[Boost-users] Re: tokenizer question