An Equal Opportunity University
CS 215-401 Fall 2014Lecture 11
Class Review, Inheritance, Polymorphism
Ismail Abumuhfouz
Slide modified from Link 1
2
Programming Concept Evolution
• Unstructured• Procedural• Object-Oriented
3
Procedural Concept
• The main program coordinates calls to procedures and hands over appropriate data as parameters.
4
Object-Oriented Concept
• Objects of the program interact by sending messages to each other
5
Objects An object is an encapsulation of both functions and data
• Objects are an Abstraction– represent real world entities– Classes are data types that define shared common properties or attributes– Objects are instances of a class
• Objects have State – have a value at a particular time
• Objects have Operations – associated set of operations called methods that describe how to carry out
operations
• Objects have Messages – request an object to carry out one of its operations by sending it a message– messages are the means by which we exchange data between objects
6
OO Perspective
Let's look at the Rectangle through object oriented eyes:• Define a new type Rectangle (a class)
– Data• width, length
– Function• area()
• Create an instance of the class (an object)• Request the object for its area
In C++, rather than writing a procedure, we define a class that encapsulates the knowledge necessary to answer the question - here, what is the area of the rectangle.
7
class Rectangle
{
private:
int width, length;
public:
Rectangle(int w, int l)
{
width = w;
length = l;
}
main(){
Rectangle rect(3, 5); cout << rect.area()<<endl;}
int area(){ return width*length;}
};
Example Object Oriented Code
8
Object-Oriented Programming Languages
• Characteristics of OOPL:–Encapsulation– Inheritance –Polymorphism–Overloading
• OOPLs support :– Modular Programming– Ease of Development– Maintainability
9
Characteristics of OOPL• Encapsulation: Combining data structure with actions
– Data structure: represents the properties, the state, or characteristics of objects– Actions: permissible behaviors that are controlled through the member functions
Data hiding: Process of making certain data inaccessible • Inheritance: Ability to derive new objects from old ones
– permits objects of a more specific class to inherit the properties (data) and behaviors (functions) of a more general/base class
– ability to define a hierarchical relationship between objects
• Polymorphism: Ability for different objects to interpret functions differently.
• Overloading
Class Definitions
Class Definitions
2
Visibility Modifiers
● Permissions for data members and member functions:
●
private: Can only be accessed by that
class protected: Can be accessed by
subclasses public: Can be accessed by
anyone
●
●
●Class members are private by
default Cannot be applied to the whole
class:
●
public class A; // Don't do this!
protected class B; // Or this!
Class Definitions
3
Example
class Box{
public:
// Class name
// Publicmembers
section
private: // Private members section
int weight;}; // Notice the semicolon
Box( int w ) { weight = w; }int getWeight( )
const{ return weight; }
Example in UML
●
Class Definitions
4
Always 3 sections– Name– Data members– Member functions
Visibility modifiers– Public (+)– Private (-)– Protected (#)
●
Box
-weight : int+getWeight() : int
5
Inline Methods
● A method that is implemented inside the class definition is called an inline method.
The compiler may choose to expand the body of the method at the point of call.
●
● The compiled code executes faster since it avoids the overhead of a function call.
Inlining can make the compiled code larger and more complex (usually not desirable properties).
●
●Use inlining only for very short methods.
Never use them with loops or recursive
calls.
●
Class Definitions
6
Class Interface
● Usually the class definition is in an interface (or header) file, and the implementation in an implementation (or source) file.
●Interface files usually have a .h extension.
Implementation files can have a .cpp, .c++, or .C.
The filename does not have to match the class
name.
●
●
● A #include statement is used to include the class definition into the implementation file:
#include “myclass.h”
Class Definitions
7
Fully Qualified Names
● Use a #ifndef ... #define ... #endif in the header file to avoid including the class definition more than once.
Methods implemented in the source file use a fully qualified function name.
●
● This avoids conflicts with other classes that have a method with the same name.
A fully qualified name consists of the class name, a double colon, and the method name:
... ClassName::methodName ...
