On Thursday 09 November 2006 19:55, Renaud Waldura wrote: > I'm thinking I might have to tokenize the document text (I have it), > then compute the intersection between the set of all terms and the set > of terms from the rewritten query. Blech. Sounds expensive. Any other > ideas?
No faster, but easier: index the Document in a RAMDirectory, open that with an IndexReader und use this reader to do the first (and only) rewrite. As this only happens when the user really clicks a match I see no problems with performance. Regards Daniel -- http://www.danielnaber.de --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
