Hi Robert, thanks for the reply. Well, I spent quite a good amount of time to understand how it works and it all points to the StandardShadowMap, the call I posted. This shadowmap technique allows you to attach shaders to it through the interface, but the program is set localy again for a local _stateset that is pushed through the CullVisitor in a cull stage. All I need is reference to this local instance of the Program. Perhaps, maybe introduce a method getProgram() which again will not hurt anyone.
Also, I can not think of any special reason to specialize the traverse call to osg::Group then forbid the extension .... Any further thoughts? Nick On Fri, May 1, 2015 at 6:50 PM, Robert Osfield <[email protected]> wrote: > Hi Nick, > > I haven't worked with osgShadow for over a year so it's not fresh in > my mind. There are mechanisms for override the state management in > osgShadow but I don't recall how widely they are implemented amongst > the shadow techniques. > > With the proposed change, with the explicit osg::Group::traverse() > method there must be a reason why this was done but without a code > review and trawl through the svn logs I can't provide the answer. In > general overriding the management of osg::Program and other > osg::StateSet setting shouldn't require lots of hacks, so if a > particular ShadowTechnique is failing in this respect then perhaps > this needs looking at. > > Robert. > > > Robert. > > On 1 May 2015 at 17:15, Trajce Nikolov NICK > <[email protected]> wrote: > > Hi Robert, > > > > I posted a while ago a question how to get the osg::Program associated > with > > the shaders from the StandardShadowMap in order to extend. And it is > localy > > defined as you can see in the code. However I found a workaround, by > > extending the ShadowingScene and catch the StateSet from the Cull > traversal. > > But, this will not work since the line below. Here is my proposed > change, it > > will not hurt anyone I think - these ShadowMap* classes are > > over-encapsulated in my opinion. > > > > void StandardShadowMap::ViewData::cullShadowReceivingScene( ) > > > > { > > > > _cv->pushStateSet( _stateset.get() ); > > > > > > #if 0 > > > > _st->getShadowedScene()->osg::Group::traverse( *_cv ); > > > > #else > > > > _st->getShadowedScene()->traverse( *_cv ); > > > > #endif > > > > > > _cv->popStateSet(); > > > > } > > > > > > It will be nice if this forcing of osg::Group::traverse is replaced by > > ordinary traverse thus anyone can re-write and extend. What you think? > > > > > > Please let me know and thanks a bunch as always! > > > > > > Nick > > > > > > -- > > trajce nikolov nick > > > > _______________________________________________ > > osg-users mailing list > > [email protected] > > > http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org > > > _______________________________________________ > osg-users mailing list > [email protected] > http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org > -- trajce nikolov nick
_______________________________________________ osg-users mailing list [email protected] http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
