From: David Abrahams (dave_at_[hidden])
Date: 2003-04-24 11:32:36
Daniel Frey <daniel.frey_at_[hidden]> writes:
> Heath Davis wrote:
>> Bravo!!! This is a truly novel approach! I agree that the
>> shared_ptr::operator< is meaningless. If boost wishes to claim STL
>> compatibility, container functions should be allowed to operate
>> directly on the base level objects, rendering smart pointers
>> completely transparent.
>
> I disagree strongly. A smart pointer is an as-transparent-as-possible
> replacement for a pointer, not a replacement for the object!
Perfectly put. I would add that just like pointer comparisons, the
semantics of shared_ptr<T>::operator< are very useful**, and if we
accepted Heath's suggested semantics, we'd be replacing functionality
that's otherwise unavailable with functionality we can easily get
another way.
-Dave
**in fact, they're even more useful than the semantics for pointers
because equivalence is not destroyed by derived<-->base class
conversions.
-- Dave Abrahams Boost Consulting www.boost-consulting.com