Introduction to JAVA

Introduction to Java
The Java architecture consists of:
• a high-level object-oriented programming language,
• a platform-independent representation of a compiled class,
• a pre-defined set of run-time libraries,
• a virtual machine.
This book is mainly concerned with the language aspects of Java and the associated
java.lang library package. Consequently, the remainder of this section provides a
brief introduction to the language. Issues associated with the other components will be
introduced as and when needed in the relevant chapters.
The introduction is broken down into the following components
• identifiers and primitive data types
• structured data types
• reference types
• blocks and exception handling
• control structures
• procedures and functions
• object oriented programming, packages and classes
• inheritance
• interfaces
• inner classes.
1 Identifiers and primitive data types
Identifiers Java does not restrict the lengths of identifiers. Although the language does allow the
use of a “_” to be included in identifier names, the emerging style is to use a mixture of
upper and lower case characters. The following are example identifiers:
http://professional-guru.com

2 Introduction to Java
Note that Java is case sensitive. Hence, the identifier exampleNameInJava is differ-
ent from ExampleNameInJava.
Primitive
data types
Java provides both a variety of discrete data types and a floating point type.
Discrete data types. The following discrete data types are supported:
• int — a 32-bit signed integer;
• short — a 16-bit signed integer;
• long — a 64-bit signed integer;
• byte — an 8-bit signed integer;
• boolean — the truth values, true and false;
• char — Unicode characters (16 bits).
All the usual operators for these types are available.
Enumeration types have been introduces as of Java 1.5. Although not a primitive
type it behaves like a descrete type and can be used in a switch statement. The follow-
ing example illustrates a simple enumeration type for days of the week.
Floating
point
numbers
Java provides float and double types that support the 32 and 64 bit IEEE 754 float-
ing point values respectively. Note, that floating point literals are automatically consid-
ered to be of double precision. Consequently, they must either be explicitly converted
to float or they can be followed by the letter f. For example:
exampleNameInJava
example_name_in_Java
public enum Day{SUNDAY, MONDAY, TUESDAY, WEDNESDAY,
THURSDAY, FRIDAY, SATURDAY}
float bodyTemperature = (float) 98.6;
// or
float bodyTemperature = 98.6f;

Introduction to Java 3
Note that the assignment operator is “=” and that comments are delimited by “/*” and
“*/”. Java also allows “//” for comments ending a line.
2 Structured data types
Java supports arrays; both single and multi dimensional arrays can be created; for
example:
Note that arrays are represented by objects and that indexes start at 0. The length of the
array can be determined by the length field associated with the array
Java has no record structure as such. However, the same effect can be achieved
using classes. The details of classes will be given in Section 7 but, for now, consider the
following class for the date:
It is not possible to express range constraints on the values of the date's components.
Note also that as a “record” is a class, an allocator (the new operator) must be used to
create an instance of the Date. Initialization of the object can be achieved using con-
structors (see Section 7).
Important
note
final int max = 10; // a constant
float[] reading = new float[max]; // index is 0 .. max-1
boolean[][] switches = new boolean[max][max];
class Date {
int day, month, year;
}
Date birthDate = new Date();
birthDate.day = 31;
birthDate.month = 1;
birthDate.year = 2000;
All objects created by the new operator are stored in an area of memory called the
heap. An activity called garbage collection will free up any associated memory
when the object is no longer referenced.

3 Reference types
All objects in Java are references to the actual locations containing the encapsulated
data, hence no additional access or pointer type is required. Furthermore, no forward
declaration is required. For example:
Note that as a result of representing objects as reference values, comparing two objects
is a comparison between their references not their values. Hence,
will compare the locations of the objects not the values encapsulated by the objects (in
this case, the values of the value and next fields). To compare values requires a
class to implement an explicit equals method. A similar situation occurs with object
assignment. The assignment operator assigns to the reference. To create an actual copy
of an object requires the class to provide methods to clone the object.
4 Blocks and exception handling
In Java, a block (or compound statement) is delimited by “{“ and “}”, and usually has
the following structure
class Node {
int value;
Node next;
// The type node can be used even though
// its declaration has not been completed yet.
}
Node ref1 = new Node();
Node ref2 = new Node();
. . .
if(ref1 == ref2) { ... }

