[GRASS-user] kml topology... ?

[Dylan Beaudette]

On Thursday 28 February 2008, Nikos Alexandris wrote:
> On Thu, 2008-02-28 at 09:15 -0800, Dylan Beaudette wrote:
>
> Thank you for your reply Dylan!
>
> > On Tuesday 26 February 2008, Nikos Alexandris wrote:
> > > Continuing my small quest with kml files... and bombing you with simple
> > > questions ;-p
> > >
> > > are kml files carrying topology information besides coordinates?
> >
> > I doubt it. As far as I know they are just "simple features".
> >
> > > I needed to manually correct a kml file which looked very bad after
> > > importing in GRASS (with -c  otherwise it was not useful!)
> >
> > i am not surprised that a KML file resulted in a topologically broken
> > file. These days everyone and their brother are creating KML via all
> > sorts of methods-- most of which have little notion/concern for
> > topological-correctness.
>
> +1
>
> > > Automatic cleaning is really not useful even after trying to feed with
> > > logical thresholds the various tools.
> >
> > Try opening the file with v.digit and look for oddities.

Hi,

> Had to correct (almost) everything. Tha data are supposed to come from a
> GPS measurement session.

I would think that the results from such as session be points or lines...

> All polygons were open after importing in GRASS. Yet there were
> centroids inside this virtually closed boundaries. And the strange thing
> is that one (call it) centroid was out of the virtual boundaries...

I wonder where the polygons came from -- could it be that one of the import 
steps mistook the geometry in KML file as polygons? Line data that are 
interpreted as polygons---especially in the presence of overlapping 
features---would definitely cause some craziness.  

> Don't know, maybe my mistake for this last statement about the centroid
> falling out.
>
> Cheers!

Have the original data on hand?

Dylan


-- 
Dylan Beaudette
Soil Resource Laboratory
http://casoilresource.lawr.ucdavis.edu/
University of California at Davis
530.754.7341