Lesson 20 - Review

Subtract the zero character from it.

int num = character – ‘0’;

See lesson 03

Dynamic memory is allocated on the heap, whereas static memory is allocated on the stack.

See lesson 11

void function(int size) {
	int* arr = new int[size];
}

The size parameter is static, the pointer (“arr” itself) is static, and the actual array of integers pointed to by “arr” is dynamic.

See lesson 11

The two statically allocated variables are freed at the closing bracket (end of the scope). In this example, the dynamic array is not actually freed at all. To prevent a memory leak, it should be freed before you lose access to it at the end of the function.

See lesson 11

Creating an object and specifying the copy constructor, passing by value, and returning by value.

MyClass a(b);				// Call the copy constructor explicitly
void func(MyClass c) {		// Called to copy in parameter
	Return d;				// Called to return value
}

See lesson 14

If the parameter to the copy constructor is passed by value, it will call the copy constructor, which will then call the copy constructor…etc. You must pass the parameter by reference to prevent an infinite loop.

See lesson 14

Default, parameterized, and copy.

See lesson 13 See lesson 14

One with no parameters at all, and one with all default parameters.

See lesson 13

student(int = 5, int = 10);

student();
student(100);			// Uses only the 1st default parameter
student(100,50);

See lesson 13

A friend function is not actually a part of a class, but has access to the private members of that class. This means that it does not have a calling object, as it is essentially a normal function with special access, so you must pass an instance of your class as a parameter. A member function is actually part of a class, and automatically has access to the data members of that class. Member functions also always have a calling object, so they automatically know which instance in particular you are talking about.

object.function(); 		// Member function
function(object);		// Friend function

See lesson 14

Only one type of constructor can be called once for any single object, and it is called when that object is created, or instantiated to be exact. The destructor is also called only once, when the object is either deleted (if dynamic), or goes out of scope.

See lesson 13

The compiler automatically generates one for you, but it will probably not do what you want—it will not default values and will not delete your memory.

See lesson 13

The default copy constructor does what’s called a “shallow” copy, meaning it essentially sets each data member equal to the source with the assignment operator. However, if you have a pointer as a data member, it will cause both pointers to point to the same data. This may be what you want, but you usually want to actually make a copy of that data, and have the new pointer point to the copy. To do this, you must implement your own copy constructor.

See lesson 13

A memory leak occurs when you lose access to dynamically allocated memory, which means it can never be freed. This is especially an issue when the code that leaks memory is called more than once, say in a function or a loop, as every time that code is run the program will leak more memory, until it eventually runs out of memory and crashes. A dangling pointer is a pointer that does not point to valid data, for example a garbage pointer. Always avoid dangling pointers by setting them to NULL when you delete dynamically allocated memory.

See lesson 11

Abstract Data Types (ADTs).

See lesson 12

student* s = new student[100];
delete[] s;

100, once for each “student” in the array.

See lesson 11

student *s1 = new student;

delete s1;

Note no brackets, as we allocated a single object rather than an array.

See lesson 11

Arrays of characters in general do not need to end with the null character, but this is the defining feature of C-Style strings, which must always end with the null character.

See lesson 07

Private makes data members only accessible from within member functions of the class, whereas public allow them to be accessed from anywhere.

class MyClass {
public:
	int x;
private:
	int y;
};

int main() {
	MyClass o;
	o.x = 5;			// Valid, x is public
	o.y = 5;			// Invalid, y is private
}

See lesson 13

You make a variable constant when you know that the value should never change, as making it constant will assure that. You make a member function constant when it will never modify the data members of that class. Finally, you can’t make non-member functions constant, as they have no associated object and that doesn’t make sense.

See lesson 14

A constant object can only call member functions if they are constant themselves, as explained in question 20.

const MyClass c;
c.function();		// Function must be a constant member function

See lesson 14

 *ptr, &, ptr->, ptr[]

The star dereferences a pointer, the ampersand takes the address of a variable, the arrow dereferences a pointer and then uses the dot operator, and the brackets offset from a pointer and then dereference it.

*pointer = 5;
pointer = &variable;
classPointer->function();
arrayPointer[3] = 5;

See lesson 10 See lesson 12