Hi Sebastian, I have just re-read this thread and feel the time is right for me to do a review of your changes. Could you provide me with your latest changes?
As a general note, the current work I'm doing is focussing on passing #define's into Shaders to toggle on/off features and provide defaults via osg::StateSet::setDefine(..) style API, as well also allow GLSL Shaders themselves to specify what they require via #pragam statements in the GLSL shaders. The new functionality could reduce the need for shader composer. I'm optimistic for many users the approach I'm working on will be simple to use and fit in well with how osg::Program/osg::Shader's have been used in application so far. I will start a new thread today or tomorrow on the topic of my work, but first I need to flesh out a bit more of the design and implementation so I can illustrate what I'm talking about with actual OSG and GLSL code. Cheers, Robert. On 5 February 2014 at 12:01, Sebastian Messerschmidt < [email protected]> wrote: > Hi Robert, > maybe this one has gone below your radar. Can you please check on it? > I've developed this a bit more and found out that I had to provide my own > compare function to get rid of the additional program being constructed. > If there are no objections from your side I'd like to prepare a submission. > > cheers > Sebastian > > Hi Robert, > > I've done my best to follow your instructions. See attached changes vs. > trunk. > The shadercomposition example still works as expected. > Comparing the outputs however, an additional program is build, and I'm > quite clueless, if my refactoring is correct (see attached outputs) > > I'd appreciate some advices here. > > Hi Sebastian, > > On 27 November 2013 11:26, Sebastian Messerschmidt < > [email protected]> wrote: > >> Okay I can try to refactor it, but I'm not 100% sure if I got you >> right. Would the caching attribute (last_applied_shadercomponent) still be >> used then? >> > > I'm thinking last_applied_shadercomponent would be redundent and could > be removed. > > >> Also do you mean by "get the last applied StateAttribute's from the >> AttributeStack" to use the last_applied_attribute member? >> > > Yes, use last_applied_attribute instead > of last_applied_shadercomponent. The code associated with the later could > be completely removed from osg::State so would simply it a little. > > Simplifying code and providing greater extensibility sounds like a good > design change to me :-) > > Robert. > > > _______________________________________________ > osg-users mailing > [email protected]http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org > > > > > _______________________________________________ > osg-users mailing > [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 > >
_______________________________________________ osg-users mailing list [email protected] http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
