Re: Full-text index: in japanese?

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

Hi,

> I am creating a database that will hold japanese text. I would like to
> be able to do a full-text search. I have found information on how to use
> the fulltextindex package but I don't think it will work with japanese.

RDBMS may be used to handle regularized data, which can be
recognized by simple indexes. The system of information that
you have to place a full-text search has high entropy, and
it is not suitable for RDBMS in general.

I was wondering if You could try to reglarize the data to avoid
needs of a full-text search. Regular expression, pattern matching
and several string search feature are available in PostgreSQL
otherwise.

>
> Can anyone offer any solutions/packages that I could use to implement
> full-text searches of a database containing japanese?
>

I used to adopt Japanese full-text search engine called "NAMAZU"
when document of full-text search system (like Yahoo) is required.

http://www.namazu.org/
(It is Japanese only. I hope your colleague can help for it)

Regards,

- --
----------------------------------------------------
Hiroki Takada <takada(at)rh(dot)xdsl(dot)ne(dot)jp>

My public key is available at the public key
servers. The message was signed as iso-2022-jp
char-set document in no PGP/MINE (RFC 2015)
format.
----------------------------------------------------

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.0.6 (GNU/Linux)
Comment: KUHASIKU WA http://www.gnupg.org/ WO GORANKUDASAI

iD8DBQE9R8fiyTl8Jc+E3sERAqdKAJ903lrOmyQZNVu5sWECA1E969x6PACgh38g
8Kobwga2jemAWc/1Ad1Gbi4=
=u6XQ
-----END PGP SIGNATURE-----