Re: Target lists can have at most 1664 entries?

On Sat, 5 Jul 2008 08:17:37 -0700
David Fetter <david(at)fetter(dot)org> wrote:

> On Wed, Jul 02, 2008 at 09:22:50AM +0200, Bjørn T Johansen wrote:
> > On Wed, 02 Jul 2008 03:04:04 -0400
> > Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> wrote:
> > > =?UTF-8?Q?Bj=C3=B8rn?= T Johansen <btj(at)havleik(dot)no> writes:
> > > > What does this mean and how can it be fixed?
> > >
> > > Reduce the number of columns in your SELECTs?
> > >
> > > This whiffs to me of excruciatingly bad schema design. How could
> > > you possibly need upwards of a thousand columns in a query result?
> > > IMHO reasonable column counts are O(10), not O(bignum).
> >
> > Well, I do agree but it is not my design and a "fix" in PostgreSQL
> > would be quicker than fixing the design....
>
> That's where you're badly mistaken. Your application is completely
> broken, and trying to adjust everybody else's Postgres to accommodate
> *your* broken application is both selfish and short-sighted. It's
> selfish because you're asking others to do work they don't need to do
> just so you can avoid doing work you need to do, and it's
> short-sighted because your application is guaranteed to be broken in
> lots of other ways if it's broken this way.
>
> Fix the application, and if you can't, find another job where they're
> not being idiots. There are plenty of Postgres-related jobs out
> there.
>
> Cheers,
> David.

Actually, this discussion was finished a long time ago (we are already looking at the Hibernate config and domain modell)..
And btw, I wasn't proposing a change in PostgreSQL, only if there were some config that could be changed to accomodate this...

BTJ