PMC meeting summary

[Alessandro Diaferia]

Hello list,
here follows a short resume of what discussed during the yesterday's meeting
:)

- *Media Fetcher Backends:*
Hayri is responsible for this with his GSoC. Currently we have 2
dataengines, called *picture *and *video*.
Both were intended as a bridge among different fetching backends and the PMC
browsing view.
Their API is really similar except for little specific stuff and so we
decided to merge them into one, unique,
WebMedia DataEngine (the name can be, of course, changed to a better one):
the common data structure we'll use to return the fetched information from
web services will be wrapped into something like:
MediaResource and MediaResourceCollection. The DataEngine and its backend
will all be written in JS given the nature of the data: web-driven.
We currently have everything to provide such kind of DataEngine except for
the *add-on finder* which is something Aaron will likely work on soon.
Also there's something we can eventually share with the kde-silk project, so
maybe we should have a talk with them in case we already have
tools for such stuff.
Additional for Hayri: as stated by Aaron: Network  access is already in the
bindings while xml parsing can be done via http://xmljs.sourceforge.net/


- *QML into PMC:
*We're gonna make use of QML for layouts in PMC and specifically in our
State Machine.
Christopher Blauvelt is contributing a lot with QML into PMC.
Currently the aim is to have a new view for the mediabrowser, inheriting its
AbstractItemView,
that takes advantage of QML in order to show pictures in a fancy way :p
For those who are curious some stuff is there under the declarative/ folder
in the root directory.

- *QGestures and input abstraction in PMC*
This is a bit tricky part and probably needs some further discussion. The
aim is to have a sort of layer that maps different input events
into common Media Center events like *play, pause, next-picture,
previous-picture, volume-up *and so on..
This way we can have different input devices to be recognized and ready to
be used (think about remotes and/or keyboards for example).
As discussed we'll probably use the QGestureEvent and eventually create our
own PMC events in order to ease the mapping.
Later events will be sent to the components loaded into PMC that will be
able to perform the proper action.
I also think that we can have virtual event methods in our base MCApplet
classes like playEvent, volumeUpEvent and such.
In each applet the gesture recognition would call the proper virtual event
method.
Of course i might be wrong so, please, point out your thoughts about this.

I think this is all; please add what i eventually missed from the meeting.

Cheers!

P.S.: How to upload the IRC log to plasma.kde.org?

-- 
Alessandro Diaferia
KDE Developer
KDE e.V. member
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.kde.org/pipermail/plasma-devel/attachments/20100613/48f0eff2/attachment.htm