●
Class Definitions
8
Example
● box.h#ifndef#define
BOX_H BOX_H
class Box{public:Box( int w);int getWeight( ) const;
private: intweight;
};#endif
● box.c#include“box.h”Box::Box( int w ){
weight = w;}
int Box::getWeight({
returnweight;
}
) const
Class Definitions
12
Constructors
● Constructors serve two purposes: they create and initialize an object
A constructor is a method with the same name as the class, and does not have a return type
There are three types of constructors:
●
●
● A default constructor takes no arguments
An ordinary constructor has some arguments
A copy constructor is used to make copies (clone)
●
●
Class Definitions
13
Copy Constructor
● A copy constructor is used to make a copy of an object value.
● It takes an instance of the same class as a constant reference argument:
Box( const Box & b );
A copy constructor is often called implicitly, such as when passing by value:
●
doStuff( a ); // Copy constructor called.
Box a; // Default constructor gets// called implicitly, too.
Class Definitions
14
Example
class Box {public:Box( ) // Default constructor{ weight = 0; }
private: intweight;};
Box( int w ) // Ordinary constructor
{ weight = w; }Box( const Box & b ) // Copy constructor
{ weight = b.weight; }
Class Definitions
22
Destructors
● The destructor is implicitly called when an object is deleted
● Object may have been explicitly deleted using delete
An object could also be automatically deleted at the end of a function if the object is stack-resident
The destructor is never called directly
●
●
● The destructor is defined using a tilde followed by the class name and takes no arguments:
~Box( );
Class Definitions
23
Destructors cont.
● The destructor usually deletes any heap-resident memory the object may have allocated:
class Storage
{ public:Storage(
int
s
)
{
space
=
new
int[s];
}int &
operator[](int i )
{returnspace[i];
}
~Storage( ) { delete []
space; } private:int * space;
};
Class Definitions
24
The keyword this
● Every method has a pointer named this which points to the object the method was invoked on
class Box {public:Box( int w ) : weight( w ){ }
Box &doStuff( ) { this->weight
= 73; return *this;
}
private: intweight;
};
Class Definitions
26
Friends
● A class can have friends that are allowed to access its private data members and functions:
classBox{ public:
Box(
int
w
)
:
weight(
w
)
{
}
// Allow access for global functionoperator<<friend ostream & operator<<( ostream & out);
// Allow class Crate to access weightfriend class Crate;
private: intweight;
};
Friend Functions Example#include <iostream>using namespace std; class Rectangle { int width, height; public: Rectangle() {} Rectangle (int x, int y) : width(x), height(y) {} int area() {return width * height;} friend Rectangle duplicate (const Rectangle&);}; Rectangle duplicate (const Rectangle& param){ Rectangle res; res.width = param.width*2; res.height = param.height*2; return res;} int main () { Rectangle foo; Rectangle bar (2,3); foo = duplicate (bar); cout << foo.area() << '\n'; return 0;}
Source: http://www.cplusplus.com/doc/tutorial/inheritance/
Friend Class Example
Source: http://www.cplusplus.com/doc/tutorial/inheritance/
#include <iostream>using namespace std; class Square; class Rectangle { int width, height; public: int area () {return (width * height);} void convert (Square a);}; class Square { friend class Rectangle; private: int side; public: Square (int a) : side(a) {}}; void Rectangle::convert (Square a) { width = a.side; height = a.side;} int main () { Rectangle rect; Square sqr (4); rect.convert(sqr); cout << rect.area(); return 0;}
More About Friends
In the previous example:• Rectangle is considered a friend class by
Square, but Square is not considered a friend by Rectangle.
• Therefore, the member functions of Rectangle can access the protected and private members of Square but not the other way around.
• Of course, Square could also be declared friend of Rectangle, if needed, granting such an access.
Another property of friendships is that they are not transitive: The friend of a friend is not considered a friend unless explicitly specified.
