Re: Re: new patch of MERGE (merge_204) & a question about duplicated ctid

On Mon, Jan 3, 2011 at 1:18 PM, Kevin Grittner
<Kevin(dot)Grittner(at)wicourts(dot)gov> wrote:
> Robert Haas <robertmhaas(at)gmail(dot)com> wrote:
>
>> Users hate having to do explicit locking (especially users whose
>> names rhyme with Bevin Bittner)
>
> :-)
>
> Before you decide to taunt me again, I guess I should point out a
> few things here.

Sorry, that was intended as good-natured humor, not taunting. I think
that the work you are doing on the serializability stuff is *exactly*
the right fix for the concurrency issues associated with MERGE.
Coming up with a fix that is specific to MERGE doesn't impress me
much. I don't believe that hacking up MERGE will lead to anything
other than an ugly mess; it's just a syntax wrapper around an
operation that's fundamentally not too easy to make concurrent. SSI
will handle it, though, along with, well, all the other cases that are
worth worrying about. I don't have quite as much of an allergy to
explicit locking as you do, but I'm quite clear that it isn't nearly
as good as "it just works".

> Should SSI and MERGE both make it into 9.1, [...]

So far the thread on large patches has lead to a status report from
most of the people working on large patches, and no volunteers to take
the lead on reviewing/committing any of them. Although I think both
of those patches are worthwhile, and although I intend to spend a
very, very large amount of time doing CF work in the next 43 days, I
don't foresee committing either of them, and I probably will not have
time for a detailed review of either one, either. I feel pretty bad
about that, but I just don't have any more bandwidth. :-(

--
Robert Haas
EnterpriseDB: http://www.enterprisedb.com
The Enterprise PostgreSQL Company