Bug#551807: openbox: xy-maximized windows uncorrectly y-sized with xfce4-panel + matchbox-keyboard
Nico Golde
reassign 550221 matchbox-keyboard
thanks
Hi,
* Pini <gilles.f...@free.fr> [2009-10-21 12:59]:
> I use xfce4 + openbox + matchbox-keyboard on an Openmoko FreeRunner[0].
> xfce4 is configured with 1 bottom xfce4-panel.
>
> All applications are xy-maximized using this configuration snippet in
> ..config/openbox/rc.xml:
>
> <application class="*">
> <decor>no</decor>
> <maximized>true</maximized>
> </application>
>
> With either xfce4-panel[1] or matchbox-keyboard[2] displayed, the
> xy-maximization is fine.
>
> When both are displayed the xy-maximized windows are too high. See [3]:
> the dialog's buttons are hidden behind the matchbox-keyboard.
>
> I guess this is an openbox bug since windows' sizes are ok with other
> window managers - e.g. matchbox.
I am reassigning this and #550221 to matchbox keyboard now. I don't know this
tool nor do I have the time to invest its x properties but it seems all
unexpected behaviour recently involves this tool. I see no wrong behaviour in
how openbox is handling ICCCM and EWMH compliance.
Cheers
Nico
--
Nico Golde - http://www.ngolde.de - ni...@jabber.ccc.de - GPG: 0xA0A0AAAA
For security reasons, all text in this mail is double-rot13 encrypted.
Gilles Filippini
thanks
Hi,
Nico Golde a écrit :
> * Pini <gilles.f...@free.fr> [2009-10-21 12:59]:
>> I use xfce4 + openbox + matchbox-keyboard on an Openmoko FreeRunner[0].
>> xfce4 is configured with 1 bottom xfce4-panel.
>>
>> All applications are xy-maximized using this configuration snippet in
>> ..config/openbox/rc.xml:
>>
>> <application class="*">
>> <decor>no</decor>
>> <maximized>true</maximized>
>> </application>
>>
>> With either xfce4-panel[1] or matchbox-keyboard[2] displayed, the
>> xy-maximization is fine.
>>
>> When both are displayed the xy-maximized windows are too high. See [3]:
>> the dialog's buttons are hidden behind the matchbox-keyboard.
>>
>> I guess this is an openbox bug since windows' sizes are ok with other
>> window managers - e.g. matchbox.
>
> I am reassigning this and #550221 to matchbox keyboard now. I don't know this
> tool nor do I have the time to invest its x properties but it seems all
> unexpected behaviour recently involves this tool. I see no wrong behaviour in
> how openbox is handling ICCCM and EWMH compliance.
I've done some testing around this bug and set up a simple testcase to
reproduce it (attached).
AFAIUI, all lies in how STRUTs are handled by the window manager. IMO
they should be cumulative: when win1, win2, ..., winN set STRUTs, the
resulting workarea should be cropped by STRUT1 + STRUT2 + ... + STRUTN.
The testcase demonstrates that openbox currently crops the workarea with
max(STRUT1, STRUT2, ..., STRUTN) instead of their sum. This way is wrong
- IMHO - since only one STRUT can be set at once per display side then.
Moreover, setting - say - a right STRUT + a bottom STRUT results in an
overlapping in the workarea's bottom right corner.
=> Reassigning to openbox.
To reproduce:
$ gcc -lX11 -c testcase.c -o testcase
$ ./testcase &
$ ./testcase &
...
Each testcase instance defines a bottom STRUT of 1/3 of the current
workarea.
Hope this helps,
_Gilles.
Dana Jansens
Openbox. Openbox does not use rounded titlebars at this time, I would
guess it is Metacity from the look of it.
--
To UNSUBSCRIBE, email to debian-bugs-...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listm...@lists.debian.org
Nico Golde
* Dana Jansens <dan...@orodu.net> [2009-11-05 19:45]:
> The window manager shown in the pictures documenting the bug is not
> Openbox. Openbox does not use rounded titlebars at this time, I would
> guess it is Metacity from the look of it.
I guess that comes from xfce...
Gilles Filippini
Hash: SHA256
Hi,
Nico Golde a écrit :
> Hi,
> * Dana Jansens <dan...@orodu.net> [2009-11-05 19:45]:
>> The window manager shown in the pictures documenting the bug is not
>> Openbox. Openbox does not use rounded titlebars at this time, I would
>> guess it is Metacity from the look of it.
>
> I guess that comes from xfce...
As I've just explained on the upstream's BTS follow up, the confusion
comes from the fact I've taken VNC snapshots of my FreeRunner - running
openbox - from my laptop - running metacity.
I can assure that I've experienced this bug with openbox.
Running the provided testcase should be enough to convince you.
Thanks,
_Gilles.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.10 (GNU/Linux)
iQEcBAEBCAAGBQJK8zkQAAoJEO/obGx//s+DGVAH/jLafrPc4OOjcSgXvvZrXNZV
GTK8GM+sflAy33xTnu1RpXOt1aFOXezd2mTuGu8J3rI+mzkUSlyOExOjZKiMrcfB
GENuxuNG4qKG+6ev5lOeJf1VhLs27dshGYGQrMR90FG95CoMMWczLi4IARiCx9ef
rX5h95EvS+rnQOTMZPDwdpNijVKY7kEcMJI/+qlRj2tIG2y4ipTr92Ey15XS7OjB
/ofG28wAFPsMccteAj0FIq0GNf8QngMPSR64LG5KUECRoKgYGI3b9gkmEsYVxqna
ZbqPg5dfIrAeQUZDrxbA6yxzLsX6RTp/KpOWNQA9+0gtq45u62PbCAXOsQt3lhg=
=Etd7
-----END PGP SIGNATURE-----