Object Oriented Design, Why?

An Object Oriented system software is usually hierarchical, highly cohesive and loosely coupled.
The key advantages of this approach are following: 1. It is more closer to human cognition 2. Reusable code is generated through provision of generalization and inheritance. 3. Reduced effort for maintenance and enhancements resulting from low coupling, higher cohesion and inheritance.

The Object and the Class

An object is an individual unit which exists in the system.
It has: 1. Identity 2. State 3. Behavior

State

The state is a collection of properties the object possesses along with their values.
These are encapsulated within the object.

Behavior

These are the actions performed by the object in response to a stimuli from the clients through the methods the object possesses.

Structure

A class is a template for objects of similar type.
It has a public view (from outside) called its interface.
It also has a private view (from inside) called its implementation.

Classification

This stage is the most critical one.
When we look at the problem from different angles, we can classify it into different categories.

The Object Model

The elements of object-oriented-design are collectively called object-model.
It consists of principles of: 1. Abstraction 2. Encapsulation 3. Hierarchy of Inheritance

Abstraction

This focuses on the external interface the class provides and it is separated from how it is internally implemented.

Encapsulation

This focuses on the fact that external entities cannot look into the implementation of the class.

Relationship among Objects

There are patterns of interaction between objects.
These relationships are: 1. Inheritance 2. Association 3. Aggregation

Inheritance

inheritance defines a “kind of” hierarchy among classes.
There are 2 classes involved. 1. super class: From which the general properties are derived 2. sub class: The derived class which can either extend the functionality of restriction the functionality (specialization).

Association

association defines a relationship in which object A “uses” object B.

Aggregation

aggregation defines a relationship in which object A is “contained by” object B.

Aggregation and Association

Basic Differences

The objects in aggregation exist in a tighter coupling.
They exist in a way that they cannot exist independently.
However, if they are usually independent, then they are in association relationship.

Object Creation and Life Time

In terms of aggregation, the objects have to be created at same time and have to be destroyed at same time as well.
Whereas, in case of association, objects have to be alive only when a message needs to be communicated between them.

Coupling and Linkages

aggregation has strong coupling as compared to association.
The links between the objects in aggregation are permanent meanwhile for association, they only exist as long as the services are required from different objects.
After the services have been done, the links can be discontinued.

Ownership and Visibility

For aggregation, since it is a container - contained relationship, the container has encapsulated its parts which cannot be accessed outside of it.
For association, the objects can be directly accessed by other objects.

Database Persistence

For aggregation, the parts have to be stored as well but for association, only reference to the objects will be enough.