Re: Stale comment in guc.h; publish listing of setting sources?

From: Michael Paquier <michael(at)paquier(dot)xyz>
To: "David G(dot) Johnston" <david(dot)g(dot)johnston(at)gmail(dot)com>
Cc: pgsql-hackers(at)lists(dot)postgresql(dot)org
Subject: Re: Stale comment in guc.h; publish listing of setting sources?
Date: 2025-09-10 23:48:47
Message-ID: aMIOX7WybtV4Lj3s@paquier.xyz
Views: Whole Thread | Raw Message | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

On Wed, Sep 10, 2025 at 01:02:46PM -0700, David G. Johnston wrote:
> * NB: see GucSource_Names in guc.c if you change this.
>
> The constant was moved to guc_tables.c

I would just remove the file reference. Knowing that GucSource_Names
matters is enough to grep for it. Documenting its location does not
matter.

> Also, these seem like they should be documented for pg_settings.source
>
> Per recent complaint:
> https://www.postgresql.org/message-id/04f101dc2288$121b2de0$365189a0$@msym.fr

Yeah, why not. Listing all of them GucSource_Names directly in the
docs may help if that clarifies things regarding some of the values.
The last additions in the list were 9475db3a4eb5 and f480e2944985, if
my grepping gets it right, pointing that this has a low risk of
rotting when a new one is added to the code, and forgotten in the
docs.
--
Michael

In response to

Responses

Browse pgsql-hackers by date

  From Date Subject
Next Message Mihail Nikalayeu 2025-09-10 23:56:00 Logical replication: lost updates/deletes and invalid log messages caused by SnapshotDirty + concurrent updates
Previous Message Michael Paquier 2025-09-10 23:38:44 Re: Only one version can be installed when using extension_control_path