> On Sept. 24, 2014, 5:10 p.m., Martin Gräßlin wrote:
> > Please watch coding style and please also have a look at the frameworks 
> > variant. It still needs porting to MacOS *hint,hint* and that would be 
> > very, very appreciated. I recently sent a mail to frameworks-devel 
> > concerning moving the daemon into the framework. Would be great if you 
> > could have a look at it and help that we have MacOS support soon in the 
> > framework.
> 
> René J.V. Bertin wrote:
>     "It" being kglobalaccel? Or the change to set "agentship" from the 
> running app, preferably by adding an API call somewhere so that end 
> applications don't have to dabble in CoreFoundation code?
>     The former I can't help you with because as you know I cannot run 
> anything KF5 in the foreseeable future. The latter is something I could work 
> on, but again without testing.
>     
>     Oh, wait, you're asking me to port to MacOS. Hmmm, I doubt Qt ever 
> existed even for version 9 of that venerable old pile of cultural luggage O:-)
>     (MacOS is what was used before Mac OS X came out. You're good at spotting 
> whitespace differences, so I don't have to explain any further :P )
>     
>     As to kglobalaccel: I've taken this app as a demonstration case because 
> it's one of a handful daemons that cluttered my Dock (the others being 
> nepomukfilewatch and nepomukstorage, not sure patches to those are of any 
> relevance?).
>     But I have no actual idea to what kglobalaccel does and to what extent it 
> does that on OS X too. I haven't noticed any loss in (shortcut) functionality 
> after quitting it... is there something I could try to see if it is required? 
>     
>     Also, the KDE4 version is the same on OS X and on Linux, why would the 
> KF5 version require porting ?
> 
> Martin Gräßlin wrote:
>     > Also, the KDE4 version is the same on OS X and on Linux, why would the 
> KF5 version require porting ?
>     
>     well the X11 version needed porting due to the switch to QPA. So I assume 
> that other platforms need porting, too.

QPA? So we all have to have green hair now?... :-) See 
http://qt-project.org/videos/watch/qpa-the-qt-platform-abstraction or is there 
some other "QPA"?... :-)

@Martin: On a serious note, can you give us the exact link on 
kde-frameworks-devel archives that you are talking about? I think only Marko is 
a member of that list. Could your post have any connection with upcoming 
discussions I am about to start on kde-core-devel, re kdeinit4, klauncher, 
kded4, kwrapper and friends?

Our priority is to get KDE background processes working properly (or at all) in 
KDE 4 on Apple OS X, because KF5 is a distant mirage for the KDE-Mac group and 
the MacPorts users (except for Marko). However, if KDE-Mac work has spinoffs 
for the Frameworks/KF5 effort, well and good. We are already starting to see 
that happen. If we could get more dialog going with KDE core developers, all of 
us could proceed a lot faster and we could all benefit.


- Ian


-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://git.reviewboard.kde.org/r/120354/#review67376
-----------------------------------------------------------


On Sept. 24, 2014, 5:23 p.m., René J.V. Bertin wrote:
> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> https://git.reviewboard.kde.org/r/120354/
> -----------------------------------------------------------
> 
> (Updated Sept. 24, 2014, 5:23 p.m.)
> 
> 
> Review request for KDE Software on Mac OS X, KDE Runtime and kdelibs.
> 
> 
> Repository: kde-runtime
> 
> 
> Description
> -------
> 
> See https://bugs.kde.org/show_bug.cgi?id=339333 for more detailed discussion.
> 
> KDE helper applications that need to be able to present widgets or otherwise 
> "talk with the GUI layer" require special attention on OS X, if one doesn't 
> want them to appear in the Dock or task switcher nor present a bare-bones 
> menubar when made active.
> 
> Applications that live in an app bundle can set LSUIElement="1" in their 
> Info.plist to signal the window server that they're "agents" (and thus don't 
> want the aforementioned visual presence). This feature is already in use (see 
> Info.plis.template for apps like kded4 and kdeinit4, and the corresponding 
> code in their respective CMake files).
> 
> kglobalaccel is a different case as it's built as a standard *n*x app 
> (`/opt/local/bin/kglobalaccel` in a standard MacPorts install) and thus has 
> no Info.plist. It is however possible to set the corresponding bit via 
> CoreFoundation, and that's what this patch does.
> 
> Suggestion: a member function I'd tentatively call `appIsService` would be 
> welcome, but one could also make this the default behaviour when starting a 
> `GUIenabled=false` application on OS X.
> That's actually the main reason for submitting this RR: see if we can come to 
> a consensus if and how to use this new knowledge.
> 
> 
> Diffs
> -----
> 
>   kglobalaccel/main.cpp 4d230b8 
> 
> Diff: https://git.reviewboard.kde.org/r/120354/diff/
> 
> 
> Testing
> -------
> 
> On OS X 10.6.8 with kdelibs 4.14.1 (git/kde4.14).
> 
> 
> Thanks,
> 
> René J.V. Bertin
> 
>

Reply via email to