Getting into plasma

Marco Martin notmart at gmail.com
Mon Mar 15 13:00:34 CET 2010


On Monday 15 March 2010, Christophe Olinger wrote:
> Hey plasma people,
> 
> My name is Christophe Olinger from Luxembourg (nick binarylooks) and I
> am following KDE and especially plasma for a few years now. I learned
> C++ a looong time ago (10 years?) and basically I fear it’s all gone.
> Nevertheless, I am trying to get into it again especially using Qt.
> If somebody is interested in these kind of things: I am 29.9 years old
> and have a PhD in molecular biology and worked in the bioinformatics
> field. Soon to be father and I hope my immense motivation in coding
> will continue.
> 
> My main interest lies in getting plasma bling (am I allowed to say
> bling?). Onto netbooks and later tablets. I really like the current

you -have- to say bling :p

> plasma-netbook workspace and right now I am thinking about creating a
> sort of really simple picture viewer. Since I can basically be

great :)

> considered a C++ newbie I am a bit afraid of jumping into an existing
> project yet. I need to learn. I am currently dissecting the frame
> plasmoid to learn about the painter and QPixmaps. (I know about plasma
> media center and will have a look at it. If I learn enough with my pet
> project, I will surely start contributing to it)

not too difficult to dive in its code i think.

> What the viewer should do:
> v0.1: Display an image, allow swiping to switch image. Allow a single
> image mode and a browse mode. The browse mode will be a grid of images
> and you can pan through the grid. Tapping an image will switch to
> single picture mode. All this should be fullscreen and feel natural

I think if you want to go immediately in this direction it will be a bit hard, 
because doing a grid of images properly is reeeally hard (basically, 
model/view over and over again)
so, you should really start to look in the mediacenter code and try to improve 
it from here, without trying to reinvent the wheel over again ;)

> 
> v0.2 use nepomuk and other stuff to filter your collection, tag images
> and rotate them. No additional manipulation
> 
> I would appreciate some intial help to get started so here my questions
> *Should I learn Qt quick for this and do all this in declarative?

this is still an open question for all of us, we are wondering if it's the 
case to start converting the pieces of the mediacenter we already have in qml 
or not, it will depend if it will turn out too limited or not i guess.

> *Should I use plasmalibs as basis for this or start it as a new qt
> (kdelibs) app?
> *Should this be a C++ plasmoid that I can later tie to a new
> containment to make it fullscreen?

details

> *Should I code this as a new containment type? like SAL or newspaper
> and this would be picture viewer?

i don't think a containment is the right way to go

> *Could I do this with a javascript plasmoid? I guess not easily
> *Should a start with gwenview as basis? and add a new viewmode to gwenview?

you can also try to improve gwenview, would e a nice thing to do too.
but there will be never much possibilities to easily mix a qwidget based 
applicatin with a qgraphicswidget based one like plasma

you should -really- look into the mediacenter shell that already have some 
beginning of this stuff (and desperately needs hands;)), rather than start 
your own.

Cheers,
Marco Martin


More information about the Plasma-devel mailing list