Develop Reference

Immutable Matrix ADT [duplicate]

How to create immutable objects in Java?
Which objects should be called immutable?
If I have class with all static members is it immutable?

Below are the hard requirements of an immutable object.
Make the class final
make all members final, set them
explicitly, in a static block, or in the constructor
Make all members private
No Methods that modify state
Be extremely careful to limit access to mutable members(remember the field may be final but the object can still be mutable. ie private final Date imStillMutable). You should make defensive copies in these cases.
The reasoning behind making the class final is very subtle and often overlooked. If its not final people can freely extend your class, override public or protected behavior, add mutable properties, then supply their subclass as a substitute. By declaring the class final you can ensure this won't happen.
To see the problem in action consider the example below:
public class MyApp{
/**
* #param args
*/
public static void main(String[] args){
System.out.println("Hello World!");
OhNoMutable mutable = new OhNoMutable(1, 2);
ImSoImmutable immutable = mutable;
/*
* Ahhhh Prints out 3 just like I always wanted
* and I can rely on this super immutable class
* never changing. So its thread safe and perfect
*/
System.out.println(immutable.add());
/* Some sneak programmer changes a mutable field on the subclass */
mutable.field3=4;
/*
* Ahhh let me just print my immutable
* reference again because I can trust it
* so much.
*
*/
System.out.println(immutable.add());
/* Why is this buggy piece of crap printing 7 and not 3
It couldn't have changed its IMMUTABLE!!!!
*/
}
}
/* This class adheres to all the principles of
* good immutable classes. All the members are private final
* the add() method doesn't modify any state. This class is
* just a thing of beauty. Its only missing one thing
* I didn't declare the class final. Let the chaos ensue
*/
public class ImSoImmutable{
private final int field1;
private final int field2;
public ImSoImmutable(int field1, int field2){
this.field1 = field1;
this.field2 = field2;
}
public int add(){
return field1+field2;
}
}
/*
This class is the problem. The problem is the
overridden method add(). Because it uses a mutable
member it means that I can't guarantee that all instances
of ImSoImmutable are actually immutable.
*/
public class OhNoMutable extends ImSoImmutable{
public int field3 = 0;
public OhNoMutable(int field1, int field2){
super(field1, field2);
}
public int add(){
return super.add()+field3;
}
}
In practice it is very common to encounter the above problem in Dependency Injection environments. You are not explicitly instantiating things and the super class reference you are given may actually be a subclass.
The take away is that to make hard guarantees about immutability you have to mark the class as final. This is covered in depth in Joshua Bloch's Effective Java and referenced explicitly in the specification for the Java memory model.

Just don't add public mutator (setter) methods to the class.

Classes are not immutable, objects are.
Immutable means: my public visible state cannot change after initialization.
Fields do not have to be declared final, though it can help tremendously to ensure thread safety
If you class has only static members, then objects of this class are immutable, because you cannot change the state of that object ( you probably cannot create it either :) )

To make a class immutable in Java , you can keep note of the following points :
1. Do not provide setter methods to modify values of any of the instance variables of the class.
2. Declare the class as 'final' . This would prevent any other class from extending it and hence from overriding any method from it which could modify instance variable values.
3. Declare the instance variables as private and final.
4. You can also declare the constructor of the class as private and add a factory method to create an instance of the class when required.
These points should help!!

From oracle site, how to create immutable objects in Java.
Don't provide "setter" methods — methods that modify fields or objects referred to by fields.
Make all fields final and private.
Don't allow subclasses to override methods. The simplest way to do this is to declare the class as final. A more sophisticated approach is to make the constructor private and construct instances in factory methods.
If the instance fields include references to mutable objects, don't allow those objects to be changed:
I. Don't provide methods that modify the mutable objects.
II. Don't share references to the mutable objects. Never store references to external, mutable objects passed to the constructor; if necessary, create copies, and store references to the copies. Similarly, create copies of your internal mutable objects when necessary to avoid returning the originals in your methods.

