Type conversion in C++

Type conversion is the process of changing the type of variables in a program. The ultimate goal of type conversions is to allow variables of a single data type operate with variables of another. This is done so that we may take use of various features of type representation and hierarchies.

Brief introduction:

Type conversions could be used to ensure that the right kind of mathematical computation is executed. Calculating the addition of two variables, one of type int and the other of type float, is a nice example of type conversion in action. We must convert the int variable to float to get the summation of variables.

Conversion of C++ types:

Type conversion can have two forms:

• Conversion of implicit types
• Conversion of explicit types

Conversion of implicit types:

The compiler performs implicit type conversion, often known as automated type conversion, without the requirement for just a user-initiated action. It occurs when an expression contains more than one data type, in which case type conversion is performed to prevent data loss.

The data type of each variable is changed to those of variable with the greatest data type.

The following is the automated type conversion order:

bool -> char -> short int -> int -> 


unsigned int -> long -> unsigned -> 


long long -> float -> double -> long double

Information like signs is lost whenever a signed type is directly transformed to an unsigned type, but when a long is directly transformed to a float, overflow occurs.

Example:

#include <iostream>
using namespace std;


int main()
{
  int i = 500; // integer value of y
  char j = 'y'; // character value of x


  // converting b implicitly to int. ASCII
  // the value of 'i' here is 1200
  a = a + b;


  // i is implicitly converted to float
  float a = a + 30.0;


  cout << "a = " << a << endl
    << "b = " << b << endl
    << "z = " << z << endl;


  return 0;
}

Output:

a = 1700
b = y
z = 1730

Explanation:

Initially, we created an int i to have value of y integer and char j for character value of x. Then, by using sequence of automatic type conversion given above, we saw how b is implicitly changed to int and i to float in the program above. Finally, we returned the values which were required as the output.

Conversion of explicit types:

User-initiated explicit type conversion is also called type casting. The user can use explicit type conversion to transform a variable from one data type to another.

Explicit type conversion is possible in C++ in two ways:

1. Conversion with the cast operator
2. Conversion with assignment operator

Conversion with assignment operator:

In this sort of conversion, the needed type is supplied before the parentheses. Data loss is caused by explicit type casting. "Forced casting" is the term for it.

Syntax:

(type) exp

The keyword type here is used to mention the data type of final result.

Example:

#include <iostream>
using namespace std;


int main()
{
  double num = 6.0;


  // Explicitly converting double to int
  int sum = (int)num + 2;


  cout << "Ans = " << sum;


  return 0;
}

Output:

Ans = 8

Explanation:

Explicit type conversion is demonstrated in the program above.  First the double num was given and then it was explicitly converted to an int to perform the operation sum. Since the result was a typecast, a double was transformed to an int. Finally, the ans was given as the output.

Conversion with the cast operator:

A cast operator is a unary operator that converts one data type to another by default. The C++ language has four different forms of casts.

These are:

• Static cast - The most basic type of cast is the static cast. Upcasts and downcast are both possible. It takes a very long time to put together the cast. There are no checks conducted all throughout conversion process to ensure that the object being converted is a full object of a target type.
• Dynamic cast - Dynamic cast ensures that the type conversion results in a full and valid object of a target pointer type.
• Const cast - It specifies whether the item must be constant or non-constant. This implies that the constant must be set or deleted.
• Reinterpret cast - Any pointer type, even those from distinct classes, can be transformed to another pointer type. It does not check if the data indicated by the pointer and the type of pointer are the same.

Let us have a look at how the cast operator for conversion works:

Static cast:

This is the most fundamental type of cast. It works throughout the compilation process. It may also do implicit type conversions such as int into float or pointer into void* by calling explicit conversion routines.

Example:

#include <iostream>
using namespace std;
int main()
{
  float num = 9.5;


  // using the cast operator here
  int ans = static_cast<int>(num);


  cout << ans;
}

Output:

9

Explanation:

In the above example, we created a program to show the use of a static cast operator for conversion in C++. We gave a float value to num as 9.5. On using the cast operator for num and stored its value in int ans. The output was int value because of the use of the cast operator.  

