Re: is metacity _really_ ICCCM-compliant ?

[Elijah Newren <newren gmail com>]

I don't fully understand the specs yet, but I'll take my best guess...

On Tue, 7 Sep 2004 09:53:35 +0200, Michael Boccara <boccara sgi com> wrote:
> When querying properties of metacity's windows, I noticed they are not
> really compliant with the specs described in
> http://freedesktop.org/Standards/wm-spec/1.3

That's the EWMH, not the ICCCM.  The ICCCM can be found, for example,
at http://tronche.com/gui/x/icccm/.

> For example, when querying the _NET_WM_WINDOW_TYPE of a popup menu,
> I should get _NET_WM_WINDOW_TYPE_MENU and have the
> WM_TRANSIENT_FOR atom set to my application window.  Instead, the type
> is _NET_WM_WINDOW_TYPE_NORMAL, and WM_F|TRANSIENT_FOR is not set.

Could you specify the case where this is happening?  Note that the
EWMH states, "[_NET_WM_WINDOW_TYPE] SHOULD be set by the Client before
mapping to a list of atoms indicating the functional type of the
window. This property SHOULD be used by the window manager in
determining the decoration, stacking position and other behavior of
the window."  I believe what this means is that the Window Manager is
/not/ the one that sets these properties--the application is.  So if
you're getting the wrong type, I believe it is probably the
application that is at fault.

> Is the lack of ICCCM-compliancy a known issue for metacity ?
> Because the following page
> http://cvs.gnome.org/viewcvs/metacity/COMPLIANCE?view=markup
> pretends that metacity is compliant, at least concerning the
> _NET_WM_WINDOW_TYPE property

I would be surprised if there was a bug that made Metacity not be
compliant with the EWMH, at least in the locations where Metacity
claims compliance.  If you could specify the application, we might be
able to check further.

> Thanks,
> Michael

Hopefully if any of my statements are wrong, someone else can clear them up...

Cheers,
Elijah