Wednesday, June 30, 2010

Object Oriented Programming Part 2

In earlier post we have already discovered Object oriented programming, orientation, its fundamentals and features. Here we will take up next remaining questions.

These are the questions whose answer are we seeking here in these blogs…

ü  What is Object Oriented Programming?
ü  What are the fundamental concepts/Features of OOP?
ü  What is Object Based Programming?
ü  What is Procedural programming?
ü  What are the features of Procedural programming?
ü  Compare basic features of the OOP and PP?

First 2 questions have already been answered in the first blog. Here I will start with the third question.

What is Object Based Programming?

Object based programming is associated with objects, objects which consist fields and methods similar to OOP. In object based programming few features(Abstraction, Polymorphism, Encapsulations, Inheritance) of OOP are included. It is not wrong to say, OBP is a subset of OOP where subset has all the features except few.

As I am not concentrating more on this paradigm so I am not going to invest much of your time here.

What is Procedural programming?

In Procedural programming instead of data; structure of the program is the determining feature. Here programmer concentrate on structure of the program instead the behaviour and state of the object.

Edsger Dijkstra's structured programming, where the logic of a program is a structure composed of similar sub-structures in a limited number of ways. This reduces understanding a program to understanding each structure on its own, and in relation to that containing it, a useful separation of concerns.

At low level. structure programming composed of simple, hierarchical program flow structures. These are Sequence(Ordered sequence of statements), selection (Decision making conditions) and repetition(repetitive statements).

In current scenario Fortron, Basic are using structured programming. Branch of structures programming used by Procedural programming which is implemented in C.

Procedural programming is a branch of structural programming where procedure are created on the basis of structures. Procedures are routines, subroutines, functions and their sequence.

What are the features of Procedural programming?

As stated above the poles of the procedural programming are methods, modules, functional calls, variables whereas procedural programming does not have features like abstraction, encapsulation, inheritance and polymorphism.

Methods: These can be understand as the basic feature. We create the structure of the program, divide it into sequence, selection, repetition and accordingly divide one routine to different subroutines and achieve the desired functionality.

Modules: n number of subroutines creates a module where all subroutines are scope specific.

functional calls: As oops uses messages pass between object here it is achieved by functional calls.

Variables: As PP does not belong to object, so each routine has its own variables associated with function’s scope.

Compare basic features of the OOP and PP?

Till now it is understood that Procedure based programming is much closer to deal with structure on the other side OOP only deals with objects, and blueprint of the objects.

Pros and Cons of PP: Procedure based programs tends to run quickly and use system resources efficiently. Whereas the drawback is that it does not fir gracefully with certain type of problems, such as in this paradigm programmer is forced to use structure approach to reach to the conclusion.

Pros and Cons of OOP: OOP is very similar to the perception of world, so it is less difficult to visualize things there. Code reuse is another big advantage of OOPS which is one among the hard hitting drawbacks of PP. OOP is runtime inefficient. It needs more memory and resource to run the program.

If we summarize the whole story, it is evident that none of the paradigm is inferior or superior. It is just our requirements which makes one preferred to us. To finalize the paradigm there are few factors which should be taken into account. First we should look for the data structure of the problem and accordingly should decide the paradigm. Benchmarking is another aspect which effects mostly testing efforts a lot. These 2 factors could highlight the preferences of selecting the programming structure.

I will take a break now. Again I am open to discussions, please post me your queries in comments.

Disclaimer: All the material here is the result of research and understanding of other material. If there is some content which is either incorrect or inappropriate to the context of the article I am always open for discussion.

Sunday, June 27, 2010

Marker Interface



This post is not in the continuation of the last post. But this week I was searching about the marker interfaces. So I thought of documenting my learning about Marker Interfaces.
As it is known, Java interface defines contract between real world and objects, which implies when a class in implementing a java interface it has to implement all the methods defined in the interface.
What is a marker interface?
Marker interface is an interface which does not have any field and method associated with it.
/** 
* This interface is a marker interface. 
* For JVM it is just another interface without any method and field.
*/
public interface MarkerInterface {
}

Here as in the example above there is an interface which do not define any method and field. But this interface will compile without any issues.

