Interesting tool. Only downside after a bit of testing: the formatting of unicode characters that take more than 8 bits to encode are not correctly splitted. ‡ is such an example.