although the declarative part can be dispersed throughout the block.
Java can have exception handlers at the end of a block if the block is labeled as a
try block. Each handler is specified using a catch statement. Consider, the following:
The catch statement is like a function declaration, the parameter of which identifies the
exception type to be caught. Inside the handler, the object name behaves like a local
variable. A handler with parameter type T will catch a thrown object of type E if:
• T and E are the same type,
• or T is a parent (super) class of E at the throw point.
It is this last point that integrates the exception handling facility with the object-ori-
ented programming model. In the above example the Exception class is the root
class of all application exceptions. Hence, the catch clause will catch all application
exceptions. Here, getMessage is a method declared in the Exception class. A call
to E.getMessage will execute the appropriate routine for the type of object thrown.
If no exception handler is found in the calling context of a function, the calling
context is terminated and a handler is sought in its calling context. Hence, Java supports
exception propagation.
Finally
clauses
Java also supports a finally clause as part of a try statement. The code attached to
this clause is guaranteed to execute whatever happens in the try statement irrespective
{
< declarative part >
< sequence of statements >
}
try {
//code that might throw an exception
} catch (Exception err) {
// Exception caught, print error message on
// the standard output.
System.out.println(err.getMessage());
}

of whether exceptions are thrown, caught, propagated or, indeed, even if there are no
exceptions thrown at all.
Checked and
unchecked
exceptions
In Java, all exceptions are subclasses of the predefined class java.lang.Throw-
able. The language also defines other classes, for example: Error, Exception,
and RuntimeException. The relationship between these (and some common
exceptions) is depicted in Figure 1.
FIGURE 1. Part of the The Java Predefined Throwable Class Hierarchy
Throughout this book, the term Java exception is used to denote any class derived from
Exception. Objects derived from Error describe internal errors and resource
exhaustion in the Java run-time support system. Although these errors clearly have a
major impact on the program, there is little that the program can do when they are
try {
...
} catch(...) {
...
} finally {
// code that will be executed under all circumstances
}
Throwable
Error Exception
RunTimeException
IllegalArgumentException
IllegalThreadStateException
SecurityException
InterruptedException
NullPointerException
checked
unchecked
IllegalMonitorStateException

thrown (raised) as no assumptions can be made concerning the integrity of the system.
Objects derived from the Exception hierarchy represent errors that programs can
handle or can throw themselves. RuntimeExceptions are those exceptions that are
raised by the run-time system as a result of a program error. They include errors such as
those resulting from a bad cast (ClassCastException), array bounds error
(IndexOutOfBoundException), a null pointer access (NullPointerExcep-
tion), integer divide by zero (ArithmeticException) etc.
Throwable objects which are derived from Error or RuntimeExceptions
are called unchecked exceptions, the others are termed checked exceptions. Checked
exceptions must be declared by the function that can throw them. The compiler will
check that an appropriate handler can be found. The compiler makes no attempt to
ensure that handlers for unchecked exceptions exist.
5 Control structures
Control structures can be grouped together into three categories: sequences, decisions
and loops.
Sequence
structures
Most languages implicitly require sequential execution, and no specific control struc-
ture is provided. The definitions of a block in Java indicate that between { and } there
is a sequence of statements. Execution is required to follow this sequence.
Decision
structures
The most common form of decision structure is the if statement; for example:
In general, a multiway decision can be more explicitly stated and efficiently imple-
mented, using a switch structure.
if(A != 0) {
if(B/A > 10) {
high = 1;
} else {
high = 0;
}
}

