Boost logo

Boost :

From: Robert Kawulak (kawulak_at_[hidden])
Date: 2006-04-27 05:18:24


> From: Spencer Collyer

> If I read your suggestion right, you are thinking that we
> could replace the bounds policy with one of your constrained
> types? Because of the need to reserve the one-past-the-end
> value for the end() iterator, I'm not sure that would be
> feasible. We need to be able to store the one-past-the-end
> value but not allow the user to specify it as a valid
> Does that make sense?

If I understand you correctly, you want the lower bound to be
included in the allowed range, but the upper bound to be
excluded. If so, then the Constrained Types library allows for
such one-side-opened ranges (I hope I express myself precisely
enough, I don't know math english that well ;-).

However, there may be a problem - the library is designed to work
only with random access iterators. Does your library use random
access iterators only or any kind of iterator is allowed?

Best regards,

Boost list run by bdawes at, gregod at, cpdaniel at, john at