Fixes issue #25. (char 255 aliased to -1 and missing tests for end of input).
1) All layers of input scanning now share a single EndOfInput value.
This avoids translation of it across layers of encapsulation.
2) Some places looking for end of line were not stopping on EndOfInput.
3) Use of "char" for the input made char values > 127 be negative numbers.
This allowed for aliasing of 255 to -1, etc. This is fixed by using
unsigned char.
This simplification is a prelude to eliminating what I appear unnecessary
symbol inserts into tables when tokenizing in the preprecessor, which
show up as taking notable time. (Performance issue.) It also simply makes
the preprocessor easier to understand, which it is badly in need of.
- consistently dealing with EOF and its effect on error recovery (bug 11444, #1)
- turning a simulated OO hierarchy of function pointers and typecasting into a real C++ class hierarchy
- correctly handling '\' everywhere, in all classes of tokens, as a layer before preprocessing itself
- conditionally handling '\n' in macro expansion input, depending on whether it is in a macro expression or not
- delete some unused "memory cleanup" code
git-svn-id: https://cvs.khronos.org/svn/repos/ogl/trunk/ecosystem/public/sdk/tools/glslang@24626 e7fa87d3-cd2b-0410-9028-fcbf551c1848
- macro expansion of hexidecimal numbers
- give errors instead of warnings/silence on extra tokens after #endif, #else, etc.
- give errors on reserved macro name use, reuse of argument, and redefinition with different whitespace presence
- detect and give error for all cases of #elif and #else after #else
git-svn-id: https://cvs.khronos.org/svn/repos/ogl/trunk/ecosystem/public/sdk/tools/glslang@23982 e7fa87d3-cd2b-0410-9028-fcbf551c1848
