[GRASS-dev] [GRASS GIS] #2637: Get direction raster in clockwise degrees starting from the North

[GRASS GIS]

#2637: Get direction raster in clockwise degrees starting from the North
-------------------------+--------------------------------------------------
 Reporter:  cgravelm     |       Owner:  grass-dev@…              
     Type:  enhancement  |      Status:  new                      
 Priority:  normal       |   Milestone:  7.0.1                    
Component:  Raster       |     Version:  7.0.0                    
 Keywords:               |    Platform:  MacOSX                   
      Cpu:  Unspecified  |  
-------------------------+--------------------------------------------------

Comment(by cmbarton):

 Replying to [comment:3 neteler]:
 > Replying to [comment:1 annakrat]:
 > > You can just use r.mapcalc "angle_cw = -angle_ccw + 450", this will
 give you what you need. It outputs angles from 90 to 450, but that's
 typically not a problem.
 >
 > ... or you solve it with an if() condition.
 >
 > Shell script solution:
 >
 > {{{
 > rotate_angle()
 > {
 >  is_negative=`echo "$1" | awk '{printf "%d\n", $1 < 0. ? 1 : 0}'`
 >
 >  if [ $is_negative -eq 0 ] ; then
 >    tmp=`echo "$1"   | awk '{printf "%f\n", 360. - $1 + 90.}'`
 >    tmp=`echo "$tmp" | awk '{printf "%f\n", $1 >= 360. ? $1 - 360. :
 $1}'`
 >    echo "$tmp"
 >  else
 >    echo "NA"
 >  fi
 > }
 >
 > rotate_angle 270
 > #[1] 180
 > rotate_angle 180
 > #[1] 270
 > }}}
 >
 > Likewise you could use r.mapcalc and its eval() function.
 >
 > > I guess it could be implemented in r.slope.aspect.
 >
 > Note the (old) patch:
 >
 > raster/r.slope.aspect/r_sl_asp_northangle_diffs.tar.gz
 >
 > Important: the output of r.slope.aspect can be used as input elsewhere,
 hence a flag would increase the risk to mess up subsequent calculations.


 I don't see how a flag would mess up other calculations. Different other
 routines use different ways of representing aspect. So you need to know
 what is needed and select that even now. There are also potential uses
 that could benefit by having aspect in the cardinal directions. It is a
 pain to always have to convert the output to make it readable for humans
 in normal ways too. A simple flag seems like a nice improvement here.

 Michael

-- 
Ticket URL: <http://trac.osgeo.org/grass/ticket/2637#comment:5>
GRASS GIS <http://grass.osgeo.org>