It's just a simple Xpath Query using "contains".
I also implemented a case insensitive search by translate.
I guess it's the most unperformant and ugliest way, but it does its job
for me.
I search for all "names" of "things" which contain the searchstring in
one of its nodes.
It looks like this:
"/things/name[contains(translate(../descendant-or-self::*,'ABCDEFGHIJKLM
NOPQRSTUVWXYZ���','abcdefghijklmnopqrstuvwxyz���'), translate('" +
searchstring +
"','ABCDEFGHIJKLMNOPQRSTUVWXYZ���','abcdefghijklmnopqrstuvwxyz���'))]";
Case sensitive it would simply be:
"/things/name[contains(../descendant-or-self::*, searchstring)]"
I hope this helps.
Wolfram
-----Urspr�ngliche Nachricht-----
Von: Mark J. Stang [mailto:[EMAIL PROTECTED]
Gesendet: Dienstag, 26. Februar 2002 00:32
An: [email protected]
Betreff: Re: AW: Full Text Index
Wolfram,
What did your code look like?
Did you do it in Xindice using
"contains" or externally?
thanks,
Mark
"W. Peschke" wrote:
> Hi,
>
> I implemented full-text search using a xpath query.
>
> In a 500 documents database it is acceptable fast.
>
> But of course it would be nice to have such a functionallity directly
> from the database system.
>
> Wolfram
>
> -----Urspr�ngliche Nachricht-----
> Von: Mark A. Richman [mailto:[EMAIL PROTECTED]
> Gesendet: Montag, 25. Februar 2002 03:42
> An: [email protected]
> Betreff: Full Text Index
>
> What is the effort required to add full-text index & search to
> Xindice?
>
> - Mark
