Re: Aggregates with non-commutative transition functions

Tom Lane wrote:
> Emmanuel Charpentier <charpent@bacbuc.dyndns.org> writes:

[ A nice optimisation allowing non-commutative aggregates to do what
they're intended to do ... ]

>>Do you plan incorporation in some forthcoming 7.3.x ? Or push it back
>>to 7.4 ?
>
>
> No, I would not risk back-patching this into 7.3.*.  It's not a bug fix.

Right !

> (But having said that, you could get the diffs from the CVS server and
> back-patch to create your own private 7.3 variant, if you can't wait
> for 7.4.  Offhand I do not think there'd be any great difficulty in
> applying the change to 7.3 branch.)

Schuck no ! I'm not good enough at this ... My coding days are behind me,
and I'm currently just a user. BTW, I tend to stay on stable versions. I'm
experimenting with 7.3 on a development machine, but my main production
machine is still 7.2 (I'm staying with Debian woody). The recent backport
of 7.3 to Woody is a Good Thing (TM), but the current 7.3.1 has problems
with ODBC. Oliver Elphick is currently backporting 7.3.2, but it(s not here
yet ...

I'll wait patiently 7.4 ...

Thanks a lot !

                    Emmanuel Charpentier

PS : Did you (or anyone) considered my suggestion for an "alter table alter
column retype" extension ? Thinking of it again, I guess it could have
other benefits and be something else than simple syntactic sugar ...