Received: from sss.pgh.pa.us (unknown [192.204.191.242]) by postgresql.org (Postfix) with ESMTP id 3DA384758DA for ; Wed, 10 Apr 2002 22:10:10 -0400 (EDT) Received: from sss2.sss.pgh.pa.us (tgl@localhost [127.0.0.1]) by sss.pgh.pa.us (8.11.4/8.11.4) with ESMTP id g3B2A7F25097; Wed, 10 Apr 2002 22:10:07 -0400 (EDT) To: Damon Cokenias Cc: pgsql-general@postgresql.org Subject: Re: View INSERTs not returning OID in Postgres 7.2.1 In-reply-to: References: <20020319155803.G37657-100000@megazone23.bigpanda.com> Comments: In-reply-to Damon Cokenias message dated "Wed, 10 Apr 2002 14:40:15 -0700" Date: Wed, 10 Apr 2002 22:10:07 -0400 Message-ID: <25094.1018491007@sss.pgh.pa.us> From: Tom Lane X-Archive-Number: 200204/430 X-Sequence-Number: 23146 Damon Cokenias writes: > I recently moved from 7.1.3 to 7.2.1. In doing so, my application > broke. It relies on INSERT returning the OID of the inserted row, > even if the INSERT was on a view. I'm afraid you were relying on a coincidental artifact of the old implementation, namely that you got back the command tag associated with the last command to be physically executed. The current code is careful to return the command tag associated with the original query (here, the insert into shadview) regardless of execution order of additional queries executed by rules. I'm not sure that we can fix your problem without breaking other cases. regards, tom lane