[mkgmap-dev] polygon type 0x4a

Fri May 17 07:48:49 BST 2013

Uups, sorry,

--transparent=false generates polygon type 0x4b. 
Type 0x4a seems to be reserved for polygons used in the overview map:
 * In addition to a low detail map, the overview map contains a number of
type
 * 0x4a polygons.  These definition areas a labeled after and correspond to
 * the detail map img files. 

I think that means that 0x4a should not be used in the style if an overview
map is
wanted ?

Gerd

GerdP wrote
> Hi all,
> 
> I was able to reproduce the error messages reported by Felx :
> http://gis.19327.n5.nabble.com/SeaGenerator-in-overview2-branch-tp5761296p5761396.html
> 
> when I add something like this in the style\polygons file:
> natural=land [0x4a resolution 10] 
> and use --precomp-sea.
> 
> As I already wrote, the polygon type 0x4a is hardcoded in mkgmap:
> 1) mkgmap generates 0x4a polygons for the whole map when you don't use
> --transparent
> 2) polygons with type 0x4 are NOT divided by the
> PolygonSubdivSizeSplitterFilter
> 
> I am not sure what to make of this. 
> The default style doesn't assign 0x4a to a polygon, it doesn't even have a
> rule for the 
> natural=land polygons generated by the SeaGenerator. I was already
> wondering why 
> we generate them.
> 
> I assume the special treatment of 0x4a in PolygonSubdivSizeSplitterFilter
> was added
> because of the bug that was fixed in r2612. 
> This is what happens if I remove the special case for 0x4a:
> - With r2613 and the (unchanged) default style I  see no change in any
> output file
> - With r2613 and the above modification I see that the error messages are
> gone 
> 
> So I guess I should just remove the special case?
> 
> Gerd
> P.S. Felix, get well soon!
> 
>  		 	   		  
> _______________________________________________
> mkgmap-dev mailing list

> mkgmap-dev at .org

> http://lists.mkgmap.org.uk/mailman/listinfo/mkgmap-dev

--
View this message in context: http://gis.19327.n5.nabble.com/polygon-type-0x4a-tp5761430p5761432.html
Sent from the Mkgmap Development mailing list archive at Nabble.com.