> On Jan. 9, 2015, 12:25 p.m., Àlex Fiestas wrote: > > Where do you plan to use this? We are not maintaining compatibility (so > > far) in our dbus apis, so why add this at all? > > Daniel Vrátil wrote: > KScreen. For now we are listening to UPower, which IMO sucks and we > should use PowerDevil instead (as it provides abstraction for alternative > backends). This makes it just easier to monitor changes. > > Àlex Fiestas wrote: > This has a few problems: > 1-It couples KScreen more to PLasma by adding a runtime dependency (which > is ok if you decide to do so) > 2-It might create a deadlock since kscreen is a kded module asking to > another module (powerdevil) for things. > 3-We will depend on Powerdevil if we use kscreen in other places (sddm > desperatly needs something like kscreen, or kscreen itself) > 4-Adds a dependency to an api that is not stable (so far it has not > been), we have changed it many times and I am sure we will change it again > > My recommendation for this will be to merge the new power management api > in Solid and make kscreen depend on it. Solid already offers backends (so we > don't have to implement this in kscreen) and we don't have to expose > powerdevil internals. > > So this is a -1 from my side (if the only reason to add this is KSCreen). > > Martin Klapetek wrote: > > 2-It might create a deadlock since kscreen is a kded module asking to > another module (powerdevil) for things. > > As dfaure explained in https://git.reviewboard.kde.org/r/121885/ QtDbus > is really smart and in-process DBus calls are made into direct methods and > should never deadlock. Which is cool :) > > Àlex Fiestas wrote: > Oh wow! One less problem to worry about then. > We still need to be careful not to deadlock by doing sync calls between > the same kded modules (happened already).
That should still work no? I mean if sync QtDBus works in-process, then it doesn't matter which modules it calls to as it is all the same process...right? - Martin ----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: https://git.reviewboard.kde.org/r/121915/#review73569 ----------------------------------------------------------- On Jan. 8, 2015, 1:04 p.m., Daniel Vrátil wrote: > > ----------------------------------------------------------- > This is an automatically generated e-mail. To reply, visit: > https://git.reviewboard.kde.org/r/121915/ > ----------------------------------------------------------- > > (Updated Jan. 8, 2015, 1:04 p.m.) > > > Review request for Plasma and Solid. > > > Repository: powerdevil > > > Description > ------- > > There's no way to detect when lid has been closed other than listening to > `changed` signal org.freedesktop.UPower and then polling the Powerdevil for > new values. This patch adds a new signal to the PowerDevil interface to > notify about the change and provide new value right away. Makes it much > easier to use. > > > Diffs > ----- > > daemon/org.kde.Solid.PowerManagement.xml 53f77e5 > daemon/powerdevilbackendinterface.h 702b66b > daemon/powerdevilbackendinterface.cpp 6dd8c71 > daemon/powerdevilcore.h e255703 > daemon/powerdevilcore.cpp d51ab19 > > Diff: https://git.reviewboard.kde.org/r/121915/diff/ > > > Testing > ------- > > Tested with qdbus-monitor, signal is emitted when laptop lid is closed/opened. > > > Thanks, > > Daniel Vrátil > >
_______________________________________________ Plasma-devel mailing list Plasma-devel@kde.org https://mail.kde.org/mailman/listinfo/plasma-devel
