Summary:
the Lookand Feel kcm already applies decorations to kwin, allow to
read them as defaults for the case of distribution customization
where a different lnf with a different deco is wanted
Test Plan: tested to start kwin with a different lnf which loads an aurorae decoration
Reviewers: #plasma, #kwin, davidedmundson
Reviewed By: #plasma, #kwin, davidedmundson
Subscribers: davidedmundson, kwin
Tags: #kwin
Differential Revision: https://phabricator.kde.org/D18261
Summary:
kcm.osdTextOnly is the inverse of "show desktop layout indicators"
The save was inversed, the load was not.
Also port from onCheckedChanged to onToggled as that's emitted only on
user changes and not on binding re-evaluation.
BUG: 403150
Test Plan:
Toggled it
Reopened KCM
Checked kwinrc
Reviewers: #kwin, ngraham
Reviewed By: ngraham
Subscribers: ngraham, kwin
Tags: #kwin
Differential Revision: https://phabricator.kde.org/D18272
Summary:
We need to return the absolute path instead of base name so decoColorToCombo
can return the correct index.
Test Plan:
* Go to System Settings > Window Management > Window Rules;
* Create a new rule for Konsole to force the Oxygen color scheme,
close the dialog, and click the Apply button;
* Open the rules dialog (you don't have to modify anything), click OK button, then Apply button.
Without this patch, Breeze color scheme will be forced after the
last step.
Reviewers: #kwin, davidedmundson
Reviewed By: #kwin, davidedmundson
Subscribers: graesslin, broulik, kwin
Tags: #kwin
Differential Revision: https://phabricator.kde.org/D17655
Summary:
So far we didn't try to do high DPI on kwin internal windows, such as
the user context menu and tab bars and whatever.
Due to wayland scaling they were the correct phyiscal size but upscaled.
This patch fixes our QPA to enable Qt's high-dpi support.
BUG: 402853
Note icons are still low res. This is because the global
QGuiApplication::devicePixelRatio which is the max of all connected
screens is static for the duration of the app. QIcon uses this when
determining the DPR to use. This will require a Qt change.
Test Plan:
Ran at 2x on my normal DPI screen (as that's easier to see anything)
* User action menu is high DPI
* Window deco tooltips are still fine
* Tab switcher is high DPI
* Overlay in present windows Desktop grid are still ok
Reviewers: #kwin
Subscribers: zzag, kwin
Tags: #kwin
Differential Revision: https://phabricator.kde.org/D18042
Summary:
We initialize all fields of Phase2Data at once, so the constructors seem
to be redundant.
Reviewers: #kwin, graesslin
Reviewed By: #kwin, graesslin
Subscribers: kwin
Tags: #kwin
Differential Revision: https://phabricator.kde.org/D18207
Summary:
We know that the maximum size is the number of windows in stacking
order. It might be that some are skipped as they are not ready for
painting, but overall it's better to reserve a little bit too much
memory than to have a growing QVector.
Test Plan: Compiles
Reviewers: #kwin
Subscribers: kwin
Tags: #kwin
Differential Revision: https://phabricator.kde.org/D18208
Summary:
Instead we depend on the surfaceless platform for which we recently
added support. Thus the plugin does not need to use gbm and udev
anymore. So simplifies a lot.
Test Plan: ctest (prior to breaking change) passes
Reviewers: #kwin
Subscribers: kwin
Tags: #kwin
Differential Revision: https://phabricator.kde.org/D18160
Summary:
The SceneOpenGLES test was starting to randomly crash on my system. On
investigation I observed that there was a graphics reset ("Attempting to
reset compositing.") which triggered a delayed restart of the
compositor. This even was only processed while waiting for XWayland to
terminate. Which resulted in a crash in KWin::getXServerVersion as the
xcb connection broke.
It makes no sense to setup the compositor again during application
shutdown. Thus the dtors set a flag that they are terminating the
application and Compositor::setup is checking for it.
Test Plan: Test no longer crashes, although it goes through the crashing path
Reviewers: #kwin
Subscribers: kwin
Tags: #kwin
Differential Revision: https://phabricator.kde.org/D18015
Summary:
With EGL_MESA_platform_surfaceless we don't need a dri device anymore.
So we don't need to skip the tests if the device is missing. Instead the
tests verify that OpenGL compositing is used if requested.
Test Plan: ctest passes
Reviewers: #kwin
Subscribers: kwin
Tags: #kwin
Differential Revision: https://phabricator.kde.org/D18014
Summary:
TabGroup is nowadays in AbstractClient so the cast to Client is
incorrect.
