PROGRAMMING LANGUAGE

HANDBOOK

Anthony Potts
David H. Friedel, Jr.

Chapter

Java Classes
and Methods

Java Classes
and Methods
Classes are the key Java components that
give the language its object-oriented
personality.

If you have some experience programming in a language like C++, you are

probably familiar with the power and flexibility that classes provide. They are
ideal for plugging general information into a template-like structure for reusing
over and over. For example, if you are developing an interactive drawing package, you could create standard classes for some of the fundamental drawing
operations and then use those classes to create more sophisticated drawing tasks.
If you are new to the world of object-oriented programming (OOP), youll soon
discover that classes are the essential building blocks for writing OOP applications. At first glance, Java classes look and operate like C++ classes; but there are
some key differences which well address in this chapter.
Well start by looking at the basics of classes. Youll quickly learn how classes are
defined and used to derive other classes. The second half of the chapter covers
methodsthe components used to breathe life into classes.

Understanding Classes
In traditional structured programming languages like C or Pascal, everything
revolves around the concepts of algorithms and data structures. The algorithms
are kept separate from the data structures, and they operate on the data to perform actions and results. To help divide programming tasks into separate units,
components like functions and procedures are defined. The problem with this
programming paradigm is that it doesnt allow you to easily create code that can
be reused and expanded to create other code.
115

116 Chapter 5

To solve this problem, object-oriented programming languages like Smalltalk

and C++ were created. These languages introduced powerful components called
classes so that programmers could combine functions (operations) and data under
one roof. This is a technique called encapsulation in the world of object-oriented
programming. Every language that uses classes defines them in a slightly different
way; however, the basics concepts for using them remain the same. The main
advantages of classes are:

They can be used to define abstract data types

Data is protected or hidden inside a class so other classes cannot access it
Classes can be used to derive other classes
New classes derived from existing classes can inherit the data and methods
already defineda concept called inheritance.

As youll learn in this chapter, the techniques for defining and using Java classes are
adapted from techniques found in the C++ language. In some cases, the Java syntax will look very similar to C++ syntax, but in other cases youll find a number of
differences, including new keywords that have been added to Java for declaring
classes and methods; restrictions, such as the elimination of pointers; and different
scoping rules that determine how classes can be used in an application.

Declaring a Class
If you recall from Chapter 2, we created a class named TickerTape, which controlled how text scrolled across the screen. Lets take a step back and look at the
full declaration used to define classes in Java:
[Doc Comment] [Modifier] class Identifier
[extends Superclassname]
[implements Interfaces] {
ClassBody;
}

Of course, keep in mind that you wont always use all of the clauses, such as Doc
Comment, Modifier, extends, and so on. For example, heres an example of the
worlds smallest class definition:
class Atom_ant {
int a = 1;
}

Java Classes and Methods

117

This class has an identifier, Atom_ant, and a body, int a = 1;. Of course, dont
try to compile this at home as is because it will only result in an error. Why?
Well, even though it is a valid class, it is not capable of standing on its own. (You
would need to set it up as an applet or a main program to make it work.)
A class declaration provides all of the information about a class including its
internal data (variables) and functions (methods) to be interpreted by the Java
compiler. In addition, class declarations provide:

Programmer comments
Specifications of the other classes that may reference the class
Specifications of the superclass the class belongs to (the classs parent)
Specifications of the methods the class can call

Using a Class
Before we move on and look at all of the other components used to declare
classes, lets return to our simple class declaration to see how classes are used in
Java programs. Once a class has been declared, you need to use it to create an
object. This process is called making an instance of a class. In a Java program
it requires two steps. First, you declare an object variable using a syntax that
looks just like a variable declaration, except the class name is used instead of the
name of a primitive data type. For example, this statement would use the
Atom_ant class we defined earlier to declare an object from the class definition:
Atom_ant crazyant;

Once the object has been declared, in this case crazyant, you then create an
instance of it in a Java application by using the new operator:
crazyant = new Atom_ant();

Now the object crazyant can access all of the components in a Atom_ant class,
thus making it an instance of an Atom_ant class. To see how this works in
context, lets expand our example:
class Atom_ant {
int a = 1;
}

// Simple class

118 Chapter 5
public class Bug {
int i = 10;
Atom_ant crazyant;

// Declare an object

public static void main (String args[]) {

// Create an instance of Atom_ant called crazyant
crazyant = new Atom_ant();
System.out.println("There are " + bug.i + " bugs here but only " +
crazyant.i + " atom ant.");
}

The output produced by this example would be:

There are 10 bugs here but only 1 atom ant.