An immutable object is an object that will not change its internal state after creation. They are very useful in multithreaded applications because they can be shared between threads without synchronization.
To create an immutable object you need to follow some simple rules:
1. Don't add any setter method
If you are building an immutable object its internal state will never change. Task of a setter method is to change the internal value of a field, so you can't add it.
2. Declare all fields final and private
A private field is not visible from outside the class so no manual changes can't be applied to it.
Declaring a field final will guarantee that if it references a primitive value the value will never change if it references an object the reference can't be changed. This is not enough to ensure that an object with only private final fields is not mutable.
3. If a field is a mutable object create defensive copies of it for
getter methods
We have seen before that defining a field final and private is not enough because it is possible to change its internal state. To solve this problem we need to create a defensive copy of that field and return that field every time it is requested.
4. If a mutable object passed to the constructor must be assigned to a
field create a defensive copy of it
The same problem happens if you hold a reference passed to the constructor because it is possible to change it. So holding a reference to an object passed to the constructor can create mutable objects. To solve this problem it is necessary to create a defensive copy of the parameter if they are mutable objects.
Note that if a field is a reference to an immutable object is not necessary to create defensive copies of it in the constructor and in the getter methods it is enough to define the field as final and private.
5. Don't allow subclasses to override methods
If a subclass override a method it can return the original value of a mutable field instead of a defensive copy of it.
To solve this problem it is possible to do one of the following:
Declare the immutable class as final so it can't be extended
Declare all methods of the immutable class final so they can't be overriden
Create a private constructor and a factory to create instances of the immutable class because a class with private constructors can't be extended
If you follow those simple rules you can freely share your immutable objects between threads because they are thread safe!
Below are few notable points:
Immutable objects do indeed make life simpler in many cases. They are especially applicable for value types, where objects don't have an identity so they can be easily replaced and they can make concurrent programming way safer and cleaner (most of the notoriously hard to find concurrency bugs are ultimately caused by mutable state shared between threads).
However, for large and/or complex objects, creating a new copy of the object for every single change can be very costly and/or tedious. And for objects with a distinct identity, changing an existing objects is much more simple and intuitive than creating a new, modified copy of it.
There are some things you simply can't do with immutable objects, like have bidirectional relationships. Once you set an association value on one object, it's identity changes. So, you set the new value on the other object and it changes as well. The problem is the first object's reference is no longer valid, because a new instance has been created to represent the object with the reference. Continuing this would just result in infinite regressions.
To implement a binary search tree, you have to return a new tree every time: Your new tree will have had to make a copy of each node that has been modified (the un-modified branches are shared). For your insert function this isn't too bad, but for me, things got fairly inefficient quickly when I started to work on delete and re-balance.
Hibernate and JPA essentially dictate that your system uses mutable objects, because the whole premise of them is that they detect and save changes to your data objects.
Depending on the language a compiler can make a bunch of optimizations when dealing with immutable data because it knows the data will never change. All sorts of stuff is skipped over, which gives you tremendous performance benefits.
If you look at other known JVM languages (Scala, Clojure), mutable objects are seen rarely in the code and that's why people start using them in scenarios where single threading is not enough.
There's no right or wrong, it just depends what you prefer. It just depends on your preference, and on what you want to achieve (and being able to easily use both approaches without alienating die-hard fans of one side or another is a holy grail some languages are seeking after).

Don't provide "setter" methods — methods that modify fields or
objects referred to by fields.
Make all fields final and private.
Don't allow subclasses to override methods. The simplest way to do this is to declare the class as final. A more sophisticated approach is to make the constructor private and construct instances in factory methods.
If the instance fields include references to mutable objects, don't allow those objects to be changed:
Don't provide methods that modify the mutable objects.
Don't share references to the mutable objects. Never store references to external, mutable objects passed to the constructor; if necessary, create copies, and store references to the copies. Similarly, create copies of your internal mutable objects when necessary to avoid returning the originals in your methods.

First of all, you know why you need to create immutable object, and what are the advantages of immutable object.
Advantages of an Immutable object
Concurrency and multithreading
It automatically Thread-safe so synchronization issue....etc
Don't need to copy constructor
Don't need to implementation of clone.
Class cannot be override
Make the field as a private and final
Force callers to construct an object completely in a single step, instead of using a no-Argument constructor
Immutable objects are simply objects whose state means object's data can't change after the
immutable object are constructed.
please see the below code.
public final class ImmutableReminder{
private final Date remindingDate;
public ImmutableReminder (Date remindingDate) {
if(remindingDate.getTime() < System.currentTimeMillis()){
throw new IllegalArgumentException("Can not set reminder" +
" for past time: " + remindingDate);
}
this.remindingDate = new Date(remindingDate.getTime());
}
public Date getRemindingDate() {
return (Date) remindingDate.clone();
}
}