/**

* MarkedClass is implementing MarkerInterface, 
* which means class MarkedClass is associated with MarkerInterface.
* This class will compile without any issues.
* */
public class MarkedClass implements MarkerInterface{
}
As mentioned in the example MarkedClass has been tagged as a MarkerInterface.
To compile this example you can copy the interface’s code to MarkerInterface.java file and copy class’s code to MarkedClass.java file.


What is the actual use of marker interfaces? Why will I prefer marker interface instead of an interface which has some methods?
Purpose of marker interfaces is to create a special type of Object, in such a way to restrict the classes that can be cast to it without placing ANY restrictions on the exported methods that must be provided. In another way these are used to provide metadata information to the classes.

To explain it further we can use an example…
There are quite a few IT organizations which primarily concentrate on product whereas there services are limited to their products and beside these organizations there are organizations which concentrate on services of the others products. These are classified as Product Organization and service organization.
A product company may sell their products either to Customer directly or to Service Company.  On the basis of the type of customer a product company wants to provide different set of contracts with the company. In general a product company has millions of customers and few service companies associated with it. So to distinguish between these two types of organization we mark Service companies with the marker interface.


This organization interface defines the contract.
/**

* This interface will have common utility methods that will be used across all the 
* customer as well as service organizations.  
*/ 
public interface Organization {
      String[] getProduct();     
      void setProduct(String[] products);
      //Many other such methods.      
}


This ServiceOrganization interface is a marker interface which will state the organization as a service organization.
/**
This interface is to mark an Organization as a Service Organization.


**/
public interface ServiceOrganization extends Organization{
}

Here you can see customer is implementing Organization, i.e. it is stating this Customer is not a special customer to me.

/**


* Customer is implementing Normal organization.
*/
 
public class Customer implements Organization {
@Override
public String[] getProduct() {
return null;
}
@Override
public void setProduct(String[] products) {
}
}
Now ServiceOrganization interface has been implemented by the Service customer class which will distinguish service customer from normal customer.
/**

* Service Customer has been tagged as service Organization but it does not need 
* to implement something other than method which will be defined by a
* common customer. 
*/
public class ServiceCustomer implements ServiceOrganization {
@Override
public String[] getProduct() {
return null;
}
@Override
public void setProduct(String[] products) {
}
}

As both types of customers have been separated with the use of marking, now it is not difficult to judge the customers.
Does Java have by default marker interface? If Yes, what all marker interfaces are there and what are their uses?
Yes, Java has few marker interfaces. Those are as follows:

java.util.RandomAccess
java.io.Serializable
java.rmi.Remote
java.util.EventListner
java.lang.Clonable


java.io.Serializable
When a class is implementing this interface this internally suggests to JVM that object of this class may be serialized for persistence purposes.
Source code for the interface is as mentioned below.
public interface Serializable {
}


So it is clear, writeObject and readObject methods are not part of the interface, but rather part of the serialization process, and help determine how to serialize an object.
java.lang.Clonable
Clonable interface is necessary to use clone method associated with each object. If clone is used without implementing clonable interface the code will throw a runtime exception.
java.util.EventListner
This interface also does not have any specific method but it is necessary for a class to implement the interface to achieve the functionality of listening the events.
java.util.RandomAccess
Implementing this interface suggests data accessing in the list is fast. So it is just marking list as fast random access.
 

java.rmi.Remote
The remote interface serves to identify interfaces whose methods may be invoked from a non-local virtual machine. Any object that is a remote object must directly or indirectly implement this interface. Only those methods specified in a "remote interface", an interface that extends java.rmi.Remote are available remotely.
Here again marker interface has made sure if something has to access remotely then it should be marked with remote interface.
List of marker interfaces is long but the interfaces mentioned above are the widely known interfaces.

Like me You must be wondering, this all is fine but still there is one important question
.

How does JVM deal with specific functionality of each marker interfaces?
I could not find something very specific to this question. But as most of the experts suggests Java does not do anything specific to identify the marker interface. instanceOf operator is used to identify whether the functionality is supported for the object or not.

