Does anyone have any experience/guidelines for running Solr w/Carrot2 on top of Jackrabbit? I know Jackrabbit will crack open a number of files (via Tika) and index their content (via Lucene), but I'd like to add search result clustering a la Carrot2.
I'd also be interested in anyone using Mahout on top of Jackrabbit. I know Mahout can use Lucene indexes, so it would be interesting to hear how to re-use the automatic Lucene indexes from Jackrabbit. Rob
