Martin Landa wrote: > I am not sure if G__ structure from G.h should be completely > initialized by GIS or Raster library or the structures from G.h should > be split and initialized separately by both libraries?
AFAICT, there are two options: 1. The current region belongs to libraster; in which case, G__ belongs in libraster, along with anything which uses it. 2. The current region belongs to libgis; in which case, G__ needs to have the window and window_set fields moved back to libgis, and everything which uses the remaining fields needs to be moved to libraster. As it stands, G.h has been moved to lib/raster, yet G__ is still defined by libgis.Nothing in libgis should be including "../raster/G.h". This needs to be fixed as a priority (personally, I wouldn't have started committing anything until this was sorted out). BTW, I intend to rename Rast.h to raster.h, to match (almost) everything else in GRASS. Vect.h was a mistake, which I see no reason to repeat. -- Glynn Clements <gl...@gclements.plus.com> _______________________________________________ grass-dev mailing list grass-dev@lists.osgeo.org http://lists.osgeo.org/mailman/listinfo/grass-dev
