: : I attach a junit test which shows strange behaviour of the inOrder : parameter on the SpanNearQuery constructor, using Lucene 2.9.4.
De-Ja-Vu? ... didn't we just talk about this on java-user recently? Here was my comment... http://markmail.org/search/?q=SpanNearQuery+inOrder#query:SpanNearQuery%20inOrder+page:1+mid:zhhlhojw55bf43r6+state:results ...i guess this is actually the same thread now on the dev list? As i said before: i don't really see how this is bug. when you specify inOrder=true, you are requiring that the positions of the subspans must be "in order" so position_span_1 < position_span_2. when inOrder=false there is nothing stoping position_span_1 == position_span_2 ... the fact that span_1 == span_2 doesn't matter. -Hoss --------------------------------------------------------------------- To unsubscribe, e-mail: dev-unsubscr...@lucene.apache.org For additional commands, e-mail: dev-h...@lucene.apache.org