As with C and C++, it is necessary to insert statements to break out of the switch once
the required command has been identified. Without these, control continues to the next
option (as with the case of A).
Loop
(iteration)
structures
Java supports the usual for and while statements. The following example illustrates
the for statement; the code assigns into the first ten elements of array, A, the value of
their positions in the array:
The free use of the loop control variable (i here) in this manner can be the cause of
many errors, consequently, Java also allows a local variable to be declared inside the
for loop.
An example while statement is given below:
switch(command) {
case 'A' :
case 'a' : action1(); break; /* A or a */
case 't' : action2(); break;
case 'e' : action3(); break;
case 'x' :
case 'y' :
case 'z' : action4(); break; /* x, y or z */
default : /* no action */
}
for(i = 0; i <= 9; i++) {
/* i must be previously declared */
A[i]= i; /* i can be read/written in the loop */
} /* the value of i is defined */
/* after the loop */
for(int i = 0; i <= max; i++) {
A[i]= i;
}

Java also supports a variant where the test occurs at the end of the loop:
The flexibility of an iteration statement is increased by allowing control to pass out of
the loop (that is, the loop to terminate) from any point within it (break) or for control
to pass directly to the next iteration (continue):
6 Procedures and functions
Procedures and functions can only be declared in the context of a class, they are known
collectively as methods. In fact, strictly Java only supports functions; a procedure is
considered to be a function with no return value (indicated by void in the function
definition).
Java passes primitive data type (int, boolean, float etc.) parameters (often called
arguments) by value. Variables of class type are reference variables. Hence, when they
are passed as arguments they are copied, but the effect is as if the object was passed by
reference. Consider a function that calculates the roots of a quadratic equation:
while(<expression>) {
/* Expression evaluating to true or false. */
/* False implies loop termination. */
<sequence of statements>
}
do {
< sequence of statements>
} while (<expression>);
while(true) {
...
if(<expression1>) break;
if(<expression2>) continue;
...
}

Note Java requires the roots of the equation to be passed as a class type, as primitive
types (including double) are passed by copy and there are no pointer types.
Function
bodies
The bodies of functions are illustrated by completing the quadratic definitions
given above.
The invoking of a function merely involves naming the function (including any object
or class name) and giving the appropriately typed parameters in parentheses.
7 Object-oriented programming, packages and
classes
Objects have four important properties [Wegner, 1987]. They have
• inheritance (type extensibility)
public class Roots {
double R1, R2;
}
// the following must be declared in a class context
boolean quadratic(double A, double B, double C,
Roots R);
boolean quadratic(double A, double B,
double C, Roots R) {
double disc;
disc = B*B - 4.0*A*C;
if(disc < 0.0 || A == 0.0) { // no roots
R.R1 = 0; // arbitrary values
R.R2 = 0;
return false;
}
R.R1 = (-B + Math.sqrt(disc)) / (2.0*A);
R.R2 = (-B - Math.sqrt(disc)) / (2.0*A);
return true;
}

• automatic object initialization (constructors)
• automatic object finalization (destructors)
• dynamic binding of method calls (polymorphism — sometimes called run-time
dispatching of operations).
All of which are supported, in some form, by Java’s class and interface mechanisms.
Related classes and interfaces can be grouped together into larger units called pack-
ages.
Inheritance is perhaps the most significant concept in an object abstraction. This
enables a type (class) to be defined as an extension of a previously defined type. The
new type inherits the “base” type but may include new fields and new operations. Once
the type has been extended then run-time dispatching of operations is required to
ensure the appropriate operation is called for a particular instance of the family of
types.
Earlier in this section, a simple class for the date type was introduced.
This example only illustrates how data items can be grouped together. The full facilities
of a class allow the data items (or instance variables or fields as Java calls them) to be
encapsulated and, hence, abstract data types to be defined. As an example, consider the
class for a queue abstraction.
First, a package to contain the queue can be declared (if there is no named pack-
age, then the system assumes an unnamed package). Items from other packages can be
imported. Then classes to be declared inside the package are given. One of these
classes Queue is declared as public, and it is only this that can be accessed outside
the package. The keyword public is termed a modifier in Java, the other ones for
classes include abstract and final. An abstract class is one from which no
objects can be created. Hence, the class has to be extended (to produce a subclass) and
that subclass made non-abstract before objects can exist. A final modifier indicates
that the class cannot be subclassed. No modifier indicates that the class is only accessi-
ble within the package. A class can have more than one modifier but certain combina-
tions, such as abstract and final, are meaningless.
Each class can declare local instance variables (or fields), constructor methods
and ordinary (member) methods. A static field is a class-wide field and is shared
between all instances of the owning class. The field is accessible even if there are no
class Date {
int day, month, year;
}

