[GRASS5] r.reclass, r.rescale, NULL bugs

[Andreas Lange]

Hi,

The bug in r.rescale (tagged as release critical) that cell values
outside the from range are not set to "0" but to NULL is not a bug IMHO.
It should be corrected in the documentation that all modules write out
NULL for missing or unknown values instead of zero. I think that this is
the desired behaviour (e. g. for later patching maps or displaying them
as overlays).

I think that all modules that calculate on raster maps should get a flag
"-z" that switches from NULL output (default) to zero output for user
convinience. This would be r.reclass and r.rescale (any other?). 

But r.rescale calls r.reclass and r.reclass core dumps if the default
rule "* = NULL" is applied. This is not related to NULL, but to the "*".
The docs do not state this default-rule feature, but it is clearly in
the source. And its not a CPU-related problem, occurs on IRIX too. 
Is this the same segmentation fault (signal 11) problem than in r.binfer
etc. ?

Another question: Has anyone expirience with using gdb on grass modules?
I recompiled the whole grass source tree with "-g" and no symbol
stripping in the library building, but i can not get the symbols from
the post-mortem debugging. 

cu,

Andreas 

-- 
Andreas Lange, 65187 Wiesbaden, Germany, Tel. +49 611 807850
Andreas.Lange at Rhein-Main.de - A.C.Lange at GMX.net


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