The main class, Bug, creates an instance of the Atom_ant classthe crazyant
object. Then it uses the object to access the data member, a, which is assigned a
value in the Atom_ant class. Notice that the dot operator (.) is used to access a
member of a class.

Object Declaration Time Saver

In Java, you can both declare an object variable and create an
instance all in one statement. Heres an example of how it is done:
Atom_ant crazyant = new Atom_ant();

Notice that the class Atom_ant is used to declare the object

variable crazyant and then the new operator is used to create
an instance of Atom_ant.

Components of a Class Declaration

Lets look at the components of the class declaration in a little more detail. As
you recall from our first example, the only really necessary part of a class declaration is its name or identifier. However, whenever you need to reference your
class in your program to reuse it, youll need to reference it by its fully qualified
name. This name is the package name, or group of classes from which it came,
followed by the identifier. For example, if Atom_ant is the class name and it
belongs to a package named molecule, its fully qualified name would be
molecule.Atom_ant.

Java Classes and Methods

119

Documentation Comment
The Doc Comment clause of the class declaration is provided as an aid to help
other programmers who might need to use your class. It allows you to write your
documentation while youre writing the code. The comments you include as
part of this clause can easily be converted into easy to read HTML pages. However, keep in mind that your HTML pages will only be as good as your comments. (To brush up on how to write comments for Java programs, make sure
you read Chapter 3.)
Lets look at an example to see how the Doc Comment feature works. This class
definition
/**
* Atom ant is the world's smallest super hero,
so we gave him a class by himself.
* @author Dave Friedel
*/
class Atom_ant {
int i = 1;
}

uses Doc Comment style comments to produce the HTML page shown in Figure
5.1. Notice how the comments are formatted and used to document the class.
In this case, Atom_ant is a subclass under the java.lang.Object classthe default parent for all classes.
In case youre wondering, the @author notation is a special type of comment tag
that allows you to personalize your class. These tags are explained in more detail
in Chapter 3.

Class Modifiers
Modifiers define the rules for how classes are used in Java applications. They
determine how other packages, or classes of other groups can interact with
the current class. There are three kinds of modifiers that can be used in a
class declaration:

public
abstract
final

120

Chapter 5

Figure 5.1
The HTML documentation created for the Atom_ant class.

If you dont use one of these modifiers when declaring a class, Java will automatically decide that only other classes in the current package may access the
class. Lets look at how each of these modifiers are used.

PUBLIC CLASS
The public modifier is used to define a class that can have the greatest amount
of access by other classes. By declaring a class as public, you allow all other
classes and packages to access its variables, methods, and subclasses. However,

Java Classes and Methods

121

only one public class is allowed in any single Java applet or a single source code file.
You can think of the one public class in an applet as serving the role that the
main() function does in a C/C++ program.
The source code for an applet must be saved as ClassName.java, where ClassName
is the name of the single public class defined in the applet. Recall that when we
created the TickerTape applet in Chapter 2, the single public class was defined as
public class TickerTape extends Applet implements Runnable {...

and the name of the file was TickerTape.java.

Lets look at another example of how the public modifier is used to define a
Java class:
// Filename: Atom_ant.java
public class Atom_ant {
public static void main (String args[]) {
System.out.println("Hello World");
}
}

In this case, Atom_ant is the name of the class and the filename for the applet is
Atom_ant.java.