instances of the class. Constructor methods have the same name as their associated
class. An appropriate constructor method is automatically called when objects of the
class are created. All methods of the object can also have associated modifiers. These
dictate the accessibility of the method; public, protected and private are
allowed. Public methods allow full access, protected allows access only from
within the same package or from with a subclass of the defining class and private
allows access only from within the defining class. Instance variables can also have
these modifiers. If no modifier is given, the access is restricted to the package. Static
methods (class-wide methods) can be invoked without reference to an object. They,
therefore, can only access static fields or other static methods.
Member methods and instance variables can have other modifiers which will be
introduced throughout this book.
The code for the Queue class can now be given.
package queues; // package name
import somepackage.Element; // import element type
class QueueNode { // class local to package
Element data; // queued data item
QueueNode next; // reference to next QueueNode
}
public class Queue {
// class available from outside the package
public Queue() { // public constructor
front = null;
back = null;
}
public void insert(Element E) { // visible method
QueueNode newNode = new QueueNode();
newNode.data = E;
newNode.next = null;
if(empty()) {
front = newNode;
} else {
back.next = newNode;
}
back = newNode;
}

Java 1.5 has introduced extra facilities for generic programming. The above example
assumed that the queue was for the Element class. To generalise this class so that it
deals with any element, the Element is defined as a generic parameter:
The queue can now be instantiated for a particular class.
public Element remove() { //visible method
if(!empty()) {
Element tmpE = front.data;
front = front.next;
if(empty()) back = null;
return tmpE;
}
// Garbage collection will free up the QueueNode
// object which is now dangling.
return null; // queue empty
}
public boolean empty() { // visible method
return (front == null);
}
private QueueNode front, back; // instance variables
}
package queues; // package name
class QueueNode<Element> { // class local to package
Element data; // generic queued data item
QueueNode next; // reference to next QueueNode
}
public class Queue<Element> {
// class available from outside the package
public Queue() { // public constructor
... // as before
}
.. // as before
}

8 Inheritance
Inheritance in Java is obtained by deriving one class from another. Multiple inheritance
is not supported although similar effects can be achieved using interfaces (see below).
Consider an example of a class which describes a two dimensional coordinate
system.
This class can now be extended to produce a new class by naming the base class in the
declaration of the derived class with the extends keyword. The new class is placed in
the same package1
Queue<String> dictionary = new Queue<String>();
package coordinates;
public class TwoDimensionalCoordinate {
public TwoDimensionalCoordinate(
float initialX, float initialY) { // constructor
X = initialX;
Y = initialY;
}
public void set(float F1, float F2) {
X = F1;
Y = F2;
}
public float getX() {
return X;
}
public float getY() {
return Y;
}
private float X;
private float Y;
}

A new field Z has been added and the constructor class has been defined, the set func-
tion has been overridden, and a new operation provided. Here, the constructor calls the
base class constructor (via the super keyword) and then initializes the final dimen-
sion. Similarly, the overloaded set function calls the base class.
All method calls in Java are potentially dispatching (dynamically bound). For
example, consider again the above example:
1. In Java, public classes must reside in their own file. Hence, a package can be distrib-
uted across one or more files, and can be continually augmented.
public class ThreeDimensionalCoordinate
extends TwoDimensionalCoordinate {
// subclass of TwoDimensionalCoordinate
float Z; // new field
public ThreeDimensionalCoordinate(
float initialX, float initialY,
float initialZ) { // constructor
super(initialX, initialY);
// call superclass constructor
Z = initialZ;
}
public void set(float F1, float F2, float F3) {
//overloaded method
set(F1, F2); // call superclass set
Z = F3;
}
public float getZ() { // new method
return Z;
}
}