Dynamic cast:

This casting takes care of polymorphism. When casting to a derived class, it's just essential to utilise. This should only be used in inheritance when type cast from a parent to a derived class.

Because the real type of the object referred to isn't the intended subclass' type, the dynamic cast will fail.

There are two different kinds of dynamic casts:

1. In dynamic cast of the pointer if a cast fails, NULL is returned. It is a quick and simple approach to determine if an object belongs to a specific dynamic type.

Syntax:

<type> *xsubclasses = dynamic_casts<<type> *>( xobjects );

• dynamic reference cast When casting a reference, it is not allowed to return a NULL pointer to indicate failure; a dynamic cast of a reference variable would cause the std::bad cast (from the typeinfo> header) exception.

Syntax:

<type> subclasses = dynamic_casts<<type> &>( ref_objct );

Example (Dynamic cast):

#include<iostream>
using namespace std;
class Class1 {
   public:
      virtual void display()const {
         cout << "It’s from Class1\n";
      }
};
class Class2 {
   public:
      virtual void display()const {
         cout << "It’s from Class2\n";
      }
};
class Class3: public Class1, public Class2 {
   public:
      void display()const {
         cout << "It’s from Class3\n";
      }
};
int main(){
   Class1* a = new Class1;
   Class2* b = new Class2;
   Class3* c = new Class3;
   a -> display();
   b -> display();
   c -> display();
   b = dynamic_cast< Class2*>(a); //The mentioned cast will fail
   if (b)
      b->display();
   else
      cout << "No Class2\n";
   a = c;
   a -> display(); //Showing from the Class3
   b = dynamic_cast< Class2*>(a); // The casting done here will be successful
   if (b)
      b -> display();
   else
      cout << "No Class B\n";
}

Output:

It’s from Class1
It’s from Class2
It’s from Class3
no Class2
It’s from Class3
It’s from Class3

Explanation:

In the above example we used the dynamic cast method. Here, the dynamic cast of the pointer, if a cast fails, returned the new classes i.e. new Class1 or new Class2 or new Class3. But when the dynamic cast pointer does not fail, the public classes are returned i.e. public Class1 or public Class2 or public Class3. It is a quick and simple approach to determine if an object belongs to a specific dynamic type or not. Hence we got the required output.

Const cast:

This method is used to remove variables' immutability. Non-const class members could be edited using the cast within a const member function.

Example:

#include <iostream>
using namespace std;


class employee
{
private:
  int roll;
public:
  // constructor
  employee(int n):roll(n) {}


  // A const function that changes roll with the help of const_cast
  void fun() const
  {
    ( const_cast <employee*> (this) )->roll = 20;
  }


  int getRolls() { return roll; }
};


int main(void)
{
  employee x(15);
  cout << "Old roll no: " << x.getRolls() << endl;


  x.func();


  cout << "New roll no: " << x.getRolls() << endl;


  return 0;


}

Output:

Old roll no: 15
New roll no: 20

Explanation:

This is treated as a const employee by the compiler, thus const this inside the const member function func(). Because this is a constant pointer to a constant object, the compiler prevents this reference from modifying the data members. Const cast changes the format of this pointer to employee const this.

Reinterpret cast:

This cast is being used to transform a pointer from one type to another, irrespective of whether the two types are related. It does not verify whether the pointer type as well as the data it points to are same. It does not have a return type. It just changes the pointer's type. Only the source pointer variable is accepted as an argument.

Example:

#include <iostream>
using namespace std;


int main()
{
  int* x = new int(75);
  char* c = reinterpret_cast<char*>(x);
  cout << *x << endl;
  cout << *c << endl;
  cout << x << endl;
  cout << c << endl;
  return 0;
}

Output:

75
A
0xe5es80
A

Explanation:

In the above example, the cast is being used to transform a pointer from one of the forms or data type to other. First an int value was introduced which was further casted using the reinterpret cast to a char value and hence printed. Here, only the pointer's type is changed and not the value.

Conclusion:

This concludes our discussion of type conversion in C++. In this article, we learned about implicit and explicit conversions in C++.

However, in order to avoid data loss and other complications, conversions or typecasting should only be used when absolutely necessary.