The QWidget of the window decoration is otherwise still thinking that
the button is pressed and waits for a release. Thus the next click on
the decoration doesn't trigger the move mode.
If the user actions menu is closed Qt looks for a QWidget at the mouse
position. If it finds one it tries to activate and raise it. If the
QWidget at the mouse position is a window decoration, it gets raised
above the Client. This makes the window unfortunately unusable.
To prevent this from happening we listen for the ZOrderChange event
in our event filter on the decoration widget and unconditionally lower
the decoration widget again - we never want the decoration widget to
be above our Client, so we can just always lower it. We have to use
the low level functionality and cannot use QWidget::lower as that would
result in a loop.
There is a check in Client::buttonReleaseEvent() for the state of the
mouse buttons compared to the button masks for button 1, 2, 3 (X11
button indices).
The check was:
if ((state & (Button1Mask & Button2Mask & Button3Mask)) == 0) { ... }
<=> if (state & 0 == 0) <=> if (true)
This change assumes what the check was supposed to be and fixes the problem. The correct fix was proposed by Thomas.
REVIEW: 113359
1. when adding a full damange, that must not replace existing (larger) repaints
2. emit geometryChanged before invoking and to update shadowGeometry through addRepaintFull
BUG: 324560
FIXED-IN: 4.11.2
During startup we only create the request, the reply will be fetched
once the atom is needed.
To make proper use of this async behavior the creation of Atoms is
moved directly to the claim of the manager selection, so they can be
fetched while we wait for the previous manager selection to give up
on it.
If the eventWindow is none the check for InputIdMatchPredicate will find
a matching Client and pass the even through the windowEvent filter which
returns true for all not handled events and thus filters out all events
processed later on in KWin.
This explains why some events were eaten...
Using a QTimer to move the resize/move related code to the end of the
event queue. In case there is another motion event in the queue it will
cancel the timer.
Instead of inspecting the XEvent queue we create a Timer with a
singleshot of 0 msec to move the setActive(false) call to the end of the
event handling. In case there is a matching FocusIn event this will be
handled before the timer fired and can cancel the timer.
the present check can break quick everything depending
on screen snapping settings.
It so far. only worked for quick maximization due to horizontal
"judder" and failed whenever that judder was not possible
(due to screen snapping on left and/or right edge)
BUG: 322852
FIXED-IN: 4.11
REVIEW: 111740
We get the missing send event attribute from anding the event's
response type with 0x80.
This fixes the issue that all open windows got unmapped. The reason for
this was that the existing new Client got reparented to the wrapper which
caused the UnmapNotify event which we used to release the Client and by
that unmapping the window completely.
Focus out handling used to check the event queue for a matching focus in
event to prevent short flickers when no window is active. This is not
possible with XCB and needs a replacement. Maybe a short timer event.
Interestingly the attribute send_event from XUnmapEvent does not exist in
xcb_unmap_notify_event_t and also the X protocol doesn't know anything
about send event.
Button Press/Release do no longer fall through to motion notify as
there is no shared mouse event in xcb. Also the methods in Effects and
TabBox are adjusted to process only button press/release or motion
notify.
ScreenEdges are no longer checked for button press/release. They don't
interact on button press/release so there is no need to check it.
Forwards all xcb events to Workspace::workspaceEvent() which got changed
to process xcb events instead of XEvents. So far all handling is
disabled. Will be re-enabled step by step in the following patches.
Unfortunately the Xcb::Window wrapper cannot be used for the client
window as the client should not be destroyed by KWin.
All the API calls except XSelectInput are changed to xcb and the name is
adjusted to m_client to follow the naming scheme.
The Xcb::Window nicely encapsulates the created wrapper window. As
almost all code is adjusted, the variable is also renamed to
m_wrapper to follow the normal naming scheme.
KWin::Cursor can track changes to the cursor image. It supports a
start/stop tracking to not handle these events if nobody is interested in
them. When enabled and the cursor image changes a signal is emitted with
the serial number of the new cursor image.
To track cursor image changes xcb_xfixes_select_cursor_input is used (see
XFixes Version 5.0 protocol, section 7).
This could be useful for the zoom effect when it replaces the cursor.
REVIEW: 110519
It's not a typical singleton as the ctor is not taking a Workspace* and
needs addtional data to be passed to NETRootInfo.
All the initialization code is moved to RootInfo::create() and the tear-
down code is moved to RootInfo::destroyed(). This includes the support
window which used to be a member of Workspace. It's only needed by
RootInfo, so there is no need to have the ownership inside Workspace.
Instead of using a QWidget we just create a normal window through xcb.
It gets destroyed again in the tear-down code after the RootInfo got
destroyed.
REVIEW: 110238
Main motivation for this change is that it's unhandy to have the class
definition in workspace.h and client.h while the implementation is in
events.cpp although nothing in events.cpp uses it directly.
By getting it out of workspace.h we get the header a little bit smaller
which should improve compile time given that it's included almost
everywhere.
In events.cpp the enum usage is changed to NETWinInfo as that's the class
where they are defined.
RootInfo does no longer hold a workspace pointer. Where it's needed it
uses the singleton accessor of Workspace.
REVIEW: 110199
Following the approaches of other split out functionality Screens is a
singleton class created by Workspace.
The class takes over the responsibility for:
* screenChanged signal delayed by timer
* number of screens
* geometry of given screen
* active screen
* config option for active screen follows mouse
The class contains a small abstraction layer and has a concrete subclass
wrapping around QDesktopWidget, but the idea is to go more low level and
interact with XRandR directly to get more detailed information.
All over KWin the usage from QDesktopWidget is ported over to the new
Screens class.
REVIEW: 109839
This code never worked (git blame ends at 05507ee2). The event type is
ColormapNotify (32) and not ColormapChangeMask (1<<23).
