I will need to double check this. I know there was a dispute early on with Neo4j Spatial because the JTS library orders bbox params in one way, and GeoTools does it another way, so you might be seeing the results of that. I believed we sorted that all out, but perhaps not. I have have just checked the RTree code, and it does the right thing, however it is possible that code outside is passing in bbox parameters in a different order than expected.
Perhaps you have some sample test code I can use to check this with? On Thu, Mar 24, 2011 at 7:01 PM, Robert Boothby <[email protected]> wrote: > Thanks for the response Craig. You're right I had mixed up the > latitude and longitude coordinates. I've now got the expected > answers... > > However the transposition of the elements of the spatial index's root > node geometry envelope definitely occurs. I just wouldn't have spotted > it if I hadn't mixed up the coordinates and debugged it. The envelopes > of the result nodes do not have transposed elements. > > The index root node element bounding box envelope looks like: > Env[-1.1907455 : 51.0852483, 0.3909055 : 52.2274931], my point > envelope looks like Env[-0.812988 : -0.812988, 51.796726 : 51.796726] > and the other node envelopes looks like Env[-0.8577898 : -0.7723875, > 51.7930991 : 51.8378447], Env[-0.8502576 : -0.8028032, 51.786247 : > 51.8211053], Env[-0.8516907 : -0.7686481, 51.7932282 : > 51.8335107],Env[-0.8574819 : -0.7733667, 51.7921887 : > 51.8394132],Env[-0.9067499 : -0.6564949, 51.6661276 : 51.9251302], > Env[-0.8159025 : -0.8114534, 51.7961756 : 51.7997272], . > > It appears that the maxx and miny values have been swapped in the > spatial index root node. > > There may be certain scenarios when valid coordinates are excluded > because the spatial index root node has the incorrect envelope. > > Robert. > _______________________________________________ > Neo4j mailing list > [email protected] > https://lists.neo4j.org/mailman/listinfo/user > _______________________________________________ Neo4j mailing list [email protected] https://lists.neo4j.org/mailman/listinfo/user