ABSTRACT CLASS
The abstract modifier is used to declare classes that serve as a shell or placeholder for implementing methods and variables. When you construct a hierarchy of classes, your top most class will contain the more general data definitions
and method implementations that represent your programs features. As you
work your way down the class hierarchy, your classes will start to implement
more specific data components and operations. As you build your hierarchy,
you may need to create more general classes and defer the actual implementation
to later stages in the class hierarchy. This is where the abstract class comes in.
This approach allows you to reference the operations that you need to include
without having to restructure your entire hierarchy.
The technique of using abstract classes in Java is commonly referred to as single
inheritance by C++ programmers. (By the way, limited multiple inheritance techniques can also be implemented in Java by using interfaces. Well cover this
topic in more detail in Chapter 6.)

122 Chapter 5

Any class that is declared as an abstract class must follow certain rules:

No objects can be instantiated from an abstract class

Abstract classes must contain at least one declaration of an abstract method
or variable
All abstract methods that are declared in an abstract class must be implemented in one of the subclasses beneath it
Abstract classes cannot be declared as final or private classes

Lets look at an example of how an abstract class is defined and used to help
create other classes:
abstract class Quark extends Atom_ant {
...
abstract void abstract_method1();
abstract void abstract_method2();
void normal_method();
...
}
public class Aparticles extends Quark {
public void abstract_method1() {
... // Definition of the method
}
}
public class Bparticles extends Quark {
public void abstract_method2() {
... // Definition of the method
}
}

Here, the class Quark is declared as an abstract class and it contains two methods that are declared as abstract methods. The subclassesAparticles and Bparticles
are located beneath the class Quark in the hierarchy of classes. Each one defines
a method based on one of the abstract methods found in the Quark class. A
compile-time error would occur if we had failed to define both of the abstract
methods in the Quark class. All abstract methods must be defined in the subclasses that are derived from abstract classes.

Java Classes and Methods

123

Restrictions in Declaring Abstract Classes

An abstract class cannot be defined as a final class (using the
final keyword) because the Java compiler will always assume
that the abstract class will be used to derive other classes
other subclasses will follow it. (As youll see in the next section, a final class defines the end of the line for a class hierarchy.)
Furthermore, you cannot used a private modifier in an abstract classs method declarations because this modifier restricts
methods from being used by any other classes except the class
they are defined in.

FINAL CLASS
The final modifier is used to declare a class that will not be used to derive any
other classes. The final class is like the last station on a railway line. By its position in a class hierarchy, a final class cannot have any subclasses beneath it. In
final class declarations, you cannot use the extends clause because the Java compiler always assumes that a final class cannot be extended. Heres an example of
what would happen if you tried to declare a final class and then use it in another
class declaration:
final class Molecule extends Element {
static String neutron = "molecule";
}
class Atom_ant extends Molecule {
static String proton = "atom_ant";
}

Compiling...
E:\java\jm\element.java
E:\java\jm\element.java:12: Can't subclass final classes: class
Moleculeclass Atom_ant extends Molecule {
^1 errorsCompile Ended.

In this case, Molecule has been defined as a final class. But notice that the second class definition, Atom_ant, attempts to use Molecule as its parent. The Java
compiler catches this illegal declaration and provides the appropriate warning.

124 Chapter 5

Class Identifiers
Each class you define in a Java program must have its own unique identifier. The
classs identifier or name directly follows the class keyword. The rules for naming classes are the same as those used to name variables. To refresh your memory,
identifiers should always begin with a letter of the alphabet, either upper or
lower case. The only exception to this rule is the underscore symbol (_) and the
dollar sign ($), which may also be used. The rest of the name can be defined
using characters, numbers, and some symbols.
Since class names are also used as file names, you need to create names that will
not cause problems with your operating system or anyone who will be using
your program.

