On Fri, Jun 24, 2011 at 3:08 AM, SBS <jturn...@uow.edu.au> wrote: > I am new to Lucene 3.2 and have it working well in most cases. However, when > I do wildcard searches such as abc*, all the resulting matched documents > have exactly the same score regardless of the words that matched, their > position within the document or the size of the document. This makes > presenting them in some meaningful order on the screen very difficult. > > Is this the expected behaviour or have I done something wrong with my > indexing or searching technique? Searches not involving wildcards produce a > range of scores as expected. > that is expected behavior. WildcardQuery is a MultiTermQuery and for MTQ you can / have to specify a RewriteMethod that decides what kind of primitive query is used to execute the MTQ. if you want to use scoring you can pass MultiTermQuery.SCORING_BOOLEAN_QUERY_REWRITE to your WildcardQuery when you create it via WildcardQuery#setRewriteMethod(...); Yet, make sure you expect a TooManyBooleanClauses exception if your WildcardQuery expands to too many terms. Something like a* is not a good idea :).
simon > Thanks, > > -sbs > > -- > View this message in context: > http://lucene.472066.n3.nabble.com/Wilcard-document-matches-all-have-the-same-score-tp3102450p3102450.html > Sent from the Lucene - General mailing list archive at Nabble.com. >
