Horton

Re: Problem with default partition pruning

yuzuko <yuzukohosoya@gmail.com>

View thread View on postgr.es
From: yuzuko <yuzukohosoya@gmail.com>
To: Ibrar Ahmed <ibrar.ahmad@gmail.com>
Cc: PostgreSQL Hackers <pgsql-hackers@lists.postgresql.org>
Date: 2019-03-05T09:13:32Z
Lists: pgsql-hackers 
Hi Ibrar,

On Tue, Mar 5, 2019 at 2:37 AM Ibrar Ahmed <ibrar.ahmad@gmail.com> wrote:
>
> Hi  Yuzuko Hosoya,
>
> Ignore my last message, I think this is also a legitimate scan on default partition.
>
Oh, I got it.  Thanks a lot.

>
> On Mon, Mar 4, 2019 at 10:29 PM Ibrar Ahmed <ibrar.ahmad@gmail.com> wrote:
>>
>> Hi
>>
>> Patch work fine to me, but I have one test case where default partition still scanned.
>>
>> postgres=# explain select * from test1 where (id < 10) and true;
>>                             QUERY PLAN
>> -------------------------------------------------------------------
>>  Append  (cost=0.00..55.98 rows=846 width=36)
>>    ->  Seq Scan on test1_1  (cost=0.00..25.88 rows=423 width=36)
>>          Filter: (id < 10)
>>    ->  Seq Scan on test1_def  (cost=0.00..25.88 rows=423 width=36)
>>          Filter: (id < 10)
>> (5 rows)
>
>
>
> --
> Ibrar Ahmed

-- 
Best regards,
Yuzuko Hosoya
NTT Open Source Software Center

Commits

  1. Don't constraint-exclude partitioned tables as much

  2. Apply constraint exclusion more generally in partitioning

  3. Improve pruning of a default partition

  4. Doc: Fix event trigger firing table

  5. Remove obsolete nbtree insertion comment.