Hi Manuel,

Thanks for the references to your CEGUI drawable.  I am was suprised to see MatrixTransform and Projection node embedded in it.  Is it that CEGUI doesn't actually set the projection and modeview matrices itself?

Design wise I think you should be able to use a local event callback than needing to directly wire it to the viewer.

Robert.

On 9/5/06, Manuel A. Fernandez Montecelo <[EMAIL PROTECTED]> wrote:
Hi,

BTW, you can see my implementation here:

http://svn.gna.org/viewcvs/fearann/trunk/src/client/cegui/cltceguidrawable.h?rev=346&view=markup
http://svn.gna.org/viewcvs/fearann/trunk/src/client/cegui/cltceguidrawable.cpp?rev=346&view=markup

Apart from that, you only have to add it to the viewer, like this:

        CltCEGUIDrawable* ceguiDrawable =
                new CltCEGUIDrawable(mWindowWidth, mWindowHeight);
        mScene->addChild(ceguiDrawable->getNode());


The event handler for CEGUI is called from "higher level" EH, which calls the
different handlers of the application in turn depending on their priority
(the GUI goes first).  I haven't found major problems with this
implementation with mouse positions, keys injected and so on.

And yes, CEGUI falls short in many aspects compared to what OSG might need as
a general windowing system for many types of applications, but it seems like
the only more-or-less mature alternative.

Regards.


On Tuesday 05 September 2006 18:09, Robert Osfield wrote:
> Hi Marc,
>
> Thanks for the feedback, this is very useful.
>
> How much restriction is there on using CEGUI w.r.t requiring the graphics
> context?  I'm think of things like loading schemes, fonts as well as input
> injection.  Obvious rendering will need a graphics context.
>
> Personally I'm a bit disappointed with the restrictions that CEGUI sets in
> terms of set up, putting this alongside the lack of support for multiple
> windows and graphics contexts I don't see it as a great fit for a scalable
> arhitecture like the OSG.  Perhaps these problems could be fixed, but it'll
> my guess is that it'll require a substational refactor.
>
> Robert.

--
Manuel A. Fernandez Montecelo < [EMAIL PROTECTED]>
_______________________________________________
osg-users mailing list
[email protected]
http://openscenegraph.net/mailman/listinfo/osg-users
http://www.openscenegraph.org/

_______________________________________________
osg-users mailing list
[email protected]
http://openscenegraph.net/mailman/listinfo/osg-users
http://www.openscenegraph.org/

Reply via email to