[GRASSLIST:4307] Re: r.contour

[Gordon Keith]

On Sun, 5 Sep 2004 17:34, Hamish wrote:
> > On grass-5.7.cvs-i686-pc-linux-gnu-28_08_2004 on debian
> > I tried to use r.contour on a kernel raster but I have this result:
> >
> > GRASS 5.7.cvs:~ > r.contour input=fagiano_kernel step=0.00001 cut=0
> > output=fagiano_kernel
> > WARNING: The vector 'fagiano_kernel' already exists and will be
> >          overwritten.
> > Reading data.
> > Percent complete:  100%
> > FPRange of data:    min =  0.000000 max = 0.003658
> > Minimum level will be 0.000000
> > Maximum level will be 0.003650
> > Continue?(y/n) [y] y
> > Displacing data.
> > Percent complete:  100%
> > Total levels:  367   Current level:(it counts levels and then...)
> > Segmentation fault
>
> This works fine for me, can you make it fail with something from the
> spearfish dataset?

FWIW I've also had problems with r.contour segfaulting when producing large 
numbers of contours on reasonably large datasets.

I don't remember the details, and it wasn't worth tracking it down, but I 
think the work around was just to ask for less contours, then it would work.

My untested suspicion is that perhaps there is a list of nodes maintained in a 
short array and it segfaults when you get more than 32K nodes.

Regards
Gordon

-- 

Gordon Keith
Programmer/Data Analyst
Marine Acoustics
CSIRO Marine Research
http://www.marine.csiro.au

In theory there is no difference between theory and practice. 
In practice there is.