Minimize mutability
An immutable class is simply a class whose instances cannot be modified. All of the information contained in each instance is provided when it is created and is fixed for the lifetime of the object.
JDK immutable classes: String, the boxed primitive classes(wrapper classes), BigInteger and BigDecimal etc.
How to make a class immutable?
Don’t provide any methods that modify the object’s state (known as mutators).
Ensure that the class can’t be extended.
Make all fields final.
Make all fields private.
This prevents clients from obtaining access to mutable objects referred to by fields and modifying these objects directly.
Make defensive copies.
Ensure exclusive access to any mutable components.
public List getList() {
return Collections.unmodifiableList(list); <=== defensive copy of the mutable
field before returning it to caller
}
If your class has any fields that refer to mutable objects, ensure that clients of the class cannot obtain references to these objects. Never initialize such a field to a client-provided object reference or return the object reference from an accessor.
import java.util.Date;
public final class ImmutableClass {
public ImmutableClass(int id, String name, Date doj) {
this.id = id;
this.name = name;
this.doj = doj;
}
private final int id;
private final String name;
private final Date doj;
public int getId() {
return id;
}
public String getName() {
return name;
}
/**
* Date class is mutable so we need a little care here.
* We should not return the reference of original instance variable.
* Instead a new Date object, with content copied to it, should be returned.
* */
public Date getDoj() {
return new Date(doj.getTime()); // For mutable fields
}
}
import java.util.Date;
public class TestImmutable {
public static void main(String[] args) {
String name = "raj";
int id = 1;
Date doj = new Date();
ImmutableClass class1 = new ImmutableClass(id, name, doj);
ImmutableClass class2 = new ImmutableClass(id, name, doj);
// every time will get a new reference for same object. Modification in reference will not affect the immutability because it is temporary reference.
Date date = class1.getDoj();
date.setTime(date.getTime()+122435);
System.out.println(class1.getDoj()==class2.getDoj());
}
}
For more information, see my blog:
http://javaexplorer03.blogspot.in/2015/07/minimize-mutability.html

an object is called immutable if its state can not be changed once created. One of the most simple way of creating immutable class in Java is by setting all of it’s fields are final.If you need to write immutable class which includes mutable classes like "java.util.Date". In order to preserve immutability in such cases, its advised to return copy of original object,

Immutable Objects are those objects whose state can not be changed once they are created, for example the String class is an immutable class. Immutable objects can not be modified so they are also thread safe in concurrent execution.
Features of immutable classes:
simple to construct
automatically thread safe
good candidate for Map keys and Set as their internal state would not change while processing
don't need implementation of clone as they always represent same state
Keys to write immutable class:
make sure class can not be overridden
make all member variable private & final
do not give their setter methods
object reference should not be leaked during construction phase

The following few steps must be considered, when you want any class as an immutable class.
Class should be marked as final
All fields must be private and final
Replace setters with constructor(for assigning a value to a
variable).
Lets have a glance what we have typed above:
//ImmutableClass
package younus.attari;
public final class ImmutableExample {
private final String name;
private final String address;
public ImmutableExample(String name,String address){
this.name=name;
this.address=address;
}
public String getName() {
return name;
}
public String getAddress() {
return address;
}
}
//MainClass from where an ImmutableClass will be called
package younus.attari;
public class MainClass {
public static void main(String[] args) {
ImmutableExample example=new ImmutableExample("Muhammed", "Hyderabad");
System.out.println(example.getName());
}
}

