because if one timestamp is reused for activating window A and then
immediatelly window B, and A will want to set focus to one
of its subwindows after receiving WM_TAKE_FOCUS, and will lag a bit,
it will actually set the focus after B, and will win. Fixes#71704.
If somebody still gets the warning, I'd be interested in the backtrace.
svn path=/trunk/kdebase/kwin/; revision=278208
don't use Workspace::activeClient() as the active client, but instead
the most recent should_get_focus client - that may make a difference
if a window was just activated, but the matching FocusIn event that'll
make it to be activeClient() wasn't yet processes. Visible with apps
that show several windows at the same time after startup.
svn path=/trunk/kdebase/kwin/; revision=270831
windows while embedding them, allowing KWin to figure out it's being used
and that it's not going away. This hack avoids the reparenting fight
between KWin and QXEmbed where QXEmbed started to loose after the recent
fixes.
In order to make systray really work, QXEmbed still needs some fixes
related to not destroying the embedded window.
svn path=/trunk/kdebase/kwin/; revision=263178
but at least now the code doesn't look like uncommented random something.
And who knows, maybe I even fixed it ;).
CCMAIL: staikos@kde.org
What the hell does "Show unmanaged windows on:" mean in the kcm dialog?
svn path=/trunk/kdebase/kwin/; revision=257723
the focus and just are there. Therefore, as a part of focus stealing
prevention, let's also block raising of windows. Also added special
request for Kicker, as taskbar requests for raising should be obeyed.
(Not that this does anything to KPassivePopup.)
svn path=/trunk/kdebase/kicker/taskmanager/; revision=255070
screen and noborder there is a good idea. How does an application get notified
that it was switched to full screen mode? How do you revert these settings?
svn path=/trunk/kdebase/kwin/; revision=254230
a hack in Qt that was trying to work around something that Qt-2.2.4 changelog
claims to be Xlib bug and that's no longer in Qt anyway.
Fixes#24087.
svn path=/trunk/kdebase/kwin/; revision=220339
Correct detection of minimized windows for KWin HEAD and kwin_iii.
Correct detection of window type even if the list includes unsupported ones.
svn path=/trunk/kdebase/kwin/; revision=219298
clients by adding an unnecessary 'enable-bidi' checkbox which was never agreed
upon on the kwin mailing list.
The use of --reverse is sufficient without needing this checkbox.
Please discuss largish patches like these first, especially during a feature
freeze!
(Everyone also seems to forget the kwin-styles in kdeartwork)
- Make clients look the same with --reverse instead of the broken state they
were in previously. If the button ordering is not to your liking in reverse
mode, just change the button positions. (Or send patches for those clients
that have not got these features yet).
- Clean up popupMenu handling somewhat.
- Remove a global static object in keramik.
svn path=/trunk/kdebase/kwin/; revision=174079
Plugin developers should use only Workspace::showWindowMenu() instead
of calling Workspace::clientPopup() and playing with the popup.
svn path=/trunk/kdebase/kwin/; revision=173296
This patch should fix such interesting things like closing the toplevel
menubar with Alt+F4 or moving Kicker to a different desktop using
keyboard shortcuts.
svn path=/trunk/kdebase/kwin/; revision=170646
Prefix all the #ifndef #define ... #endif symbols with KWIN_ , just in case
somebody uses CLIENT_H or so too.
And don't include config.h .
svn path=/trunk/kdebase/kwin/; revision=168940
Edit kwinrc and use either of:
Placement=StupidlyCentered
Placement=StupidlyZeroCornered
and then, at a prompt:
dcop kwin KWinInterface reconfigure
svn path=/trunk/kdebase/kwin/; revision=163755
