[GRASS-dev] [GRASS GIS] #2917: r.mapcalc neighbour modifier uses wrong row

GRASS GIS trac at osgeo.org
Mon Feb 22 07:54:20 PST 2016


#2917: r.mapcalc neighbour modifier uses wrong row
--------------------------+---------------------------------
  Reporter:  marisn       |      Owner:  grass-dev@…
      Type:  defect       |     Status:  new
  Priority:  blocker      |  Milestone:  7.0.4
 Component:  Raster       |    Version:  svn-releasebranch70
Resolution:               |   Keywords:  r.mapcalc
       CPU:  Unspecified  |   Platform:  Unspecified
--------------------------+---------------------------------

Comment (by mlennert):

 Replying to [comment:2 marisn]:
 > I confirm based on the test case - issue is solved for 7.1.
 > Now we have only a "small" problem - it has been present in all releases
 since 6.3.1 thus potentially compromising any computation based on it.
 >
 > 1) It must be backported to 7.x and 6.x


 I backported to release70.

 The issue is not present in 6.x.


 >
 > 2) A clarification on potential impact should be provided to allow
 endusers to evaluate potentially affected results of any analysis
 performed with 6.3-7.0.

 What I don't really understand (without having taken the time to delve
 into the code) is why


 {{{
 r.mapcalc "nd = if(landclass96[1,0]!=landclass96, landclass96[1,0],
 null())"
 }}}

 provokes the bug, but


 {{{
 r.mapcalc "nd_temp = landclass96[1,0]"
 r.mapcalc "nd2 = if(nd_temp!=landclass96, nd_temp, null())"
 }}}

 doesn't.

 This would be important to understand in order to explain the impact.
 Also: where do want such an explanation ? A mail to grass-user and grass-
 dev ?

--
Ticket URL: <https://trac.osgeo.org/grass/ticket/2917#comment:3>
GRASS GIS <https://grass.osgeo.org>



More information about the grass-dev mailing list