[GRASS-dev] Re: [GRASS GIS] #1088: r.fillnulls: support other interpolation methods

[GRASS GIS]

#1088: r.fillnulls: support other interpolation methods
-------------------------+--------------------------------------------------
 Reporter:  kyngchaos    |       Owner:  grass-dev@…              
     Type:  enhancement  |      Status:  new                      
 Priority:  normal       |   Milestone:  6.4.3                    
Component:  Raster       |     Version:  svn-develbranch6         
 Keywords:  r.fillnulls  |    Platform:  All                      
      Cpu:  All          |  
-------------------------+--------------------------------------------------

Comment(by mmetz):

 Replying to [comment:23 glynn]:
 > Replying to [comment:14 mmetz]:
 >
 > > PS: I fixed r.buffer in trunk, it could now be used (again) by
 r.fillnulls instead of r.grow.

 > What was the problem?
 metric=geodesic does not return squared distances, and for other metrics
 the distances as returned by r.grow.distance were in map units, not
 meters, but r.buffer requires meter as unit.

 > Because the fix is needlessly inefficient. The efficient way to compare
 Euclidean distance against a fixed value is "distance!^2 < limit!^2"
 rather than "sqrt(distance!^2) < limit". Calculating Euclidean distance
 invariably involves calculating distance-squared as an intermediate value;
 if you only need the value for a comparison, squaring one value is faster
 than taking the square root of the other.

 Ok, r.buffer now (r50830) uses squared distance for metric=squared and
 regular distance in meters for metric=geodesic.

 Markus M

-- 
Ticket URL: <https://trac.osgeo.org/grass/ticket/1088#comment:24>
GRASS GIS <http://grass.osgeo.org>