Extending Classes
In most Java applets and programs you write, you will have a number of classes
that need to interact each otherin many cases classes will be derived from other
classes creating hierarchies. The keyword that handles the work of helping you
extend classes and create hierarchies is named appropriately enough, extends.
In a class hierarchy, every class must have a parentexcept the class that is at
the top. The class that serves as a parent to another class is also called the superclass of the class it derivesthe class that takes the position immediately above
the class. Lets look at an example. As Figure 5.2 indicates, the classes 911, 944,
and 928 all belong to the superclass Porsche. And Porsche belongs to the superclass sportscar, which in turn belongs to the superclass automobile.
When you derive a class from a superclass, it will inherit the superclasss data and
methods. (For example, 911 has certain characteristics simply because it is derived
from Porsche.) To derive a class from a superclass in a class declaration hierarchy,
you will need to use the extend clause followed by the name of the superclass. If no
superclass is defined, the Java compiler assumes that you are deriving a class using
Javas top-level superclass named Object. Here is an example:
public class Element extends Object {
public static void main() {
Atom_ant ATOMOBJ = new Atom_ant();
Molecule MOLEOBJ = new Molecule();
System.out.println(ATOMOBJ.proton);

Java Classes and Methods

125

Automobile

Sports car

Porsche

911

944

928

Figure 5.2
A sample class hierarchy.
}
}
class Molecule extends Element {
static String neutron = "molecule";
}
class Atom_ant extends Molecule {
static String proton = "atom_ant";
}

In this class declaration section, the top-level class defined is Element. Notice
that it is derived or extended from Objectthe built-in Java class. The first
line of the declaration of Element could have also been written as
public class Element {
...

since the Java compiler will assume that a class is automatically derived from the
Object class if the extends clause is omitted. The second class, Molecule, is
derived from Element and the third class, Atom_ant, is derived from Molecule.
As Figure 5.3 shows, both Molecule and Atom_ant inherit the components of
the Element class.

126 Chapter 5

Using the implements Clause to Create

Class Interfaces
When classes are used to derive other classes, the derived classes can access the
data and methods of the classes higher up in the hierarchy chain. Fortunately,
Java provides a mechanism called interfaces so that classes that are not part of a
hierarchy can still access components of other classes. An interface is created for
a class by using the implements clause. A class can implement as many interfaces as it wishes, but all the interfaces introduced must have all their methods
defined in the body of the class implementing it. Thus, all the subclasses that
follow from that point on will inherit the methods and variables defined.
Lets develop the Atom_ant class we introduced in the previous section to see
how an interface can be coded:

Element

Molecule

Figure 5.3
Using the extends keyword to derive a series of classes.

Atom_ant

Java Classes and Methods

127

class Atom_ant extends Molecule implements Protons, Neutrons, Electrons {

static int proton = 45378444;
void Proton_function() {
... // definition of the Proton_function()
}
void Neutron_function() {
... // definition of the Neutron_function()
}
void Electron_function() {
... // definition of the Electron_function()
}
}

Here we are making the assumption that the interfaces Protons, Neutrons, and
Electrons only have one method declared in each of the interfaces. For example,
Protons may be set up as follows:
Public interface Protons {
void Proton_function(); // declares the method that will be used
}

As you can see, setting up the interface is a two step process. The class where the
methods are defined uses the implements clause to indicate which interfaces
can have access to the methods. Then, interface statements are used to declare
the method that will be used.
If you recall from Chapter 2, the TickerTape class implemented the interface
Runnable from the package java.lang. The Runnable interface has only one
method declared in it, which is run(). This method is then defined in the class
that is implementing it. In this case, the applet TickerTape has defined run() to
instruct the thread to sleep, call the setcoord() method, and rerun the paint()
method every time the applet calls the run() method. This happens in situations
where the screen is resized or, in this case, where the applet is instructed to move
the text across the screen and the run() method is called.
// TickerTape Applet
import java.applet.*;
import java.awt.*;

128 Chapter 5
// TickerTape Class
public class TickerTape extends Applet implements Runnable {
...
public void run() {
while(ttapeThread != null){ // verifies the thread is still active
try {Thread.sleep(50);} catch (InterruptedException e){}
setcoord(); // changes the placement of the text
repaint();
// repaints the screen by activating the paint()
// method
}
}
...
} // End TickerTape

This allows the ability to effectively encapsulate(hide) the classes and all its methods that actually support the run() method. Interfaces allow for distinct behaviors, defined by the programmer, to be used without exposing the class(es) to
everyone. Well discuss these techniques in more detail in Chapter 6.

