I have a couple comments on your code after briefly looking through it. * If you want to work with miles, initialize SimpleSpatialContext with DistanceUnits.MILES. I see you chose KM but your leading statistics are in miles. Perhaps that is the reason for the discrepency in your numbers. * You are not supposed to instantiate Shape instances directly, the spatial context is a factory for them. It will also do some important argument normalization (e.g. ensuring lat-lon is within proper boundaries). * You are initializing GeohashPrefixTree using the maximum detail possible (roughly double precision). I suspect you don't actually need that level of accuracy. Look at SpatialPrefixTreeStrategy.initMaxLevels() for code that chooses a value based on a target fraction of your chosen units (e.g. miles). If you want to hard-code a specific value only good for geohashes, you could alternatively look up a value in a table of precision on geohashes on wikipedia.
Keep in mind that LSP is subject to change at will; it hasn't been officially released yet. The remaining issues I see for a first release are enumerated here: http://code.google.com/p/lucene-spatial-playground/wiki/Tasks The "Extras / demo related" part arguably doesn't count. ~ David Smiley ----- Author: http://www.packtpub.com/apache-solr-3-enterprise-search-server/book -- View this message in context: http://lucene.472066.n3.nabble.com/Spatial-Search-tp3623494p3629480.html Sent from the Lucene - Java Users mailing list archive at Nabble.com. --------------------------------------------------------------------- To unsubscribe, e-mail: java-user-unsubscr...@lucene.apache.org For additional commands, e-mail: java-user-h...@lucene.apache.org
