From: | "Ibrahim Tekin" <itekin(at)gmail(dot)com> |
---|---|
To: | "Scott Marlowe" <smarlowe(at)g2switchworks(dot)com>, "Ibrahim Tekin" <itekin(at)gmail(dot)com>, pgsql-performance(at)postgresql(dot)org |
Subject: | Re: LIKE query on indexes |
Date: | 2006-02-21 20:28:09 |
Message-ID: | e4dcba670602211228u2900be53mb0de28bd5e78b93f@mail.gmail.com |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-performance |
this trick did the job.
thanks.
On 2/21/06, Alvaro Herrera <alvherre(at)commandprompt(dot)com> wrote:
>
> Scott Marlowe wrote:
> > On Tue, 2006-02-21 at 09:57, Ibrahim Tekin wrote:
> > > hi,
> > > i have btree index on a text type field. i want see rows which starts
> > > with certain characters on that field. so i write a query like this:
> > >
> > > SELECT * FROM mytable WHERE myfield LIKE 'john%'
> > >
> > > since this condition is from start of the field, query planner should
> > > use index to find such elements but explain command shows me it will
> > > do a sequential scan.
> > >
> > > is this lack of a feature or i am wrong somewhere?
> >
> > This is an artifact of how PostgreSQL handles locales other than ASCII.
> >
> > If you want such a query to use an index, you need to back up your
> > database, and re-initdb with --locale=C as an argument.
>
> ... or you can choose to create an index with the text_pattern_ops
> operator class, which would be used in a LIKE constraint regardless of
> locale.
>
> http://www.postgresql.org/docs/8.1/static/indexes-opclass.html
>
> --
> Alvaro Herrera
> http://www.CommandPrompt.com/
> The PostgreSQL Company - Command Prompt, Inc.
>
From | Date | Subject | |
---|---|---|---|
Next Message | Jim C. Nasby | 2006-02-22 00:14:56 | Re: How to optimize a JOIN with BETWEEN? |
Previous Message | Ibrahim Tekin | 2006-02-21 20:12:17 | Re: LIKE query on indexes |