MIME-Version: 1.0
References: <CAAJ_b96Jit0hu0KYNDUM6L7wg2hYRjXu2VOGSBgZT8LGZOTcJA@mail.gmail.com>
 <202502271118.3jopzlc5yfkn@alvherre.pgsql>
In-Reply-To: <202502271118.3jopzlc5yfkn@alvherre.pgsql>
From: Amul Sul <sulamul@gmail.com>
Date: Fri, 28 Feb 2025 10:27:58 +0530
Message-ID: <CAAJ_b94OUAXbPLJ--N4-k7BU9yLDvpvnFXjVRVG0zNcamcEAgw@mail.gmail.com>
Subject: Re: NOT ENFORCED constraint feature
To: =?UTF-8?Q?=C3=81lvaro_Herrera?= <alvherre@alvh.no-ip.org>
Cc: Peter Eisentraut <peter@eisentraut.org>, Ashutosh Bapat <ashutosh.bapat.oss@gmail.com>, 
	jian he <jian.universality@gmail.com>, 
	PostgreSQL-development <pgsql-hackers@postgresql.org>, Joel Jacobson <joel@compiler.org>, 
	Suraj Kharage <suraj.kharage@enterprisedb.com>
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
Archived-At: <https://www.postgresql.org/message-id/CAAJ_b94OUAXbPLJ--N4-k7BU9yLDvpvnFXjVRVG0zNcamcEAgw%40mail.gmail.com>
Precedence: bulk

On Thu, Feb 27, 2025 at 4:48=E2=80=AFPM =C3=81lvaro Herrera <alvherre@alvh.=
no-ip.org> wrote:
>
> On 2025-Feb-27, Amul Sul wrote:
>
> > Attached is the rebased patch set against the latest master head,
> > which also includes a *new* refactoring patch (0001). In this patch,
> > I=E2=80=99ve re-added ATExecAlterChildConstr(), which is required for t=
he main
> > feature patch (0008) to handle recursion from different places while
> > altering enforceability.
>
> I think you refer to ATExecAlterConstrEnforceability, which claims
> (falsely) that it is a subroutine to ATExecAlterConstrRecurse; in
> reality it is called from ATExecAlterConstraintInternal or at least
> that's what I see in your 0008.  So I wonder if you haven't confused
> yourself here.  If nothing else, that comments needs fixed.  I didn't
> review these patches.
>

Yeah, that was intentional. I wanted to avoid recursion again by
hitting ATExecAlterChildConstr() at the end of
ATExecAlterConstraintInternal(). Also, I realized the value doesn=E2=80=99t
matter since recurse =3D false is explicitly set inside the
cmdcon->alterEnforceability condition. I wasn=E2=80=99t fully satisfied wit=
h
how we handled the recursion decision (code design), so I=E2=80=99ll give i=
t
more thought. If I don=E2=80=99t find a better approach, I=E2=80=99ll add c=
learer
comments to explain the reasoning.


Regards,
Amul