Content-Type: text/plain;
	charset=utf-8
Mime-Version: 1.0 (Mac OS X Mail 16.0 \(3696.120.41.1.2\))
Subject: Re: GUC for temporarily disabling event triggers
From: Daniel Gustafsson <daniel@yesql.se>
In-Reply-To: <CA+TgmoYJdpN6AJ87q0sEY+NEee9SkMyLp-W-kSdkYpwi6yP3+w@mail.gmail.com>
Date: Wed, 5 Apr 2023 20:43:07 +0200
Cc: Michael Paquier <michael@paquier.xyz>,
 Justin Pryzby <pryzby@telsasoft.com>,
 Mikhail Gribkov <youzhick@gmail.com>,
 pgsql-hackers@lists.postgresql.org
Content-Transfer-Encoding: quoted-printable
Message-Id: <7781ABFD-D273-4CC8-A4D6-BC6ABB111058@yesql.se>
References: <9140106E-F9BF-4D85-8FC8-F2D3C094A6D9@yesql.se>
 <CALte62x7c9R0ohKcQSr5N9PFB-D6pCb_THhwWzTscR7ZB6OHvg@mail.gmail.com>
 <167482801126.1114.1095311628314619931.pgcf@coridan.postgresql.org>
 <8D26ACC2-7669-460D-833A-582EAD6CC09E@yesql.se>
 <ZCncJGbFLex2f9GP@telsasoft.com>
 <E241BCDE-A8BA-483A-B368-F758CB8E0602@yesql.se>
 <CA+Tgmoa7ODAMvJn8sncmMzxkHL=hKmqXa3TLqh-SCLmiB64qzQ@mail.gmail.com>
 <54E8A181-D611-4581-9CC4-62188D1379C4@yesql.se>
 <CA+TgmoZDjc04aPD0LM4LKZ_GstggvAzxmmhWdeX0iHrnC1V2Yg@mail.gmail.com>
 <AEFDB346-50F7-45BB-AE0C-3DCA931ACA8C@yesql.se>
 <ZC0s+BRMqRupDInQ@paquier.xyz>
 <1FF33910-88DB-43B6-AA77-4D04804289FC@yesql.se>
 <CA+TgmoYJdpN6AJ87q0sEY+NEee9SkMyLp-W-kSdkYpwi6yP3+w@mail.gmail.com>
To: Robert Haas <robertmhaas@gmail.com>
Archived-At: <https://www.postgresql.org/message-id/7781ABFD-D273-4CC8-A4D6-BC6ABB111058%40yesql.se>
Precedence: bulk

> On 5 Apr 2023, at 16:30, Robert Haas <robertmhaas@gmail.com> wrote:
>=20
> On Wed, Apr 5, 2023 at 4:57=E2=80=AFAM Daniel Gustafsson =
<daniel@yesql.se> wrote:
>>> - Being able to write a list of event triggers working would be much
>>> more interesting than just individual elements.
>>> - There may be an argument for negated patterns?  Say,
>>> a "!sql_drop,!ddl_command_start" would cause sql_drop and
>>> ddl_command_start to be disabled with all the others enabled, and =
one
>>> should not ne able to mix negated and non-negated patterns.
>>=20
>> I'm not convinced that it's in our interest to offer a GUC to =
configure the
>> cluster by selectively turning off SQL features.  The ones we have =
for planner
>> tuning which is a different beast.  At the very least it should be in =
a thread
>> covering that topic, as it might be a bit hidden here.
>=20
> But, isn't that exactly what you're proposing?

Yeah, but it didn't really strike me until after typing and sending that =
email.

> I mean if this was just event_triggers =3D on | off it would be =
exactly
> like row_security and as far as I'm concerned there would be nothing
> to debate.

Which is what v1-v3 did until I changed it based on review input, but I =
agree
with the reasoning here so will revert back (with some internal changes =
too).
Moving this to the next CF for another stab at it when the tree =
re-opens.

--
Daniel Gustafsson