pg.ddx.io pgsql-bugs@postgresql.org mailing list archive
help / color / mirror / Atom feed回复：Re: Incorrect sort result caused by ROLLUP and WHERE operation
2+ messages / 1 participants
[nested] [flat]
* 回复：Re: Incorrect sort result caused by ROLLUP and WHERE operation
@ 2025-01-03 01:12 谭忠涛 <zhongtao.tan@seaboxdata.com>
0 siblings, 0 replies; 2+ messages in thread
From: 谭忠涛 @ 2025-01-03 01:12 UTC (permalink / raw)
To: Zhang Mingli <zmlpostgres@gmail.com>; pgsql-bugs <pgsql-bugs@lists.postgresql.org>
"Where a = 1" generates a "Equivalenceclass" containing "a" and "1" ,
but in "grouping sets", a maybe generate nulls, so don't generate a "Equivalenceclass"
when a is not in all subsets of grouping sets.
select a,b,sum(10) as s from t1 where a = 1 group by grouping sets((a,b),(a)) order by a, s desc; -- ok,can generate Equivalenceclass
select a,b,sum(10) as s from t1 where a = 1 group by grouping sets((a,b),(b)) order by a, s desc; -- error, can't generate Equivalenceclass
原谅我蹩脚的英文，我们分析了原因，大致就是“a=1”这个条件生成等价类了，把a当成1处理了，但是实际上在带有grouping sets的sql中，
如果grouping sets的子集中有不包含a的group项，则a会被替换成null值，因此结果会错误；
我们的修改方案是生成等价类的时候，判断一下a是否在所有的grouping sets的子集都存在，如果有不存在的则不生产等价类。
谢谢。
Regards.
Zhang Mingli<zmlpostgres@gmail.com> 在 2025年1月2日 周四 18:50 写道：
Hi,
Zhang Mingli www.hashdata.xyz
Do you mean order of null value (column a) is wrong?
Use NULLS FIRST/LAST to order null values.
On Jan 2, 2025 at 18:19 +0800, 谭忠涛 <zhongtao.tan@seaboxdata.com>, wrote:
Case:
create table t1(a int, b int);
insert into t1 values(1,1),(1,2);
select a,b,sum(10) as s from t1 where a = 1 group by rollup(a,b) order by a, s desc;
^ permalink raw reply [nested|flat] 2+ messages in thread
* 回复：Re: Incorrect sort result caused by ROLLUP and WHERE operation
@ 2025-01-03 01:17 谭忠涛 <zhongtao.tan@seaboxdata.com>
0 siblings, 0 replies; 2+ messages in thread
From: 谭忠涛 @ 2025-01-03 01:17 UTC (permalink / raw)
To: Tom Lane <tgl@sss.pgh.pa.us>; +Cc: pgsql-bugs <pgsql-bugs@lists.postgresql.org>
Thank you for your feedback.
regards
Tom Lane<tgl@sss.pgh.pa.us> 在 2025年1月3日 周五 0:04 写道：
"=?utf-8?B?6LCt5b+g5rab?=" <zhongtao.tan@seaboxdata.com> writes:
> create table t1(a int, b int);
> insert into t1 values(1,1),(1,2);
> select a,b,sum(10) as s from t1 where a = 1 group by rollup(a,b) order by a, s desc;
Yeah. This is fixed in HEAD (v18-to-be) but the fix is too invasive
to consider back-patching. Basically the problem is that older
versions don't understand that the post-ROLLUP value of "a" can be
different from the pre-ROLLUP value, so they think the "where a = 1"
clause removes any need to sort by "a".
As a workaround you could write something like "order by a+0, s desc"
to fool the optimizer into considering the ordering column to be
different from the value that's constrained by WHERE.
regards, tom lane
^ permalink raw reply [nested|flat] 2+ messages in thread
end of thread, other threads:[~2025-01-03 01:17 UTC | newest]
Thread overview: 2+ messages (download: mbox mbox.gz follow: Atom feed)
-- links below jump to the message on this page --
2025-01-03 01:12 回复：Re: Incorrect sort result caused by ROLLUP and WHERE operation 谭忠涛 <zhongtao.tan@seaboxdata.com>
2025-01-03 01:17 回复：Re: Incorrect sort result caused by ROLLUP and WHERE operation 谭忠涛 <zhongtao.tan@seaboxdata.com>
This inbox is served by DDX for PostgreSQL; see mirroring instructions
for how to clone and mirror all data and code used for this inbox