Date: Thu, 12 Mar 2026 11:15:56 +0100
From: Viktor Holmberg <v@viktorh.net>
To: Andrew Dunstan <andrew@dunslane.net>, jian he
 <jian.universality@gmail.com>
Cc: PostgreSQL-development <pgsql-hackers@postgresql.org>
Message-ID: <86e2c66e-f07b-4285-b42e-e3d821e0f0d7@Spark>
In-Reply-To: 
 <CACJufxHomghtudwKpxqbVyy2n+cBPrahJjnpAo2OvRRXN-UE4w@mail.gmail.com>
References: 
 <CACJufxE_+iZBR1i49k_AHigppPwLTJi6km8NOsC7FWvKdEmmXg@mail.gmail.com>
 <229413.1741144420@sss.pgh.pa.us>
 <CACJufxGZqTUqOCmwVveq4TGG6UXrMnODkMw7x5LV4DM4RkNRcw@mail.gmail.com>
 <CACJufxHtbn6pUPt8RRXv8BAFX67K5gyqR5SGwu8P1nSEF=ts9g@mail.gmail.com>
 <CACJufxECb4nVjssRPknepNmqF2CR0ycPzTmJZPxoWYK4JD-mFQ@mail.gmail.com>
 <CACJufxGs_A0mrq9StOFAkVnK9x1U_s0+KuybGKUt2VmspC90Hw@mail.gmail.com>
 <CACJufxGNCw51aeVxXOtgzgbE9J710e91umNzH-56rG5tmtK8mw@mail.gmail.com>
 <CACJufxFdzChzGn659CabeYSJA2fN_NNuZLShqKN5grbxLLK5LA@mail.gmail.com>
 <CACJufxGFP5VJJwPqTZJ_fQq_BKLqD_90JQ49AJH6fPqSmdFSnQ@mail.gmail.com>
 <CACJufxG-NQmEutUnsA3q1RV1iT5MWpaj3MSOH4aVo_UNZUA86g@mail.gmail.com>
 <bba5e471-d122-45e2-8c8b-6421804c2043@dunslane.net>
 <CACJufxE=VV5xvRLUhVP2Oej3SpZs1+Ksr=QUACm4ZUhE9TjSng@mail.gmail.com>
 <5a8c55ea-81d3-47f4-aff9-a89fabcc2304@Spark>
 <31d12caf-17d8-4ab7-af58-ec3c5a310a92@dunslane.net>
 <CACJufxHomghtudwKpxqbVyy2n+cBPrahJjnpAo2OvRRXN-UE4w@mail.gmail.com>
Subject: Re: support fast default for domain with constraints
MIME-Version: 1.0
Content-Type: multipart/alternative; boundary="69b29261_721da317_91cf"
Archived-At: 
 <https://www.postgresql.org/message-id/86e2c66e-f07b-4285-b42e-e3d821e0f0d7%40Spark>
Precedence: bulk

--69b29261_721da317_91cf
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: quoted-printable
Content-Disposition: inline

On 12 Mar 2026 at 04:36 +0100, jian he <jian.universality=40gmail.com>, w=
rote:
> On Thu, Mar 12, 2026 at 3:50=E2=80=AFAM Andrew Dunstan <andrew=40dunsla=
ne.net> wrote:
> > Also added some tests.
>
> V11 looks good to me.
>
>
> On Wed, Mar 11, 2026 at 6:34=E2=80=AFPM Viktor Holmberg <v=40viktorh.ne=
t> wrote:
> >
> > I=E2=80=99ve been burned my this issue in the past so would be great =
if this could get in.
> >
> > + /*
> > + * If the domain has volatile constraints, we must do a table rewrit=
e
> > + * since the constraint result could differ per row and cannot be
> > + * evaluated once and cached as a missing value.
> > + */
> > + if (has=5Fvolatile)
> > + =7B
> > + Assert(has=5Fdomain=5Fconstraints);
> > + tab->rewrite =7C=3D AT=5FREWRITE=5FDE=46AULT=5FVAL;
> > + =7D
> >
> > I'm not sure. But seems to me this makes the pre-existing guard for v=
irtual columns
> > redundant=3F
> > I mean this code on line 7633:
> > if (colDef->generated =21=3D ATTRIBUTE=5FGENERATED=5FVIRTUAL)
> > tab->rewrite =7C=3D AT=5FREWRITE=5FDE=46AULT=5FVAL;
> >
>
> If the first =60if (has=5Fvolatile)=60 is false, then
> > if (colDef->generated =21=3D ATTRIBUTE=5FGENERATED=5FVIRTUAL)
> > tab->rewrite =7C=3D AT=5FREWRITE=5FDE=46AULT=5FVAL;
> is still reachable.
>
> > Also, perhaps virtual generated columns could use a test=3F
>
> Virtual generated columns based on domain are not currently supported.
> I have a patch for it: https://commitfest.postgresql.org/patch/5725
> but it's not doable now because of
> https://git.postgresql.org/cgit/postgresql.git/commit/=3Fid=3D0cd69b3d7=
ef357f2b43258de5831c4de0bd51dec
> You may also be interested in https://commitfest.postgresql.org/patch/5=
907
>
> --
> jian
> https://www.enterprisedb.com/
Nice, v11 looks good to me. I=E2=80=99ll change the status of the commitf=
est entry to ready for committer.

