Lecture01 object oriented-programming

Tip to Save Printer Ink
 If you prefer to save your printer ink, follow the
steps below to print the lectures in black text on
white background.
1. In PowerPoint, select File > Print.
2. In the Settings section, locate an item with
value "Color", change it to "Pure Black and
While".

Introduction to
Object-Oriented
Programming

Lecture 1

TCP1201 OOPDS

Learning Objectives
 To understand object & class
 To understand abstraction in C++
 To understand encapsulation in C++
 To categorize similar objects
 To categorize objects by Composition
 To understand object behaviors
 To construct UML Class Diagram
 To differentiate between Procedural
Programming and Object-Oriented
Programming

What is an “object”?
An object is a computer representation of
real-world person, place, event or anything in
the problem that we are solving.

What is an “object”?
 An object consists of:
– attributes/data/states/fields/variables (typically
noun), e.g. name, date, balance, size, mark, etc.
– behaviors/procedures/methods/operations/functions
(typically verb), e.g. eat, drive, set, get, push, etc.
 We refer to a group of similar objects with the
same attributes and behaviors as a class.
 Same attribute does not mean same value for
the attribute, e.g. 2 students have attribute
'name', but the name of each student can be
different.

Object and Class
Class: Student.
Attributes: name, student_ID.
Behavior/method: do_homeworks().
Objects of Student class: steve, victor.

Object and Class
Steve and victor share the same attributes (name &
student_id), but each object has its own value for the
attributes.

Object and Class
 We refer to a group of similar objects with the
same attributes and behaviors as a class.
 Thus, we can define a class as:
– “a set of objects comprised of the same
attributes and behaviors”.
 Thus, we can define object as “a particular
instance of a class”.

Object-Oriented Programming
 The idea is to design a program to solve a
problem by combining both the data and
procedures (member functions) that operate on
that data into a singe unit called object.
Object2
Data

Object1 Member Object3
Function
Data Data

Member Member
Function Function
Messages

 Popular OOP languages: C++, Java, C#, VB, etc.
 C++ in TCP1201 focuses on OOP.

Steps to Use OOP
1. Identify the objects in the problem.
2. Identify the data/attributes and
operations/methods in each object.
3. Determine how the objects interact with one
another (messaging).
Object2
Data

Object1 Member Object3
Function
Data Data

Member Member
Function Function
Messages

4 Principles of OOP
1. Abstraction – identify the properties that are
important to the user in the problem, and
create representations that are similar to its
original meaning.
2. Encapsulation – combine data and its
operations in a single unit, and hide the
implementation from user.
3. Inheritance – create new classes from existing
classes (Lecture 3).
4. Polymorphism – use the same expression to
denote different operations (Lecture 4).

OOP Principle – Abstraction
 Abstraction is the process of identifying
important logical properties (object, attributes,
methods) that simplifies the modeling and
working of the problem.
 If you are designing a mobile phone, example
properties that are important to the user are the
screen, keypad, UI, cover, etc., but not the
internal working of the processor, how the
screen is actually rendered, etc.

OOP in Practice – Abstraction
 Assume that we are developing simple system to
keep track of subjects registered by student. It
won't be difficult to identify the following Student
class, its attributes and behaviors/methods.
– We keep the class simple for the purpose of discussion.

Class
Student
Attributes
id
subjects
Methods
show_subjects
register_subject
withdraw_subject

OOP in Practice – Abstraction
 Based on what we collected and consider
appropriate data type, the following class definition
can be declared:
class Student {
int id;
vector<string> subjects;
void show_subjects();
void register_subject();
void withdraw_subject();
};

 The next step is to consider encapsulation – the
2nd OOP principle.

OOP Principle – Encapsulation
 Encapsulation is the idea that the internal
workings of an object can be hidden from the
outside world.
 Encapsulation is performed in 2 ways:
1. We encapsulate how attributes are
accessed via access privileges
(private, public, protected, friend).
2. We encapsulate how methods are
implemented by separating interface and
implementation into different files.

Access Privileges in C++
 Access privileges allow us to restrict the access
to an object’s members.
 4 types of access privileges in C++:
1. private
2. public
3. protected (Lecture 3 Inheritance)
4. friend (Lecture 5 Operator Overloading)

private and public Access
Privileges
private members public members
 Not accessible from  Accessible from
anywhere except for the anywhere including
object itself. outside the class.
 We generally declare  We generally declare
attributes as private. methods as public.
 By default, all members  By default, all members
of a C++ class are of a C++ struct are
declared as private if declared as public if
unspecified. unspecified.

OOP Principle – Encapsulation
 To maintain data integrity, attribute is usually set
hidden/private inside a class to prevent direct
modification from outside the class.
 To access or modify an object' attribute from
outside the class, we provide public get or set
methods.
 Get method – a method that returns the value of
an attribute but does not modify it.
 Set method – a method that modifies the value of
an attribute.

