Re: PG 18 release notes draft committed

From: Bruce Momjian <bruce(at)momjian(dot)us>
To: "David G(dot) Johnston" <david(dot)g(dot)johnston(at)gmail(dot)com>
Cc: Noah Misch <noah(at)leadboat(dot)com>, PostgreSQL-development <pgsql-hackers(at)lists(dot)postgresql(dot)org>
Subject: Re: PG 18 release notes draft committed
Date: 2025-06-04 21:52:19
Message-ID: aEDAE0590fqvU0Ud@momjian.us
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

On Wed, Jun 4, 2025 at 02:02:38PM -0700, David G. Johnston wrote:
> On Wed, Jun 4, 2025 at 1:45 PM Bruce Momjian <bruce(at)momjian(dot)us> wrote:
>
> Now, if we do want to mention it, it should be done in a way that makes
> it clear to readers whether they are affected by this change.  We can
> try text like:
>
>         Execute non-SECURITY-DEFINER AFTER triggers as the role that was
>         active at the time the trigger was fired
>
>         Previously such triggers were run as the role that was active at
>         commit time.
>
>
> "Deferred constraint triggers now run as the role active when the trigger was
> fired: previously they used the role active when execution began."
>
> The timing is not only at commit, and it makes more sense to me to focus on
> "deferred constraint" instead of the more general "after" trigger.

Ah, yes, I see your point --- it is really only deferred after triggers,
not before triggers, and deferrability is only available for constraint
triggers.

> The trigger doesn't have a security definer clause - the function does and
> would of course take effect during execution.  Not strongly opposed to keeping
> the note.

I think we have to keep the non-SECURITY-DEFINER designation to keep the
text accurate, but you are right it is part of the function, not the
trigger:

Execute deferred constraint triggers attached to
non-SECURITY-DEFINER functions as the role that was active at
the time the trigger was fired

Previously such triggers were run as the role that was active at
commit/execution time.

As you can see, this is going to be hard to read, and I don't know if a
sufficient number of people will care.

--
Bruce Momjian <bruce(at)momjian(dot)us> https://momjian.us
EDB https://enterprisedb.com

Do not let urgent matters crowd out time for investment in the future.

In response to

Responses

Browse pgsql-hackers by date

  From Date Subject
Next Message Bruce Momjian 2025-06-04 21:53:38 Re: PG 18 release notes draft committed
Previous Message Nathan Bossart 2025-06-04 21:42:01 Re: like pg_shmem_allocations, but fine-grained for DSM registry ?