I have a similar need to disable Lucene indexing, at least for a particular file/mimetype. My application generates a Word docx file that is adequate for Word to open and work with but not for Lucene. When trying to index my file Lucene goes into a death spiral in which it consumes all of the heap memory and requires a server restart. This is repeated every time the repository is accessed to display the problem file. I've tried changing the mimetype recorded with the file to "application/octet-stream" but, apparently, Lucene checks the file internals or uses the file extension and ignores what is set on the node.
If anyone has a recommendation for bypassing the indexing it would be greatly appreciated. -- View this message in context: http://jackrabbit.510166.n4.nabble.com/How-to-disable-Lucene-in-Jackrabbit2-0-0-tp1589117p2315091.html Sent from the Jackrabbit - Users mailing list archive at Nabble.com.
