It's based on concept of calling procedure/funtions
A technique that consists of well-structured & separated modules
Follows logic of a program - it requires more discipline at the design and logical structuring stage
In procedural programming, we needed to only write an application program (the main function and some other functions) to use objects of built-in types
C++ has many predefined (built-in) types such as int, double, char, & bool
Example; C
OOP brings together data & functions
Supports encapsulation, abstraction, inheritance, polymorphism
Includes data hiding feature
This model is based on real life entities
Example; C++, Java, Python, C#
Procedural/Structured | Object Oriented |
|---|---|
| Program section are called functions | Program section are called objects |
| Follows top down approach | Follows bottom up approach |
| No access specifier | Access specifiers like private, public, protected |
| Not easy to add new data & function is not easy | Easy to add new data & function |
| Less secure - no data hiding | Secure - data hiding |
| No overloading | Allows overloading |
| Function is more important than data | Data is more important than function |
| Based on unreal world | Based on real world |
A type is a concept from which an instance is created or is an abstraction
An instance of that type is a concrete entity
The relationship between a type & its instances is one-to-many
An attribute is a characteristic of an instance that we are interested in
For example, an employee is an instance, and the attributes are name, address, position, and salary
A behavior is an operation that we assume an instance can perform on itself
In C++, a user-defined type can be created using a construct named class
An object is an instance of a class. This means that we use type and class
In OOP, attributes & behaviors of an object are represented as data members & member functions
Types of classes; Concrete, abstract & mixin
A data member of an object is a variable whose value represents an attribute
Data members in a class must not depend on each other
In OOP, a function is an entity that can do something
A member function in OOP is a function that simulates one of the behaviors of an object
In C++ OOP, Programmers can create new types, using classes
To write object-oriented programs, we need to create a class, as a type, and then instantiate objects as instances of that type
OOP requires 3 sections;
- declare the data members and member functions - define all member functions - instantiate objects and apply the member function to those objectsC++ program in object-oriented paradigm
An example; Lets create two circle objects
Access modifier determines how a class can be accessed
When there is no access modifier for a member, it is private by default
A private data member is visible, but it cannot be accessed except through a member function
The data members of a class are normally set to private
To operate on the data members, the application must use member functions, which means that the declaration of member functions usually must be set to public
The instance member functions of a class are normally set to public
C language uses structs
A struct in the C++ language is actually a class with one difference;
We can always create a class to simulate a struct as shown below
struct{string first; char middle; string last; };
class{public: string first; char middle; string last; };
In OOP, an instance is an object that encapsulates the data members defined in the class definition
A constructor is a special member function that creates and initializes an object
A destructor is a special member function that cleans and destroys an object
A constructor has 2 characteristics;
Constructor overload - is having more than one constructors in a class
Types of constructors in a class
Bitbucket: Constructor Declaration
Supports overload, which means that we can have several parameter constructors each with a different signature
The advantage of the parameter constructor is that we can initialize the data members of each object with a specific value
For example; the radius of one circle can be initialized to 3.1, another one to 4.6
Bitbucket: Paramater Constructor
A constructor with no parameters
It is used to create objects with each data member for all objects set to some literal values or default values
No overload because it has no parameter list and therefore is not eligible for overloading
Bitbucket: Default Constructor
Copies the data member values of the given object to the new object just created
Has only one parameter that receives source object by reference
The const modifier guarantees that the pass-by-reference cannot change the source object
No overload because the parameter list is fixed and we cannot have an alternative form
A destructor has two special characteristics;
(not even void) because it returns nothing.A destructor is guaranteed to be automatically called and executed by the system when the object instantiated from the class goes out of scope
No overloads
A destructor is a special-purpose member function with no parameter and is designed to clean up and recycle an object
Bitbucket: destructor Definition
Bitbucket: Constructors & Destructors
When designing a class, we can have two groups of members;
An instance data members is normally private and can be accessed only through instance member functions
An instance member function of a class must be public to be accessed from outside the class
Member selector operators;
A class type can also have two types of members;
A static data member is a data member that belongs to all instances; it also belongs to the class itself
A static member function enables access to the corresponding static data member when an instance wants to access it or when the application needs to access it
A static member function cannot be used to access instance data members because it has no this pointer parameter. i.e. A static member function has no host object because it is not associated with any instance
C++ as an OOP language, has normally 3 code sections;
Interface file contains the class definition i.e.,
It gives the general picture of the class to be used by the other files; i.e., it defines the type that is used by the other two files
The name of this file is normally the name of the class with an h extension, such as circle.h. The letter h designates it as a header file
Implementation file contains the definition of member functions
It is the code for all member function declarations given in the interface file
The name of this file is normally the name of the class with a cpp extension, such as circle.cpp, although the extension may vary in different C++ environments
Application file includes the main function that is used to instantiate objects and let each object perform operations on themselves
The application file must also have the extension cpp, but the name of the file is usually chosen by the user
The name we use is app.cpp, although the extension may vary in different C++ environments
After creating 3 separate files, we must compile them to create an executable file. In C++, the process is referred to as separate compilation
Queries about this Lesson, please send them to:
*References*
- C++ Programming - An Object-Oriented Approach, 2019
Behrouz, Richard et.al- Accelerated C++ - Practical Programming by Example, 2000
Andrew, Barbara et.al
Courtesy of …