Hello
Does anybody know why r.resamp.rst creates uncompressed output raster
maps? Line 85 of src/libes/rst_gmsl/interp_float/resout2d.c contains a
call to G_open_fp_cell_new_uncompressed() for the elevation file whereas
all the other calls are to G_open_fp_cell_new(). Is it all right to change
this to save space?
I also fixed a bug in r.resamp.rst where it had been really quite unusable
with maps that contained null cells. I think this might be the bug in
segmenting procedure referred to in src/raster/r.resamp.rst/BUGS. I'm not
sure though about the other reasons this module was disabled for.
Paul
Paul Kelly wrote:
Hello
Does anybody know why r.resamp.rst creates uncompressed output raster
maps? Line 85 of src/libes/rst_gmsl/interp_float/resout2d.c contains a
call to G_open_fp_cell_new_uncompressed() for the elevation file whereas
all the other calls are to G_open_fp_cell_new(). Is it all right to change
this to save space?
please do that. I am not sure how it got there.
I also fixed a bug in r.resamp.rst where it had been really quite unusable
with maps that contained null cells. I think this might be the bug in
segmenting procedure referred to in src/raster/r.resamp.rst/BUGS. I'm not
sure though about the other reasons this module was disabled for.
I have somewhat abandoned it so nobody is really watching it and making
sure
that it gets updated when changes are made (that is probably the reason
for nulls
not working). However, others (mostly Jaro) have been updating it
occasionally so it is
probably functional. I don't use it myself, when I need to resample I
randomly
sample the DEM and create site file and then use s.surf.rst.
It would be worth doing it with r.resamp.rst if it was much faster, but
it was
not implemented in a way that I had in mind - because the distances
between
the points are always the same, the system of equations that is being
solved
has the same matrix, only the z values change, but the program does not
take
advantage of it so it is essentially doing the same thing as s.surf.rst
(but it does not have all the updates that involve new input, such as
anisotropy).
Helena
Paul
_______________________________________________
grass5 mailing list
grass5@grass.itc.it
http://grass.itc.it/mailman/listinfo/grass5