-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On Tue, 08 Oct 2002 07:15, Dave Thomas wrote:
> Nat Pryce <nat.pryce / b13media.com> writes:
> > Why are you using MySQL?  If it's just to index articles by article
> > numbers, then you might be better off storing the FAQ in plaintext files
> > named by article number.  If lookup performance is too slow, use a
> > filesystem that is designed to lookup filenames in huge directories very
> > fast, such as ReiserFS.  You can then use normal file system tools to
> > search the text, or map the text files into memory, etc. etc.
>
> Something I'd like is queries such as
>
>   article contains:  +matz +irb -python
>
> I know I can do this using egrep n times and combining the results,
> but I'd really rather not.
>
> I've tried Glimpse, but I currently can't get 'and' searches to
> work. I've also tried Swish, but it's currently segfaulting on me if I
> just index a single file...

At work we used Swish-e for that task, but one of our developers had to modify 
the spider to get it to work as we wanted it to (something to do with 
controlling the depth of site that gets indexed). This is running on a 
win2000 machine. Unfortunately, I can't give you the details of what was 
modified in the spider because I wasn't a part of this project. The spider 
was written in perl. 

- -- 
Signed,
Holden Glova
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.0.7 (GNU/Linux)

iD4DBQE9opob0X8w8X71zPcRAvMUAJ9kYCKm+Y3oLk3sgQj3vdbWMbP7iACXVU8I
Rifm7TMMPGfLxpYT0egQXQ==
=hZbo
-----END PGP SIGNATURE-----