So what’s a good use case for references? Let’s take a look.

Previously, when we passed parameters to a function, we used normal variables and that’s known as pass-by-value. But because the variables passed into the function are out of scope, we can’t actually modify the value of the arguments.

Pass-by-reference refers to passing parameters to a function by using references. When called, the function can modify the value of the arguments by using the reference passed in.

This allows us to:

• Modify the value of the function arguments.
• Avoid making copies of a variable/object for performance reasons.

The following code shows an example of pass-by-reference. The reference parameters are initialized with the actual arguments when the function is called:

void swap_num(int &i, int &j) {

  int temp = i;
  i = j;
  j = temp;

}

int main() {

  int a = 100;
  int b = 200;

  swap_num(a, b);

  std::cout << "A is " << a << "\n";
  std::cout << "B is " << b << "\n";

}

Notice that the int &i and int &j are the parameters of the function swap_num().

When swap_num() is called, the values of the variables a and b will be modified because they are passed by reference. The output will be:

A is 200
B is 100

Suppose we didn’t pass-by-reference here and have the parameters as simply int i and int j in the swap_num() function, then i and j would swap, but a and b wouldn’t be modified.

And the output will then be:

A is 100
B is 200

To reiterate, using references as parameters allows us to modify the arguments’ values. This can be very useful in a lot cases.