The Mighty C++ Template

The Mighty C++ Template - Understanding function overloading
(Page 2 of 9 )

To understand where templates have come from though, you must first understand function overloading. Let's say for an example, that we wanted to create a function that would work with numeric values and return the average of those values. The function should be able to accept integers, floats and doubles. To accomplish this, we could create three functions to handle each different type of numeric value:

To work with integers:

int GetAverage(int num1, int num2, int num3)

{

return static_cast<int>((num1 + num2 + num3) / 3);

}

To work with floating point values:

float GetAverage(float num1, float num2, float num3)

{

return static_cast<float>((num1 + num2 + num3) / 3);

}

To work with double values:

double GetAverage(double num1, double num2, double num3)

{

return static_cast<double>((num1 + num2 + num3) / 3);

}

Then, whenever we called the GetAverage() function from our program, the C++ compiler would decide which function it should call based on the types of the parameters passed to that function. If an un-handled type, such as long was passed to the function, then the compiler would automatically cast it to the type that it most closely resembled (In this case, long would be converted to int).

As you can see, we end up with three functions that do exactly the same thing. This is bad, because it creates a lot of redundant code, and our executable file will be bigger than it needs to be.

There has to be a better way... and there is... it's called a template.
blog comments powered by Disqus