If the start Client is not part of the focus chain the call to
nextClientFocusChain() cannot return the Client again. So the loop break
condition is never reached and as the focus chain is not empty the call
always returns a not null Client which means KWin is caught in an endless
loop.
This change checks that the starting Client is in the focus chain and if
not the first Client of the focus chain is used.
BUG: 306260
BUG: 306275
FIXED-IN: 4.9.2
So far the first Client to be shown in the list (that is the
currently active window) was inserted as the last client into
the list by prepending it to the list.
This meant that if another Client actually blocks the inclusion
of the currently selected Client (e.g. only one window per app)
the currently active Client never got included in this list.
This change ensures that the recently used model switching has
the starting Client as the first Client in the list and also
simplifies the code.
Stacking order switching mode is not adjusted as it seems rather
broken, like the comment already says.
BUG: 304950
FIXED-IN: 4.9.1
REVIEW: 106139
The recently used mode of TabBox uses the active window as the entrance
into the focus chain. If there is no active window it does not find any
Clients. To solve this issue the ClientModel now uses the first entry
of the focus chain in case there is no active window.
BUG: 305449
FIXED-IN: 4.9.1
REVIEW: 106088
If the ClientModel does not contain any Clients, which can
happen if there is no desktop window, accessing the data of a
ModelIndex returns an invalid QVariant. Because of that it
needs to be ensured that the QVariant is valid before trying to
cast it to a Client Pointer.
BUG: 304620
FIXED-IN: 4.9.1
REVIEW: 105935
Since the QML port the LayoutMode had been hardcoded to vertical
layout making it a completely useless code-path.
MinWidth/Height are nowadays completely controlled by the QML
theme. They were not read anywhere except in the kcm, but there
not even bound to a ui element.
The selectedItemLayoutName is also not used anymore with the
new themes.
* KDE/4.9:
Fix callback race in the session save and shutdown code.
Update for 4.9.0 release
SVN_SILENT made messages (.desktop file)
SVN_SILENT made messages (.desktop file)
Fix regression in QML shutdown menu. Ampersands should be removed from menu entries' text by default.
Fix typos in a message that broke string freeze.
Fix potential parallel build failure.
Conflicts:
CMakeLists.txt
The method was missing a check whether the weak pointers in the
internal list got deleted. This could in very unlikely cases
lead to a crash.
In order to verify that adding the null pointer check fixes the
crash a unit test is added to simulate the situation of a
pointer being deleted. This required to add a mock a few
classes of TabBox. A MockTabBoxHandler and MockTabBoxClient are
added implementing the specific interfaces. The DeclarativeView
is completely mocked to make the linker happy. Including the
actual implementation is not possible as it pulls in half of
KWin core.
BUG: 303840
FIXED-IN: 4.9.0
REVIEW: 105645
Client holds a SharedPointer to the TabBoxClient and only
provides access to a WeakPointer which is passed to TabBox.
ClientModel is adjusted to hold a list of WeakPointers instead
of the direct pointers.
This fixes the following reproducable crash:
1. Configure both primary and secondary TabBox with different
layouts
2. Use primary TabBox
3. Close a window, best the one which used to be active
4. Use secondary TabBox
-> Crash
The reason is that the ClientModel still contains the pointer
to the deleted TabBoxClient in step 3 and while creating the
layout access to the TabBoxClient is needed to get the Client's
icon.
By using the weak pointer it can be ensured that we don't try
to dereference the deleted pointer and prevent the crash.
CCBUG: 290482
CCBUG: 285747
CCBUG: 237345
REVIEW: 105000
Flicking has to stop at the bounds in the window switcher
layouts. It's not natural on the desktop that they are
flickable in the first place.
Window Strip is unchanged as this layout is intended to be used
on touch and flickable user experiences.
REVIEW: 105001
cppcheck complained about that piece of code and it was not
really good. A TabBoxClient belongs to exactly one Client which
means the setter should for Client should not be there in the
first place and was only used during construction.
REVIEW: 104909
A CMake variable is used to specify the name of the binary.
By default this is "kwin" but building for PA changes the
name to "kwinactive". The variable adjusts all names, e.g.
kwinnvidiahack becomes kwinactivenvidiahack.
The remaining usage of kwinrc in core and libs is replaced
by a cmakedefine for the configuration name and all data
installations are moved to the defined name. Dynamic loading
for scripts & co is adjusted for loading based on defined name.
This change allows the side-by-side installation of both kwin
for desktop and kwin for Plasma Desktop without the known
issues like conflicts in config files or missing build options
if kwin desktop is used for Plasma Active.
Likewise the KCMs are not adjusted as they are not intended to
be used for kwinactive.
REVIEW: 104299
BUG: 296084
FIXED-IN: 4.9.0
CCMAIL: active@kde.org
Martin Gräßlin