Feedback-ID: i0fe9450f:Fastmail
Date: Wed, 11 Sep 2024 15:00:31 +0900
From: Michael Paquier <michael@paquier.xyz>
To: Laurenz Albe <laurenz.albe@cybertec.at>
Cc: Greg Sabino Mullane <htamfids@gmail.com>,
	pgsql-docs <pgsql-docs@lists.postgresql.org>
Subject: Re: Table rewrite supporting functions for event triggers
Message-ID: <ZuEx_09-ohYExkJG@paquier.xyz>
References: 
 <CAKAnmmL+Z6j-C8dAx1tVrnBmZJu+BSoc68WSg3sR+CVNjBCqbw@mail.gmail.com>
 <718fcf4a067b576b501bd5b18b54a1fdc4fda639.camel@cybertec.at>
 <CAKAnmm+SfSEg+bmWqT5Q9BSTU3FGaSquVebrQ+8jg3Dq-_1vxg@mail.gmail.com>
 <CAKAnmm+NKVhZMjNN8okbPHXMxSb0Ngqhxgt4hEMGU=RWq9fnKA@mail.gmail.com>
 <c264f34792b17f5bd30133071cfa252919bf5323.camel@cybertec.at>
MIME-Version: 1.0
Content-Type: multipart/signed; micalg=pgp-sha512;
	protocol="application/pgp-signature"; boundary="MteqK8VgG19KIocY"
Content-Disposition: inline
In-Reply-To: <c264f34792b17f5bd30133071cfa252919bf5323.camel@cybertec.at>
Archived-At: 
 <https://www.postgresql.org/message-id/ZuEx_09-ohYExkJG%40paquier.xyz>
Precedence: bulk


--MteqK8VgG19KIocY
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable

On Tue, Sep 03, 2024 at 09:34:02PM +0200, Laurenz Albe wrote:
> On Tue, 2024-09-03 at 11:54 -0400, Greg Sabino Mullane wrote:
>> How about something like this?
>=20
> This patch looks good to me.

-        Returns a code explaining the reason(s) for rewriting.  The exact
-        meaning of the codes is release dependent.
+        Returns a code explaining the reason(s) for rewriting. The value is
+        a bitmap built from the following values: 1 (the table has changed
+        persistence), 2 (a column has changed a default value), 4 (a column
+        has a new data type), and 8 (the table access method has changed).

Agreed that the user experience with this function is poor and that
the documentation should be improved.  Still, I am not sure that this
is optimal.  On top of the values, how about adding the variable names
and also mention that these are defined in event_trigger.h?

Putting the documentation change aside for a bit, could it be better
to redesign this function and return a text value rather than an
integer?  We could directly return the names, minus "AT_REWRITE_", for
instance.
--
Michael

--MteqK8VgG19KIocY
Content-Type: application/pgp-signature; name="signature.asc"

-----BEGIN PGP SIGNATURE-----

iQIzBAABCgAdFiEEG72nH6vTowiyblFKnvQgOdbyQH0FAmbhMf8ACgkQnvQgOdby
QH1Chw//cC36cSgsFmfItU7jvyLNDw/h+GvSnilTwmtndscQn3UWgne6m2ypNW/Y
1YiW3hhWf0UuCE7d/iJ3lyhYJ4WdkSBFVWkcBgODR3IC4H4uyhgJoT0gSG010lG+
+oQwL+4uTNV0CL619hD+MdyHgqnz1GzFrWE98zlEnhUqf0wAG6TKFP2l3WWYzYsT
GjkrrhywR0nqxDJD80WAHju9uBD3pQ+1dQZR8TyiaZ86UogsYtvqxUPFFY/CNY1z
Fz8jGX/KkVVR1pKF1bttQVNp0hy/vgQ3dsQub9lpTT/Pj9C2ST55F78A141rD/u+
un9HMkdbHDe9W+bkjl/tKepSKAjGyfn0f2cu80pJIjg9g94qWC6ZLs6WHxkDOUQw
LR3ooAf4rV8c8AZ4UQUws6aR+eeYFMsCvqnxFKPue0f+i+RuwTHjqM+JHi+bX20y
8DgiDLEwx8pzjnVIEtWQfOFAyMqh0FdKd1iXVsA6lYXzDgy4bspjDJ9hwka0GcaG
n3wm2OIGWIeywh2QwnHGi8Rpyw2daKsQpWKOuZ0W0eAaOW9SyyYUBA9tJpBDgRrI
clh3WtCQtSJwru0aZzounXewMPTQCQepV+QLDFTbFuPqZsIEgcqjiF3EHuJf55RO
mKOnV/1dNoCV9JGidR1+Mhu89nHlC+XT0MNsj67oiNP9UQFkM8w=
=/f/j
-----END PGP SIGNATURE-----

--MteqK8VgG19KIocY--