If you look at the code of ObjectOutputStream.java, it is quite clear when serialization is supported by java.
//Piece of code of method - writeObject0(Object obj, boolean unshared){

.............

} else if (cl.isArray()) {
writeArray(obj, desc, unshared);
} else if (obj instanceof Serializable) {
writeOrdinaryObject(obj, desc, unshared);
} else {
throw new NotSerializableException(cl.getName());
}
.................

Now it is clear what are marker interfaces and at what instances do we use them. Java 5 release sun has introduced annotations, a way which will replace marker interface in future designs. Annotations are more descriptive and powerful mechanism to tag a class. I will talk about annotations in coming blogs..

Disclaimer: All the material here is the result of research and understanding of other material. If there is some content which is either incorrect or inappropriate to the context of the article I am always open for discussion.

Thursday, June 17, 2010

Object Oriented Programming

Java is called Object Oriented Programming where as C called Procedure oriented programming. There are always confusion and enquiry. It is always a question why did new paradigm evolve when there was already existing paradigm which did not have much hassles.

There are few questions which always click to mind. Few are as following

ü  What is Object Oriented Programming?
ü  What are the fundamental concepts/Features of OOP?
ü  What is Object Based Programming?
ü  What is structured programming?
ü  What are the features of structured programming?
ü  Compare OOP and SP?

These are the few questions that have been troubling me for past few years. Now when I am four years senior in IT then it feels best to start looking for answers and understand them.

What is Object Oriented Programming?

Object Oriented Programming known as OOP is a programming paradigm which relies on data. It is a paradigm which represents data and its use. So basically an object is a representation of data structure which consist fields and methods, which provides a way a interaction between each other.

OOP concepts were mainly introduced for maintainability of code. Earlier people were having function oriented approach, which was becoming difficult to maintain in the long term. To get rid of this tedious process orientation moved to data and communication of data, which was introduced in OOP. OOP is a collection of interacting object having communicating facilities between objects. The technology focuses on data rather than processes/procedure/functions.

Simula was the first language which introduced the concepts of OOP, it was followed by LISP. Well known language to us is JAVA.


public class DemoClass {

                private int test = 0;

                public int getTest() {
                                return test;
                }

                public void setTest(int test) {
                                this.test = test;
                }
               
}

Here class DemoClass has a variable which has an integer as field who is deterministic factor to define the state of the object of the class (We will discuss private and public modifier later). getTest method will provide the state of the variable whereas setTest method will change the existing state of the test variable.

This class uses oops fundamental where it is designed around data instead of procedures/functions.


class InteractWithDemo{
               
                DemoClass testDemo = new DemoClass();
                public void useTestDemo(){
                                testDemo.setTest(10);
                                testDemo.getTest();
                }
               
}


PS: Please ignore the syntax if you are not aware.

InteractWithDemo class works as interacting class with the class created above. It creates an object of DemoClass and change the state of the DemoClass's object. This is the way objects interact with each other.

What are the fundamental concepts and Features of OOP?

Foundation of every entity is kept on fundamental concepts and it is appreciated on the basis of features. OOP also have few fundamental concepts and few others important and fascinating features.

Following are the fundamental concepts and Features of OOP

ü  Object
ü  Class
ü  Instance
ü  Method
ü  Interface
ü  Package
ü  Message Passing
ü  Abstraction
ü  Encapsulation
ü  Polymorphism
ü  Inheritance

These are the fundamental concepts along with that features of OOPS. Let’s discuss them in details further.

Object

Object is any entity who has characteristics, behaviours and state. In our day to day we see lots of object. Every human in itself is an object. How? Each individual has few characteristics which can be used to uniquely identify him. E.g. A person has name, DOB, parents name which are used to uniquely identity him. The same guy will have some habits. His state can be his age, marital status, job status, property balance and many more things. So a human being is an object. Similarly we have software concepts.
There are few advantage listed to have objects:
ü  Modularity
ü  Information Hiding
ü  Code Re-Use
ü  Pluggability and Debugging Use Case.

