[GRASS5] r.patch questions

[Rich Shepard]

  According to the man page, r.patch replaces NULL values in input map #1
with non-NULL values from input map #2 on a cell-by-cell basis, and writes
the output to the specified map file. I've been using this fairly heavily
the past couple of days and I've noticed two behaviors inconsistent with the
description in the man page.

  I hope that someone knowledgable here will let me know if it's again a
case of user error or the program not behaving as advertised.

  First, the few times I tried multiple input maps (greater than two, that
is), the results were not what i expected. I cannot tell you what they were,
exactly, because the displayed map coverd the entire monitor window
regardless of how the region was set. (I was tiling topographic quad DEMS,
and the default window covered the maximum extent of all of them.) However,
when I added them one at a time the results were what I expected. That is, I
could see the single addition on the previous map.

  Second, after interpolating elevation values in the dem "holes", I used
r.patch to fit the small, filled section back onto the full dem. According
to the samples on the bottom of the r.patch man page, this should not have
worked. But, it did. That is, it replaced non-NULL values in input map #1
with non-NULL values in input map #2. Actually, I like this behavior. :-)

  Third, would it be possible to have r.patch replace non-NULL values in
input map #1 with NULL values of input map #2? This would allow the user to
clip out a defined section of a raster map, whether sticking out from the
edge (as I now have) or inside somewhere.

Thanks,

Rich

Dr. Richard B. Shepard, President

                       Applied Ecosystem Services, Inc. (TM)
            2404 SW 22nd Street | Troutdale, OR 97060-1247 | U.S.A.
 + 1 503-667-4517 (voice) | + 1 503-667-8863 (fax) | rshepard at appl-ecosys.com
                 Making environmentally-responsible mining happen.