Date: Thu, 12 Nov 2020 10:40:43 -0300
From: Alvaro Herrera <alvherre@alvh.no-ip.org>
To: "David G. Johnston" <david.g.johnston@gmail.com>
Cc: Matthieu Garrigues <matthieu.garrigues@gmail.com>,
	Michael Paquier <michael@paquier.xyz>,
	Dave Cramer <davecramer@postgres.rocks>,
	PostgreSQL-development <pgsql-hackers@postgresql.org>,
	Craig Ringer <craig@2ndquadrant.com>,
	Vaishnavi Prabakaran <vaishnaviprabakaran@gmail.com>,
	Andres Freund <andres@anarazel.de>,
	Daniel Verite <daniel@manitou-mail.org>,
	Michael Paquier <michael.paquier@gmail.com>
Subject: Re: PATCH: Batch/pipelining support for libpq
Message-ID: <20201112134043.GA19787@alvherre.pgsql>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: 
 <CAKFQuwZTNgNwnHdxQnQnOpKfyX5KW6twgj23ZNUZByDVgmdi7w@mail.gmail.com>
User-Agent: Mutt/1.10.1 (2018-07-13)
Precedence: bulk

(Adding previous reviewers to CC)

On 2020-Nov-03, David G. Johnston wrote:

> Given the caveats around blocking mode connections why not just require
> non-blocking mode, in a similar fashion to how synchronous functions are
> disallowed?

This is a very good question.  Why indeed?  Does anybody have a good
answer to this?  If not, I propose we just require that non-blocking
mode is in use in order for batch mode to be used.

I've been doing a review pass over this patch and have an updated
version, which I intend to share later today (after I fix what appears
to be a misunderstanding in the "singlerow" test in testlibpqbatch.c)