Commonly ignored but important properties on immutable objects
Adding over to the answer provided by #nsfyn55, the following aspects also need to be considered for object immutability, which are of prime importance
Consider the following classes:
public final class ImmutableClass {
private final MutableClass mc;
public ImmutableClass(MutableClass mc) {
this.mc = mc;
}
public MutableClass getMutClass() {
return this.mc;
}
}
public class MutableClass {
private String name;
public String getName() {
return this.name;
}
public void setName(String name) {
this.name = name;
}
}
public class MutabilityCheck {
public static void main(String[] args) {
MutableClass mc = new MutableClass();
mc.setName("Foo");
ImmutableClass iMC = new ImmutableClass(mc);
System.out.println(iMC.getMutClass().getName());
mc.setName("Bar");
System.out.println(iMC.getMutClass().getName());
}
}
Following will be the output from MutabilityCheck :
Foo
Bar
It is important to note that,
Constructing mutable objects on an immutable object ( through the constructor ), either by 'copying' or 'cloing' to instance variables of the immutable described by the following changes:
public final class ImmutableClass {
private final MutableClass mc;
public ImmutableClass(MutableClass mc) {
this.mc = new MutableClass(mc);
}
public MutableClass getMutClass() {
return this.mc;
}
}
public class MutableClass {
private String name;
public MutableClass() {
}
//copy constructor
public MutableClass(MutableClass mc) {
this.name = mc.getName();
}
public String getName() {
return this.name;
}
public void setName(String name) {
this.name = name;
}
}
still does not ensure complete immutability since the following is still valid from the class MutabilityCheck:
iMC.getMutClass().setName("Blaa");
However, running MutabilityCheck with the changes made in 1. will result in the output being:
Foo
Foo
In order to achieve complete immutability on an object, all its dependent objects must also be immutable

From JDK 14+ which has JEP 359, we can use "records". It is the simplest and hustle free way of creating Immutable class.
A record class is a shallowly immutable, transparent carrier for a fixed set of fields known as the record components that provides a state description for the record. Each component gives rise to a final field that holds the provided value and an accessor method to retrieve the value. The field name and the accessor name match the name of the component.
Let consider the example of creating an immutable rectangle
record Rectangle(double length, double width) {}
No need to declare any constructor, no need to implement equals & hashCode methods. Just any Records need a name and a state description.
var rectangle = new Rectangle(7.1, 8.9);
System.out.print(rectangle.length()); // prints 7.1
If you want to validate the value during object creation, we have to explicitly declare the constructor.
public Rectangle {
if (length <= 0.0) {
throw new IllegalArgumentException();
}
}
The record's body may declare static methods, static fields, static initializers, constructors, instance methods, and nested types.
Instance Methods
record Rectangle(double length, double width) {
public double area() {
return this.length * this.width;
}
}
static fields, methods
Since state should be part of the components we cannot add instance fields to records. But, we can add static fields and methods:
record Rectangle(double length, double width) {
static double aStaticField;
static void aStaticMethod() {
System.out.println("Hello Static");
}
}

Why are not volatile final fields permitted?

I'm designing a thread-safe container class called ConcurrentParamters. here is what I tend to write:
Interfaces:
public interface Parameters {
public <M> M getValue(ParameterMetaData<M> pmd);
public <M> void put(ParameterMetaData<M> p, M value);
public int size();
}
public interface ParameterMetaData<ValueType> {
public String getName();
}
Implementation:
public final class ConcurrentParameters implements Parameters{
private final Map<ParameterMetaData<?>, Object> parameters;
private final volatile AtomicInteger size; //1, Not compile
{
size = new AtomicInteger();
parameters = new ConcurrentHashMap<>();
}
public static Parameters emptyParameters(){
return new ConcurrentParameters();
}
#Override
public <M> M getValue(ParameterMetaData<M> pmd) {
M value = (M) parameters.get(pmd);
return value;
}
#Override
public <M> void put(ParameterMetaData<M> p, M value){
parameters.put(p, value);
size.incrementAndGet();
}
#Override
public int size() {
return size.intValue();
}
}
I tried to make the AtomicInteger field representing the size final, to ensure that no method can change the field poinitng to another object, as well as initialize it duriong consruction.
But since the container will be accessed concurrently, I need that any thread observes changes made by other. So, I tried to declared it volatile as well in order ot avoid unnecessary synchronization (I don't need mutual-exclusion).
I didn't compile. Why? Is there any reason? Does it make no sense to decalre a field that way? I thought it would be sesnsible... Maybe it is inheritly not safe?

The answer is simple:
All guarantees volatile makes, are done by final already. So it would be redundant.
Take a look at the answer here from axtavt for more detailed information:
Java concurrency: is final field (initialized in constructor) thread-safe?

volatile means that reads and writes to the field have specific synchronization effects; if you cannot write to the field, volatile makes no sense, so marking a field final volatile is forbidden.
You don't need volatile, and it wouldn't help. Mutations to an AtomicInteger are not assignments to the field that holds the AtomicInteger, so they wouldn't be affected by volatile anyway. Instead, reads and modifications of an AtomicInteger's value already have the appropriate thread-safety mechanisms applied by the AtomicInteger implementation itself.

