[boost] Re: Any good reason not to have == for shared_ptr?

From: ivan66_at_[hidden]
Date: 1999-11-09 21:09:56

darin adler <dari-_at_[hidden]> wrote: 
original article:http://www.egroups.com/group/boost/?start=910
> > Well, what would you want it to do, compare the pointers
> > or the contained objects? That's a good enough reason as
> > far as I'm concerned.
> 
> Hmm. I had assumed that given this
> 
>     shared_ptr<X> a;
>     shared_ptr<X> b;
> 
> that this
> 
>     a == b
> 
> would compare the pointers and this
> 
>     *a == *b
> 
> would compare the objects.
> 
> It seems a bit extreme to leave this up to each individual programmer
to
> decide. But I'd like to hear others' thoughts on the matter.
> 
>     -- Darin
> 

That sounds reasonable.  shared_ptr has pointer-like semantics, so its
operator== should have pointer-like semantics too.  If for some reason
you did want to compare objects and not pointers, the '*a == *b' syntax
is clean and already supported.

I assume that it would be legal to compare a shared_ptr<Derived> with a
shared_ptr<Base>?  Also, that a shared_array<Base> could be compared to
a shared_array<Base>, but attempting comparison with a
shared_array<Derived> would trigger a compile-time error?

--
Best regards,
Ivan J. Johnson