#2512: r.out.gdal wrongly messes with SetColorInterpretation
------------------------+---------------------------------------------------
 Reporter:  rouault     |       Owner:  grass-dev@…              
     Type:  defect      |      Status:  new                      
 Priority:  normal      |   Milestone:  6.4.5                    
Component:  Raster      |     Version:  svn-trunk                
 Keywords:  r.out.gdal  |    Platform:  All                      
      Cpu:  All         |  
------------------------+---------------------------------------------------

Comment(by mmetz):

 Replying to [ticket:2512 rouault]:
 > r.out.gdal export_band.c currently contains
 >
 {{{
     CPLPushErrorHandler(CPLQuietErrorHandler);
     GDALSetRasterColorInterpretation(hBand, GPI_RGB);
     CPLPopErrorHandler();
 }}}
 >
 > This is a wrong usage of the GDAL API.
 GDALSetRasterColorInterpretation() expects an enumerated value from the
 GDALColorInterp enumeration, whereas GPI_RGB is a value from the
 GDALPaletteInterp enumeration. Consequently, due to the fact that GPI_RGB
 = 1 and GCI_GrayIndex=1, the above snippet will effectively force the
 color interpreation to be Gray level.
 >
 > On GeoTIFF this was without effect before GDAL 1.11, since in those
 versions, SetColorInterpration() in the GeoTIFF driver has no effect. But
 in GDAL 1.11, the SetColorInterpration() in the GeoTIFF driver can modify
 the default PHOTOMETRIC TIFF tag. For example, for a 3 band GeoTIFF file,
 at creation time, the GeoTIFF driver defaults to PHOTOMETRIC=RGB, but if
 later, SetRasterColorInterpretation(GCI_GrayIndex) is called, it will undo
 that setting to fallback to PHOTOMETRIC=MINISBLACK.
 >
 > All in all, I suggest just to remove those 3 lines. The
 CPLPushErrorHandler(CPLQuietErrorHandler) / CPLPopErrorHandler()
 surrounding is a hint that this call is wrong.

 Done in r63682,3 (trunk, relbr70).

-- 
Ticket URL: <http://trac.osgeo.org/grass/ticket/2512#comment:2>
GRASS GIS <http://grass.osgeo.org>

_______________________________________________
grass-dev mailing list
[email protected]
http://lists.osgeo.org/mailman/listinfo/grass-dev

Reply via email to