[GRASS-dev] Re: [GRASS GIS] #1276: r.null after v.to.rast in winGRASS does not work properly

[GRASS GIS]

#1276: r.null after v.to.rast in winGRASS does not work properly
-----------------------------------------+----------------------------------
 Reporter:  helena                       |       Owner:  grass-dev@…              
     Type:  defect                       |      Status:  new                      
 Priority:  critical                     |   Milestone:  6.4.2                    
Component:  Raster                       |     Version:  6.4.1 RCs                
 Keywords:  r.null, v.to.rast, wingrass  |    Platform:  MSWindows 7              
      Cpu:  Unspecified                  |  
-----------------------------------------+----------------------------------

Comment(by hamish):

 Replying to [comment:18 hamish]:
 > > is r50497 "SF_CLOSE_DESCRIPTOR should actually close the
 > > descriptor" related? should it be backported?

 Replying to [comment:20 glynn]:
 > Maybe. It's possible that the DBMI driver is inheriting the client
 > ends of the pipes (i.e. the write end of the driver's stdin and the
 > read end of its stdout). If that is the case, it would prevent the
 > driver from receiving EOF on stdin when the client dies.

 backported to devbr6 in r50529, but no luck.

 the driver now being closed (or not opened) the 6.4 and 6.5 nightly
 wingrass builds allows r.null to work as it should; and the primary
 subject of this ticket is resolved.

 but if the module exits with a G_fatal_error (e.g. if you typo col=speed2
 which doesn't exist) the dbf.exe and cmd.exe are left running, even when
 you exit grass.


 Hamish

-- 
Ticket URL: <https://trac.osgeo.org/grass/ticket/1276#comment:24>
GRASS GIS <http://grass.osgeo.org>