A volatile variable means that it may be accessed by unsynchronized threads so its value should always be written back to memory after each and every change to it. But then again, you can't change a variable declared as final, thus volatile is irrelevant in your example.

Creating instances by static field

I have the following two classes:
public class A{
private String s;
public A(String s){
this.s = s;
}
}
public class B{
private static final String STR = "String";
public void doAction(){
A a = new A(STR); //Does it look really wierd?
}
}
I've never passed the static final field as a constructor parameter, so can it lead to potential bugs? Should we avoid it or we can do that if it seems concise.

I've never passed the static final field as a constructor parameter, so can it lead to potential bugs?
This cannot lead to a bug, because doAction is an instance method. All static fields with initializers will be initialized before the first instance method is called, so you are safe.
Should we avoid it or we can do that if it seems concise?
Using a static final field, which is effectively a String constant, inside an instance method, is а perfectly valid choice.

You shouldn't worry if the field is immutable (like in your case: String is immutable data structure). With mutable objects (for example arrays) you should consider that all changes to this field in one object will be visible to other objects with the same field whether it is static or not. These code doesn't look weird.

String is a reference type, but it works like a primitive type sometimes. If you use a String object as an argument, it will be copied, like other primitive variables. I don't understand the reason you pass a class variable as an argument of other object. But it will not occur any problem caused by sharing of an attribute between objects of this class.

In which situations we make variables as public and methods as private?

Currently I am learning basics of java and C++. I have read in book Let Us C++, that in almost every case we we make instance variables private and methods public for the security purposes. But it is also mentioned in this book that in some cases we make variables public and methods private..
I am continuously thinking, in which cases we will do so. Can anyone please explain this.

Private methods (or private member functions in C++ terminology) are mostly useful as helper functions. For example, think of the case that you want to implement fractions, but want to ensure that your fraction is always normalized. Then you could use a private member function normalize() which normalizes your fraction, and which is called after each operation which might result in a non-normalized fraction, for example (C++ code):
class Fraction
{
public:
Fraction(int num, int den = 1);
Fraction operator+=(Fraction const& other);
Fraction operator*=(Fraction const& other);
// ...
private:
int numerator, denominator;
};
Fraction::Fraction(int num, int den):
numerator(num),
denominator(den)
{
normalize();
}
Fraction Fraction::operator+=(Fraction const& other)
{
int new_den = denominator*other.denominator;
numerator = numerator*other.denominator + denominator*other.numerator;
denominator = new_den;
}
Fraction Fraction::operator*=(Fraction const& other)
{
numerator *= other.numerator;
denominator *= other.denominator;
normalize();
}
void Fraction::normalize()
{
int factor = gcd(numerator, denominator);
numerator /= factor;
denominator /= factor;
}
Another, C++ specific use of private functions is based on the fact that in C++ private is only about access control, not about visibility. This enables to do unoverridable pre-post-condition checking in the base class while making the actual function virtual:
class Base
{
public:
foo frobnicate(some arguments);
private:
virtual foo do_frobnicate(some arguments) = 0;
};
foo Base::frobnicate(some arguments)
{
check_precondition(arguments);
foo result = do_frobnicate(arguments);
check_post_condition(foo);
return foo;
}
Classes derived from Base will override do_frobnicate, while users will call frobnicate which always checks the pre/postconditions no matter what the derived class does.

Generally static final variables are public in a class. If you don't need to change the value of that variable and want other classes to access it then you make it public static final.
Private methods are used only within the class for doing the task, which is internal to that class. Like a utility method or some business calculation method. Or simply to break the code of public method into multiple private methods, so that methods don't grow too big.

When a method is to be used by other methods(public) of the class and you do not want the object to access that method directly, we make that method as private.
And in some cases, if you want to access your variable directly from the class object, then make it public.

If you don't need the varibale or methode in other classes don't make it public. This goes for methodes and variables.

private methods are for the internal use of the class. They can be called from other public classes. Those are private because you encapsualted from outer world.
For example
public void method1(){
method2();
}
private void method2(){
// for internal use
}
Public variables are mainly used for class variables in which cases there is no harm of direct accessing the variables from outside. For example
public static final int FLAG = true;
You can directly call the variable from outside.

