Return-Path: postarch
Received: by postgres.Berkeley.EDU (5.61/1.29)
	id AA15535; Thu, 28 May 92 09:26:35 -0700
Message-Id: <9205281626.AA15535@postgres.Berkeley.EDU>
From: postarch (Postgres Mailing Archive)
Subject: Re: Postgres and Textual data
To: postgres@postgres.berkeley.edu
Sender: pg_adm@postgres.berkeley.edu
Reply-To: mer@postgres.berkeley.edu
In-Reply-To: Your message of "Tue, 26 May 92 15:50:40 PDT."
             <9205262250.AA23197@postgres.Berkeley.EDU> 
Date: Thu, 28 May 92 09:26:17 PDT

you write:

>I was wanting to use postgres in a library cataloging system- essentially
>all the fields (Title, Author, Description) will be textual data and
>I will be performing 'textregexeq' kind of searches on the fields.

>1. Any plans on using the Henry Spencer collection of regex routines instead
>   of the system's C library version ? Henry Spencer's colleciton is much more
>   flexible (and used in Perl, etc. also) and is perhaps faster than the
>   library call.

No, but Postgres is an exstensible system and you can add the functions and
operators that use whatever regular expression implementation you prefer.

>2. How do I create indexes on the words of the various fields ? I am sorry
>   but I am sketchy on the use of indexes on words and regular expressions.

btree indices are no good when it comes to regular expressions.  I'm not sure
a good indexing scheme exists for this.


Jeff Meredith
mer@postgres.berkeley.edu