Class Body
The class body contains the code that implements the class. This is where you
provide the detail for the actions the class needs to perform (methods) and the
data it needs to use (variables). The body can also contain constructors (special
methods) and initializers. The basic format for a class body is:
{
Variable-declarations;
Method-declarations;
}

The variable declarations can be any standard Java declaration (see Chapter 3
and the material presented at the end of this chapter if you need a review). Later
in this chapter well discuss how methods are declared and used. Heres an example of a class with a body:
public class TickerTape extends Applet implements Runnable {
// Beginning of class body
String inputText;
String animSpeedString;
Color color = new Color(255, 255, 255);

Java Classes and Methods

129

int xpos;
...
// Methods
public void paint(Graphics g) {
paintText(osGraphics);
g.drawImage(im, 0, 0, null);
}
...
// End of Class Body
}

NAME SPACE
Every method and variable defined in a class is recorded into an area called a
name space. This name space is then inherited by the other classes in a class
hierarchy which are derived from the class. If a variable or method has been
previously defined elsewhere in the structure with the same name, a shadowing
effect occurs for that level. To access the value of a variable that supersedes the
current value, you need to put the prefix clause super in front of the variable
name. This clause instructs the expression to take the value of the superclass.
To access the value of the current variable, you use the prefix this. Lets look at
an example:
public class House extends Object {
void main() {
Room();
}
}

static int tvamount = 8; // Variable

public class Room extends House {

static int tvamount = 5; // Variable
int Child = this.tvamount; // Child equals 5same as saying tvamount
int Parent = super.tvamount;
// Parent equals 8
}

In this example the House class is derived from the standard Object class. Then,
the Room class is derived from House. Now notice that each class defines a
variable named tvamount and assigns it a value. In the second assignment statement in Room, the variable Child is assigned the value 5 because this is used to
access the classs local copy of the tvamount variable. In the next assignment
statement, notice how super is used to access the value tvamount was assigned
in Housethe superclass.

130 Chapter 5

Methods
As weve seen, the mechanisms used to implement operations in classes are called
methods. This terminology is borrowed directly from object-oriented languages
like Smalltalk and C++. Methods define the behavior of a class and the objects
created from the class. A method can send, receive, and alter information to
perform a task in an application. Java requires that every method be defined
within a class or interface, unlike C++ where methods (functions) can be implemented outside of classes.
Lets refer to the car class hierarchy we presented earlier in this chapter to get a
better understanding of the role methods play. All of the cars we introduced
have doors and we could define two methods to operate on these doors: open
and close. These same methods could be designed to perform operations on
other car components such as windows, trunks, hoods, and so on. A component
like a door can be viewed as an object. Of course, a car would be made up of
many objects and many methods would be required to process all of the objects.
As a programmer, it would be up to you to decide how to arrange the objects
you need and what methods must be implemented.

Declaring a Method
If you recall from Chapter 2, our TickerTape applet included a number of methods. The first one defined was the init() method, which was responsible for
initializing the applet upon loading. Lets take a step back and look at the full
declaration used to define a Java method:
[Modifier] ReturnType Identifier([ParameterList]) [Throws]
{
MethodBody;
}

The Modifier and Throws clauses are optional. They are used to specify how the
method needs to be accessed and which exceptions should be checked for. (For
more information on exceptions and how to catch errors, refer to Chapter 7.)

