Function Pointers

Function pointers are an advanced topic, and this section can be safely skipped or skimmed by those only looking for C++ basics.
In the lesson on pointers, you learned that a pointer is a variable that holds the address of another variable. Function pointers are similar, except that instead of pointing to variables, they point to functions!
Consider the case of an array:

1	int nArray[10];

As you now know, nArray is actually a constant pointer to a 10 element array. When we dereference the pointer (either by *nArray or nArray[nIndex]), the appropriate array element is returned.
Now consider the following function:

1	int foo();

If you guessed that foo is actually a constant pointer to a function, you are correct. When a function is called (via the () operator), the function pointer is dereferenced, and execution branches to the function.
Just like it is possible to declare a non-constant pointer to a variable, it’s also possible to declare a non-constant pointer to a function. The syntax for doing so is one of the ugliest things you will ever see:

1	// pFoo is a pointer to a function that takes no arguments and returns an integer

2	int (*pFoo) ();

The parenthesis around *pFoo are necessary for precedence reasons, as int *pFoo() would be interpreted as a function named pFoo that takes no parameters and returns a pointer to an integer.
In the above snippet, pFoo is a pointer to a function that has no parameters and returns an integer. pFoo can “point” to any function that matches this signature.
Assigning a function to a function pointer
There are two primary things that you can do with a pointer to a function. First, you can assign a function to it:

01

int foo()

02

{

03

}

04

05

int goo()

06

{

07

}

08

09	int main()

10

{

11	int (*pFoo)() = foo; // pFoo points to function foo()

12	pFoo = goo; // pFoo now points to function goo()

13

return 0;

14

}

One common mistake is to do this:

1	pFoo = goo();

This would actually assign the return value from a call to function goo() to pFoo, which isn’t what we want. We want pFoo to be assigned to function goo, not the return value from goo(). So no parenthesis are needed.
Note that the signature (parameters and return value) of the function pointer must match the signature of the function. Here is an example of this:

01	// function prototypes

02	int foo();

03	double goo();

04	int hoo(int nX);

05

06	// function pointer assignments

07	int (*pFcn1)() = foo; // okay

08	int (*pFcn2)() = goo; // wrong -- return types don't match!

09	double (*pFcn3)() = goo; // okay

10	pFcn1 = hoo; // wrong -- pFcn1 has no parameters, but hoo() does

11	int (*pFcn3)(int) = hoo; // okay

Calling a function using a function pointer
The second thing you can do with a function pointer is use it to actually call the function. There are two ways to do this. The first is via explicit dereference:

1	int foo(int nX)

2

{

3

}

4

5	int (*pFoo)(int) = foo; // assign pFoo to foo()

6

7	(*pFoo)(nValue); // call function foo(nValue) through pFoo.

1	int foo(int nX)

2

{

3

}

4

5	int (*pFoo)(int) = foo; // assign pFoo to foo()

6

7	pFoo(nValue); // call function foo(nValue) through pFoo.

As you can see, the implicit dereference method looks just like a normal function call — which is what you’d expect, since normal function names are pointers to functions anyway! However, some older compilers do not support the implicit dereference method, but all modern compilers should.
Why use pointers to functions?
There are several cases where pointers to function can be of use. One of the most common is the case where you are writing a function to perform a task (such as sorting an array), but you want the user to be able to define how a particular part of that task will be performed (such as whether the array is sorted in ascending or descending order). Let’s take a closer look at this problem as applied specifically to sorting, as an example that can be generalized to other similar problems.
All sorting algorithms work on a similar concept: the sorting algorithm walks through a bunch of numbers, does comparisons on pairs of numbers, and reorders the numbers based on the results of those comparisons. Consequently, by varying the comparison (which can be a function), we can change the way the function sorts without affecting the rest of the sorting code.
Here is our selection sort routine from a previous lesson:

01	void SelectionSort(int *anArray, int nSize)

02

{

03	using namespace std;

04	for (int nStartIndex= 0; nStartIndex < nSize; nStartIndex++)

05

{

06	int nBestIndex = nStartIndex;

07

08	// Search through every element starting at nStartIndex+1

09	for (int nCurrentIndex = nStartIndex + 1; nCurrentIndex < nSize; nCurrentIndex++)

10

{

11	// Note that we are using the user-defined comparison here

12	if (anArray[nCurrentIndex] < anArray[nBestIndex]) // COMPARISON DONE HERE

13	nBestIndex = nCurrentIndex;

14

}

15

16	// Swap our start element with our best element

17	swap(anArray[nStartIndex], anArray[nBestIndex]);

18

}

19

}

Now, let’s replace that comparison with a function to do the comparison. Because our comparison function is going to compare two integers and return a boolean value, it will look something like this:

1	bool Ascending(int nX, int nY)

2

{

3	return nY > nX;

4

}

And here’s our selection sort routine using the Ascending() function to do the comparison:

01	void SelectionSort(int *anArray, int nSize)

02

{

03	using namespace std;

04	for (int nStartIndex= 0; nStartIndex < nSize; nStartIndex++)