28
Class Definition
Data MembersCan be of any type, built-in or user-definednon-static data member
Each class object has its own copy
static data memberActs as a global variableOne copy per class type, e.g. counter
29
class Rectangle
{
private:
int width;
int length;
static int count;
public:
void set(int w, int l);
int area();
}
Static Data Member
Rectangle r1;Rectangle r2;Rectangle r3;
widthlength
widthlength
widthlength
r1
r3
r2
count
Static Data Member#include <iostream>using namespace std;class Box{ public: static int objectCount; // Constructor definition Box(double l=2.0, double b=2.0, double h=2.0){ cout <<"Constructor called." << endl; length = l; breadth = b; height = h; // Increase every time object is created objectCount++; } double Volume(){ return length * breadth * height; } private: double length; // Length of a box double breadth; // Breadth of a box double height; // Height of a box};// Initialize static member of class Boxint Box::objectCount = 0;int main(void){ Box Box1(3.3, 1.2, 1.5); // Declare box1 Box Box2(8.5, 6.0, 2.0); // Declare box2 // Print total number of objects. cout << "Total objects: " << Box::objectCount << endl; return 0;}
Source: http://www.tutorialspoint.com/cplusplus/cpp_static_members.htm
Inheritance
32
class Rectangle{
private:
int numVertices;
float *xCoord, *yCoord;
public:
void set(float *x, float *y, int nV);
float area();
};
Inheritance Concept
RectangleTriangle
Polygon
class Polygon{
private:
int numVertices;
float *xCoord, *yCoord;
public:
void set(float *x, float *y, int nV);
};
class Triangle{
private:
int numVertices;
float *xCoord, *yCoord;
public:
void set(float *x, float *y, int nV);
float area();
};
33
RectangleTriangle
Polygonclass Polygon{
protected:
int numVertices;
float *xCoord, float *yCoord;
public:
void set(float *x, float *y, int nV);
};
class Rectangle : public Polygon{
public:
float area();
};
class Rectangle{
protected:
int numVertices;
float *xCoord, float *yCoord;
public:
void set(float *x, float *y, int nV);
float area();
};
Inheritance Concept
34
RectangleTriangle
Polygonclass Polygon{
protected:
int numVertices;
float *xCoord, float *yCoord;
public:
void set(float *x, float *y, int nV);
};
class Triangle : public Polygon{
public:
float area();
};
class Triangle{
protected:
int numVertices;
float *xCoord, float *yCoord;
public:
void set(float *x, float *y, int nV);
float area();
};
Inheritance Concept
35
Inheritance Concept
Point
Circle 3D-Point
class Point{
protected:
int x, y;
public:
void set (int a, int b);
};
class Circle : public Point{
private:
double r;
};
class 3D-Point: public Point{
private:
int z;
};
xy
xyr
xyz
36
Augmenting the original class
Specializing the original class
Inheritance Concept
RealNumber
ComplexNumber
ImaginaryNumber
Rectangle Triangle
Polygon Point
Circle
realimag
real imag
3D-Point
37
Why Inheritance ?
Inheritance is a mechanism for
building class types from existing class types
defining new class types to be a specialization augmentation
of existing types
38
Define a Class Hierarchy
Syntax:class DerivedClassName : access-level BaseClassName
where access-level specifies the type of derivation
private by default, orpublic
Any class can serve as a base classThus a derived class can also be a base class
39
Class Derivation
Point
3D-Point
class Point{
protected:
int x, y;
public:
void set (int a, int b);
};
class 3D-Point : public Point{
private:
double z;
… …
};
class Sphere : public 3D-Point{
private:
double r;
… …
};
Sphere
Point is the base class of 3D-Point, while 3D-Point is the base class of Sphere
40
What to inherit?
In principle, every member of a base class is inherited by a derived class
just with different access permission
41
Access Control Over the Members
Two levels of access control over class members
class definitioninheritance type
base c lass / superc lass /pa ren t c lass
deriv ed c lass / subc lass /ch ild c lass
deriv
e fro
m
mem
bers
goe
s to
class Point{
protected: int x, y;
public: void set(int a, int b);
};
class Circle : public Point{
… …
};
42
The type of inheritance defines the access level for the members of derived class that are inherited from the base class
Access Rights of Derived Classes
private protected public
private - - -
protected private protected protected
public private protected public
Type of Inheritance
Access Controlfor M
embers
43
class daughter : --------- mother{
private: double dPriv;
public: void mFoo ( );
};
Class Derivationclass mother{
protected: int mProc;
public: int mPubl;
private: int mPriv;
};
class daughter : --------- mother{
private: double dPriv;
public: void dFoo ( );
};
void daughter :: dFoo ( ){
mPriv = 10; //error
mProc = 20;
};
private/protected/publicint main() {
/*….*/
}
class grandDaughter : public daughter {
private: double gPriv;
public: void gFoo ( );
};
44
What to inherit?
