Quoting Costin Manolache: > On 08 Sep 2004 11:03:12 +0200, Sven Neumann <[EMAIL PROTECTED]> wrote: > > Hi, > > > > sorry for the late reply. I only just found your mail in my inbox. > > > > Costin Manolache <[EMAIL PROTECTED]> writes: > > > > > When you create a widget ( button, label, etc ), you can specify the > > > absolute position and size. You may have overlaping widgets - and they > > > should be displayed in the z-order of the widget. By default the > > > zorder is given by the order you add childs to the parent - newest > > > child is above. This can be controlled with moveAbove(swt) which is > > > mapped to gdk_window_raise(). > > > > I am sorry but as far as I know GTK+ doesn't support any overlapping > > widgets. You might be able to create them using some rather unusual > > GTK+ and GDK calls but there is no defined behaviour for them. > > Do you have any pointer to gtk+ documentation regarding (non)support > of overlapping > widgets ? SWT ( eclipse.org ) does support z-order, and their GTK > implementation > seem to rely on GTK/GDK - I assume it works because X11 supports this. > > If GDK/GTK+ explicitely doesn't support this - then probably SWT is > the right layer to > implement this.
Did you considerate an SWT implementation directly on top of DirectFB? I wrote an AWT 1.1 implementation from scratch, specially designed for DirectFB from the beginning. It's *very* fast, even on an embedded system. With nearly 200 MIPS and hardware acceleration I wrote a game in Java using this AWT, producing >50 frames per second in a 600x300 pixel portion of the screen with about 60% CPU load. Measurements have shown that the blitter engine is utilized 100%. So the frame limit is the memory bandwidth limit of the accelerator. -- Best regards, Denis Oliver Kropp .------------------------------------------. | DirectFB - Hardware accelerated graphics | | http://www.directfb.org/ | "------------------------------------------" Convergence GmbH