OOP in Practice – Encapsulation
 Since all our methods are meant to be used from
outside the class, we should declare them as
public.
class Student {
// private by default
int id;
public: // public from now on
int getId(); // get method for id
void setId (int id); // set method for id
};

 The next step is to consider where to place the
implementation.

Placing Implementation
 There are 2 ways to place the implementation
(method body) in C++.

Inside of class declaration Outside of class declaration
class Student { class Student {
int id; int id;
... ...
int getId() { int getId(); //prototype
return id; void setId (int id); //prototype
} ...
void setId (int id) { }; // End of class
this->id = id;
} int Student::getId() {
... return id;
}; // End of class }
void Student::setId (int id) {
this->id = id;
}

Placing Implementation Outside
of Class Declaration
 "::" is a scope resolution operator.
 "Student::" indicates that the function is a
method of Student class, not a global function.
class Student {
int id;
...
int getId(); //prototype
void setId (int id); //prototype
...
}; // End of class

int Student::getId() {
return id;
}
this->id = id;
}

The this Pointer
 We use this pointer to refer to a member of a
class. class Student {
int id;
...
...
}; // End of class
...
this->id = id;
}

id here is attribute id here is function
parameter, not attribute

The this Pointer
 We use this pointer to refer to a member of a
class.

class Student {
int id;
...
...
}; // End of class
...
id = id; // Wrong, "parameter = parameter".
// Attribute id is not updated.
this->id = id; // Correct, "attribute = parameter".
Student::id = id; // Correct, "attribute = parameter".
}

Maintaining Data Integrity with
Encapsulation
 By declaring attributes as private and providing
public set methods, we can prevent erroneous
data being entered into the object.
...
// Ensure id entered is within range.
if (id < 1000000000 || id > 9999999999) {
cout << "Error: Id out of range.n";
this->id = 0;
}
else
this->id = id;
}

Placing Interface and
Implementation in Separate Files
 To bring encapsulation to the next level, we
separate the interfaces and its implementations
into different files.
 For every class:
– Place its interface in a .hpp file – called header file, e.g.
Student.hpp.
– Place its implementation in a .cpp file – called source
file, e.g. Student.cpp.
– In every cpp that needs to refer to the class, "#include"
the hpp.

HPP Header File
 The use of #ifndef, #define & #endif is to
prevent multiple inclusion.
// Student.hpp header file
#ifndef STUDENT_HPP // Prevent multiple inclusion.
#define STUDENT_HPP
#include <iostream>
#include <string>
using namespace std;

class Student {
int id;
public:
int getId();
void setId (int id);
}; // End of class
#endif

CPP Source File
 Every cpp source file that needs to refer to the
class should "#include" the hpp.
// Student.cpp source file
#include "Student.hpp" // Use "", not <>.

int Student::getId() { return id; }
void Student::setId (int id) { this->id = id; }
void Student::show_subjects() { ... }
void Student::register_subject() { ... }
void Student::withdraw_subject() { ... }

// main.cpp source file
#include "Student.hpp" // Use "", not <>.

int main() {
Student s;
s.register_subject();
...
}

Why Encapsulation?
1. To maintain data integrity
– By limiting direct access to attributes, we prevent class
users from providing invalid data to the object.
2. To reduce the need for class users to worry
about the implementation
– By separating the interface from the implementation
(shorter class declaration), class users can focus on
using the class instead of being bothered by the
implementation of the class.
3. To improve program maintenance
– Separating the interface from the implementation
enables us to change the implementation without the
class users ever being aware (provided that the
interface remains the same).

Categorizing Objects
 There might be many objects.
 2 common ways to categorize objects are:
1. Categorize similar objects that have the same
attributes and behaviors, e.g. the student
example.
– We have covered this one up until now.
2. Categorize objects by composition, that is
when an attribute of a class is a class by
itself, e.g. a faculty has many students.

Categorizing by Composition
 When categorizing objects by composition, an attribute of
a class is a class by itself.
 Is also referred to as a “has-a” relationship.
 Example 1: A faculty has many students (both faculty and
student are classes).
 Example 2: Typical corporate organization:
– 3 classes can be identified: Dept, Staff, Data.
– A Dept has Staff and Data.

Personnel Data Personnel Dept.

