jackrabbit-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Marcel Reutegger <marcel.reuteg...@gmx.net>
Subject Re: Searching Binary Nodes
Date Thu, 16 Jun 2005 07:59:01 GMT
Hi Simon,

Simon Gash wrote:
> I've set up a binary document filter for MSWord documents, the Apache
> Slide stuff. JackRabbit seems to be picking up the filter alright but
> I'm having trouble querying the index. This is the XPath statement that
> I'm using
>  
> //*[jcr:like(@jcr:data,'%document%')]
>  
> I'm assuming that this will search any of the nodes underneath the root
> that have a 'jcr:data' type property and use the like function to filter
> out those that contain the text item 'document' or am I missing
> something ?

This depends on how the text filter extracts the text and assigns it to 
index fields. E.g. the TextPlainTextFilter adds a fulltext field to the 
node that gets indexed. This means the binary content can only be 
searched with the jcr:contains() function. I assume the same applies to 
the Word filter you are refering to.

regards
  marcel

Mime
View raw message