Reviewers: #kwin
Subscribers: kwin
Tags: #kwin
Differential Revision: https://phabricator.kde.org/D17898
Summary:
The default implementations just return false/nullptr. The advantage of
having this in AbstractClient is that we can reduce the needed casts
from AbstractClient to Client in core as can be seen in this change.
There are more cases which can be improved thanks to this refactoring
which will follow in dedicated commits.
Test Plan: ctest passes
Reviewers: #kwin
Subscribers: kwin
Tags: #kwin
Differential Revision: https://phabricator.kde.org/D17890
Summary:
Given that QList<T> will allocate items on the heap when sizeof(T) is
bigger than sizeof(void *), which is the case for Phase2Data, use
QVector instead.
Reviewers: #kwin, davidedmundson
Reviewed By: #kwin, davidedmundson
Subscribers: kwin
Tags: #kwin
Differential Revision: https://phabricator.kde.org/D18198
Summary:
Phase2Data already includes the scene window, so we don't have to store
it again in QPair.
Reviewers: #kwin, davidedmundson
Reviewed By: #kwin, davidedmundson
Subscribers: kwin
Tags: #kwin
Differential Revision: https://phabricator.kde.org/D18188
Summary: This matches the behaviour of workspace.cpp on initial load.
Test Plan: Test now passes
Reviewers: #kwin, zzag
Reviewed By: #kwin, zzag
Subscribers: kwin
Tags: #kwin
Differential Revision: https://phabricator.kde.org/D18183
Summary:
Currently, our implementation of zwp_idle_inhibitor_v1 is not fully
spec-compliant. If the associated surface is not visible, we still
honor the idle inhibitor object.
This change addresses those spec-compliance issues. If a surface is
minimized or it's not on the current virtual desktop, then the
associated inhibitor object won't be honored by KWin.
Reviewers: #kwin, davidedmundson
Reviewed By: #kwin, davidedmundson
Subscribers: graesslin, kwin
Tags: #kwin
Differential Revision: https://phabricator.kde.org/D17343
Summary:
virtual desktops can be destroyed, active clients update, but deleted
keeps a cache.
Someone needs to do cleanup to avoid dangly pointers.
Test Plan:
I couldn't find a case of someone calling desktops mid way through an
animation, so it's only a hypothetical bug.
Reviewers: #kwin, zzag
Reviewed By: #kwin, zzag
Subscribers: graesslin, zzag, kwin
Tags: #kwin
Differential Revision: https://phabricator.kde.org/D17247
Summary:
If a user saves a decoration that has a plugin but no theme we currently
delete the entry saving just the plugin/library.
This is an issue as on load we call:
const QString plugin = config.readEntry("library", s_defaultPlugin);
const QString theme = config.readEntry("theme", s_defaultTheme);
That would give us a non-default library with the default theme name,
which would be invalid.
If compiled with Breeze s_defaultTheme is blank so this ends up being
effectively the same thing. In the normal case it will neither break
nor fix anything.
Test Plan:
None.
Just happened to see it when doing a code review.
Reviewers: #kwin, vpilo
Reviewed By: vpilo
Subscribers: kwin
Tags: #kwin
Differential Revision: https://phabricator.kde.org/D18146
Summary:
KWin::Script::loadScriptFromFile ran in it's own thread and accessed
member variables of KWin::Script without any guards.
Potentially script could be destroyed whilst the file is loading.
Rather than adding mutexes everywhere, this patch scopes the QFile
object to be local to the threaded function making it independent.
BUG: 403038
Test Plan: Ran a script from a file
Reviewers: #kwin, graesslin
Reviewed By: #kwin, graesslin
Subscribers: kwin
Tags: #kwin
Differential Revision: https://phabricator.kde.org/D18126
Summary:
We no longer need to include xcb/xfixes.h in cursor.cpp because
X11Cursor moved to its own file.
Reviewers: #kwin, davidedmundson
Reviewed By: #kwin, davidedmundson
Subscribers: kwin
Tags: #kwin
Differential Revision: https://phabricator.kde.org/D18133
Summary:
* save changes to the config files when the layout is saved
* :name() works even if netrootinfo isn't there
* as soon a rootinfo is set, connect all the desktops with name changes
Test Plan:
* tested with the kcm to add, remove and rename desktops, all of that works
* setting the number of rows still only partly works: kwin notices it but
the pager doesn't notice, a plasma restart is needed
Reviewers: #plasma, #kwin, hein, davidedmundson, graesslin
Reviewed By: #plasma, #kwin, hein, davidedmundson, graesslin
Subscribers: zzag, graesslin, davidedmundson, ngraham, kwin
Tags: #kwin
Maniphest Tasks: T4457
Differential Revision: https://phabricator.kde.org/D17265
Summary: This change reformats the documentation so it's more easier to read it.
