Date: Mon, 5 Sep 2005 08:15:45 -0600
From: Michael Fuhr <mike@fuhr.org>
To: Aldor <an@mediaroot.de>
Cc: pgsql-sql@postgresql.org
Subject: Re: POSIX Regular Expression question
Message-ID: <20050905141545.GA20793@winnie.fuhr.org>
References: <431C4EB2.80401@mediaroot.de>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <431C4EB2.80401@mediaroot.de>

On Mon, Sep 05, 2005 at 02:57:06PM +0100, Aldor wrote:
> 
> I want to get out a string only with characters A-Za-z.
> 
> I tried really a lot of things with substring and read many POSIX docs,
> I'm also familiar with the Perl RegEx but right now, I'm giving up... ;-(
> 
> Any idea how to do this in Postgres with POSIX Regex?

Match the beginning of the string with ^.

Match one or more characters in the set A-Za-z with [A-Za-z]+ (or
with just [A-Z]+ or [a-z]+ if you're doing a case-insensitive match).
Using [[:alpha:]]+ should also work.

Match the end of the string with $.

Examples:

SELECT 'abcd' ~ '^[A-Za-z]+$';
 ?column? 
----------
 t
(1 row)

SELECT 'ABCD' ~* '^[a-z]+$';
 ?column? 
----------
 t
(1 row)

SELECT 'ABC123' ~* '^[a-z]+$';
 ?column? 
----------
 f
(1 row)

-- 
Michael Fuhr