It depends how much security you want for each class.
For example, if you have a Vector class, that only has 3 variables x, y and z, you should make them public. Many classes will probably use the Vector class and it's fine if they change values in it.
If you have a Person class that stores credit card number, background record, address etc, you should make them private to avoid security issues.
However, if you have all variables as private, and you provide accessors and mutators for all of them, you're effectively making them just like public (but with more work).
EDIT:
All constant variables should be public, because you cannot change them anyway.
Static variables could be both, depending on a situation. Probably better to have static get and set functions for static variables.

Private variables or functions can be use only in the class where they are declarated.
Public variables or functions can be use everywhere in your application.
So you should declarate private all those variables and functions that you are going to use ONLY in the class where they belong.
Example:
public class Car {
private String model;
public setModel(String model) {
if (model != null)
this.model = model;
}
public getModel() {
return model;
}
private doSomething() {
model = "Ford";
}
}
In the class Car we declarate the String model as private because we are going to use it only in the class Car, doing this we assure that other classes couldn't change the value of this String without using the function setModel.
The functions setModel and getModel are public, so we can access the private variable model from other classes ONLY using those methods.
In this example, the function setModel checks if the value its null, in which case it doesn't set the value. Here you can see that if you had declarated the String model as public, you wouldn't have control over what value it's being recorded.
The function doSomething is private and other classes can't use it. For other side, like this function is private and it belong to the same class where is the String model, it can change its value without using the method setModel.

A rule of thumb, you make methods public when it is okay for other classes to access them. internal methods or helper methods should either be protected or private.Protected if you want the method to be extendable by those extending your class however if you don't want this just mark them private.

Adding parameterless constructor to non-serializable, extendable class

What is special about addid a parameterless constructor to a non serializable, extendable class.
In Effective java , the author talks about this topic.
Naively adding a parameterless constructor and a
separate initialization method to a class whose remaining constructors establish its
invariants would complicate the state space, increasing the likelihood of error.
The following code is copied from Effective Java 2nd Edition[Page 292-293]
public class AbstractFoo {
private int x, y; // Our state
// This enum and field are used to track initialization
private enum State {
NEW, INITIALIZING, INITIALIZED
};
private final AtomicReference<State> init = new AtomicReference<State>(
State.NEW);
public AbstractFoo(int x, int y) {
initialize(x, y);
}
// This constructor and the following method allow
// subclass's readObject method to initialize our state.
protected AbstractFoo() {
}
protected final void initialize(int x, int y) {
if (!init.compareAndSet(State.NEW, State.INITIALIZING))
throw new IllegalStateException("Already initialized");
this.x = x;
this.y = y;
// ... // Do anything else the original constructor did
init.set(State.INITIALIZED);
}
// These methods provide access to internal state so it can
// be manually serialized by subclass's writeObject method.
protected final int getX() {
checkInit();
return x;
}
protected final int getY() {
checkInit();
return y;
}
// Must call from all public and protected instance methods
private void checkInit() {
if (init.get() != State.INITIALIZED)
throw new IllegalStateException("Uninitialized");
}
// ... // Remainder omitted
}
All public and protected instance methods in AbstractFoo must invoke
checkInit before doing anything else. This ensures that method invocations fail
quickly and cleanly if a poorly written subclass fails to initialize an instance. Note
that the initialized field is an atomic reference (java.util.concurrent.
atomic.AtomicReference). This is necessary to ensure object integrity in
the face of a determined adversary. In the absence of this precaution, if one thread
were to invoke initialize on an instance while a second thread attempted to use
it, the second thread might see the instance in an inconsistent state.
Why are we doing this? I did not fully understand this. Can anyone explain ?

I was having the same issue when reading the book. I was bit confused in that exact place. With bit of research I found out this.
http://docs.oracle.com/javase/7/docs/api/java/io/Serializable.html
Read this. According to that
"During deserialization, the fields of non-serializable classes will be initialized using the public or protected no-arg constructor of the class. A no-arg constructor must be accessible to the subclass that is serializable. The fields of serializable subclasses will be restored from the stream"
I think this answers your question.
Hope this would be helpfull.
If there any thing wrong in this comment please feel free to correct it.

Separate initialization method is very useful when you have multi-threading issue. You can see good article : http://www.ibm.com/developerworks/java/library/j-jtp0618/index.html