Is there anyway that we could have a setting for the material domain within a material function?
What this would be especially useful for is being able to use Scene Texture:Scene Depth nodes within a Post Process Material Function, that could then output the captured image and a separate depth-based mask. This would allow for key-less masking of 3D assets rendered in front of a UI background.
Currently, we don’t have alpha channel support in post process materials, so the entire image must be contained within the capture itself. We would have to go back to depth-generating a chroma key color in the background and masking by that. This is all fine and dandy assuming you’re not making a project that has any hint of that particular color range anywhere in any scene.
It would then be up to the user to ensure that the material which the function is nested into matches the same domain and type and should flag errors with a mismatch the same way it does on texture nodes with different sample methods.