Re: Getting ERROR with FOR UPDATE/SHARE for partitioned table.

From: Amit Langote <amitlangote09(at)gmail(dot)com>
To: amul sul <sulamul(at)gmail(dot)com>
Cc: Rajkumar Raghuwanshi <rajkumar(dot)raghuwanshi(at)enterprisedb(dot)com>, pgsql-hackers <pgsql-hackers(at)postgresql(dot)org>
Subject: Re: Getting ERROR with FOR UPDATE/SHARE for partitioned table.
Date: 2020-05-28 09:36:00
Message-ID: CA+HiwqFd4Suq6XoKHNP9iZw5POBToyOub+f7Dh7urnPZR8wZCQ@mail.gmail.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

On Thu, May 28, 2020 at 1:36 PM amul sul <sulamul(at)gmail(dot)com> wrote:
> On Wed, May 27, 2020 at 12:53 PM Amit Langote <amitlangote09(at)gmail(dot)com> wrote:
>> Actually, if you declare the cursor without FOR SHARE/UPDATE, the case
>> would fail even with traditional inheritance:
>>
>> drop table if exists p cascade;
>> create table p (a int);
>> create table c (check (a = 2)) inherits (p);
>> insert into p values (1);
>> insert into c values (2);
>> begin;
>> declare c cursor for select * from p where a = 1;
>> fetch c;
>> update p set a = a where current of c;
>> ERROR: cursor "c" is not a simply updatable scan of table "c"
>> ROLLBACK
>>
>
> I am not sure I understood the point, you'll see the same error with declarative
> partitioning as well.

My point is that if a table is not present in the cursor's plan, there
is no way for CURRENT OF to access it. Giving an error in that case
seems justified.

OTOH, when the CURRENT OF implementation has RowMarks to look at, it
avoids the error for traditional inheritance children due their
inactive RowMarks being present in the cursor's PlannedStmt. I think
that's only by accident though.

--
Amit Langote
EnterpriseDB: http://www.enterprisedb.com

In response to

Responses

Browse pgsql-hackers by date

  From Date Subject
Next Message Amit Kapila 2020-05-28 09:45:12 Re: PATCH: logical_work_mem and logical streaming of large in-progress transactions
Previous Message Kyotaro Horiguchi 2020-05-28 09:11:39 Re: SIGSEGV from START_REPLICATION 0/XXXXXXX in XLogSendPhysical () at walsender.c:2762