An interesting code about a constructor calling a constructor

Code:

---

class A
{
public:
        int i;
        A():i(0){
                cout<<"A1:"<<endl;
                cout<<"i="<<i<<endl;
                cout<<"A1 End"<<endl;
        }
        A(int x):i(x)
        {
                cout<<"A2:"<<endl;
                cout<<"i="<<i<<endl;
                A();
                cout<<"i="<<i<<endl;
                cout<<"A2 End"<<endl;
        }
};

int main()
{
    A obj(2);
}

---

It's interesting to see that when A() modifies i to 0, A2 still has i as 2. How come?!

int A::i is shared in the obj, right?

When it returns form A() it returns I to 2.

I assume the primary constructor takes precedence.

Why would you do this anyway, seems a bit asinine.

It's the same with inheritance too!

Code:

---

class B
{
    public:
    int theInt;
    B():theInt(0) { cout << "B, theInt=" << theInt << endl; };
    B(int x):theInt(x) { cout << "B, theInt=" << theInt << endl; };

};
class A : public B
{
public:
        int i;
        A():B(),i(0) {

                cout<<"A1:";
                cout<<"i="<<i<<endl;
                cout<<"A1 End"<<endl;
        }
        A(int x):B(x)
        {

                cout<<"A2:";
                cout<<"i="<<i<<endl;
                B();
                A();
                cout<<"A:B, theInt=" << B::theInt<<endl;
                cout<<"i="<<i<<endl;
                cout<<"A2 End"<<endl;
        }
};

int main()
{
    A obj(2);
    system("pause");
    return EXIT_SUCCESS;
}

---

This is not multiple inheritance. This is single inheritance.
As to why, I only know that &i inside A(int) != &i inside A().

The question to your answer is - it isn't.
But you may be confused by the actual events.
First constructor A(int) is invoked, which invokes constructor B(int).
In the body of A(int), it then invokes B(), which returns and then invokes A(), which then invokes B().
Then it returns from the constructor.

Quote:

---

Originally Posted by meili100

Code:

---

class A
{
public:
        int i;
        A():i(0){
                cout<<"A1:"<<endl;
                cout<<"i="<<i<<endl;
                cout<<"A1 End"<<endl;
        }
        A(int x):i(x)
        {
                cout<<"A2:"<<endl;
                cout<<"i="<<i<<endl;
                A();
                cout<<"i="<<i<<endl;
                cout<<"A2 End"<<endl;
        }
};

int main()
{
    A obj(2);
}

---

It's interesting to see that when A() modifies i to 0, A2 still has i as 2. How come?!

int A::i is shared in the obj, right?

---

The line "A();" creates a completely new object of type A, unrelated to obj. If you want i to be shared between different copies of A objects, make it a static member.

Quote:

---

Originally Posted by King Mir

The line "A();" creates a completely new object of type A, unrelated to obj. If you want i to be shared between different copies of A objects, make it a static member.

---

Painfully obvious... just not to me... lol

Code:

---

class A
{
public:
        int i;
        A():i(0){
                cout<<"A1:"<<endl;
                cout<<"i="<<i<<endl;
                cout<<"A1 End"<<endl;
        }
        A(int x):i(x)
        {
                cout<<"A2:"<<endl;
                cout<<"i="<<i<<endl;
                this->A(); //modify here
                cout<<"i="<<i<<endl;
                cout<<"A2 End"<<endl;
        }
};

int main()
{
    A obj(2);
}

---

Why I can't use this->A() to qualify the call?!

Because a constructor is not a method. You'll need a separate reset() method to reset an object to it's initial state.

Actually, there is another way. you can call the destructor then placement new. But you can't do that inside the constructor.

That sounds like an awful hack anyway.
Break the constructor out into a common method and call it instead.

In C++09, this is added as a new feature. The code you had is wrong because you simply instantiate a new A which disappears at the end of the (constructor) scope (like King Mir pointed out); you would expect it to work in an initializer list instead:

Code:

---

A(int x) : A() { ... }

---

This will not compile today.

The idea is that you often have a "basic" constructor which does a bunch of work, and some "specialized" ones which all do the same "basic" work plus something extra. Currently, you'll have to break out the "basic" stuff in a separate function, instead of calling the basic constructor from within the initializer list of the "specialized" ones.

--
Computer Programming: An Introduction for the Scientifically Inclined