Re: Questions about chapter 11 (Operator Overloading) of TC++PL.

From:

"Victor Bazarov" <v.Abazarov@comAcast.net>

Newsgroups:

comp.lang.c++

Date:

Tue, 23 Oct 2007 14:42:43 -0400

Message-ID:

<fflfb2$me$1@news.datemas.de>

Rajesh S R wrote:

On Oct 23, 2:09 pm, anon <a...@no.no> wrote:

Wayne Shu wrote:

Hi everyone, I am reading B.S. 's TC++PL (special edition).
When I read chapter 11 Operator Overloading, I have two questions.

1. In subsection 11.2.2 paragraph 1, B.S. wrote "In particular,
operator =, operator[], operator(), and operator-> must be nonstatic
member function; this ensures that their first operands will be
lvalues". I know that these operators must be nonstatic member
functions, but why this ensure their first operands will be lvalues?
e.g.
class foo
{
int i;
public:
foo & operator = (int ii) { i = ii; return *this; }

The first operant is ii.

};

foo bar() { return foo(); }

bar() = 10;

I wonder if this is legal?
The above statement is equivalent to:

bar().operator=(10);

bar() produces a temporary object of type foo. Its member is modified
by the above statement. Is it legal to change the value of temporary
objects created in a full expression, as pre C++ standards?

Yes. Why wouldn't it be legal? A temporary object is not constant.

Some more clarifications are:
Can I evaluate the address of such temporary objects?

No, the address-of operator only applies to lvalues, but you can return
'this' from some member function.

I mean, can I do:
std::cout << &bar();

No, but you can do

    class foo {
        ...
        foo& getAddr() { return this; }
    };

    ...
    std::cout << bar().getAddr();

Please give me the relevant sections of C++03 which deals with the
temporary objects created within a full expression.

See section 12.2.

V
--
Please remove capital 'A's when replying by e-mail
I do not respond to top-posted replies, please don't ask