r.line stops partway through

[Simon Cox]

Thanks for the pointer Olga

However, I have to report that YES, I did run r.thin first.  On further
investigation
I was starting to suspect the crowded cells, since I have now written a
r.mapcalc script
to remove the isolated cells and the problem persisted.  I zoomed in on where
it seemed to
be stopping and there is indeed a block of four cells occupied - is this a
"crowded cell"?

In which case I might dust off r.mapcalc again to trash these ...

By the way, we compared the results of r.thin with running the same thing
through a
skeletonizing operation in NIH-Image on the Macintosh, and the latter seemed to
do
a cleaner job - for example, we have some blobs on our map (representing
sink-holes)
and while r.thin turned them into crosses, Image made them into dots.  I can
understand
why crosses might emerge, but equidimensional objects really should turn into
dots, don't
you think?  Perhaps this could be added to r.thin, maybe as an option?  And
then either
r.thin or r.line should learn to cope with the crowded cells.

If we continue this chat, maybe we should go off the list?

Thanks for you time.

Simon Cox
----
___________________________________________________
      __  L     Dr Simon Cox			
   ,~'  L_|\    VIEPS Department of Earth Sciences, 
,-'         \   Monash University, Clayton Vic 3168        
(            \  Australia
\    ___     /          Phone +61 3 565 5762
 L,~'   "\_x/           Fax   +61 3 565 5062
           u    simon at cerberus.earth.monash.edu.au
___________________________________________________