In principle, every member of a base class is inherited by a derived class
just with different access permission
However, there are exceptions forconstructor and destructor operator=() member friends
Since all these functions are class-specific
45
Constructor Rules for Derived Classes
The default constructor and the destructor of the base class are always called when a new object of a derived class is created or destroyed.
class A {
public:
A ( )
{cout<< “A:default”<<endl;}
A (int a)
{cout<<“A:parameter”<<endl;}
};
class B : public A
{
public:
B (int a)
{cout<<“B”<<endl;}
};
B test(1);A:defaultB
output:
46
Constructor Rules for Derived Classes
You can also specify an constructor of the base class other than the default constructor
class A {
public:
A ( )
{cout<< “A:default”<<endl;}
A (int a)
{cout<<“A:parameter”<<endl;}
};
class C : public A {
public:
C (int a) : A(a)
{cout<<“C”<<endl;}
};
C test(1);A:parameterC
output:
DerivedClassCon ( derivedClass args ) : BaseClassCon ( baseClass args )
{ DerivedClass constructor body }
47
Define its Own Members
Point
Circle
class Point{
protected:
int x, y;
public:
void set(int a, int b);
};
class Circle : public Point{
private:
double r;
public:
void set_r(double c);
};
xy
xyr
class Circle{
protected:
int x, y;
private:
double r;
public:
void set(int a, int b);
void set_r(double c);
};
The derived class can also define its own members, in addition to the members inherited from the base class
48
Even more …
A derived class can override methods defined in its parent class.
With overriding, the method in the subclass has the identical signature to the method in the base class. a subclass implements its own version of a base class method.
class A {
protected:
int x, y;
public:
void print ()
{cout<<“From A”<<endl;}
};
class B : public A {
public:
void print ()
{cout<<“From B”<<endl;}
};
49
class Point{
protected:
int x, y;
public:
void set(int a, int b)
{x=a; y=b;}
void foo ();
void print();
};
class Circle : public Point{
private: double r;
public:
void set (int a, int b, double c) {
Point :: set(a, b); //same name function call
r = c;
}
void print(); };
Access a Method
Circle C;
C.set(10,10,100); // from class Circle
C.foo (); // from base class Point
C.print(); // from class Circle
Point A;
A.set(30,50); // from base class Point
A.print(); // from base class Point
50
Putting Them Together
Time is the base classExtTime is the derived class with public inheritanceThe derived class can
inherit all members from the base class, except the constructoraccess all public and protected members of the base classdefine its private data memberprovide its own constructordefine its public member functionsoverride functions inherited from the base class
ExtTime
Time
51
class Time Specification
class Time{
public :
void Set ( int h, int m, int s ) ;void Increment ( ) ;void Write ( ) const ;Time ( int initH, int initM, int initS ) ; // constructor Time ( ) ; // default
constructor
protected :
int hrs ; int mins ; int secs ;
} ;
// SPECIFICATION FILE ( time.h)
52
Class Interface Diagram
Protected data:
hrs
mins
secs
Set
Increment
Write
Time
Time
Time class
53
Derived Class ExtTime // SPECIFICATION FILE ( exttime.h)
#include “time.h”
enum ZoneType {EST, CST, MST, PST, EDT, CDT, MDT, PDT } ;
class ExtTime : public Time // Time is the base class and use public inheritance
{ public :
void Set ( int h, int m, int s, ZoneType timeZone ) ;void Write ( ) const; //overridden
ExtTime (int initH, int initM, int initS, ZoneType initZone ) ; ExtTime (); // default constructor
private :ZoneType zone ; // added data member
} ;
54
Class Interface Diagram
Protected data:
hrs
mins
secs
ExtTime class
Set
Increment
Write
Time
Time
Set
Increment
Write
ExtTime
ExtTime
Private data:zone
55
Implementation of ExtTime
Default Constructor
ExtTime :: ExtTime ( ){
zone = EST ;}
The default constructor of base class, Time(), is automatically called, when an ExtTime object is created.
