Re: What is the impact of "varchar_pattern_ops" on performance and/or memory

From: Edson Richter <edsonrichter(at)hotmail(dot)com>
To: Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
Cc: pgsql-general(at)postgresql(dot)org
Subject: Re: What is the impact of "varchar_pattern_ops" on performance and/or memory
Date: 2013-01-22 20:14:18
Message-ID: BLU0-SMTP19047052A40596DF3363D78CF160@phx.gbl
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-general

Em 22/01/2013 18:04, Tom Lane escreveu:
> Edson Richter <edsonrichter(at)hotmail(dot)com> writes:
>> Almost all indexed columns of kind varchar in my database require
>> "varchar_pattern_ops" op class in order to make my LIKE queries more
>> optmized.
>> Is there any way to define that this operator class is the default for
>> my database for varchar columns?
> No. (Well, you could go and hack the catalog entries to mark it as the
> default, but things would fail rather badly whenever you dump and reload
> the database, because pg_dump won't dump changes to built-in objects.)

Can I change this in template database, and then get the benefit of
having these changes propagated to newly created databases?

>
>> What would be the impact in terms of performance and memory consumption?
> It's probably cheaper, actually, than the regular strcoll-based
> comparisons.
>
> BTW, have you considered whether you could run your database in C locale
> and thus dodge the whole problem? In C locale there's no difference
> between this opclass and text_ops.

Using C locale, would I face problems with Portuguese Brazilian
characters like ã, ç, é?

Thanks, Tom!

Edson

>
> regards, tom lane
>
>

In response to

Browse pgsql-general by date

  From Date Subject
Next Message Tom Lane 2013-01-23 03:05:11 Re: Fwd: Question on Trigram GIST indexes
Previous Message Tom Lane 2013-01-22 20:04:58 Re: What is the impact of "varchar_pattern_ops" on performance and/or memory (was: What is impact of "varchar_ops")?