Re: [Pymag-devel] Unable to render Polygons
Status: Alpha
Brought to you by:
lhj
From: henrik j. <hen...@gm...> - 2009-05-20 07:46:31
|
I found the problem. If the first vertex of the polygon is the same as the lower right corner of the bounding box it's of type 3. In the Bremen data set there was one polygon of type 3 which crashed MapsendLite. I have now disabled generation of type 3 polygons (they will still be generated but of another type) so it should work now. But I still don't know why type 3 makes it crash. /Henrik On 19/05/2009, henrik johansson <hen...@gm...> wrote: > Actually they are discarded, they thing is that those warnings are not > from those that make Mapsend crash. I still haven't been able to find > out those that cause the problem. In Ingo's example from Bremen I know > that there are only polygons from parks that cause problems. Far from > all parks though because I can see most of them and they look correct. > > Maybe the easiest way is to make a osm file with only those parks and > then try to find the bad ones. > > /Henrik > > On 19/05/2009, Josh <jo...@fe...> wrote: >> Oops, I had another think about that, and realised a 3 pointed closed >> polygon is actually more like a straight line ;) >> >> Is it possible to just discard bad polygons, seeing as they don't >> render anyway, and just break the map? >> >> Josh >> >> >> >> Quoting jo...@fe...: >> >>> I've just tried the latest code from svn, with mixed results. Some of >>> my maps have polygons rendering again, but others that warn about >>> Improper polygons now crash Mapsend/Vantagepoint. >>> >>> Some of the warnings I can understand - polygons with only 2 points, >>> or polygons that are 'open', but one example I saw has me confused... >>> >>> I got this warning one one tonight - >>> WARNING Improper polygon found: [(144.844255, -38.373168800000002), >>> (144.8432851, -38.373168800000002), (144.844255, -38.373168800000002)] >>> That shows a polygon with 3 points, and two of them have the same >>> coordinates? That should be a valid polygon unless I'm missing >>> something. >>> >>> The nodes/way relating to that error are: >>> >>> <node id="172752298" timestamp="2009-02-25T17:02:53Z" >>> lat="-38.3731688" lon="144.844255"/> >>> <node id="172751478" timestamp="2009-02-25T17:02:53Z" >>> lat="-38.3731688" lon="144.8432851"/> >>> >>> <way id="16711763" timestamp="2009-02-25T17:02:53Z"> >>> <nd ref="172752298"/> >>> <nd ref="172751478"/> >>> <nd ref="172752298"/> >>> <tag k="created_by" v="Potlatch 0.5d"/> >>> <tag k="leisure" v="park"/> >>> </way> >>> >>> >>> Regards, >>> Josh >>> >>> >>> >>> >>> ------------------------------------------------------------------------------ >>> Crystal Reports - New Free Runtime and 30 Day Trial >>> Check out the new simplified licensing option that enables >>> unlimited royalty-free distribution of the report engine >>> for externally facing server and web deployment. >>> http://p.sf.net/sfu/businessobjects >>> _______________________________________________ >>> Pymag-devel mailing list >>> Pym...@li... >>> https://lists.sourceforge.net/lists/listinfo/pymag-devel >>> >> >> >> >> >> ------------------------------------------------------------------------------ >> Crystal Reports - New Free Runtime and 30 Day Trial >> Check out the new simplified licensing option that enables >> unlimited royalty-free distribution of the report engine >> for externally facing server and web deployment. >> http://p.sf.net/sfu/businessobjects >> _______________________________________________ >> Pymag-devel mailing list >> Pym...@li... >> https://lists.sourceforge.net/lists/listinfo/pymag-devel >> > |