parser: Lexer.get_token use String[j] instead of String.chars[]