From: Tom Lane <tgl@sss.pgh.pa.us>
To: Pavel Borisov <pashkin.elfe@gmail.com>
Cc: Emre Hasegeli <emre@hasegeli.com>
Cc: Michael Paquier <michael@paquier.xyz>
Cc: PostgreSQL Hackers <pgsql-hackers@lists.postgresql.org>
Subject: Re: Bogus documentation for bogus geometric operators
Date: Thu, 12 Nov 2020 17:16:33 -0500
Message-ID: <235089.1605219393@sss.pgh.pa.us> (raw)
In-Reply-To: <CALT9ZEGWzCe4CSFh1BcyZegYmxBLBgM-aUUBNizcdnqHxGTqvw@mail.gmail.com>
References: <24348.1587444160@sss.pgh.pa.us>
<CAE2gYzzFb6kPJn1qTewPbUnq9O7Rn_4A+KcwD0npk3dht4_cvA@mail.gmail.com>
<20200905014522.GA17191@paquier.xyz>
<CAE2gYzx50zddo3pbf+XN0oiTF0XOyNoLWc86SM3AbO+JKn+4qA@mail.gmail.com>
<CALT9ZEE3jHRk4be3bq6swsUimU9OKVPBJdq49Z2RkXSwVthHEQ@mail.gmail.com>
<CALT9ZEFVqr1Yk9Nixa9UH2z+7rac-wNDDUobQRX7cd2nZJ3oiw@mail.gmail.com>
<CALT9ZEG3f4dUO9oY5e_f01_YWgtca7J2Qa4ZsG-vz6KP6q7EjQ@mail.gmail.com>
<CAE2gYzzu5fozKg_o3Ns0NN2+PrAy3465QJQ18F8giuz2JJcfeA@mail.gmail.com>
<CALT9ZEGWzCe4CSFh1BcyZegYmxBLBgM-aUUBNizcdnqHxGTqvw@mail.gmail.com>
Pavel Borisov <pashkin.elfe@gmail.com> writes:
> [ v4-0001-Deprecate-and-replace-and-operators-for-points.patch ]
I made a cursory pass over this, and two things stood out to me:
1. The patch removes <^ and >^ from func.sgml, which is fine, but
shouldn't there be an addition for the new operators? (I think
probably this need only be an addition of "point" as a possible
input type for <<| and |>>.) Actually, as long we're not completely
removing these operators, I'm not sure it's okay to make them completely
undocumented. Maybe instead of removing, change the text to be
"Deprecated, use the equivalent XXX operator instead." Or we could
add a footnote similar to what was there for a previous renaming:
Note
Before PostgreSQL 8.2, the containment operators @> and <@ were
respectively called ~ and @. These names are still available, but
are deprecated and will eventually be removed.
2. I'm a bit queasy about removing these operators from the opclasses.
I'm not sure anyone will thank us for "the operator is still there, so
your query is still accepted, but it runs 1000X slower than before".
It seems like more plausible answers are either "nuke the operators
entirely, force people to change their queries now" or else "support
both old and new names in the opclasses for awhile to come". In
previous operator renamings we've generally followed the second path,
so that's what I'm inclined to think should happen here.
regards, tom lane
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Reply to all the recipients using the --to and --cc options:
reply via email
To: pgsql-hackers@postgresql.org
Cc: tgl@sss.pgh.pa.us, pashkin.elfe@gmail.com, emre@hasegeli.com, michael@paquier.xyz, pgsql-hackers@lists.postgresql.org
Subject: Re: Bogus documentation for bogus geometric operators
In-Reply-To: <235089.1605219393@sss.pgh.pa.us>
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
This inbox is served by DDX for PostgreSQL; see mirroring instructions
for how to clone and mirror all data and code used for this inbox