ExtTime et1;
hrs = 0mins = 0secs = 0zone = EST
et1
56
Implementation of ExtTimeAnother Constructor
ExtTime :: ExtTime (int initH, int initM, int initS, ZoneType initZone) : Time (initH, initM, initS) // constructor initializer
{ zone = initZone ;}
ExtTime *et2 =
new ExtTime(8,30,0,EST);hrs = 8mins = 30secs = 0zone = EST
et2
5000
???
6000
5000
57
Implementation of ExtTime
void ExtTime :: Set (int h, int m, int s, ZoneType timeZone) { Time :: Set (hours, minutes, seconds); // same name function call
zone = timeZone ;}
void ExtTime :: Write ( ) const // function overriding
{
string zoneString[8] =
{“EST”, “CST”, MST”, “PST”, “EDT”, “CDT”, “MDT”, “PDT”} ;
Time :: Write ( ) ;
cout <<‘ ‘<<zoneString[zone]<<endl;
}
58
Working with ExtTime
#include “exttime.h”… …
int main() {
ExtTime thisTime ( 8, 35, 0, PST ) ; ExtTime thatTime ; // default constructor
called
thatTime.Write( ) ; // outputs 00:00:00 EST
thatTime.Set (16, 49, 23, CDT) ; thatTime.Write( ) ; // outputs 16:49:23 CDT
thisTime.Increment ( ) ;thisTime.Increment ( ) ;thisTime.Write ( ) ; // outputs 08:35:02 PST
}
59
Inheritance Summary
Inheritance is a mechanism for defining new class types to be a specialization or an augmentation of existing types.
In principle, every member of a base class is inherited by a derived class with different access permissions, except for the constructors
ABSTRACT CLASS
Abstract Classes● An abstract class (or abstract base class) is a class
that contains pure virtual methods.
●A pure virtual method does not have a
body. It is instead assigned a null value:class
Animal{ public:
virtual
void
speak(
)
=
0;
●
};
● Abstract base classes can only be used through inheritance● It is impossible to create an instance of an
abstract class
Abstract Classes & Pure Virtual Functions
Some classes exist logically but not physically.Example : Shape
Shape s; // Legal but silly..!! : “Shapeless shape”
Shape makes sense only as a base of some classes derived from it. Serves as a “category”Hence instantiation of such a class must be prevented
class Shape //Abstract { public : //Pure virtual Function virtual void draw() = 0;}
A class with one or more pure virtual functions is an Abstract Class
Objects of abstract class can’t be created
Shape s; // error : variable of an abstract class
63
ExampleShape
virtual void draw()
Circle
public void draw()
Triangle
public void draw()
64
A pure virtual function not defined in the derived class remains a pure virtual function.Hence derived class also becomes abstract
class Circle : public Shape { //No draw() - Abstractpublic :void print(){ cout << “I am a circle” << endl;}
class Rectangle : public Shape {public :void draw(){ // Override Shape::draw() cout << “Drawing Rectangle” << endl;}
Rectangle r; // ValidCircle c; // error : variable of an abstract class
65
Pure virtual functions : Summary• Pure virtual functions are useful because they make
explicit the abstractness of a class• Tell both the user and the compiler how it was
intended to be used.• Note : It is a good idea to keep the common code as
close as possible to the root of you hierarchy
66
Summary ..continuedIt is still possible to provide definition of a pure virtual function in the base class.The class still remains abstract and functions must be redefined in the derived classes, but a common piece of code can be kept there to facilitate reuse.In this case, they can not be declared inline
class Shape { //Abstract public : virtual void draw() = 0;};
// OK, not defined inline void Shape::draw(){ cout << “Shape" << endl;}
class Rectangle : public Shape
{ public : void draw(){ Shape::draw(); //Reuse cout <<“Rectangle”<< endl;}
Polymorphism
68
Polymorphism – An Introduction
noun, the quality or state of being able to assume different forms - Webster.
