Re: Performance for seq. scans

From: Jules Bean <jules(at)jellybean(dot)co(dot)uk>
To: Mitch Vincent <mitch(at)venux(dot)net>
Cc: pgsql-general(at)postgresql(dot)org
Subject: Re: Performance for seq. scans
Date: 2000-07-26 14:58:26
Message-ID: 20000726155825.E30047@grommit.office.vi.net
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-general

On Wed, Jul 26, 2000 at 03:44:47PM +0100, Jules Bean wrote:
>
> OTOH, are you saying that the FTI code in 7.0.2 contrib does in fact
> do the 'index all occuring 2-char substring' trick? I must go and
> investigate that straight away. Many thanks.

OK, I apologise for my skepticism.

I have just read the FTI code, and as far as a quick glance can tell
me, it appears to index, for each 'word' in my main table, all the
final segments of that word. This enables me to match any substring,
since if %bar% matches, then there must be /some/ final segment which
begins with bar, so I can search for bar% in the FTI table.

Rather cleverer than the algorithm I'd thought of ;-)

Thanks to everyone for all your help. Now it just remains to make up
my mind whether I have the disk space for this ;-)

Jules

In response to

Browse pgsql-general by date

  From Date Subject
Next Message Tom Lane 2000-07-26 15:12:08 Re: Some questions on user defined types and functions.
Previous Message Tom Lane 2000-07-26 14:46:27 Re: Is Pg 7.0.x's Locking Mechanism BROKEN?