[GRASS-dev] [GRASS GIS] #614: r.proj method=cubic_f passes nulls

GRASS GRASS developer
1

#614: r.proj method=cubic_f passes nulls
--------------------+-------------------------------------------------------
Reporter: hamish | Owner: grass-dev@lists.osgeo.org
     Type: defect | Status: new
Priority: normal | Milestone: 6.4.0
Component: Raster | Version: svn-develbranch6
Keywords: r.proj | Platform: Linux
      Cpu: x86-32 |
--------------------+-------------------------------------------------------
Hi,

Running r.proj with method=cubic_f should not propagate NULLs, yet I still
get some.

In this set of commands the Northern Hemisphere reprojects properly, but
the Southern Hemisphere gets a diagonal row of NULLs from the upper right
to lower left of the region.

  http://grass.osgeo.org/wiki/Earth_at_night

?,
Hamish

--
Ticket URL: <https://trac.osgeo.org/grass/ticket/614&gt;
GRASS GIS <http://grass.osgeo.org>

2

#614: r.proj method=cubic_f passes nulls
---------------------+------------------------------------------------------
  Reporter: hamish | Owner: grass-dev@lists.osgeo.org
      Type: defect | Status: new
  Priority: normal | Milestone: 6.4.0
Component: Raster | Version: svn-develbranch6
Resolution: | Keywords: r.proj
  Platform: Linux | Cpu: x86-32
---------------------+------------------------------------------------------
Comment (by hamish):

... happens along 180 longitude ...

--
Ticket URL: <https://trac.osgeo.org/grass/ticket/614#comment:1&gt;
GRASS GIS <http://grass.osgeo.org>

3

#614: r.proj method=cubic_f passes nulls
---------------------+------------------------------------------------------
  Reporter: hamish | Owner: grass-dev@lists.osgeo.org
      Type: defect | Status: new
  Priority: normal | Milestone: 6.4.0
Component: Raster | Version: svn-develbranch6
Resolution: | Keywords: r.proj
  Platform: Linux | Cpu: x86-32
---------------------+------------------------------------------------------
Comment (by kyngchaos):

Since cubic_f is essentially a patch of cubic,bilinear,nearest, only nulls
left from nearest should pass thru (propagate).

...testing...

Looks like it happens with cubic, bilinear AND nearest also. So it's not
the fallback logic.

--
Ticket URL: <https://trac.osgeo.org/grass/ticket/614#comment:2&gt;
GRASS GIS <http://grass.osgeo.org>