Basics OO-programming can be defined as a combination of Abstract Data Types (ADTs) with Inheritance and Dynamic Binding. Encapsulation, inheritance and polymorphism
Each practice handles a different aspect of system composition:
Encapsulation can be seen as a 2D component - public and private interface
Inheritance adds an additional dimension - the ADT picks up the characteristics of another component.
Polymorphism adds to inheritance - postponing implementation decisions until later (perhaps even run-time).
Basics Recall that inheritance is a means of
specifying hierarchical relationships between types
C++ classes can inherit both data and function members from other (parent) classes
Terminology: "the child (derived or subclass) type inherits (or is derived from) the parent (base or superclass) type."
The derived type is just the base type plus:
Added specializations Change implementation
without changing the base class interface
Added Generalizations
/Extensions new operations and/or
data
What a derived class inherits Every data member defined in the parent
class (although such members may not always be accessible in the derived class!)
Every ordinary member function of the parent class (although such members may not always be accessible in the derived class!)
What a derived class doesn't inherit
The base class's constructors and destructor
The base class's assignment operator The base class's friends
What a derived class can add New data members New member functions (also overwrite existing
ones)
New constructors and destructor New friends
When a derived-class object is created & destroyed
Space is allocated (on the stack or the heap) for the full object (that is, enough space to store the data members inherited from the base class plus the data members defined in the derived class itself)
The base class's constructor is called to initialize the data members inherited from the base class
The derived class's constructor is then called to initialize the data members added in the derived class
The derived-class object is then usable When the object is destroyed (goes out of scope or is deleted)
the derived class's destructor is called on the object first Then the base class's destructor is called on the object Finally the allocated space for the full object is reclaimed
Inheritance in C++ The class header is modified to allow a
derivation list consisting of base classes (C++ allows multiple inheritance)
class Foo { };
class Bar : public Foo { };
class More : public Foo, public Bar { };
Key Properties of C++ Inheritance
The “is-a” relationship is maintained A pointer to the base type may point to a derived
type object The above relationship combined with
dynamic binding - promotes a type-secure, polymorphic style of programming The programmer need not know the actual type of
an object at compile-time (dynamic-binding)
Simple Base Class (Screen class) Derived from Screen (Window class)Derived from Screen (Window class)
Derived from Window (Menu class)
Inheritance HierarchyInheritance Hierarchy
Multiple Levels of Derivation
A pointer to a derived class can be assigned to a pointer of any of its public base classes without requiring an explicit cast: Menu m;
Window &w = m; Screen *ps1 = &w; Screen *ps2 = &m;
Public vs private inheritance The public keyword in the inheritance syntax
means that publicly accessible members inherited from the base class stay publicly accessible in the derived class
But sometimes its preferable to inherit the public members of a parent in such a way that they become private in the child
Private inheritance Public members of base class become
private members of derived class Public and protected members are only
available to derived-class member functions - not to a derived object.
Protected inheritance private members of the base class are not
accessible in the derived class (to preserve encapsulation)
Protected qualification allows encapsulated data members which are not publicly accessible to be accessible by derived classes
("Protected" members are not accessible from outside the class, except in derived classes)
Initializer lists Derived class constructor automatically
calls the "no argument" base class constructor(s)
But what if a base class has a non-void constructor which needs to be called? specify which base-class constructor gets
called, as part of the definition of the derived-class constructor
Initializer lists
class Coefficient{public: Coefficient(void) { myValue = 0; myAccesses = 0; }
Coefficient(double initval) { myValue = initval; myAccesses =
0; }
// ETC. };
class StatusCoefficient : public Coefficient{public: StatusCoefficient(void) { myStatus = OverStatus; }
StatusCoefficient(double initval, Status initStatus) : Coefficient(initval) { myStatus = initStatus; }
// ETC. };
class StatusCoefficient : public Coefficient{public: StatusCoefficient(void) { myStatus = OverStatus; }
StatusCoefficient(double initval, Status initStatus) : Coefficient(initval) { myStatus = initStatus; }
// ETC. };
Using Initializer lists to initialize values
Initializer lists can also be used to initialize class members with specific value (instead of assigning to them in the body of the constructor)
class First{ public: First() : _foo( "initialize foo first" ), _bar( "then bar" ) { }
private: string _foo; string _bar;};
Polymorphism Meaning: some code or operations or objects
behave differently in different contextsExample: the + operation behaves differently depending on the
operands (overloading)
member function with the same name can me implemented in different classes
Use :: to refer to a function in the base class with the same name - baseCL::function()
appropriate version of function is determined at runtime – dynamically
virtual member functions used to implement polymorphism
Polymorphism Dynamic binding:
At runtime the C++ program selects which member function to execute - if the function is virtual and exists at the different levels within a inheritance hierarchy
Contrasts with Static Binding
emp.displayEmployeeInfo();