From: | Adrian Klaver <adrian(dot)klaver(at)gmail(dot)com> |
---|---|
To: | pgsql-general(at)postgresql(dot)org |
Cc: | Sim Zacks <sim(at)compulab(dot)co(dot)il> |
Subject: | Re: update with from |
Date: | 2012-01-23 17:10:19 |
Message-ID: | 201201230910.20127.adrian.klaver@gmail.com |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-general |
On Monday, January 23, 2012 7:32:35 am Sim Zacks wrote:
> On 01/23/2012 05:13 PM, Adrian Klaver wrote:
>
> When I throw in code to make the select only return the correct rows
> The select statement takes 9 secs by itself:
> select a.partid,a.deliverywks
> from poparts a where popartid in (
> select b.popartid from poparts b
> join pos c using(poid)
> join stock.lastrfqdateperpart d using(partid)
> where c.isrfq and c.issuedate > d.issuedate-7
> AND b.unitprice > 0::numeric AND b.quantity >= 100::numeric AND
> c.postatusid = ANY (ARRAY[40, 41])
> and b.partid=a.partid
> order by b.partid,b.unitprice, b.deliverywks
> limit 1
> )
To clarify what I posted earlier, my suggestion was based on rewriting the
second query as:
select b.partid,b.deliverywks b.popartid from poparts b
join pos c using(poid)
join stock.lastrfqdateperpart d using(partid)
where c.isrfq and c.issuedate > d.issuedate-7
AND b.unitprice > 0::numeric AND b.quantity >= 100::numeric AND
c.postatusid = ANY (ARRAY[40, 41])
order by b.partid,b.unitprice, b.deliverywks
limit 1
I may be missing the intent of your original query, but I think the above gets
to the same result without the IN.
--
Adrian Klaver
adrian(dot)klaver(at)gmail(dot)com
From | Date | Subject | |
---|---|---|---|
Next Message | Maxim Boguk | 2012-01-23 23:14:58 | Re: Question about (probably wrong) index scan cost for conditional indexes |
Previous Message | Adrian Klaver | 2012-01-23 16:52:06 | Re: update with from |