[GRASS-dev] [GRASS GIS] #1728: r.slope.aspect: possible data type bug?

GRASS GIS trac at osgeo.org
Mon Sep 10 16:58:28 PDT 2012


#1728: r.slope.aspect: possible data type bug?
----------------------------+-----------------------------------------------
 Reporter:  hamish          |       Owner:  grass-dev@…              
     Type:  defect          |      Status:  new                      
 Priority:  normal          |   Milestone:  7.0.0                    
Component:  Raster          |     Version:  svn-trunk                
 Keywords:  r.slope.aspect  |    Platform:  All                      
      Cpu:  All             |  
----------------------------+-----------------------------------------------
 Hi,

 I noticed what might be a bug in the r.slope.aspect code.

 If output map type precision (CELL,FCELL,DCELL) is not given, the
 data_type variable defaults to DCELL*. According to the code comments,
 data_type should be wrt the input map, not the output map.
 (* while the output map type defaults to FCELL)

 If output map type is given, the data_type is set to the same. The
 intention is a bit clearer in the legacy code, a just-now change to
 today's trunk removed some old code around that.

 should it be using Rast_map_type() instead? Which type should the allocs
 called later with data_type be using?

 applies to all branches.

 a DCELL input map with prec=int specified on the command line does not
 cause a segfault in my tests, but none the less the code seems a bit
 muddled...


 thanks,
 Hamish

-- 
Ticket URL: <https://trac.osgeo.org/grass/ticket/1728>
GRASS GIS <http://grass.osgeo.org>



More information about the grass-dev mailing list