| From: | Jean-Christian Imbeault <jc(at)mega-bucks(dot)co(dot)jp> |
|---|---|
| To: | pgsql-general <pgsql-general(at)postgresql(dot)org> |
| Subject: | Re: IN vs EXIIST |
| Date: | 2002-09-19 11:04:53 |
| Message-ID: | 3D89AF55.4000107@mega-bucks.co.jp |
| Views: | Whole Thread | Raw Message | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-general |
Henshall, Stuart - WCP wrote:
>
> AND invoice_li.invoice_id=sq_inv_li.invoice_id
> is the bit that tells it to check the id of the sub query and parent are
> the same.
> If you don't have this bit I believe it makes the subquery kind of
> piontless
You are right. I had "ordered" my data set in between running the
queries. When I added some new data running the EXISTS query without the
AND a.id=b.id bit the results were fast *and* wrong ;)
I'm still scratching my head as to why EXISTS isn't faster *and* why
EXPLAIN says it should be ...
Jc
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Jan Weerts | 2002-09-19 11:09:26 | Re: IN vs EXIIST |
| Previous Message | Jochem van Dieten | 2002-09-19 10:59:55 | Re: IN vs EXIIST |