From: Tom Lane <tgl@sss.pgh.pa.us>
To: onni@keksi.io
cc: Laurenz Albe <laurenz.albe@cybertec.at>, pgsql-novice@lists.postgresql.org
Subject: Re: How can I create a feature request for QUALIFY clause into
 PostgreSQL?
In-reply-to: <95FC2266-B9FF-4977-BA27-B5C70CC691D3@keksi.io>
References: <1701225.1664830660@sss.pgh.pa.us>
 <95FC2266-B9FF-4977-BA27-B5C70CC691D3@keksi.io>
Comments: In-reply-to onni@keksi.io
	message dated "Tue, 04 Oct 2022 07:52:03 +0300"
MIME-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Content-ID: <1770927.1664862226.1@sss.pgh.pa.us>
Content-Transfer-Encoding: quoted-printable
Date: Tue, 04 Oct 2022 01:43:46 -0400
Message-ID: <1770928.1664862226@sss.pgh.pa.us>
Archived-At: 
 <https://www.postgresql.org/message-id/1770928.1664862226%40sss.pgh.pa.us>
Precedence: bulk

onni@keksi.io writes:
> This still doesn't answer the original question of how to ask for new fe=
atures in Postgres.

Well, you post on the mailing lists ;-)

I don't think this particular request is going anywhere.  We're not
terribly receptive to non-SQL-standard syntax if there's already
other ways to get the same thing done.  New syntax has large costs:
not only the initial implementation, but documentation, ongoing
maintenance, and incremental slowdown of the parser.  Plus, if
it's something that's not in the SQL standard, there's a big risk
of getting blindsided by incompatible future extensions of the
standard.  Having said all that, really compelling ideas might
seduce us anyway ... but this specific idea seems ugly and
nonintuitive as well as nonstandard.

			regards, tom lane