Test Plan:
Generated Doxygen documentation, started Python's Simple HTTP server, navigated
to AnimationEffect's documention, haven't noticed any serious issues.
Reviewers: #kwin, davidedmundson
Reviewed By: #kwin, davidedmundson
Subscribers: graesslin, davidedmundson, kwin
Tags: #kwin
Differential Revision: https://phabricator.kde.org/D17832
Summary:
Even though the buffer is passed through a side channel some properties
we might use are double-buffered on the surface commit.
Test Plan: Used by unit test InternalWindowTest::testScale in linked commit
Reviewers: #kwin, graesslin
Reviewed By: #kwin, graesslin
Subscribers: kwin
Tags: #kwin
Differential Revision: https://phabricator.kde.org/D18085
Summary: For unit test purposes
Test Plan: Used by unit test InternalWindowTest::testScale in linked commit
Reviewers: #kwin, graesslin
Reviewed By: #kwin, graesslin
Subscribers: kwin
Tags: #kwin
Differential Revision: https://phabricator.kde.org/D18086
Summary:
BUG: 346222
BUG: 399992
BUG: 356076
FIXED-IN: 5.15
The QML changes make the tab look and feel less awful:
* make the fake window borders stand out less
* add margins to drop area to make it easier to drop buttons on the title bar (fixing 346222)
* change cursor to a non-pointing one [1]
[1] Drag&drop: When dropping, we receive the position of the button being dropped,
not of the cursor. The two can be far (when starting drag by moving the cursor fast):
which makes the whole experience very confusing, as the user means to drop at cursor location.
Test Plan: Tested drag/drop: same (messy) behavior as before, but can now drop more loosely around titlebar items
Reviewers: #vdg, #kwin, ngraham
Reviewed By: #vdg, ngraham
Subscribers: davidedmundson, ngraham, kwin
Tags: #kwin
Differential Revision: https://phabricator.kde.org/D17982
Summary:
The "new" animation option no longer uses hard coded effects, which
means one could install a third party virtual desktop switching animation,
for example from store.kde.org, and it will be displayed in the KCM.
Test Plan: {F6503565}
Reviewers: #kwin, #vdg, ngraham, davidedmundson
Reviewed By: #kwin, #vdg, ngraham, davidedmundson
Subscribers: davidedmundson, hein, ngraham, kwin
Tags: #kwin
Differential Revision: https://phabricator.kde.org/D17766
Summary:
The main motivation for moving it into a shared library is to share the
model between Desktop Effects and Virtual Desktops KCM.
The extracted model is quite the same as the one in Desktop Effects KCM,
except some minor changes, e.g. rename loadEffect and syncConfig to more
convenient names, add comments, some whitespace changes, fix coding
style in some parts, etc.
Test Plan: effectModelTest passes, Desktop Effects KCM works.
Reviewers: #kwin, davidedmundson
Reviewed By: #kwin, davidedmundson
Subscribers: hein, kwin
Tags: #kwin
Differential Revision: https://phabricator.kde.org/D17765
Summary:
The new test is similar to testToplevelOpenCloseAnimation. The
main purpose of this test is to verify that the Fading Popups effect
animates all sorts of popups, e.g. popup menus, tooltips, etc.
testAnimateUserActionsPopup is skipped for now because stuff
that animates unmapped clients is broken.
Reviewers: #kwin, davidedmundson
Reviewed By: #kwin, davidedmundson
Subscribers: kwin
Tags: #kwin
Differential Revision: https://phabricator.kde.org/D17992
Summary:
7834bec52a missed to port minimize and
unminimize to dynamic dispatch. In addition to that, we don't benefit
from QMetaObject::invokeMethod so port addRepaint and addLayerRepaint
to dynamic dispatch as well.
Reviewers: #kwin, davidedmundson
Reviewed By: #kwin, davidedmundson
Subscribers: davidedmundson, kwin
Tags: #kwin
Differential Revision: https://phabricator.kde.org/D18049
Summary: Be nice and destroy all created windows in testDontAnimatePopups.
Test Plan: The test still passes.
Reviewers: #kwin, davidedmundson
Reviewed By: #kwin, davidedmundson
Subscribers: kwin
Tags: #kwin
Differential Revision: https://phabricator.kde.org/D18037
Summary:
This patch ports the Effects KCM to QQC2, which yields the following benefits:
1. General performance enhancements of only using QQC2 rather than a mix of 1 and 2
2. Some code simplification is possible
3. Improves the appearance of the checkable menu items in the dropdown menu, fixing https://bugs.kde.org/show_bug.cgi?id=402701
4. Improves the appearance when using a fractional scale factor, fixing https://bugs.kde.org/show_bug.cgi?id=396725
5. Fixes the incorrect size when opened from `kcmshell5`
Along the way, two visual changes are introduced as a by-product of porting:
1. The scrollbar is inline, so it overlaps some of the buttons in the content. This is somewhat undesirable, but adopting a Kirigami scrollview would fix this.
