Falko, just an idea : for theory 2, perhaps can you try to clean either a or b map, by snapping boundaries vertices over each other : v.edit map=a bgmap=b type=boundary tool=snap snap=vertex thres=...
Vincent. Le lundi 12 mars 2012 à 17:11 +0100, Falko Engel a écrit : > Dear List, > > I have encountered some problems when using v.overlay on polygon maps > (and also v.patch): There are some polygons missing in the output map > that should not be missing. And there are polygons there that should not > be there. > > > I have two theories for this problem: > > 1. The input data (either ainput or binput or both) has geometry errors > > 2. Data set ainput and binput "almost" share some boundaries. > This can happen, when someone creates an new vector map and uses an > existing one as a basis. He would then digitize manually along a > boundary of the base map and produce an inexactly overlaying new boundary. > My theory in this case is that when using v.overlay on these two data > sets the inexactly created boundaries result in many very small > intersections which again result in wrong geometries. > I'm not sure about this, it's just a theory. > > > In my case I have used the following command to overlay my vector maps: > > v.overlay ainput=veg atype=area binput=plot btype=area > output=res_overlay operator=and > > You can find the data I have used at: > > http://dl.dropbox.com/u/1542545/data_overlay.zip > > My system components are: > GIS: GRASS GIS 6.4.1 > OS: openSUSE 11.3 (x86_64) > > > I checked the geometries using ftools in QGIS. There seem to be no > geometry errors in these data sets. So I am focused on my second theory. > I use GRASS GIS in a quite large project where I can't choose the data > sets I am working with. I came across this problem more than once so far > and I hope that I am missing something or that there is a solution to this. > > I would be thankful if someone could spare the time and have a look at > my data. > > Falko > > > > > > _______________________________________________ > grass-user mailing list > [email protected] > http://lists.osgeo.org/mailman/listinfo/grass-user > _______________________________________________ grass-user mailing list [email protected] http://lists.osgeo.org/mailman/listinfo/grass-user
