I think that having two separate windows isn't going to work out. There are no Rast_get_window(), Rast_window_rows() etc functions, which means that a lot of code is examining G__.window when it's actually R__.window which matters. If the two windows ever differ, lots of stuff will break.
I'm not sure whether it's feasible to change all of the code to deal with this. On one hand, there isn't much code which actually needs to use G__.window, so we could change calls to the G_* functions to Rast_* except for the specific cases which actually need to use G__.window. Or we could just remove R__.window and make the raster library use G__.window. We still need Rast_set_window(), as this updates the column mappings. Comments? -- Glynn Clements <[email protected]> _______________________________________________ grass-dev mailing list [email protected] http://lists.osgeo.org/mailman/listinfo/grass-dev
