True, I was confusing versions. The "normal" syntax does not work with 2.1.4.
This is weird, because you need to indicate the algorithm to use...so without that 0 it shouldn't work. Another one of this annoying SAGA mysteries.... I will fix that call to gdal_io. Other modules, however, seems to accept the algorithm index/name correctly. I tried the export passing the algorithm name instead of 0, and it also fails Thanks for your help! BTW, did anyone find SAGA 2.1.4 to be extremely slow?. It takes a lot of time to launch SAGA before it starts doing the actual computing, there is a huge overhead. With several calls per Processing algorithm, the time it takes to run is rather long... 2015-01-20 11:55 GMT+01:00 Giovanni Manghi <giovanni.man...@gmail.com>: >> No problem. Actually, that "None" version shouldn't appear, i ahve >> added a small fix to improve that version detection >> >> The error you are getting seems to be caused by SAGA believing that a >> layer is already exported, so it is not calling the export routines >> before calling the algorithm itself. I had seen this before, and I >> have now added a fix for it. >> >> All this changes are applied in this commit >> >> https://github.com/qgis/QGIS/commit/e9228b67c934f4eaeae0beb8a033916f5317318f >> >> Hope this helps > > > Hi Victor, > I tested also under Window, from the command line (no Processing > involved, but of course affected) and SAGA 2.1.4 io_gdal also fails > when > > saga_cmd io_gdal 0 -TRANSFORM... > > while > > saga_cmd io_gdal -TRANSFORM... > > > works > > > cheers > > > -- G -- _______________________________________________ Qgis-developer mailing list Qgis-developer@lists.osgeo.org http://lists.osgeo.org/mailman/listinfo/qgis-developer