Nobody noticed for years so it seems like unneeded to be fixed.
REVIEW: 109912
There is only one instance hold by Workspace which means it should follow
the common approach with ::self and ::create.
The hasTabBox is completely removed as it's rather useless and the same
as the ifdef around the usages any way.
REVIEW: 109851
this broke re-embedding clients
XReparentWindow causes an unmap of mapped clients, currently leading into releaseClient()
This will (among other) eg. reparent the client to the root and this is (usually?) executed after the original XReparentWindow, so the client does not end up where it's supposed to be.
REVIEW: 109484
Where possible it is changed to Cursor::pos(), where we cannot use the
Cursor class (e.g. Aurorae) we can at least try to limit the usage to
prevent roundtrips to the X server.
REVIEW: 109178
The code is rewritten in a way to replace the local event loop with event
filtering done in the normal way through events.cpp. Therefore instead of
creating a KillWindow whenever it's needed, there is one available in
Workspace which will be reused on next invocation.
The responsible events are passed from events.cpp to KillWindow for
processing.
In order to port the keycode to symbol to XCB, KWin now finds
xcb_keysyms and links it.
To get the right cursor KWin now links the XCursor library which is
unfortunatelly an XLib based library, but there seems to be no XCB
replacement.
The new class FocusChain manages two different kind of focus chains.
First of all there is a most recently used focus chain which is primarily
used for TabBox.
Then there is one focus chain per virtual desktop. These chains are used
to determine which Client needs to be activated when e.g. switching to a
virtual desktop.
The individual chains are implemented as a simple QList of Client* with
the most recently used Client as the last element. That way one can see
it as a LIFO like structure.
The desktop focus chains are internally represented as a hash with the id
of the virtual desktop as the key and a list as described as the value.
FocusChain is a singleton which provides some methods to manipulate the
chains and to get a specific Client for a task (e.g. TabBox).
While splitting out the code some unused code inside TabBox got removed
as well as some activities related code (windows cannot be moved while
switching activities).
REVIEW: 107494
In fact it already used to be a Singleton as there is just one object
hold by the Singleton Workspace. So let's make it a proper Singleton
following our kind of standard approach of having a ::create factory
method called from Workspace ctor and a ::self to get to the singleton
instance.
This rewrite is mostly motivated by the need to handle multi screen
setups correctly. That is have edges per screen and not for the combined
geometry. Also porting from XLib to XCB has been a motivation for the
rewrite.
The design of the new ScreenEdge handling is described in the
documentation of ScreenEdges in screenedge.h.
In addition the following changes have been performed:
* move configuration from Options to ScreenEdge
* add screen edge information to Workspace::supportInformation (obviously
replaces what had been read from Options)
* have Workspace hold a pointer to ScreenEdges instead of an object
* forward declaration of ScreenEdges in workspaces.h, this explains the
seemingly unrelated changes of just another include in some files
BUG: 290887
FIXED-IN: 4.11
The extension handling is removed from kwinglobals and moved into the
xcbutils in KWin core in namespace KWin::Xcb. The motivation for this
change is that the Extensions are only used in KWin core and are marked
as internal. So there is no need to have them in the library.
What remains in Extensions are the non-native pixmaps. This will be
removed once we are on Qt 5 as QPixmap can no longer reference an XPixmap.
The remaining code in kwinglobals also still initialize the XLib versions
of extensions emitting events. It seems like there are no XEvents emitted
if not done so even if the extension is correctly initialized with xcb.
This needs to be removed once the event handling is ported over to xcb.
REVIEW: 107832
The ownership for virtual desktops is moved from Workspace into a new
VirtualDesktopManager. The manager is responsible for providing the count
of virtual desktops and keeping track of the currently used virtual
desktop.
All methods related to moving between desktops are also moved from
Workspace to the new manager, though all methods related to Clients on
Virtual Desktops remain in Workspace for the time being. This is to have
the new manager as independent from KWin core as possible.
An rather important change for the handling of virtual desktops is that
the count and the id of a desktop is now an unsinged integer instead of
an integer. The reason for that is that we cannot have a negative count
of desktops as well as it is not possible to be on a desktop with a
negative identifier.
In that regard it is important to remember that a Client can be on a
desktop with a negative identifier. The special value for a Client being
on all desktops is handled by using -1 as a desktop. For the time being
this is not adjusted but instead of comparing the virtual desktop ids one
should prefer to use the convenient methods like isOnDesktop and
isOnAllDesktops. This would allow in future to internally change the
representation for on all desktops.
For most actions where the compositor needs to perform an action
(e.g. scheduling another repaint) signals were already emitted.
So it's easier to just connect the signals to the Compositor
which in turn makes the code much more readable.
All signals are connected from the Workspace when either the
Compositor gets constructed or a Toplevel gets created.
The Scene has always been created and destroyed inside what is
now the split out compositor. Which means it is actually owned
by the Compositor. The static pointer has never been needed
inside KWin core. Access to the Scene is not required for the
Window Manager. The only real usage is in the EffectsHandlerImpl
and in utils.h to provide a convenient way to figure out whether
compositing is currently active (scene != NULL).
The EffectsHandlerImpl gets also created by the Compositor after
the Scene is created and gets deleted just before the Scene gets
deleted. This allows to inject the Scene into the EffectsHandlerImpl
to resolve the static access in this class.
The convenient way to access the compositing() in utils.h had
to go. To provide the same feature the Compositor provides a
hasScene() access which has the same behavior as the old method.
In order to keep the code changes small in Workspace and Toplevel
a new method compositing() is defined which properly resolves
the state. A disadvantage is that this can no longer be inlined
and consists of several method calls and pointer checks.
