[GRASS5] [bug #1332] (grass) r.in.gdal: import of arcinfo-asciigrid: northward shift by 1 cellsize

[Request Tracker]

this bug's URL: http://intevation.de/rt/webrt?serial_num=1332
-------------------------------------------------------------------------

Subject: r.in.gdal: import of arcinfo-asciigrid: northward shift by 1 cellsize

Platform: GNU/Linux/i386
grass obtained from: Trento Italy site
grass binary for platform: Downloaded precompiled Binaries
GRASS Version: GRASS 5.0.0 (August 2002)

Hi!


Import of arcinfo asciigrid data in a lat/lon-location using

r.in.gdal -o in=testdata out=testdata_rast

where the file "testdata" consists of
 
ncols 1 
nrows 1 
xllcenter 0.000000 
yllcenter 0.000000 
cellsize 1.000000 
nodata_value -1.E20 
42.0

yields a raster file with info (obtained from "r.info map=testdata_rast"):

|   Rows:         1                                                          
|   Columns:      1                                                          
|   Total Cells:  1                                                          
|        Projection: Latitude-Longitude (zone 0)                             
|            N:      1:30N    S:      0:30N   Res:     1                     
|            E:      0:30E    W:      0:30W   Res:     1                     
|   Range of data:    min =  42.000000 max = 42.000000                       

Number of rows, cells, longitude-info and range of data are correctly determined. However, the correct latitude information should instead be 
  N:      0:30N    S:      0:30S   Res:     1
shouldn't it?

The erroneous (?) northward shifting by one cellsize seems to occur even with more complex arcinfo-asciigrid indata files and for other location types.
Is this an index problem in r.in.gdal for arcinfo-asciigrid data?


Help/comments appreciated, thanks in advance,

Thomas Klein



-------------------------------------------- Managed by Request Tracker