contextview themeing: plasmification?

[Nikolaj Hald Nielsen]

Other attendes at aKademy spent a lot more time time looking into
plasma than me, but I will give you my thoughts on the issue.

I like the general idea, but as I mentioned to the other devs at
aKademy, I think we need to start seriously narrowing the scope of
features we will try to implement for Amarok 2.0 in order to have at
least a usable beta by the time KDE 4.0 ships. This particular feature
does sound like it could eat up a LOT of time.

Also, and maybe more significantly, there are ideas floating around to
make the plasma canvas "embeddable" in applications in a future
release of KDE. This would mean not only theming support, but also the
ability to drag the context items to the desktop or wherever


- Nikolaj

On 7/7/07, Leo Franchi <lfranchi at gmail.com> wrote:
> hey everyone-
>     so as leinir mentioned to me yesterday, the guys over at akademy were
> discussing the problem of themeing in the context view, and  a solution that
> was similar to the plasma theme engine came up. now, everyone at akademy a)
> knows more about how plasma works internally than i do (even after a few
> hours of grokking the code) b) has already thrown this idea around. i just
> want to open it up to everyone else and see what people think.
>
> basically, how it works is plasma renders SVG pixmaps onto the QGraphicsView
> desktop. it works something like this: 1) widget loads the svg widget theme
> 2) when time comes to display data, the widget paints the data "through" the
> SVG-- it takes care of the layout. 3) rendered pixmap is put onto desktop.
> this is very good for many reasons, including resolution independence, but
> most importantly, artists can provide theme packages of SVGs that completely
> control the layout of the widgets. this would fit exactly with our needs.
> also, it does caching of the SVGs so apparently its quite efficient.
>
> the con  that i can see right now is this: it's quite significantly more
> complicated than the current system. not that that is necessarily bad, but
> its just going to take longer to implement/perfect.
>
> also, plasma is fast :) and right now amarok2 is really slow. especially
> (but not only) the contextview. so if this helps, i'm all for it. i'm also
> willing to code this up, but basically want to gauge your guys' reactions:
> do you think it's a good idea? comments? problems?
>
> leo
>
> --
> ______________________________________________________
> Leo Franchi                    angel666 at myrealbox.com
> 665 Channing Ave         lfranchi at gmail.com
> Palo Alto                        cell: (650) 704 3680
> CA, USA                        home: (650) 329 0125
> Junior,
> Palo Alto High School,   http://euthydemus.homelinux.net
> 65 Embarcadero Road,
> Palo Alto,
> CA, USA
>
> GPG Key Fingerprint: 713F 1C92 11E3 4696 D067  B681 72D5 EAF0 1499 8B03
> Key ID: 14998B03
> Public key: http://euthydemus.homelinux.net/pub_key.txt
>
> _______________________________________________
> Amarok-devel mailing list
> Amarok-devel at kde.org
> https://mail.kde.org/mailman/listinfo/amarok-devel
>
>