Comment by norir

When using null terminated strings, parsing can be branchless because you don't need bounds checks and can use a jump table indexed by the byte.