[GRASSLIST:7670] Re: r.statistics method=variance gives NAN

[Glynn Clements]

Martin Wegmann wrote:

> when using r.statistics with method "variance" I receive NAN as output values, 
> however if I use "average" or "stddev" the ouput  looks reasonable.
>  
> I tried it with GRASS 6.1-cvs a few weeks ago and GRASS 6.0 from the debian 
> repository but both produced the same result.
> 
> Any idea what the problem might be?

The m_var() function fails to initialise the local variable "s".

Fixed in CVS, or apply the attached patch.

-- 
Glynn Clements <glynn at gclements.plus.com>

-------------- next part --------------
--- raster/r.statistics/o_var.c~	2004-11-09 13:58:33.000000000 +0000
+++ raster/r.statistics/o_var.c	2005-07-22 06:07:18.000000000 +0100
@@ -107,6 +107,7 @@
 
    *vari = 0.0;
    ep = 0;
+   s  = 0.0;
 
 
    for(i = 0; i < n; i++)              /* First pass to get the mean     */