Here the method plot has been added. Now if plot is overridden in a child (sub)
class:
Then the following:
would plot a two dimensional coordinate; where as
public class TwoDimensionalCoordinate {
// as before
public void plot() {
// plot a two dimensional point
}
}
public class ThreeDimensionalCoordinate
extends TwoDimensionalCoordinate {
// as before
public void plot() {
// plot a three dimensional point
}
}
{
TwoDimensionalCoordinate A = new
TwoDimensionalCoordinate(0f, 0f);
A.plot();
}

would plot a three dimensional coordinate even though A was originally declared to be
of type TwoDimensionalCoordinate. This is because A and B are reference
types. By assigning B to A, only the reference has changed not the object itself.
The Object
class
All classes, in Java, are implicit subclasses of a root class called Object. The defini-
tion of this class is given below:
{
TwoDimensionalCoordinate A =
new TwoDimensionalCoordinate(0f, 0f);
ThreeDimensionalCoordinate B =
new ThreeDimensionalCoordinate(0f, 0f, 0f);
A = B;
A.plot();
}
package java.lang;
public class Object {
public final Class getClass();
public String toString();
public boolean equals(Object obj);
public int hashCode();
protected Object clone()
throws CloneNotSupportedException;
public final void wait()
throws InterruptedException;
// throws unchecked IllegalMonitorStateException
public final void wait(long millis)
public final void wait(long millis, int nanos)

There are six methods in the Object class which are of interest in this book. The three
wait and two notify methods are used for concurrency control and will be consid-
ered in Chapter 3.
The sixth method is the finalize method. It is this method that gets called just
before the object is destroyed. Hence by overriding this method, a child class can pro-
vide finalization for objects which are created from it. Of course, when a class over-
rides the finalize method, its last action should be to call the finalize method
of its parent. However, it should be noted that finalize is only called when garbage
collection is about to take place, and this may be some time after the object is no longer
in use; there are no explicit destructor methods as in, say, C++. Furthermore, garbage
collection will not necessarily take place before a program terminates. In the System
class there are two methods that can be used to request finalization:
The first of these methods requests the JVM to complete all outstanding finalization
code. The second method (when called with a true parameter) indicates that all out-
standing finalization be completed before the program exits; however, it has now been
deprecated as its use was found to lead to potential deadlocks.
public final void notify()
// throws unchecked IllegalMonitorStateException;
public final void notifyAll()
// throws unchecked IllegalMonitorStateException;
protected void finalize() throws Throwable;
}
package java.lang;
public class System {
...
public static void runFinalization();
public static void runFinalizeOnExit(boolean value);
// deprecated
}

9 Interfaces
Interfaces augment classes to increase the reusability of code. An interface defines a
reference type which contains a set of methods and constants. The methods are by defi-
nition abstract, so no instances of interfaces can be constructed. Instead, one or more
classes can implement an interface, and objects implementing interfaces can be passed
as arguments to methods by defining the parameter to be of the interface type. What
interfaces do, in effect, is to allow relationships to be constructed between classes out-
side of the class hierarchy.
Consider, a “generic” algorithm to sort arrays of objects. What is common about
all classes that can be sorted is that they support a less than, <, or greater than, >,
operator. Consequently, this feature is encapsulated in an interface. The Ordered
interface defines a single method lessThan that takes as an argument an object
whose class implements the Ordered interface. Any class which implements the
Ordered interface must compare its object’s ordering with the argument passed and
return whether it is less than the object or not.
The class for complex numbers is such a class.
package interfaceExamples;
public interface Ordered {
boolean lessThan (Ordered O);
}
import interfaceExamples.*;
class ComplexNumber implements Ordered {
// the class implements the Ordered interface
public boolean lessThan(Ordered O) {
// the interface implementation
ComplexNumber CN = (ComplexNumber) O; // cast O
if((realPart*realPart + imagPart*imagPart) <
(CN.getReal()*CN.getReal() +
CN.getImag()*CN.getImag())) {
return true;
}
return false;
}

