[GRASS5] r.null bug or gis library bug?

[Markus Neteler]

On Fri, Sep 08, 2000 at 05:18:58AM +0900, GRASS wrote:
> hi all,
> 
> r.null has bug:
> 
> 	r.mapcalc "tmp=2.0"
> 	r.null map=tmp setnull=2.0
> 
> Floating exception error occurred.
Hi Huidae Cho,

on my Linux box there is no floating point exception.
 
> 1. Size of f_range file changes to zero while running r.null.
Yes, here, too:
ls -la $LOCATION/cell_misc/tmp/
insgesamt 10
drwxr-xr-x   2 neteler  user         1024 Sep  8 14:30 .
drwxr-xr-x  26 neteler  user         1024 Sep  8 14:30 ..
-rw-r--r--   1 neteler  user           52 Sep  8 14:30 f_format
-rw-r--r--   1 neteler  user            5 Sep  8 14:30 f_quant
-rw-r--r--   1 neteler  user            0 Sep  8 14:30 f_range
-rw-r--r--   1 neteler  user         5712 Sep  8 14:30 null
 
> 2. G_number_of_cats()	called on line 316 in src/raster/r.null/null.c
> 
> 3.  G_read_range()      called on line 395 in src/libes/gis/cats.c
> 
> 4.   G_read_fp_range()  called on line 209 in src/libes/gis/range.c
> 
> 5.                      returns 2 on line 156 in src/libes/gis/range.c
> 				because f_range size is 0.
> 
> 6.  G_get_fp_range_min_max() set drange->min, drange->max to NULL
> 				as drange->first_time == 1
> 
> 7.  Now, trying to assign double NULL value(means NaN) to integer var.
> 	==> BUG
Yes, this looks like a bug (at least to me).

Yours
 
 Markus

---------------------------------------- 
If you want to unsubscribe from GRASS Development Team mailing list write to:
minordomo at geog.uni-hannover.de with
subject 'unsubscribe grass5'