MIME-Version: 1.0
In-Reply-To: <5B4F2BE9.3000304@lab.ntt.co.jp>
References: <5AFC0865.8050802@lab.ntt.co.jp>
 <CAFjFpRfGL9w-4C5N-ioaG3m+iheC1ey7_RMZ4vJijGn=wWJyxw@mail.gmail.com>
 <5AFD6580.5090308@lab.ntt.co.jp> <5AFE81CC.8000508@lab.ntt.co.jp>
 <5B17C5EC.1090602@lab.ntt.co.jp>
 <CAFjFpRffA-BfL0RXcUVLLbcOBF1ie7oObTB-RE2q_o45X1HTYg@mail.gmail.com>
 <5B1E5071.1000704@lab.ntt.co.jp> <5B23A964.9020800@lab.ntt.co.jp>
 <5B28FB19.3090809@lab.ntt.co.jp>
 <CA+TgmoZboTszGQk3OecBOZUFNNjzKE-SOo_Q46xtTJ4Yj95-RQ@mail.gmail.com>
 <20180703163519.tsigyic5xefunv74@alap3.anarazel.de>
 <5B3C2A4E.5030901@lab.ntt.co.jp>
 <5B4F2BE9.3000304@lab.ntt.co.jp>
From: Robert Haas <robertmhaas@gmail.com>
Date: Fri, 20 Jul 2018 08:38:09 -0400
Message-ID: 
 <CA+TgmobR9XWEhmF3s_WsoLNnnjaFyS_xtbrVG3dCGiUgof20Nw@mail.gmail.com>
Subject: Re: Expression errors with "FOR UPDATE" and postgres_fdw with
 partition wise join enabled.
To: Etsuro Fujita <fujita.etsuro@lab.ntt.co.jp>
Cc: Andres Freund <andres@anarazel.de>,
 Ashutosh Bapat <ashutosh.bapat@enterprisedb.com>,
	Rajkumar Raghuwanshi <rajkumar.raghuwanshi@enterprisedb.com>,
	pgsql-hackers <pgsql-hackers@postgresql.org>
Content-Type: text/plain; charset="UTF-8"
Precedence: bulk

On Wed, Jul 18, 2018 at 8:00 AM, Etsuro Fujita
<fujita.etsuro@lab.ntt.co.jp> wrote:
> [ new patch ]

+ /*
+ * If the child plan is an Append or MergeAppend, the targetlists of its
+ * subplans would have already been adjusted before we get here, so need
+ * to work here.
+ */
+ if (IsA(subplan, Append) || IsA(subplan, MergeAppend))
+ return;
+
+ /* The child plan should be a scan or join */
+ Assert(is_projection_capable_plan(subplan));

This looks like a terrible design to me.  If somebody in future
invents a new plan type that is not projection-capable, then this
could fail an assertion here and there won't be any simple fix.  And
if you reach here and the child targetlists somehow haven't been
adjusted, then you won't even fail an assertion, you'll just crash (or
maybe return wrong answers).

I'm going to study this some more now, but I really think this is
going in the wrong direction.

-- 
Robert Haas
EnterpriseDB: http://www.enterprisedb.com
The Enterprise PostgreSQL Company