manolo gouy wrote: > Thanks Albrecht, your idea of placing the traverse call earlier > indeed saves an offset calculation that was effectively done twice. > I have further improved that by putting the traverse call after > the fl_pop_clip() call, so we have no deeper clip stack cost > (when the subwindow is printed, it will be clipped by print_widget), > and by removing a useless origin() call that's immediately followed > by another one.
Very good. I had seen the useless origin() after posting my patch, but didn't want to do any more modifications. Thanks for removing that as well. And you're right about clipping. It's a pleasure to work with you on this project. > The printer demo was modified a few days ago including a sub-sub-window > in it to test for this issue. You can check that both our codes > handle that correctly. Ah, I see. But the more general case would be a hierarchy of window - group - window - group - window with non-zero x/y coordinates of each. Anyway, the current code is _designed_ to cope with this. I hope it will do. Albrecht _______________________________________________ fltk-dev mailing list fltk-dev@easysw.com http://lists.easysw.com/mailman/listinfo/fltk-dev
