. You can work harder on fixing things in the
>> hope they change their minds, but again, that isn't addressing their
>> request.

> If Tom or Andres still feel that their concerns have not been
> addressed over the last few days, I am happy to revert the patch with
> no further discussion from me in this cycle.

FWIW, I still vote to revert. Even if the patch were now perfect,
there is not time for people to satisfy themselves of that, and
we've got lots of other things on our plates.

I'd be glad to participate in a proper review of this when v12
opens. But right now it just seems too rushed, and I have little
confidence in it being right.

regards, tom lane

PS: If you do revert, please wrap it up as a single revert commit,
not a series of half a dozen. You've already put several
non-buildable states into the commit history as a result of this
patch, each one of which is a land mine for git bisect testing.
We don't need more of those. Also, looking at the reverse of the
reversion commit will provide a handy way of seeing the starting
point for future discussion of this patch.