--69b29261_721da317_91cf
Content-Type: text/html; charset="utf-8"
Content-Transfer-Encoding: quoted-printable
Content-Disposition: inline

<html xmlns=3D=22http://www.w3.org/1999/xhtml=22>
<head>
<title></title>
</head>
<body>
<div name=3D=22messageReplySection=22>
<div dir=3D=22auto=22>On 12 Mar 2026 at 04:36 +0100, jian he &lt;jian.uni=
versality=40gmail.com&gt;, wrote:</div>
<blockquote style=3D=22border-left-color: rgb(26, 188, 156); margin: 5px;=
 padding-left: 10px; border-left-width: thin; border-left-style: solid;=22=
>On Thu, Mar 12, 2026 at 3:50=E2=80=AFAM Andrew Dunstan &lt;andrew=40duns=
lane.net&gt; wrote:<br />
<blockquote style=3D=22border-left-color: rgb(230, 126, 34); margin: 5px;=
 padding-left: 10px; border-left-width: thin; border-left-style: solid;=22=
>Also added some tests.</blockquote>
<br />
V11 looks good to me.<br />
<br />
<br />
On Wed, Mar 11, 2026 at 6:34=E2=80=AFPM Viktor Holmberg &lt;v=40viktorh.n=
et&gt; wrote:<br />
<blockquote style=3D=22border-left-color: rgb(230, 126, 34); margin: 5px;=
 padding-left: 10px; border-left-width: thin; border-left-style: solid;=22=
><br />
I=E2=80=99ve been burned my this issue in the past so would be great if t=
his could get in.<br />
<br />
+ /*<br />
+ * If the domain has volatile constraints, we must do a table rewrite<br=
 />
+ * since the constraint result could differ per row and cannot be<br />
+ * evaluated once and cached as a missing value.<br />
+ */<br />
+ if (has=5Fvolatile)<br />
+ =7B<br />
+ Assert(has=5Fdomain=5Fconstraints);<br />
+ tab-&gt;rewrite =7C=3D AT=5FREWRITE=5FDE=46AULT=5FVAL;<br />
+ =7D<br />
<br />
I'm not sure. But seems to me this makes the pre-existing guard for virtu=
al columns<br />
redundant=3F<br />
I mean this code on line 7633:<br />
if (colDef-&gt;generated =21=3D ATTRIBUTE=5FGENERATED=5FVIRTUAL)<br />
tab-&gt;rewrite =7C=3D AT=5FREWRITE=5FDE=46AULT=5FVAL;<br />
<br /></blockquote>
<br />
If the first =60if (has=5Fvolatile)=60 is false, then<br />
<blockquote style=3D=22border-left-color: rgb(230, 126, 34); margin: 5px;=
 padding-left: 10px; border-left-width: thin; border-left-style: solid;=22=
>if (colDef-&gt;generated =21=3D ATTRIBUTE=5FGENERATED=5FVIRTUAL)<br />
tab-&gt;rewrite =7C=3D AT=5FREWRITE=5FDE=46AULT=5FVAL;</blockquote>
is still reachable.<br />
<br />
<blockquote style=3D=22border-left-color: rgb(230, 126, 34); margin: 5px;=
 padding-left: 10px; border-left-width: thin; border-left-style: solid;=22=
>Also, perhaps virtual generated columns could use a test=3F</blockquote>=

<br />
Virtual generated columns based on domain are not currently supported.<br=
 />
I have a patch for it: https://commitfest.postgresql.org/patch/5725<br />=

but it's not doable now because of<br />
https://git.postgresql.org/cgit/postgresql.git/commit/=3Fid=3D0cd69b3d7ef=
357f2b43258de5831c4de0bd51dec<br />
You may also be interested in https://commitfest.postgresql.org/patch/590=
7<br />
<br />
--<br />
jian<br />
https://www.enterprisedb.com/</blockquote>
<div dir=3D=22auto=22>Nice, v11 looks good to me. I=E2=80=99ll change the=
 status of the commitfest entry to ready for committer.</div>
</div>
</body>
</html>

--69b29261_721da317_91cf--