Hi, I thought again about this patch and I think it is not correct, but I am not able to produce a correct solution. Why is it not correct? The removeShortArcsByMergingNodes() method in ElementSaver iterates multiple times over all ways, a change in the "last" way may trigger a change in ways that come before during the next pass. My patch changes that so that a change in the last way doesn't trigger anything in previously processed ways. I tried to change the patch so that all roads added in StyledConverter.addConvertedWay() are first placed in a HashMap, together with a copy of the corresponding GType object and all these roads are processed later in the end() routine, but I see different results when I do so and don't understand why. Any ideas ? Gerd -- View this message in context: http://gis.19327.n5.nabble.com/removeShortArcs-v1-patch-tp5743043.html Sent from the Mkgmap Development mailing list archive at Nabble.com.