Better patch to replace !218.
- Auto and quick detection of previous D-Bus instance;
- Remove private D-Bus compile definition, only use it on macOS without an existing D-Bus instance;
- Safe reboot after crashes because the indicator is not relating on the kdeconnectd to run a D-Bus session;
- Safe exit after clicking on `Quit` in the systray.
More details in commit logs:
Only enable private D-Bus on macOS because the other platforms do not
need them.
The app should be able to easily detect the session bus from the env
DBUS_LAUNCHD_SESSION_BUS_SOCKET from launchd through launchctl.
Because https://gitlab.freedesktop.org/dbus/dbus/-/blob/master/dbus/dbus-sysdeps-unix.c#L4392
shows that it is the only probing method on macOS with launchd.
The D-Bus session bus can be easily found from launchd/launchctl
with DBUS_LAUNCHD_SESSION_BUS_SOCKET env. It can be an external one
(installed from HomeBrew) or an internal one (launched by a previous
instance followed by a crash).
The indicator helper on macOS can now automatically detect whether we can use a potentially
(with launchd/launchctl env set, or KDE Connect macOS
private_bus_address set) existed and usable session bus.
If previous bus is usable, just try to launch the kdeconnectd with us.
Otherwise, launch a private D-Bus daemon, export the launchd/launchctl
env, and run a kdeconnectd instance.
Everything works better and quicker now :)
Ensures the reply window is raised using Qt::WindowActive state. This
properly raises the window reliably (including from plasmoid, which
didn't work at all before) and focuses the reply textbox.
On Plasma qqc2-desktop-style is used automatically, but not on other
DEs. Set it as a default and allow overriding it via environment
variables.
Also add it as a runtime dep so distros actually ship it.
Ever since joining the release service we never increased our version number on release.
I've seen distros automatically apply the release service version scheme (20.0X) to the package,
but when running e.g. kdeconnect-cli -v we'd still get 1.4.
This way the version is automatically adjusted by the release tooling so we don't need to care about and cannot forget it any longer.
We use kdeconnect-version.h in several places and therefore it needs to be in the include path. We currently do this by setting target_include_path in a few places. Replace this with an interface library that we can link against that sets up the correct include path. IMO it is cleaner this way.
This automatizes the generation of logging categories so a
kdeconnect-kde.categories is generated and installed to
/usr/share/qlogging-categories5/ so kdebugsettings can use it.
Also, sets the default logging level to Warning. So now the logs
of users won't be filled with debug messages but they can
modify the configuration easily with kdebugsettings.
## Summary
The core idea is as follows:
1. When a Link loads the BatteryPlugin, we query Solid for a list of batteries.
1. If the list is empty, we print a warning message and return quickly
2. Otherwise, we connect *two signals* to every object in that list
2. We send out a single new NetworkPacket as soon as we've processed that list
3. When either of those two signals emits, we send another new NetworkPacket
### Multi-battery Support
BUG: 357193
To handle devices with multiple batteries (requested in that bug), we average
together the battery percentages. This also includes a new field in the packet for
'number of batteries' called `batteryQuantity`. For backwards compatibility, we can
assume it has a default value of one.
This should ensure we support
- devices with no batteries at all (like many desktop machines)
- devices with hot-pluggable batteries (like those laptops with detachable screens)
### Concerns
Note that the implementation isn't perfect.
We'll need some new localizable text to make it clear that we now support sending
battery status information.
Then there's a rather significant question: maybe we should have two battery plugins
on each client, like we do for the `findmyphone`/`findthisdevice` plugins?
## Test Plan
We need to ensure that other clients (including those using the Android codebase)
will respond correctly. The main things to look at are
1. are these new packets sent when the plugin is enabled, and not sent when it's disabled?
2. is the charge percentage accurate?
3. is the charge state (charging, discharging, or full) accurate?
and
4. do we see the correct number of warnings for low-battery?
