Date: Tue, 20 Nov 2012 11:55:08 -0300
From: Alvaro Herrera <alvherre@2ndquadrant.com>
To: Marcin Krawczyk <jankes.mk@gmail.com>
Cc: "pgsql-sql@postgresql.org" <pgsql-sql@postgresql.org>
Subject: Re: regexp_replace behavior
Message-ID: <20121120145508.GC3948@alvh.no-ip.org>
References: 
 <CABnqL301mDLD5UwXfLcUprk2EMt7c5JzJKJz_icAHyx3fN61Vg@mail.gmail.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Content-Disposition: inline
In-Reply-To: 
 <CABnqL301mDLD5UwXfLcUprk2EMt7c5JzJKJz_icAHyx3fN61Vg@mail.gmail.com>
User-Agent: Mutt/1.5.21 (2010-09-15)
Content-Transfer-Encoding: quoted-printable

Marcin Krawczyk escribi=F3:
> Hi list,
>=20
> I'm trying to use regexp_replace to get rid of all occurrences of
> certain sub strings from my string.
> What I'm doing is:
>=20
> SELECT regexp_replace('F0301 305-149-101-0 F0302 {x1} 12W47 0635H
> {tt}{POL23423423}', E'\{.+\}', '', 'g')
>=20
> so get rid of whatever is between { } along with these,
>=20
> but it results in:
> 'F0301 305-149-101-0 F0302 '
>=20
> how do I get it to be:
> 'F0301 305-149-101-0 F0302 12W47 0635H'
>=20
> ??
>=20
> as I understood the docs, the g flag "specifies replacement of each
> matching substring rather than only the first one"

The first \{.+\} match starts at the first { and ends at the last },
eating the {s and }s in the middle.  So there's only one match and that's
what's removed.

> what am I missing ?

You need a non-greedy quantifier.  Try

 SELECT regexp_replace('F0301 305-149-101-0 F0302 {x1} 12W47 0635H {tt}{P=
OL23423423}', E'\{.+?\}', '', 'g')

--=20
=C1lvaro Herrera                http://www.2ndQuadrant.com/
PostgreSQL Development, 24x7 Support, Training & Services