[MPlayer-dev-eng] [Fwd: Re: fullscreen problem with metacity 2.4]]

[gabor]

hi, 

i sent a mail describing the problem with metacity window manager ( if
you switch to fullscreen, you can't switch back ) to the creator of
metacity ....

he examined the mplayer sources, and sent me the following...


bye,
gabor


-----Forwarded Message-----

> From: Havoc Pennington <hp at redhat.com>
> To: gabor <gabor at realtime.sk>
> Subject: Re: [Fwd: Re: Re: [MPlayer-users] fullscreen problem with metacity 2.4]
> Date: 22 Sep 2002 17:20:31 -0400
> 
> 
> Hi,
> 
> Here is how I would change mplayer, maybe you can forward this along.
> Disclaimer: because mplayer seems to try to special-case window managers
> with vo_fsmode I'm not sure exactly which code is getting used.
> 
>  - it should not set _NET_WM_STATE_STAYS_ON_TOP, this hint does 
>    not exist in the spec
>    (http://www.freedesktop.org/standards/wm-spec/1.2/html/x224.html)
>    and _NET_WM_STATE_FULLSCREEN already implies "on top" so there's no 
>    need to set this. _NET_WM_STATE_ABOVE is only used when there's 
>    not enough info already.
>  - it should set _NET_WM_STATE_FULLSCREEN instead, I don't 
>    currently see code to set _NET_WM_STATE_FULLSCREEN
>  - in the client message for setting either, it should 
>    set data.l[2] to zero instead of leaving it uninitialized
>  - the XUnmapWindow() after sending the client message 
>    is officially speaking breaking the spec, should be
>    XWithdrawWindow(). I don't think it's doing this with metacity but 
>    I'm not sure.
>  - when setting MWM hints, resizing is disabled, but in
>    WM_NORMAL_HINTS, the min/max sizes allow resizing. 
>    If resizing is not desired, min/max size should both be 
>    set, and set to the same value.
> 
> However none of these are the problem, the base problem is that once
> the window is in fullscreen mode, its size is constrained to
> fullscreen size; so requests from mplayer to resize are ignored.  I
> will try to put in some workaround, but as this bug already results
> from working around apps not using _NET_WM_STATE_FULLSCREEN, and the
> workaround for the workaround could cause even more issues, I think
> just adding _NET_WM_STATE_FULLSCREEN support to mplayer will produce
> the best results.
> 
> Havoc
-- 
That's life for you, said McDunn.  Someone always waiting for someone 
who never comes home.  Always someone loving something more than that 
thing loves them.  And after awhile you want to destroy whatever 
that thing is, so it can't hurt you no more.
                -- R. Bradbury, "The Fog Horn"