Feedback-ID: ia2694551:Fastmail
Date: Thu, 22 Dec 2022 20:39:21 +0100
From: Alvaro Herrera <alvherre@alvh.no-ip.org>
To: Peter Eisentraut <peter.eisentraut@enterprisedb.com>
Cc: pgsql-hackers <pgsql-hackers@postgresql.org>
Subject: Re: dynamic result sets support in extended query protocol
Message-ID: <20221222193921.yaosmvvoymncxyl2@alvherre.pgsql>
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Disposition: inline
Content-Transfer-Encoding: 8bit
In-Reply-To: <20221221194120.wttku6u7cbd6e574@alvherre.pgsql>
Archived-At: 
 <https://www.postgresql.org/message-id/20221222193921.yaosmvvoymncxyl2%40alvherre.pgsql>
Precedence: bulk

On 2022-Dec-21, Alvaro Herrera wrote:

> I suppose that in order for this to work, we would have to find another
> way to "advance" the queue that doesn't rely on the status being
> PGASYNC_READY.

I think the way to make this work is to increase the coupling between
fe-exec.c and fe-protocol.c by making the queue advance occur when
CommandComplete is received.  This is likely more correct protocol-wise
than what we're doing now: we would consider the command as done when
the server tells us it is done, rather than relying on internal libpq
state.

-- 
Álvaro Herrera        Breisgau, Deutschland  —  https://www.EnterpriseDB.com/