[boost] Re: Problem solved: EBO pessimization in smart_ptr not an issue anymore!

From: David B. Held (dheld_at_[hidden])
Date: 2002-05-02 18:34:42

"Andrei Alexandrescu" <andrewalex_at_[hidden]> wrote in message
news:aas3gb$dkc$1_at_main.gmane.org...
> "David Abrahams" <david.abrahams_at_[hidden]> wrote in message
> news:133a01c1f1c7$00b96320$6401a8c0_at_boostconsulting.com...
> [...]
> > 3. Is using public inheritance at all really important? My impression
> > is that you do this to allow the policies to add interface to the
> > smart_ptr... while at the same time you eschew public member
> > functions.

I just understood what Dave was really saying.  Doh!

> Hey, I think it would be a great idea to use private inheritance. That
way,
> policies can enhance interface only through free functions, which is my
> intent anyway. What do others think?

I'm afraid I don't know how it works, now.  Consider:

template <typename T>
class MyPolicy
{
protected:
    friend inline void free_function(const MyPolicy& ???);
};

If smart_ptr inherits privately from MyPolicy, there is no way for
free_function() to know that a smart_ptr is really a MyPolicy.
Otherwise, we would need to specify the full pointer type, which
the policy shouldn't know about, should it?

Dave