Closed GoogleCodeExporter closed 9 years ago
Just wanted to report that same problem, but you were first :)
I did a little investigation and as far as I can see, xxkb relies on the wm
calling
XReparentWindow when a window is created. But XMonad doesn't call this function
(twm
does and most other wms, I suppose, too). So the question is: is this an
xmonad's
failure or a failure of xxkb to rely on a non-standard wm behavior?
Original comment by sgsoftw...@gmail.com
on 8 Apr 2008 at 2:08
xmonad is a non-reparenting window manager by design. I would expect xxkb to
fail with other non-
reparenting window managers such as ratpoison and dwm as well.
Original comment by allber...@gmail.com
on 8 Apr 2008 at 2:19
Yes indeed, you are right. xxkb doesn't work properly in dwm and ratpoison.
Original comment by sgsoftw...@gmail.com
on 8 Apr 2008 at 5:13
Closing this, as its an application issue.
Do we have a definitive document somewhere that states that applications
relying on
reparenting are in violation of the spec?
Original comment by don...@gmail.com
on 8 Apr 2008 at 5:46
http://tronche.com/gui/x/icccm/sec-4.html#s-4.2.1
(I'm not sure I want to trust that as a stable, or official, link; but x.org's
seems to be PDF)
The key is that it says *some* window managers reparent and *if* windows are
reparented.
Original comment by allber...@gmail.com
on 8 Apr 2008 at 6:18
Thanks as always, allbery.b.
The spec sez: the app is wrong.
Original comment by don...@gmail.com
on 8 Apr 2008 at 6:27
I'm not sure what you mean by 'switching troubles', but my xxkb is working good
in
xmonad.
Example steps:
1. Run xxkb.
2. Spawn two terminal windows.
3. Switch to russian keyboard layout in first terminal window.
4. Move to another terminal window. Set keyboard layout to 'us'.
5. Back to first window - layout is preserved for this window (still 'ru').
6. Move to second window - layout is preserved for this window (still 'us').
That's exactly that behaviour I'm expecting from xxkb.
Original comment by vese...@gmail.com
on 10 Apr 2008 at 6:33
Original issue reported on code.google.com by
don...@gmail.com
on 7 Apr 2008 at 6:20