Content-Type: text/plain; charset=utf-8
Mime-Version: 1.0 (Mac OS X Mail 10.3 \(3273\))
Subject: Re: BUG #14623: pg_trgm doesn't correctly process some regexp with
 negative lookahead
From: Corey Csuhta <its+postgres@csuhta.com>
In-Reply-To: <22558.1492145650@sss.pgh.pa.us>
Date: Fri, 14 Apr 2017 01:05:06 -0400
Cc: pgsql-bugs@postgresql.org
Content-Transfer-Encoding: quoted-printable
Message-Id: <606E7756-3DAE-482D-9131-A002F3F02CFD@csuhta.com>
References: <E2B01A4B-4530-406B-8D17-2F67CF9A16BA@csuhta.com>
 <22558.1492145650@sss.pgh.pa.us>
To: Tom Lane <tgl@sss.pgh.pa.us>
Precedence: bulk
Sender: pgsql-bugs-owner@postgresql.org

Does this still seem like a problem? Is it helpful if I submit a new bug # =
for tracking purposes?

> On Apr 14, 2017, at 12:54 AM, Tom Lane <tgl@sss.pgh.pa.us> wrote:
>=20
> Corey Csuhta <its+postgres@csuhta.com> writes:
>> I don=E2=80=99t know if this is related or a separate bug, but I am also=
 having this problem with the `+` operator and character classes. Very simi=
lar test case:
>> CREATE TABLE foobar (x text);
>> INSERT INTO foobar VALUES ('Trying +1/+1'),('Also doing +20/+20');
>> CREATE EXTENSION pg_trgm;
>> CREATE INDEX ON foobar USING gin (x gin_trgm_ops);
>> SET enable_seqscan TO on;
>> SELECT * FROM foobar WHERE x  ~ '\+\d+\/\+\d+'; -- Returns both rows
>> SET enable_seqscan TO off;
>> SELECT * FROM foobar WHERE x  ~ '\+\d+\/\+\d+'; -- Returns 0 rows
>=20
> Hmm ... seems unrelated, since there's no lookahead constraint in this
> example.  Also, the patch I committed earlier today doesn't fix it.
>=20
> 			regards, tom lane


--=20
Sent via pgsql-bugs mailing list (pgsql-bugs@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-bugs