Hi Micheal, I tried your data to verify that it can be processed with my patched version in the memory_optimization branch. Unfortunately it seems that splitter has another problem not related to memory or high node id values: During the 1st phase, the program stops reading nodes too early. Result: The algorithm in splittableArea.split doesn't work because it uses a far too small value for the nodes: it counts only the 1st 145 nodes and stops counting when the 1st way is processed.
I tried also a smaller sample from OSM with the same result: http://api.openstreetmap.org/api/0.6/map?bbox=11.54,48.14,11.543,48.145 Using parameter --legacy-mode=true changed it a little bit, but I think I think this is an error, so I'll try to find a fix for it. Am I right? Ciao, Gerd -- View this message in context: http://gis.638310.n2.nabble.com/Problems-with-latest-splitter-versions-tp6083054p6987585.html Sent from the Mkgmap Development mailing list archive at Nabble.com. _______________________________________________ mkgmap-dev mailing list [email protected] http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev
