Content-Type: text/plain;
  charset="iso-8859-2"
From: Richard Huxton <dev@archonet.com>
Organization: Archonet Ltd
To: Tomasz Myrta <jasiek@klaster.net>
Subject: Re: virtual table
Date: Mon, 23 Jun 2003 12:01:49 +0100
User-Agent: KMail/1.4.3
Cc: PGSQL-SQL <pgsql-sql@postgresql.org>
References: <3EF61E44.2020601@klaster.net>
	<200306230954.54185.dev@archonet.com>
	<3EF6C3D6.7080904@klaster.net>
In-Reply-To: <3EF6C3D6.7080904@klaster.net>
MIME-Version: 1.0
Content-Transfer-Encoding: quoted-printable
Message-Id: <200306231201.49969.dev@archonet.com>

On Monday 23 Jun 2003 10:09 am, Tomasz Myrta wrote:
> Dnia 2003-06-23 10:54, U=BFytkownik Richard Huxton napisa=B3:
> > Hmm - I think your problem is going to come before that. Any time you do
> > an INSERT, PostgreSQL is going to need to know the types of all the
> > columns involved.
> >
> > For this sort of thing, I try to keep all the related bits (initial
> > function, views, triggers) in the same text-file to encourage me to
> > remember to update them all together.
>
> I also do this - it isn't really difficult with well written scripts. I w=
as
> just wondering if I don't waste my time tracking all the changes.

I know what you mean - except for simple cases I've never managed to automa=
te=20
type-checking my code without running it through PG.

> > It sounds like this table1 is only being used during transfer. Is there
> > any reason why you aren't just inserting the required rows into table2.
>
> My queries are ordinary transportation problems with 0 up to 2 changes. I
> found it's better (for performance reason) to find all possible relations
> as single rows and split them at the end into simple connections. Maybe I
> should dig into this problem again and rewrite my queries...

Not saying you've got the wrong approach. Might be worth a few minutes=20
attention though

--=20
  Richard Huxton