MIME-Version: 1.0
References: <19435-3cc1a87f291129f1@postgresql.org> <CAHGQGwEEHFnH8DsZbwxdZeiHozm2LCRTLSgL8Qn=5MoN4450ZA@mail.gmail.com>
 <CAPpHfdsDQhDqvqi+-UxvyYRP72ASkdWzCn43e2Hj6WsY15opvA@mail.gmail.com> <CAHewXNmU3E0oXLgvoOtXw7kfTiAZsKu8La_hQ=pyScgxvP45iw@mail.gmail.com>
In-Reply-To: <CAHewXNmU3E0oXLgvoOtXw7kfTiAZsKu8La_hQ=pyScgxvP45iw@mail.gmail.com>
From: Tender Wang <tndrwang@gmail.com>
Date: Tue, 17 Mar 2026 21:19:45 +0800
Message-ID: <CAHewXN=LjuWz3PcyhjdbJAyo+Zs9MisPDRYnSZBUy4PMeKi+zA@mail.gmail.com>
Subject: Re: BUG #19435: Error: "No relation entry for relid 2" Triggered by
 Complex Join with Self-Referencing Tables
To: Alexander Korotkov <aekorotkov@gmail.com>
Cc: Fujii Masao <masao.fujii@gmail.com>, ammmkilo@163.com, 
	pgsql-bugs@lists.postgresql.org
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
Archived-At: <https://www.postgresql.org/message-id/CAHewXN%3DLjuWz3PcyhjdbJAyo%2BZs9MisPDRYnSZBUy4PMeKi%2BzA%40mail.gmail.com>
Precedence: bulk

Tender Wang <tndrwang@gmail.com> =E4=BA=8E2026=E5=B9=B43=E6=9C=8817=E6=97=
=A5=E5=91=A8=E4=BA=8C 20:59=E5=86=99=E9=81=93=EF=BC=9A
>
> The error was reported in rebuild_joinclause_attr_needed() when
> processing Relid =3D 1(rtindex =3D1),
> When processing its joininfo" ON tom1.col_bool IS NOT NULL",
> (gdb) pgprint rinfo->clause
> Var [varno=3D2 varattno=3D2 vartype=3D16
> varreturningtype=3DVAR_RETURNING_DEFAULT varnosyn=3D2 varattnosyn=3D2]
>
> The varno=3D2, rtindex=3D2(tom1) has been removed. In
> add_vars_to_attr_needed(), to find the base_rel, but the
> root->simple_rel_array[2] is NULL.
> So the error is reporting.
> It seems the joininfo should be replaced by rtindex =3D 3, because the
> rtindex=3D2 would be removed.
> --
(gdb) pgprint rinfo
RestrictInfo [is_pushed_down=3Dfalse can_join=3Dfalse pseudoconstant=3Dfals=
e
has_clone=3Dtrue is_clone=3Dfalse leakproof=3Dfalse
has_volatile=3DVOLATILITY_UNKNOWN security_level=3D0
              num_base_rels=3D1 rinfo_serial=3D4 eval_cost=3D{startup =3D -=
1,
per_tuple =3D 0} norm_selec=3D-1 outer_selec=3D-1 outer_is_left=3Dfalse
hashjoinoperator=3D0 left_bucketsize=3D-1
              right_bucketsize=3D-1 left_mcvfreq=3D-1 right_mcvfreq=3D-1
left_hasheqoperator=3D0 right_hasheqoperator=3D0]
[clause] Var [varno=3D2 varattno=3D2 vartype=3D16
varreturningtype=3DVAR_RETURNING_DEFAULT varnosyn=3D2 varattnosyn=3D2]
[clause_relids] Bitmapset [3]
[required_relids] Bitmapset [3 1]
[incompatible_relids] Bitmapset [7 6]
[outer_relids] Bitmapset [6 5 3]

The above is the joininfo of the rtindex=3D1(tom0),  we can see that the
required_relids is changed to [3 1], but the clause is still
rtindex=3D2(varno=3D2).
I guess the current logic in remove_self_join_rel() may forget to
process the rinfo->clause.


--=20
Thanks,
Tender Wang