Open fknfilewalker opened 2 years ago
I would second this request, however the general ecosystem around shader development seems to be going into multiple directions simultanenously. Several language implementors provide their own compilation backends to transform their host language to SPIR-V directly and aim for tighter host-language integration. This should be a preferred path.
GLSL is a bad and limited reinvention of C, with all its flaws, lousy compilation model, "type-safety" etc.. Actually fixing GLSL almost definetively means not using it in the first place.
I understand that, but this is also a sad solution
Are there plans to improve the glsl language. This could be from simple features like default parameters for functions, till more complex ones like interfaces/classes and/or templates? It seems to me that there was no progress made in this direction in the last couple of years. Features like these would make it much easier to write clean and compact code. If there is already an ongoing discussion, please refer me to it.