Re: Custom geometry, why slow?

From: Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
To: Paul Matthews <plm(at)netspace(dot)net(dot)au>
Cc: pgsql-hackers(at)postgresql(dot)org
Subject: Re: Custom geometry, why slow?
Date: 2009-08-14 14:33:16
Message-ID: 8652.1250260396@sss.pgh.pa.us
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

Paul Matthews <plm(at)netspace(dot)net(dot)au> writes:
> The problem is, according to EXPLAIN, it still wants to do a sequential
> scan and not use the index. Any pointers as to why?

Can you force it with enable_seqscan = off? If so, then positionsel
isn't estimating a small enough number of matches to make an indexscan
look useful. (Which, for a "left of" operator, is hardly surprising.)
If not, there's something wrong with your opfamily declarations.

regards, tom lane

In response to

Browse pgsql-hackers by date

  From Date Subject
Next Message Teodor Sigaev 2009-08-14 14:58:21 Re: CommitFest 2009-07: Remaining Patches
Previous Message Tom Lane 2009-08-14 14:29:27 Re: pg_hba.conf: samehost and samenet