Uses simple ranges for matching, optimization done via LLVM.
Faster performance in read and write on x86-64.
Improved code readability.