Now it is possible to write algorithms which will sort this and any other class that
implements the interface. For example, an array sort:
The sort method takes two arguments; the first is an array of objects that implement
the Ordered interface and the second is the number of items in the array. The imple-
mentation performs an exchange sort. The important point about this example is that
when two objects are exchanged, it is the reference values which are exchanged and
public ComplexNumber (float I, float J) {
// constructor
realPart = I;
imagPart = J;
}
public float getReal() { return realPart; }
public float getImag() { return imagPart; }
protected float realPart;
protected float imagPart;
}
package interfaceExamples;
public class ArraySort {
public static void sort (Ordered[] oa) {
Ordered tmp;
int pos;
for (int i = 0; i < oa.length - 1; i++) {
pos = i;
for (int j = i + 1; j < oa.length; j++) {
if (oa[j].lessThan(oa[pos])) {
pos = j;
}
}
tmp = oa[pos];
oa[pos] = oa[i];
oa[i] = tmp;
}
}
}

hence it does not matter what type the object is (as long as it supports the Ordered
interface).
To use the above classes and interfaces simply requires the following
In fact, the package java.lang already defines an interface called Comparable
with a function called compareTo which could be used in place of Ordered. Fur-
thermore, there is a static method in java.util.Arrays called sort that imple-
ments a merge sort of objects.
Interfaces have three further properties which should be noted. Firstly, like
classes, they can participate in inheritance relationships with other interfaces. However,
unlike classes, multiple inheritance is allowed. Secondly, a class can implement more
than one interface, and hence, much of the functionality of multiple inheritance can be
achieved for classes. Finally, interfaces also provide the mechanisms by which call-
backs can be implemented. This is when an object (server) needs to call one or more
methods defined in the caller's (client’s) class. The client’s class can be of any type. As
Java does not support function pointers, it is necessary to provide an alternative mecha-
nism to that used traditionally in languages like C and C++. If the class implements an
interface which defines the required functions, then the client can pass itself as a
parameter to the server. The server can then call-back on any of the methods defined in
that interface.
{
ComplexNumber[] arrayComplex = { // for example
new ComplexNumber(6f,1f),
new ComplexNumber(1f, 1f),
new ComplexNumber(1f, 7f)
};
// array unsorted
ArraySort.sort(arrayComplex);
// array sorted
}

10 Inner classes
Inner classes are classes that are declared within other classes, they allow more control
over the visibility and give added flexibility. There are various types of inner classes,
the most important for this book are member, local and anonymous classes.
Member
classes
A member class is declared in the same way that a method or a field is declared. The
code of the member class can access all fields and methods of the enclosing class.
Local
classes
A local class is declared within a block of code within an enclosing class (such as
within a method of the enclosing class). It can access everything that a member class
can access. It can also access any final variables declared locally to the block declaring
the local class (and final parameters, if the block is a method).
public class Outer {
// local fields and methods of Outer
class MemberClass {
// Fields and methods of the Member class can
// access local fields and methods of the
// Outer class.
}
}
public class Outer {
// local fields and methods of Outer
void method(/* parameters */) {
// local variables
class LocalClass {
// Fields and methods of the LocalClass
// can access local fields and methods of
// the Outer class, and final fields or
// parameters of the declaring method.
}
}
}

Anonymous
classes
An anonymous class is a local class definition that has no name and is used to immedi-
ately create an instance of that class. It is often employed to create a class that imple-
ments an interface without having to give a name to the class definition.
11 Summary
This document has provided some necessary introductory material on Java for those not
familiar with the language but have experience in C, C++ and are aware of the basic
principles of object-oriented programming.for the remainder of the book. The book
assume that the reader is familiar with sequential programming in Java.
public class Outer
// local fields and methods on Inner
Ordered method(/* parameters */) {
// local variables
return new Ordered()
{
// Here is the code for the anonymous class
// that implements the Ordered interface.
// Its fields and methods can access local fields
// and methods of the Outer class, and final
// fields or parameters of the declaring method.
}
}
}

Introduction to JAVA

More Related Content

What's hot (20)

Similar to Introduction to JAVA (20)

More from Professional Guru (20)

Recently uploaded (20)

Introduction to JAVA