05

{

06	int nBestIndex = nStartIndex;

07

08	// Search through every element starting at nStartIndex+1

09	for (int nCurrentIndex = nStartIndex + 1; nCurrentIndex < nSize; nCurrentIndex++)

10

{

11	// Note that we are using the user-defined comparison here

12	if (Ascending(anArray[nCurrentIndex], anArray[nBestIndex])) // COMPARISON DONE HERE

13	nBestIndex = nCurrentIndex;

14

}

15

16	// Swap our start element with our best element

17	swap(anArray[nStartIndex], anArray[nBestIndex]);

18

}

19

}

In order to let the caller decide how the sorting will be done, instead of using our own hard-coded comparison function, we’ll allow the caller to provide his own sorting function! This is done via a function pointer.
Because the caller’s comparison function is going to compare two integers and return a boolean value, a pointer to such a function would look something like this:

1	bool (*pComparison)(int, int);

So, we’ll allow the caller to pass our sort routine a pointer to their desired comparison function as the third parameter, and then we’ll use the caller’s function to do the comparison.
Here’s a full example of a selection sort that uses a function pointer parameter to do a user-defined comparison, along with an example of how to call it:

01	#include <algorithm> // for swap

02

03	// Note our user-defined comparison is the third parameter

04	void SelectionSort(int *anArray, int nSize, bool (*pComparison)(int, int))

05

{

06	using namespace std;

07	for (int nStartIndex= 0; nStartIndex < nSize; nStartIndex++)

08

{

09	int nBestIndex = nStartIndex;

10

11	// Search through every element starting at nStartIndex+1

12	for (int nCurrentIndex = nStartIndex + 1; nCurrentIndex < nSize; nCurrentIndex++)

13

{

14	// Note that we are using the user-defined comparison here

15	if (pComparison(anArray[nCurrentIndex], anArray[nBestIndex])) // COMPARISON DONE HERE

16	nBestIndex = nCurrentIndex;

17

}

18

19	// Swap our start element with our best element

20	swap(anArray[nStartIndex], anArray[nBestIndex]);

21

}

22

}

23

24	// Here is a comparison function that sorts in ascending order

25	// (Note: it's exactly the same as the previous Ascending() function)

26	bool Ascending(int nX, int nY)

27

{

28	return nY > nX;

29

}

30

31	// Here is a comparison function that sorts in descending order

32	bool Descending(int nX, int nY)

33

{

34	return nY < nX;

35

}

36

37	// This function prints out the values in the array

38	void PrintArray(int *pArray, int nSize)

39

{

40	for (int iii=0; iii < nSize; iii++)

41	cout << pArray[iii] << " ";

42	cout << endl;

43

}

44

45	int main()

46

{

47	using namespace std;

48

49	int anArray[9] = { 3, 7, 9, 5, 6, 1, 8, 2, 4 };

50

51	// Sort the array in descending order using the Descending() function

52	SelectionSort(anArray, 9, Descending);

53	PrintArray(anArray, 9);

54

55	// Sort the array in ascending order using the Ascending() function

56	SelectionSort(anArray, 9, Ascending);

57	PrintArray(anArray, 9);

58

59

return 0;

60

}

This program produces the result:

9 8 7 6 5 4 3 2 1
1 2 3 4 5 6 7 8 9

Is that cool or what? We’ve given the caller the ability to control how our selection sort does it’s job.
The caller can even define his own “strange” comparison functions:

01	bool EvensFirst(int nX, int nY)

02

{

03	// if nX is not even and nY is, nY goes first

04	if ((nX % 2) && !(nY % 2))

05	return false;

06

07	// if nX is even and nY is not, nX goes first

08	if (!(nX % 2) && (nY % 2))

09	return true;

10

11	// otherwise sort in Ascending order

12	return Ascending(nX, nY);

13

}

14

15	int main()

16

{

17	using namespace std;

18

19	int anArray[9] = { 3, 7, 9, 5, 6, 1, 8, 2, 4 };

20

21	SelectionSort(anArray, 9, EvensFirst);

22	PrintArray(anArray, 9);

23

24

return 0;

25

}

The above snippet produces the following result:

2 4 6 8 1 3 5 7 9

As you can see, using a function pointer in this context provides a nice way to allow a caller to “hook” it’s own functionality into something you’ve previously written and tested, which helps facilitate code reuse! Previously, if you wanted to sort one array in descending order and another in ascending order, you’d need multiple version of the sort routine. Now you can have one version that can sort any way the caller desires!
Making function pointers pretty with typedef
Let’s face it — the syntax for pointers to functions is ugly. However, typedefs can be used to make pointers to functions look more like regular variables:

1	typedef bool (*pfcnValidate)(int, int);

This defines a typedef called “pfcnValidate” that is a pointer to a function that takes two ints and returns a bool.
Now instead of doing this:

1	bool Validate(int nX, int nY, bool (*pfcn)(int, int));

You can do this:

1	bool Validate(int nX, int nY, pfcnValidate pfcn)

Which reads a lot nicer!