Sorry for my verification the REVIEW-126486
the patch git rm directory.sh and ExtraDesktop.sh
I forgot to revert those files when stash that patch, I am so sorry for that.
REVIEW: 126298
As describled in 'Self-hiding plasmoid' requirement
https://forum.kde.org/viewtopic.php?f=83&t=39501
I added Plasma::Types::HiddenStatus, then when the plasmoid's status is
SystemTray.Task.HiddenStatus, the plasmoid is just self-hiding.
REVIEW: 126298
QTimer::singleShot(time, QObject*, char*)
QTimer::singleShot(time, QObject*, PointerToFunction)
behave differently when time == 0
So the porting had an effect
This patch migrates to QMetaObject::invokeMethod which is clearer
CCBUG: 352055
There seems to be two different bug in 352055 but looks similar, the other one is fixed in Qt 5.6 according to Albert Astals Cid.
Last patch has a reggression due to QtQuick issue (only on Xcb):
QEvent::Leave is triggered after QEvent::MouseButtonPress Qt::LeftButton
BUG: 354651
REVIEW: 126331
If grabbed icons are blank, try to salvage the copied data as well as
possible while leaving setups where image grabbing works fine alone.
Based on a patch by Rakyn Barker.
BUG:355684
REVIEW: 126336
Somehow key events no longer end up in the expanded task, this could be related
to the fact that we now have a proper ScrollView instead of a plain ListView/Flickable.
REVIEW: 126293
Graphics seem more likely than anything else to be broken, if the user
explicitly selects failsafe mode from the DM we should follow the same
rationale as kwin and not trust the user's drivers.
REVIEW: 126205
Shutdown scripts are done by startkde after ksmserver quits. Which never
happens because we've told systemd to shutdown.
Old systems worked because they used to communicate with the display
manager which then closed us before shutting down, this is no longer the
case.
This moves handling the shutdown scripts into ksmserver (which already
handles startup scripts) and runs them before asking logind to shutdown.
BUG: 356190
REVIEW: 126268
It can happen that the onAppletLocationChanged() slot gets called before
init(), making init() reset the calculated position from the slot.
CCBUG: 356461
The globalConfig() in ctor returns a non-(yet-)existing KConfigGroup
which makes it always return the default value. So the config access
must be delayed for init().
Custom notification position is now correctly remembered.
BUG: 356419
FIXED-IN: 5.5.1
This slot gets triggered also on popup hiding, which is unwanted.
Should also prevent a crash when unloading notifications applet while
there is an active popup on screen.
BUG: 356402
FIXED-IN: 5.5.1
The notifications popup positioning recently regressed
by some other changes (looks like Qt) and the popups
would fly across the screen.
The proper solution is using KWin effect but given how
close the release is, this needs to be dealt with in a
different way.
The main problem is calculating the initial popup size
because as long as the Dialog is invisible, it has an
incorrect geometry, so it needs to be positioned right
after it's being displayed. The Dialog however gets the
sizes even later, so the code now calls a slot from Dialog
that ensures the sizes are correct before the initial
placement on screen. It's not ideal but I'm out of ideas
otherwise. Plus it should be only temporary until the
KWin effect will replace it.
Besides that, it moves the custom position handling from the
QML into the C++ applet class, making it easier.
REVIEW: 126221
BUG: 355069
FIXED-IN: 5.5.0
Turns out that this method is sometimes called at the time where the
virtualGeometry() is invalid -- which sounds pretty much like a Qt bug,
but it isn't exactly the first bug in screen handling in Qt5. Rather
than producing values which are surely invalid, try to cover up the
mess.
REVIEW: 126216
David Edmundson
Michael Pyne
l10n daemon script
Leslie Zhai
Kai Uwe Broulik
Yuri Chornoivan
Xuetian Weng
Anthony Fieroni
Jonathan Riddell
Martin Klapetek
Benedikt Gollatz
David Edmundson
Eike Hein
Imran Tatriev
Martin Gräßlin
Jan Kundrát