Return value without copy construct

[stereomatching]

struct String
{
  String()
  {
    std::cout<<"String::String(), this="<<this<<std::endl;
  }

  ~String()
  {
    std::cout<<"String::~String(), this="<<this<<std::endl;
  }

  String(String const &other)
  {
    std::cout<<"String::String(String const &other), this="<<this;
    std::cout<<", other="<<&other<<std::endl;
  }

  String& operator=(String const &other)
  {
    std::cout<<"String::operator=(String const &other), this="<<this;
    std::cout<<", other="<<&other<<std::endl;
    return *this;
  }	
};

String const operator*(String const &lhs, String const &rhs)
{
  String temp(rhs);
  std::cout<<"return by value(* operator)"<<std::endl;
  return temp; //without copy construct
}

int main()
{													
  std::cout<<"***************value******************"<<std::endl;
  String D;
  String E;
  String F;
  D = E * F;
  std::cout<<"***************value******************"<<std::endl<<std::endl;	
		
	
  std::cin.get();

  return 0;
}

Compiler vc2010
OS win7 64bits

I expected it would copy construct temp before calling copy assignment
What kind of optimization did it done?NRVO?
If NRVO did it, what is the limit of NRVO?
Thanks

[mike_2000_17]

That is obviously a return value optimization, I mean, it's a textbook example of NRVO. That's what you would expect any half-decent compiler to do.

I don't understand your question about the "limit" of NRVO. If you mean, how complex a situation can the compiler cope with, then I would guess that is pretty compiler-dependent.

[stereomatching]

how complex a situation can the compiler cope with, then I would guess that is pretty compiler-dependent.

That is what I mean, it is a good news that we have && now.

[mike_2000_17]

I assume you are referring to rvalue-references, not the logic "and" operator.

[stereomatching]

I assume you are referring to rvalue-references, not the logic "and" operator.

of course, next time I will say rvalue-references rather than &&