KDM Plasma: Good News Everyone!

After a long "absence", lack of blog entries, and too much code not being written...I am finally able to resume (to some extent), those things...or...the opposite of those things, rather.

Although I am still searching for the best migraine medications, my current combinations are working decently enough to allow me more hours instead of having to sleep for the whole day. The good news is that it there isn't anything "severely" wrong with me. Unfortunately, to search for the right medicine, there's a period where it regresses, until I discover that the medicine either works or doesn't; and it takes many weeks to know that The good news is that they're less severe generally, I'd say, as well as being less frequent. I still have not had a day without them for 9+ months or so. So, don't get the impression that I'm purposely avoiding KDE ;)

Okay, now exciting part: some updates on KDM-Plasma. First of all, not too long ago I got it to the point where it was possible to login. That was pretty sweet.
Well, now I have some infrastructure changes I am going to be working on. Most importantly, I've decided to now make KDM-Plasma support QML. By "support", I mean that it will
be using it, exclusively. (well, I suppose you could use C++ for it instead...if you're nuts).

What does it mean when I say KDM-Plasma will be using QML? Well, KDM will take one Plasmoid and stretch it to the size of the scene. This applet will be written in QML/Javascript,
and it will be able to access the DataEngines. There will be one main DataEngine, which will have the general data sources like the sessions available, the userlist, etc. etc. Then there will additionally, be n numbers of authentication plugins. The idea is to separate it from the previous concept whereby the Plasmoid effectively *is* the authentication plugin as well as the creator of the UI. This of course had drawbacks in terms ease of customization (how many times do you want to change the authentication plugin through code, as opposed to just changing the look).

So, just to sum it up: DataEngines/Services (in addition to the Main one) are now serving as authentication plugins. The QML applets can pick which one it wants to use, and it should be relatively easy and require not much intervention in terms of getting it to work. It will have to do -something-, to an extent, in logic, but 99% of this will be in the DataEngine, and the 1% that needs to talk to the DataEngine authentication will be the applet. The DataEngines are essentially a "write once and probably never worry about it again", much like the current KDM plugins are (classic, winbind, generic, (unofficially fingerprint)). Hope that makes sense. I haven't thought about how to break apart the API nicely, but I'm working on it now...

QML should make it exceptionally easy for someone to write a Plasmoid, ) which takes up the entire screen, remember) - and therefore it will be "in charge" of anything it wants to. (e.g. you can place a billion buttons all over the place and stuff like that). Right now, I have a PathView incorporated into it for the userlist, which you can flick and click. That's just for the applet I'm working on. You can do whatever you want for the QML applet you want to do. So, this allows *absolutely simple* branding, UI redesign, as well as fun animations, to make it feel more alive and less stale/static.

Well, if any of you know QML, or want to learn it...feel free to impress me with neat themes and their animations, etc. (I suppose that's the best word for them)! You can actually create one right now just through qmlviewer...and I would just port it over later. In fact, I'd love it if you did...it'd give me some more ideas as well as give me more to include with KDM). If you care to follow through with that, feel free to contact me if you need help or what not.

Bad News Everyone!

It will not be making it into KDE 4.7, it everything ends up freezing around mid-May, so there's no way I can get it in by then. And I'm too much of a perfectionist to rush it, especially with something as important as this; don't want any security implications and what not, and I don't want to give a bad impression in terms of Plasma being used in KDM. Polish, polish, polish.

By the way, I have moved this code to Git now, which I am really liking, especially because of the offline way it treats things (admittedly I have only used small features), so feel free to clone/watch/poke at it. Link: Quick Git (this is my branch of kde-workspace, and I affect the ./kdm/plasmafrontend/ directory).