[mkgmap-dev] Bug in LocationHook?
From WanMil wmgcnfg at web.de on Tue Jan 10 21:45:02 GMT 2012
> Hi WanMil, > > > > > > Any idea? > > > > I think the problem does not exist in the current quad tree because the > > areas are splitted into subareas. This also reduces the number of points > > and makes it quicker and easier to test. You could get the number of > > points from the boundary and decide if an area should be splitted into > > subareas. It would also be possible to save the splitted subareas as > > precompiled bounds. If your grid always have the same dimension this > > might improve the handling. > > Okay, it seems I have to look more closely to this. I did not understand > that our quadtree allows to save areas (shapes). > Or did I get this wrong? I think you already got it. I can add an area into the quadtree but only the points of the shapes are used for the query. So it's not a real support for areas (and lines). > Splitting into subareas seems also to be a good option. > > > > > You might also remove the merge step. For most tiles more than one > > precompiled file is loaded. The bounds are merged afterwards. This step > > sounds superfluous as the LocationHook only checks if one point of a way > > is in the boundary and not for all points. Therefore it is not necessary > > to have the boundary area as one complete area object. > > > Okay, I'll look at this as well. > > Thanks for your help! > > Gerd > > > _______________________________________________ > mkgmap-dev mailing list > mkgmap-dev at lists.mkgmap.org.uk > http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev
- Previous message: [mkgmap-dev] Bug in LocationHook?
- Next message: [mkgmap-dev] Bug in LocationHook?
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
More information about the mkgmap-dev mailing list