> -----Original Message-----
> From: Michael Meskes [mailto:meskes@postgresql.org]
> Sent: Wednesday, June 12, 2002 5:41 AM
> To: Dann Corbit
> Cc: pgsql-hackers@postgresql.org
> Subject: Re: [HACKERS] PostGres Doubt
>
>
> On Mon, Jun 10, 2002 at 02:08:22PM -0700, Dann Corbit wrote:
> > ECPG is single threading. Hence, tools written in ECPG are
> a pain in
> > the neck if you want multiple threads of execution. I
> recommend against
>
> Did he say he wants to write a multi-threaded app?
Or run concurrent queries queries at the same time? Or later discover
the need to do so?
> > using it for any purpose except porting a single threading
> project that
> > already uses embedded SQL. The embedded SQL interface for
> PostgreSQL is
> > a disaster.
>
> Oh, that's what I call constructive critizism. I cannot remember you
> filing any bug reports or asking for some special features. Wouldn't
> that be the first step? And not calling other people's work a
> disaster.
I posted the problems to this list long ago. I wanted to use ECPG and
discovered it was a joke. Do a search through the list and you will
find a half dozen complaints.
> > The libpq functions are reentrant. These will be useful
> for just about
> > any project.
>
> Well if they are (I never checked myself) it shouldn't be too
> difficult
> to make ecpg reentrant too.
Then why not do it. I looked at doing it myself, but the implementation
of embedded SQL is totally nonstandard and uses global structures and
fails to use the SQLCA and SQLDA structures properly. It would be a
nightmare to try and fix it.
> > If you are going to completely replace the data in a table, drop the
> > table, create the table, and use the bulk copy interface.
>
> Oh great! Talking about valuable comments. Ever bothered to
> even ask if
> they are using triggers, constraints, etc. before coming with such a
> proposal?
I would assume that they would use their brain.
