Subject: Re: insert without oids
From: Neil Conway <neilc@samurai.com>
To: Michael Stone <mstone+postgres@mathom.us>
Cc: pgsql-performance@postgresql.org
In-Reply-To: <20060113201011.GI1408@mathom.us>
References: <20060113201011.GI1408@mathom.us>
Content-Type: text/plain
Date: Fri, 13 Jan 2006 16:29:15 -0500
Message-Id: <1137187755.2509.6.camel@localhost.localdomain>
Mime-Version: 1.0
Content-Transfer-Encoding: 7bit

On Fri, 2006-01-13 at 15:10 -0500, Michael Stone wrote:
> OIDs seem to be on their way out, and most of the time you can get a
> more helpful result by using a serial primary key anyway, but I wonder
> if there's any extension to INSERT to help identify what unique id a
> newly-inserted key will get? Using OIDs the insert would return the OID
> of the inserted row, which could be useful if you then want to refer to
> that row in a subsequent operation. You could get the same result by
> manually retrieving the next number in the sequence and using that value
> in the insert, but at the cost of additional DB operations.

There's really no additional operations required:

INSERT INTO t1 VALUES (...);
INSERT INTO t2 VALUES (currval('t1_id_seq'), ...);

You need a separate SELECT if you want to use the generated sequence
value outside the database, although the INSERT ... RETURNING extension
will avoid that (there's a patch implementing this, although it is not
yet in CVS).

-Neil