On 15 Dec 2009, at 11:04, Ian Boston wrote:

> Hi,
> I am considering how to get TermVectors from Jackrabbit queries, and it looks 
> like an ExcerptProvider is the best route as it will give me access to the 
> current Lucene Index and allow me to convert NodeId's into Lucene Docs to 
> extract the terms in the FullText field of the Lucene Document representing 
> the JCR Node.
> 
> Is this the best way of doing this in Jackrabbit ?


Having looked long and hard through the JR1.6 code base I suspect the answer is 
no for the following reason.

1. ExcerptProvider appears to be non selectable, there is on one 
ExcerptProvider class per JVM instance, and its not exposed, so I could 
re-write or extend the ExcerptProvider implementation but then we would not be 
able to use rep:excerpt for highlighting.

Is there any other way of getting to the SearchIndex, so that I can get to the 
Lucene Document and the TermVector (other than AspectJ or cglib)

Ian


> Has anyone already done this under an AL2 ?
> 
> Thanks
> Ian

Reply via email to