Re: Bug in PL/pgSQL GET DIAGNOSTICS?

Manfred Koizar wrote:
> On Wed, 25 Sep 2002 21:40:03 -0400 (EDT), Bruce Momjian
> <pgman(at)candle(dot)pha(dot)pa(dot)us> wrote:
> >Item 3 is the most controversial. Some say sum all tuple counts, i.e.
> >sum INSERT/UPDATE/DELETE. That just seems to messy to me. I think
> >summing only the matching tags has the highest probability of returning
> >a meaningful number.
>
> [Trying to keep it short this time]
>
> I still believe that there is more than one correct answer; it just
> depends on what the dba intends. So I proposed a syntax change for
> letting the dba explicitly mark the statements she/he wants to affect
> tuple count and oid.
>
> -> http://archives.postgresql.org/pgsql-hackers/2002-09/msg00720.php
>
> Unfortunately I tried to summarize all other proposals and the mail
> got so long that nobody read it to the end :-(

That is an interesting idea; some syntax in the rule that marks the
items. The one downside to that is the fact the rule writer has to
make adjustments. Perhaps we could implement the behavoir I described
and add such tagging later.

--
Bruce Momjian | http://candle.pha.pa.us
pgman(at)candle(dot)pha(dot)pa(dot)us | (610) 359-1001
+ If your life is a hard drive, | 13 Roberts Road
+ Christ can be your backup. | Newtown Square, Pennsylvania 19073