Hello Martin,
Yes I know, but I won't to use the CompositeViewer. Because I will use this
feature in the gtk embedded viewer.
CompositeViewer and embedded graphics contexts are two separate things.
You can still use a CompositeViewer even if you use an embedded graphics
context.
In your case, I assume you're using a single window with multiple views
overlaid over the main one? You could create multiple osgViewer::Views,
then set the same GraphicsContext (GraphicsWindowEmbedded in your case I
guess) on all the views' main cameras. Then set a different
CameraManipulator on each view and you're set.
Having now looked at the osgviewerGTK code, I see it has a class
OSGGTKDrawingArea that derives from osgViewer::Viewer. That strikes me
as unnecessary coupling, but I've only looked at the code quickly, so I
don't know the details. You could make this derive from CompositeViewer
instead. I don't know what it uses from osgViewer::Viewer - perhaps
instead of deriving, it could have a pointer to an osgViewer::ViewerBase
instead? Then it would be independent of the viewer type.
Anyways, I can imagine it's probably more work than what you want to do,
but CompositeViewer is the tool you need, so using things as designed
will be more future-proof. Alternatively calling getInverseMatrix() and
setting that matrix on the camera you want might work, or might not. Or
just don't use a camera manipulator and write your own event handler
that will manipulate the camera's view matrix.
Hope this helps,
J-S
--
______________________________________________________
Jean-Sebastien Guay jean-sebastien.g...@cm-labs.com
http://www.cm-labs.com/
http://whitestar02.webhop.org/
_______________________________________________
osg-users mailing list
osg-users@lists.openscenegraph.org
http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
