hsp wrote:
Is this provided by jackrabbit (internally by lucene I though), or I
must to do programatically?
full text search is provided by jackrabbit as described in the JCR spec
in section 6.6.5.2.
I mean, if I want to search the documents
files, whatever the node is, that have a string "hello world" in any
part of their contents (binary), can I do this with xpath directly?
a basic query that returns nodes that contain the string 'hello world'
in any of their properties is:
//*[jcr:contains(.,'hello world')]
Please note that only binary properties of nt:resource nodes are
indexed. And that you need a text filter that is able to extract a text
representation of the binary content.
See
https://svn.apache.org/repos/asf/incubator/jackrabbit/trunk/textfilters/README.txt
regards
marcel