• An essential feature of an OO Language• It builds upon Inheritance.• Allows run-time interpretation of object type for a
given class hierarchy• Also Known as “Late Binding”• Implemented in C++ using virtual functions
69
Dynamic Binding
• Is the run-time determination of which function to call for a particular object of a derived class based on the type of the argument
• Declaring a member function to be virtual instructs the compiler to generate code that guarantees dynamic binding
• Dynamic binding requires pass-by-reference
Polymorphism in C+
● All polymorphism in C++ is done using inheritance; there is no concept of an interface
A subclass is declared using the name of the class, a colon, the visibility of the parent class, and the name of the parent class:
●
public:intx_pos; int
y_pos;};
public:int radius;
};
Class 2DObject { Class{
Circle : public 2DObject
Polymorphism
71
Virtual and Non-Virtual Overriding
● Overriding occurs when a child class has a method with the exact same type signature as one of the parent class methods
Binding is the process of deciding whether to execute the parent's version or the child's version of a method
The keyword virtual determines whether static binding or dynamic binding is used
virtual only appears in the class definition
●
●
●
72
Virtual Functions
Virtual Functions overcome the problem of run time object determinationKeyword virtual instructs the compiler to use late binding and delay the object interpretationHow ?
• Define a virtual function in the base class. The word virtual appears only in the base class
• If a base class declares a virtual function, it must implement that function, even if the body is empty
• Virtual function in base class stays virtual in all the derived classes
• It can be overridden in the derived classes• But, a derived class is not required to re-implement a
virtual function. If it does not, the base class version is used
A Class Hierarchy
● Consider the following class hierarchy:class Animal {public: virtual};
void speak( ) = 0;
class Bird : public Animal {public: virtual void speak( )
}{ cout <<};
“twitter”;
Polymorphism
73
Polymorphism
74
A Class Hierarchy cont.
class Cat : public Mammal { public:void speak( ) { cout << “meow!”; }virtual void purr( ) { cout <<“purrrrr”; }};
class Dog : public Mammal { public:virtual void speak( ) { cout <<“woof!”;}void bark( ) { cout << “woof!”; }};
class Mammal : public Animal {public: virtual void speak( ){ cout << “can't speak”;} void bark( ) { cout << “can't bark”; };
Static Binding• virtual is not used when declaring the
method:void bark() {
cout << “can't bark”; }
• The decision is made at compile time based on the type of the variable:
Dog * d = new Dog( );Mammal * m = d ;d->bark() ; //woofm->bark(); // can't bark.
Dynamic Binding
● virtual is used to declare the method:virtual void speak( ){
cout << “woof!”; }
The binding decision is made at run-time based on the type of the object:
●
d->speak(); // woof!
m->speak(); // woof!
Animal *a = d;a->speak ();// woof!
Limitations• The validity of calling a method is always static.
If a method is not defined in a class or inherited from a parent class, it cannot be called:
• Overriding only works with heap-resident values:
Dog * d = new Dog( );Animal *a =d;d->bark( ); // woof!a->bark( );// Compile error, not allowed.
Mammal m =*d;m.speak(); // can't Speak
More Limitations
● Child classes cannot change the type of binding● A method that is declared virtual in a parent class
will always be virtual in a child class, even if virtual is not used in the child class
Similarly, a method that is not declared virtual in the parent class can never be made virtual in the child class
●
● Any method that is called from a constructor cannot be overridden
Virtual methods are never inlined●
Polymorphism
79
Private and Protected Inheritance
● Usually inheritance is public
Protected inheritance changes public members in the parent to protected in the child
Private inheritance changes public and protected members to private
●
●
class Pig :{
public:
protected Mammal
void oink( ) { cout << “Oink!”; }// The speak//
and bark methods can only beaccessed by child classes.
};
Virtual Destructors
● If any virtual methods are used, the destructor should be virtual to ensure that both the parent and child destructors are called
class Bird public: virtual
: public Animal {
~Bird( ) { cout << “bird killed”; }};
class Duck public: virtual virtual
: public Bird {
void speak( ) { cout << “quack!”; }}~Duck( ) { cout << “duck killed”;
};
Polymorphism Example
Please check the following example of polymorphism
82
Polymorphism Summary:When you use virtual functions, compiler store additional information about the types of object available and createdPolymorphism is supported at this additional overheadImportant :
virtual functions work only with pointers/referencesNot with objects even if the function is virtualIf a class declares any virtual methods, the destructor of the class should be declared as virtual as well.