Sales Dept. Personnel Finance Dept.
Staff
Sales Data Finance Data

Sales Messages Finance
Staff Staff

Categorizing by Composition
 Sample code for the composition involving faculty
and student.

class Student { // Student is a class.
...
}; // End of class

class Faculty { // Faculty is a class.
string name;
vector<Student> students1; // Composition, attribute is
// a class by itself.
//vector<Student*> students2; // Also composition.
//Student students3[10]; // Also composition.
//Student* students3; // Also composition.
...
};

UML Class Diagram
 UML is a formal notation to describe models
(representations of things/objects) in sofware
development.
 Class Diagram is one type of many diagrams
in UML.
 Class Diagram contains 2 elements:
– Classes represent objects with common
attributes, operations, and associations.
– Associations represent relationships that relate
two or more classes.
32

Class Diagram

 Must have 3 sections.
 Class name is placed at the top Student
section.
-id:int
 Attributes are placed at the middle -subjects:string[*]
section.
+getId():int
 Behaviors are placed at the bottom +setId(id:int):void
section. +show_subjects():void
+register_subject():void
 '-' denotes private access privilege. +withdraw_subject():void
 '+' denotes public access privilege.
 '[*]' denotes "many" (array, vector).

Class Diagram
Faculty Student
Association
-name:string -id: int
-students:Student[*] -subjects:string[*]
1 1..n
+getId():int
+setId(id:int):void
+intake()
'n' means many +show_subjects():void
+register_subject():void
+withdraw_subject():void

 Associations shows the relationship between
instances of classes, e.g. a faculty has one or more
students, a student belongs to exactly one faculty
only.

Revisiting Procedural Programming
 C++ in TCP1101 is taught as a procedural programming
language.
 In procedural programming, the idea is to design a program
to solve a problem by concentrating on the procedures first
and data second. This approach is known as top-down
design.
 Procedures and data are 2 separate units that relate
primarily via function parameter.

Main Program
Data

Function1 Function2 Function3

Problems with Procedural Programming
 Unrestricted access to global data struct Student {
and procedures. int id;
 Poor modeling of the real world };
(data and procedures are
void register_subject
separated). (Student& a)
 Not the way that humans naturally { ... }
think about a situation.
void withdraw_subject
 Poor code reusability. (Student& s)
{ ... }

int main() {
Student s;
register_subject (s);
withdraw_subject (s);
}

Why Object-Oriented Programming?

 Solve the problems of procedural programming.
 Restrict access to data and procedures (via
encapsulation).
 Better modeling of real world objects (via
abstraction).
– Data and procedures are combined in a single unit.
– Easier to understand, correct, and modify.
 Better code reusability – existing objects can be
reused to create new objects via inheritance (Lecture
3).
 More useful for development of large and complex
systems.

Converting Procedural Program
to OOP
1. Identify the variables and the global functions
that use the variables as parameters, create a
class and include the variables and the global
functions as class members.
2. Make all attributes private. Const attributes can
opt for public.
3. For methods that use the class as
parameter(s), remove ONE such parameter.
Then update the method body to refer to the
member instead of the removed parameter.

// Procedural version Point readPoint() {
#include <iostream> Point p;
#include <string> cout << "Enter point x y : ";
#include <cmath> cin >> p.x >> p.y;
using namespace std; return p;
struct Point { }
int x, y; // public by default
}; void printPoint (Point p) {
cout << "(x = " << p.x
Point readPoint(); << ", y = "<< p.y
void printPoint (Point); << ")" << endl;
double distPoint (Point, Point); }

int main() { double distPoint (Point p, Point q)
Point p1 = readPoint(); {
cout << "p1 = "; double dx = p.x - q.x;
printPoint (p1); double dy = p.y - q.y;
cout << endl; double dsquared = dx*dx + dy*dy;
Point p2 = readPoint(); return sqrt (dsquared);
cout << "p2 = "; }
printPoint (p2);
cout << endl << endl;
double d = distPoint (p1, p2);
cout << "Distance p1 to p2 = "
<< d << endl;
}

// OOP version double Point::distPoint (Point q) {
#include <iostream> double dx = x - q.x;
#include <string> double dy = y - q.y;
#include <cmath> double dsquared = dx*dx + dy*dy;
using namespace std; return sqrt(dsquared);
}
class Point {
int x, y; // private by default int main() {
public: Point p1, p2;
void readPoint();
void printPoint(); p1.readPoint();
double distPoint (Point q); cout << "p1 = ";
}; p1.printPoint();
cout << endl;
void Point::readPoint() { p2.readPoint();
cout << "Enter point x y : "; cout << "p2 = ";
cin >> x >> y; p2.printPoint();
} cout << endl;
void Point::printPoint() {
cout << "(x = " << x double d = p1.distPoint (p2);
<< ", y = " << y cout << "Distance p1 to p2 = "
<< ")" << endl; << d << endl;
} }

Lecture01 object oriented-programming

Recommended

More Related Content

What's hot (20)

Viewers also liked (17)

Similar to Lecture01 object oriented-programming (20)

More from Hariz Mustafa (20)

Lecture01 object oriented-programming