2. The button that displays a dropdown menu no longer has a downward-pointing arrow to indicate as such. This is not my preference, but there's a benefit to being consistent, and eventually we could can change this in one place to impeove the appearance of buttons that display dropdown menus everywhere in one fell swoop.
BUG: 396725
BUG: 402701
BUG: 396076
FIXED-IN: 5.15.0
Test Plan:
All functionality still works
In System Settings, showing improved checkable menu item appearance:
{F6523385}
In `kcmshell` at 1x scale:
{F6523386}
In `kcmshell` at 1.5x scale:
{F6523387}
Reviewers: #kwin, #plasma, davidedmundson
Reviewed By: #kwin, #plasma, davidedmundson
Subscribers: davidedmundson, kwin
Tags: #kwin
Differential Revision: https://phabricator.kde.org/D17928
Summary:
Just accessing Qt::Popup WindowFlag.
BUG: 402852
Test Plan: Not tested for the described problematic case
Reviewers: #kwin
Subscribers: kwin
Tags: #kwin
Differential Revision: https://phabricator.kde.org/D17981
Summary:
Thus we don't require vgem any more which fails with latest mesa.
Kepping the vgem variant as fallback for older systems. See T10245
Test Plan:
SceneOpenGL Test passes without init vgem, not tested with
latest mesa
Reviewers: #kwin
Subscribers: kwin
Tags: #kwin
Differential Revision: https://phabricator.kde.org/D17980
Summary:
This ensures that KWin gets the same keyboard layout config as is
configured in the session and not a default config.
BUG: 402764
FIXED-IN: 5.14.5
Test Plan:
Xephyr to verify the config is applied, otherwise completely
untested. I don't have an X session.
Reviewers: #kwin
Subscribers: kwin
Tags: #kwin
Differential Revision: https://phabricator.kde.org/D17967
This reverts commit 16e904592a.
The commit has logical errors, corrupting DND on Wayland. Also the solution
with static_cast to solve the compiler warnings is not a sensible one, it just
hides the root problem.
And next time such a change has to go through review!
Summary:
Not only does the function windowsShareDesktop become a lot more complex
with the new desktop cardinality comparing if they share any desktop
isn't actually what we want.
If I have 2 windows on desktop 1 and the lower stacked window is also on
desktop 2, when I switch desktops the focus will change, but we don't
want to animate anything here as on this dekstop it has nothing to slide
in front of.
Instead this patch simply checks both windows are on the current
desktop.
Test Plan:
Loaded a few windows. Some on one, some on more.
Before if a window was on desktop 1&2 it wouldn't animate on 2.
Windows now seem to animate when they should and not when they shouldn't.
Reviewers: #kwin, zzag
Reviewed By: #kwin, zzag
Subscribers: kwin
Tags: #kwin
Differential Revision: https://phabricator.kde.org/D17943
Summary:
If a window is not on the current virtual desktop, then it won't be
painted/slided. Thus, we can replace EffectWindow::desktop by
EffectsHandler::currentDesktop.
Test Plan: The effect still works correctly on both X11 and Wayland.
Reviewers: #kwin, davidedmundson
Reviewed By: #kwin, davidedmundson
Subscribers: kwin
Tags: #kwin
Differential Revision: https://phabricator.kde.org/D17944
Summary:
Fixes the preselected decoration style on module load.
Also:
* Prevent the module state to be set to modified on resize.
* Fix QML errors in logs.
Reviewers: davidedmundson
Reviewed By: davidedmundson
Subscribers: davidedmundson, kwin
Tags: #kwin
Differential Revision: https://phabricator.kde.org/D17775
Summary:
Does something similar to our existing ServerDecoration, but based
around XDG Shell patterns and with a few subtle differneces.
We'll probably still need both in kwin for the forseeable future as GTK3
won't ever change from using the KDE Server Decoration.
Test Plan:
Relevant unit test. It's a bit simpler as spec states
toolkits must follow what the compositor configures if they
bind the interface.
Modified plasma-integration to remove ServerIntegration
(as Qt5.12 has native support) all my windows look and act the same.
Reviewers: #kwin, zzag
Reviewed By: #kwin, zzag
Subscribers: kwin
Tags: #kwin
Differential Revision: https://phabricator.kde.org/D17756
