[GRASS-user] r.contour "level" attribute

Moritz Lennert mlennert at club.worldonline.be
Fri May 16 13:25:59 EDT 2008


On 16/05/08 17:26, M S wrote:
> Hi everyone. 
> 
> I had generated some contours from a DEM with r.contour.  In the 
> resulting vector contour map, I get cat and level, where level is the 
> elevation value.  I must be confused with the level terminology, because 
> I thought levels pertained to topological (2) or non-topological data 
> (1).  However, in r.contour output, "level attribute" corresponds with 
> the contour elevation.

In GRASS the same words sometimes have different meanings according to 
context. Generally for historical reasons...

> 
> Regardless, I'm able to query the lines in GRASS monitor to get the cat 
> and level (elevation) values for the contour lines, but in QGIS I'm 
> unable to query or identify the features.  In QGIS, I am able to label 
> the lines on the "level" item though.
> 
> I refrained from cross posting to QGIS list, because it seems like 
> something I'm not understanding in GRASS vector model.
> 
> Any ideas why the features are (repeatably) unidentifiable in QGIS, but 
> are viewed and labeled without issue? 

No idea, but I can confirm with the North Carolina demo data:

- elev_contour_3m: no data found
- any other line layer (streams, busroutesall, etc): I get the relevant info

After a bit of testing, it seems to me that this is due to the fact that 
the contour lines have 3D points as vertexes, i.e.:

v.out.ascii elev_contour_3m format=standard

L  309 1
  644845.39132067 228495       58
  644850.5257363 228485       58
  644855       228476.80568172 58
  644856.10080552 228475       58


whereas the others have 2D points as vertexes, i.e.:

v.out.ascii busroutesall format=standard

L  40 1
  638759.35668041 224267.86790137
  638748.56434918 224206.16089597
  638738.04094476 224149.69241332
  638701.95659054 224155.1582149
  638712.48508636 224219.60539241
  638727.41956522 224217.30778103
  638737.10320251 224271.95933017


So it seems to me that this is a flaw in QGIS' handling of this type of 
data.

Moritz


More information about the grass-user mailing list