Philip Boulain wrote: > Paul Pogonyshev wrote: > > As an ugly workaround, try passing `*static_cast <const > > Gdk::Rectangle*> (0)' as a parameter. > > That causes me a lovely segfault. ;)
Well, that was untested and might have caused a segfault, right :) > I actually converted this (inside a Gtk::Widget subclass' on_expose_event() > method): > > get_style()->paint_box(gdk_window, get_state(), Gtk::SHADOW_IN, 0, *this, 0, > 0, 0, get_allocation().get_width(), get_allocation().get_height()); > > To this: > > gtk_paint_box(get_style()->gobj(), gdk_window->gobj(), (GtkStateType) > get_state(), GTK_SHADOW_IN, NULL, this->gobj(), NULL, > 0, 0, get_allocation().get_width(), get_allocation().get_height()); > > Which makes the (hopefully safe) assumption that the Gtk::StateType and > GtkStateType enumerations use the same values. And thanks to whichever > GTKmm developer had the sense to realise that being able to access the C > data structures being wrapped would be useful. :) If there is no gobj() method it must be the same as C enum, I think. But I have made one mistake already :) Paul _______________________________________________ gtkmm-list mailing list [email protected] http://mail.gnome.org/mailman/listinfo/gtkmm-list
