Re: Rethinking the parameter access hooks for plpgsql's benefit

On Mon, Mar 9, 2015 at 4:03 PM, Andres Freund <andres(at)2ndquadrant(dot)com> wrote:
> FWIW, I think you actually don't have much reason to complain. This work
> has probably gotten more attention in total than any other recent
> patch. Certainly, by far, more than any other in the 9.5 cycle.

That has to be true, because the patch has been around in various
forms for so long.

> So far I've not seen a single version that could be considered 'ready
> for committer'. Even if there's points to be resolved you can make one
> (presumably yours) solution ready.

Of course there isn't, since (for example, hint hint) the logical
decoding stuff isn't fully resolved (there is really only one other
issue with unique index inference). Those issues are the only two real
impediments to at least committing ON CONFLICT IGNORE that I'm aware
of, and the former is by far the biggest.

My frustration is down to not having anything I can do without
feedback on these issues...I've run out of things to do without
feedback entirely. It's not for me to decide whether or not I'm
justified in complaining about that. I don't think it matters much
either way, but as a matter of fact I am feeling very concerned about
it. It would be very unfortunate if the UPSERT patch missed the 9.5
release, and for reasons that don't have much to do with me in
particular.
--
Peter Geoghegan