83
So far, Polymorphism
Polymorphism is built upon class inheritance
It allows different versions of a function to be called in the same manner, with some overhead
Polymorphism is implemented with virtual functions, and requires pass-by-reference
Overloading
Operator Overloading
A Rational Class
● Consider this class for storing rational numbers:
class rational {
private: int top; intbottom;};
public:rational( int t = 0, int b = 1 )
:top(
t ), bottom( b ) { }
rational( const rational & r ):top(
r.top ), bottom( r.bottom ) { }int numerator( ) const{
return top; }
int denominator( ) const { returnbottom;
}
An add function
● To implement addition with two rationals, the following could be added to the class definition:
const rational add( const rational & r) const { int
intt = top * r.bottom + bottom * r.top; b = bottom * r.bottom;
return rational( t, b);
}● Now addition works:
rational a( 5, 6 );
rational b( 2, 3 );
rational c = a.add( b );
A Better add Function
● The syntax of the add function could be better. It would be nicer (and make sense) to write:Rational c = a + b;
● Operator overloading makes this possible:const rational operator+( const rational r) const& { int t = top * r.bottom + bottom *r.top; int b = bottom * r.bottom; return rational( t,b );}
Operator Overloading
89
Operator Overloading
● Operator overloading allows existing C++ operators to work with user-defined data types.
There are limits to this, however:●
● At least one operand must be a user-defined type. It is impossible to change the meaning of 2 + 2.
Cannot create new operators.
Cannot change precedence and associativity.
Don't change the meaning of an operator - operator+should always do something similar to addition.
●
●
●
Overloaded Operators
)
-> ->* new delete
+
&
-
|
*
~
/
!
%
&&
^
||
++ -- << >> , <
<= == != > >= =
+= -= *= /= %= &=
|= ^= <<= >>= [ ] (
Operator Overloading
91
Functions and Methods
● Operators can generally be overloaded as member functions or global functions.
● Unary operators can be methods with no arguments or global functions with one argument.
Binary operators can be methods with one argument or global functions with two arguments.
●
● Operators [], (), ->, and = must be methods.
If used as I/O operators (as they usually are), >> and<< must be global functions.
●
Binary Arithmetic Operators
● The result should be a new value.
The return value should be constant so it cannot be the target of an assignment:(a+b) =b; // This should be impossible
Parameters are values or constant references.
●
●
• The operands should not be modified.
• Methods should be declared constant:const rational operator/( const rational &r) const;
Binary Arithmetic Ops. cont.
●Subtraction as a method:const rational operator-( const rational &r) const{
int t= top * r.bottom – bottom * r.top; int b= bottom * r.bottom; return rational(t,b);}
●Multiplication as a global function:const rational operator*( const rational & l, const rational & r ){ return rational(l.numerator( )*r.numerator(),l.denominator( ) * r.denominator( ) );}
Comparison Operators
Work like the binary arithmetic operators, except these return a boolean.
Equals and less-than as methods:
bool operator==( const rational & r ) const{
return top * r.bottom == bottom *r.top;
}bool operator<({
const rational & r ) const
*r.top;
return top * r.bottom < bottom}
Increment and Decrement
● Can be prefix form (++i) or postfix form (i++).● Prefix form increments and returns the new
value:
● Postfix form increments but returns the original value:
int c = a++; // a = 7, c = 6● Prefix increment for the rational as a method:const
rational top = top
+
operator++( ){ bottom;
return}
int a = 5;int b = a++; // a = 6, b = 6
Increment and Decrement cont.
● To distinguish postfix from prefix, the postfix version uses a dummy integer argument:const rational operator++( int) { rational temp = *this;
top += return
bottom; temp;
}
const rational operator--( int ) {rational temp = *this;top -= return
bottom; temp;
}
Assignment Operator
● The right operand is copied to the left operand.
Should return a constant reference or a constant value to prevent a second assignment.
Assignment operator for rational as a method:
●
●
const rational & operator=( const rational & r ){
top = r.top;bottom return
= r.bottom;*this;
}
Assignment Operator cont.
• The assignment operator will be provided by the compiler if the programmer doesn't write it
• The compiler version just copies the data members
• If the class has pointers to other values that should be copied, the programmer should write the assignment
• Common mistakes:
• Not returning a value
• Not handling self-assignment
• Simply copying pointers rather than making copies of the heap-resident values the object has pointers to