C++ Virtual Functions Interview Preparation Guide

If the destructor in the base class is not made virtual, then an object that might have been declared of type base class and instance of child class would simply call the base class destructor without calling the derived class destructor.

Hence, by making the destructor in the base class virtual, we ensure that the derived class destructor gets called before the base class destructor.

class a
{
public:
a(){printf("\nBase Constructor\n");}
~a(){printf("\nBase Destructor\n");}
};

class b : public a
{
public:
b(){printf("\nDerived Constructor\n");}
~b(){printf("\nDerived Destructor\n");}
};
int main()
{
a* obj=new b;
delete obj;
return 0;
}

Output:
Base Constructor
Derived Constructor
Base Destructor

By Changing
~a(){printf("\nBase Destructor\n");}
to
virtual ~a(){printf("\nBase Destructor\n");}

Output:
Base Constructor
Derived Constructor
Derived Destructor
Base Destructor.

When a Derived Class object is assigned to Base class, the base class' contents in the derived object are copied to the base class leaving behind the derived class specific contents. This is referred as Object Slicing. That is, the base class object can access only the base class members. This also implies the separation of base class members from derived class members has happened.

class base
{
public:
int i, j;
};
class derived : public base
{
public:
int k;
};
int main()
{
base b;
derived d;
b=d;
return 0;
}
here b contains i and j where as d contains i, j& k. On assignment only i and j of the d get copied into i and j of b. k does not get copied. on the effect object d got sliced.

A virtual function is a member function that is declared within a base class and redefined by a derived class. To create virtual function, precede the function’s declaration in the base class with the keyword virtual. When a class containing virtual function is inherited, the derived class redefines the virtual function to suit its own needs.

Base class pointer can point to derived class object. In this case, using base class pointer if we call some function which is in both classes, then base class function is invoked. But if we want to invoke derived class function using base class pointer, it can be achieved by defining the function as virtual in base class, this is how virtual functions support runtime polymorphism.

Consider following program code:

Class A
{
int a;
public:
A()
{
a = 1;
}
virtual void show()
{
cout <<a;
}
};

Class B: public A
{
int b;
public:
B()
{
b = 2;
}
virtual void show()
{
cout <<b;
}
};

int main()
{
A *pA;
B oB;
pA = &oB;
pA->show();
return 0;
}

Output is 2 since pA points to object of B and show() is virtual in base class A.

A virtual table is a mechanism to perform dynamic polymorphism i.e., run time binging. Virtual table is used to resolve the function calls at runtime. Every class that uses virtual functions is provided with its own virtual functions.

Every entry in the virtual table is a pointer that points to the derived function that is accessible by that class. A hidden pointer is added by a compiler to the base class which in turn calls *_vptr which is automatically set when an instance of the class is created and it points to the virtual table for that class..

Overriding of functions occurs when one class is inherited from another class. Overloading can occur without inheritance.
Overloaded functions must differ in function signature ie either number of parameters or type of parameters should differ. In overriding, function signatures must be same.
Overridden functions are in different scopes; whereas overloaded functions are in same scope.
Overriding is needed when derived class function has to do some added or different job than the base class function.
Overloading is used to have same name functions which behave differently depending upon parameters passed to them.

Overriding of functions occurs in Inheritance. A derived class may override a base class member function. In overriding, the function names and parameter list are same in both the functions. Depending upon the caller object, proper function is invoked.

Consider following sample code:

class A
{
int a;
public:
A()
{
a = 10;
}
void show()
{
cout << a;
}
};

class B: public A
{
int b;
public:
B()
{
b = 20;
}
void show()
{
cout << b;
}
};

int main()
{
A ob1;
B ob2;
ob2.show(); // calls derived class show() function. o/p is 20
return 0;
}

As seen above, the derived class functions override base class functions. The problem with this is, the derived class objects can not access base class member functions which are overridden in derived class.

Base class pointer can point to derived class objects; but it has access only to base members of that derived class.

Therefore, pA = &ob2; is allowed. But pa->show() will call Base class show() function and o/p would be 10.

By default, matching of function call with the correct function definition happens at compile time. This is called static binding or early binding or compile-time binding. Static binding is achieved using function overloading and operator overloading. Even though there are two or more functions with same name, compiler uniquely identifies each function depending on the parameters passed to those functions.

C++ provides facility to specify that the compiler should match function calls with the correct definition at the run time; this is called dynamic binding or late binding or run-time binding. Dynamic binding is achieved using virtual functions. Base class pointer points to derived class object. And a function is declared virtual in base class, then the matching function is identified at run-time using virtual table entry.

Vtables are used for virtual functions. Its a shortform for Virtual Function Table.

It's a static table created by the compiler. Compiler creates a static table per class and the data consists on pointers to the virtual function definitions. They are automatically initialised by the compiler's constructor code.

Since virtual function pointers are stored in each instance, the compiler is
enabled to call the correct vrtual function at runtime.

When two or more objects are derived from a common base class, we can prevent multiple copies of the base class being present in an object derived from those objects by declaring the base class as virtual when it is being inherited. Such a base class is known as virtual base class. This can be achieved by preceding the base class’ name with the word virtual.

Consider following example:

class A
{
public:
int i;
};

class B : virtual public A
{
public:
int j;
};

class C: virtual public A
{
public:
int k;
};

class D: public B, public C
{
public:
int sum;
};

int main()
{
D ob;
ob.i = 10; //unambiguous since only one copy of i is inherited.
ob.j = 20;
ob.k = 30;
ob.sum = ob.i + ob.j + ob.k;
cout << “Value of i is : ”<< ob.i<<”\n”;
cout << “Value of j is : ”<< ob.j<<”\n”; cout << “Value of k is :”<< ob.k<<”\n”;
cout << “Sum is : ”<< ob.sum <<”\n”;

return 0;
}.