Hi Andrzej, Mike, I am not splitting short lines here, I talk about lines of > 50 m. If you have such a rather long line with delta_lat = 1 and delta_lon > 50 there is no point where I can split the line without creating one horizontal part and another that still has delta_lat = 1, both parts building a visible angle, so in that case I would not split. With delta_lat > 10 it is likely to find a point that is close to the straight line and not too far away from the calculated cut point.
If you take my "bed of nails" example, I am searching for a nail that is close enough (~10m) to a given point on the rubber band and very close to the rubber band itself (< 0.5m). My current solution is to raster the line with the Bresenham algo and check the points which are close to the wanted one. I think it is no problem regarding performance, but it seems not very elegent. Gerd popej wrote > Hi Gerd, > > one more notice - for splitting very short lines could be better to > calculate intermediate point using grid coordinates of line instead of > high precision like in makeBetweenPoint(). > > -- > Best regards, > Andrzej > _______________________________________________ > mkgmap-dev mailing list > [email protected] > http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev -- View this message in context: http://gis.19327.n5.nabble.com/help-needed-for-graphical-problem-tp5833271p5833365.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
