Hi Chetan, On Thu, May 22, 2014 at 6:52 AM, Chetan Mehrotra <[email protected]> wrote: > ...This might be due to OAK-1462. We had to disable the > LuceneIndexProvider form getting registered as OSGi service...
Would that mean that the LuceneIndexEditor is still called, but the result isn't used? I'm asking because when adding a PDF, LuceneIndexEditor.addOrUpdate does call context.getWriter().updateDocument with a Document that does contain the PDF's full text in a field named :fulltext, so the text extraction is working (thanks Alex for the tika-parsers hint). But the query mentioned earlier in this thread still finds only .txt documents, not .pdf. Adding a .txt also causes LuceneIndexEditor.addOrUpdate to call context.getWriter().updateDocument, but maybe the text is also indexed in another way? -Bertrand
