[Xquartz-dev] Fully self contained Xquartz.app?

[Eeri Kask]

Am 07/16/2010 12:03 AM, Jeremy Huddleston <jeremyhu at apple.com> schrieb:
> This sounds like it may be this issue:
> 
> http://xquartz.macosforge.org/trac/ticket/58
> 
> On Jul 15, 2010, at 12:03, Bob wrote:
>> Yes, I usually work in fullscreen mode on my laptop and this is a huge
>> annoyance.  It happens intermittently about once an hour for me and
>> this is using quartzwm with terms and emacs (so I don't think your
>> lack of a WM is the issue).  I seem to remember a long time ago there
>> was some discussion of this and it was stated that it is caused by
>> things outside of X11 and thus hard (seemingly impossible?) to fix.
>> Also, I think it is related to the momentary loss of focus seen in
>> rootless mode but it is far more jarring in fullscreen mode (and seems
>> to last much longer).
>>>
>>> On Jul 15, 2010, at 4:44 AM, Eeri Kask wrote:
>>>> The most I can say today, it seems window stacking order changes or
>>>> unmap/map requests result in effects which can be explained like the
>>>> misordered delivery or dropping of some enter/leave events, which
>>>> then understandably often results in a mess in focus management; let
>>>> it be focus by PointerRoot by the X-server or by assigning by some
>>>> window manager implementing sloppy-focus-follows mouse.


Yes the above ticket deals with the once-per-hour visible switches
from fullscreen-X to Aqua screen and back to X, a phenomenon I
managed to successfully defeat by removing

    /System/Library/UserEventPlugins/iCalAlarmAgent.plugin/

from the system.

The problems I pointed out above most probably are unrelated to this
known issue as they even seem to have no temporal correlations at
all as opposed to the predictability of iCal chronometrical
determinism.  (There are definitely no X->Aqua->X switches
observably occurring this time.)

The enter/leave/focus issues happen often by some client window
being mapped (or changing window stacking order) exactly under the
mouse which then doesn't get appropriate focus from server while in
PointerRoot, and sometimes don't get Enter-/LeaveNotify events.

So it is different from the above iCal because the focus apparently
doesn't leave X, the mouse/focus only appear in the wrong X-window;
due to some window-configure (raise/lower/map/unmap) side effects.

Greetings,

    Eeri Kask