From: | Łukasz Dąbek <sznurek(at)gmail(dot)com> |
---|---|
To: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
Cc: | pgsql-general(at)lists(dot)postgresql(dot)org |
Subject: | Re: Using b-tree index for >= condition when joining |
Date: | 2020-05-17 19:45:33 |
Message-ID: | b0d75b33-2006-15e3-047c-4487520b06cd@gmail.com |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-general |
On 17/05/2020 at 04:22, Tom Lane wrote:> Note that you're not really
getting the same plan that way: it's not
> a left join anymore, because you put a strict constraint on the join's
> inner relation, so the planner realizes it doesn't have to produce any
> null-extended rows. You could make it work with the desired semantics
> with something along the lines of
>
> SELECT * FROM tbl1 t1
> LEFT JOIN (select * from tbl2 where tbl2.date >= '2019-04-21') t2
> USING (date)
> WHERE t1.date >= '2019-04-21';
>
> but of course that's even less easy :-(
Thanks, I didn't realise my version of the query was incorrect.
It seems like there isn't much hope of creating a view equivalent to the
query which would behave reasonably with "date >= CONST" constraint, or
am I missing something? I think I can create a workaround using
functions but would love to know if there's something simpler I could do.
Thanks,
Lukasz
From | Date | Subject | |
---|---|---|---|
Next Message | James Sewell | 2020-05-18 00:32:42 | Re: Inherited an 18TB DB & need to backup |
Previous Message | Weatherby,Gerard | 2020-05-17 16:22:19 | NFS version for replication |