Re: [boost] [inplace_factory] Question about nplace_factory

From: Steven Watanabe (steven_at_[hidden])
Date: 2007-11-19 13:37:29

AMDG

Corrado Zoccolo <czoccolo <at> gmail.com> writes:

> I think SFINAE is not enough to replicate faithfully the rules employed by
> the compiler to select the which operator new is called for an expression
> like new T(a1,..,an) . It will look not only in T, but also in its base
> classes.

I don't understand.  Both "new T" and has_new<T> look in base
classes of T.  SFINAE seems to work on both msvc 8.0 and gcc 3.4.4

I think that what I would do is create a generic allocator
that looks up the correct operator new/delete

template<class T, bool = has_new<T>::value>
struct my_allocator;

template<class T>
struct my_allocator<T, true> {
    T* allocate(std::size_t n) {
        return(T::operator new(n * sizeof(T)));
    }
    void deallocate(T* victim, std::size_t n) {
        return(T::operator delete(victim, n * sizeof(T)));
    }
    //...
};

template<class T>
struct my_allocator<T, false> {
    T* allocate(std::size_t n) {
        return(::operator new(n * sizeof(T)));
    }
    void deallocate(T* victim, std::size_t n) {
        return(::operator delete(victim));
    }
    //...
};

In Christ,
Steven Watanabe