Well, I couldn't describe a better use case, even if I tried. Check out the free sample, which shows how to build a lexer/scanner/tokenizer.