Class

Class is the blue print of the Object. Class is to hold the state and the behaviour of the entity. In the example given above, here to define a state of a human being following is the information that may suffice to define a unique person
  • Name
  • DateOfBirth
  • Father's name
  • Mothers' Name

These 4 attribute can tell the characteristics of the human along with that we can more variable to define the state and accordingly there can be few methods which define characteristics.

Instance

Instance is an occurrence or copy of an object. Multiple instances, of a class share, same set of attributes whereas value of the attributes may differ for each instance.

Methods

Method is a subroutine which is either associated to a class or to an instance of object. Method is a sequence of the statement to perform an action on the instances of objects or classes. Method may have inputs which are either processed or used to modify the state of the object. Few methods return the required datatype.

In case of OOP, methods provide the mechanism for accessing and manipulating the encapsulated data of the object.

In OOP these are type of methods:
  • Static methods - Mainly associated with the class.
  • Instance Methods - Affiliated with the instance of the class i.e. object.
  • Constructor - Cannot be called explicitly, in general these methods are used for initialization purpose.
  • Abstract Methods - These methods do not body defined for it (Implementation of the method is not given) i.e. sequence of instructions are defined by some other class.
  • Destructor - Cannot be called explicitly, in general these methods are used for object destroy purpose.

Interface

Best example to define interfaces will be a TV Set with a remote functionality. E.g. let’s say Videocon has different types of TV, where display type, screen size are different but company wants to create a generic remote for all. So Generic Remote is an interface to interact with TV object of various types of TV classes.

Here interface remote will be implemented by each TV class to perform actions guided by user.
Interface defines the behaviour to interact with outside the world.


                interface Remote{
                                void increaseVolume();
                                void reduceVolume();
                                void nextChennal();
                                void previousChannel();
                                void changeChennalTo(int);
                }


As defined above interface remote has defined five methods. Implementations of these methods have not been provided in the interface. Now when a class implements an interface it will implement all methods.  Interfaces form a contract between the class and the outside world, whereas this contract is enforced at build time by the compiler.

Package
Package is similar to the folders in windows, which are used to manage the data better and efficient. Packages are the namespace which provide better overview and understanding of the classes and their utilization. It is recommended, while coding related objects should belong to the same package.

Message passing

Message passing is a method of communication where sender sends messages to one or more than one recipients. These forms of message are method invocation, signal and packets.
In case of OOP invocating a method from object, class or other object is called message passing. The order in which message have been passed in their reversal order method responses are received by the sender.

Abstraction

An "abstraction" (noun) is a concept that acts as super-categorical noun for all subordinate concepts, and connects any related concepts as a group, field, or category.
In OOP abstraction is used to provide bare minimum information at the user level. The information which matters the user most is provided. Abstraction asks to make sure to provide minimal information. As user does not internals of the class, object, database and design it is not even altered with the changes of inside the objects.
The best thing of abstract is that this decouples the user of the object and its implementation. So now object is easy to understand and maintain also. As if there is any change in the process of some operation. You just need to change the inner details of a method, which have no impact on the client of class.

Encapsulation

Encapsulation is data hiding or reducing the access to others. Here it is made sure that every field of the class should not be accessible from the outside world. They should be taken care by methods and their accessibility should not go beyond the boundaries of the class. Encapsulation is a tool which provides the way to implement abstraction for a class.

Inheritance

Inheritance is well known terminology used in programming fundamentals. Inheritance is child’s rights on his father’s property. i.e. anything that belongs to father will belong to child until unless father explicitly mentions his wishes. Same inheritance is applied in Java, if class is child of a class it means it will be able to use all the members other than private members.

Usage of inheritance is very important in reusability of the code, which reduces the efforts of rewriting the same code again.

Polymorphism

It is a paradigm where values of two different data type are handled by uniform interface. In other way 2 methods or fields name can co exist with different set of parameters?
Polymorphism is better known as providing similar functionality to different set of parameters. In java achieving polymorphism is possible in 2 ways. Overriding and overloading which will be analyzed in the further discussions.
 

Disclaimer: All the material here is the result of research and understanding of other material. If there is some content which is either incorrect or inappropriate to the context of the article I am always open for discussion.