Agreed folks are going to have bigger problems from Informix than just
this, and in fact I used Informix for years and didn't know they allowed
this.
However, what solution do we have for UPDATE (coll...) = (select val...)
for folks? It is awkward to repeat a query multiple times in an UPDATE.
I think it makes sense to add it only if it adds functionality.
---------------------------------------------------------------------------
Tom Lane wrote:
> Justin Clift <justin@postgresql.org> writes:
> > As a thought, will it add significant maintenance penalties or be
> > detrimental?
>
> Well, yes it will if you look at the big picture. In the past we've
> generally regretted it when we put in nonstandard features just to be
> compatible with some other database. (Tatsuo already pointed out the
> "foo = NULL" fiasco.) And we get ragged on regularly for the non-SQL-
> standard features we've inherited from Berkeley Postgres (eg, the
> implicit-FROM frammish that was under discussion yesterday).
>
> I don't think we're really doing the users any favor either. If they
> want to move to some other database after Postgres, are they likely to
> get that other database to insert a not-very-useful nonstandard syntax?
> Sooner or later they're going to have to bite this bullet, and it may
> as well be sooner. (I can hardly believe that this is the worst
> compatibility issue an ex-Informix user would face, anyhow.)
>
> This is an Informix-ism. It should stay that way.
>
> regards, tom lane
>
> ---------------------------(end of broadcast)---------------------------
> TIP 3: if posting/reading through Usenet, please send an appropriate
> subscribe-nomail command to majordomo@postgresql.org so that your
> message can get through to the mailing list cleanly
>
-- Bruce Momjian | http://candle.pha.pa.us pgman@candle.pha.pa.us | (610)
359-1001+ If your life is a hard drive, | 13 Roberts Road + Christ can be your backup. | Newtown Square,
Pennsylvania19073