Components of a Method Declaration

If you were to break down the method declaration, you would find it performs
three main tasks:

Java Classes and Methods

131

It determines who may call the method

It determines what the method can receive (the parameters)
It determines how the method returns information

Method Modifiers
Earlier in this chapter, you learned that a set of modifiers are available for defining how classes can be accessed. Methods also can be defined using modifiers,
although the method modifiers only affect how methods are used, not the class
they are defined in. Java provides eight modifiers for defining methods, but only
one modifier from each of the groups listed next may be used in a method
declaration. For example, you cannot use a public and private modifier in the
same declaration. Here is the complete set of method modifiers:

public, protected, private

static
abstract, final, native, synchronized

Keep in mind that it doesnt make sense to use some modifiers in one group with
modifiers from another group. For example, a method that is defined using the
private and abstract modifiers contradicts itself. An abstract method is one that
requires its actual code to be defined in the subclasses that follow, whereas a private
method is one that can only be accessed in the class it is defined in. The rule of
thumb when choosing and combining modifiers is that you need to make sure
that they are complementary rather than contradictory. If a modifier is not used,
the method may be accessed only by the classes that are in the current package.

PUBLIC METHOD
A method declared as public can be accessed by any class in the same package. It
can also be accessed by other classes from other packages. This modifier gives a
method the most freedom.

PROTECTED METHOD
A method declared as protected can only be used by other classes within the
same package. All the subclasses beneath the class the method is defined in may
access the method unless shadowing occurs. Shadowing involves naming a method
using a name that already exists in a superclass above the class the method is
defined in.

132 Chapter 5

PRIVATE METHOD
A method declared as private is one that can only be accessed by the class it is
defined in. This modifier gives a method the least amount of freedom.

STATIC METHOD
A method declared as static is one that cannot be changed. This type of method
is also referred to as a class method, because it belongs explicitly to a particular
class. When an instance of the class that defines the method is created, the static
method cannot be altered. For this reason, a static method can refer to any other
static methods or variables by name. Limitations of static methods to keep in
mind are that they cannot be declared as final, and they cannot be overridden.

ABSTRACT METHOD
A method declared as abstract is one that must be defined in a subclass of the
current class. However, an abstract method must be declared in the current class
with a (;) semicolon in place of the methods block of code. Methods that are
declared abstract are not required to be implemented in every subclass.

FINAL METHOD
A method declared as final is one that ends the hierarchical tree. No methods
having the same name can be defined in subclasses that exist below the class that
declares the method as final.

NATIVE METHOD
A method declared as native is one that will be implemented using outside code
code that is written in another language, to be used in conjunction with your
current program. This limits you to a specific platform and restricts you from
creating Java applets. Native methods are declared by leaving out the method
body and placing a semicolon at the end of the method declaration.

SYNCHRONIZED METHOD
A method declared as synchronized limits it from being executed by multiple
objects at the same time. This is useful when you are creating Java applets and you
could have more than one thread running at the same time accessing one central
piece of data. If the method is static (e.g., a class method), the whole class would
be locked. If you just declare a particular method as synchronized, the object containing the method would only be locked until the method finishes executing.

Java Classes and Methods

133

Return Type of a Method

Any information that is returned from a method is declared as the return type.
This assures that the information that is returned from a method call will be of
the correct type; otherwise, a compile-time error will be generated. If no information will be returned by a method, the void keyword should be placed in
front of the method name. The different data types that may be returned by
methods are covered in Chapter 4.

Parameter Lists for a Method

The parameter list consists of the ordered set of data elements passed to a method.
You can pass zero, one, or multiple parameters by listing them between the
parentheses, with each type and variable name being separated by a comma. If
no parameters are passed, the parentheses should be empty. All variables that are
passed become local for that instance of the method. Heres an example of how
methods can be declared with and without parameters:
public static void MyFirstMethod(String Name, int Number) {
...
// the String variable Name is assigned whatever is passed to it
// the integer variable Number is assigned whatever is passed to it
...
}
public static void MyFirstMethod() {
...
// Nothing is passed to it.
...
}

Method Throws
The throws clause is used to specify the type of error(s) that will be handled within
a method. In effect, it is used to help you set up an automatic error-handler. In the
event of an error, the error must be assignable to one of the exceptions in either the
Error, RunTimeException, or Exception classes. (These are special classes that
Java provides for catching compile-time and run-time errors. Well cover them in
more detail in Chapter 7.) Each method you declare does not need to use the
throws clause in its declaration, but in the event of an error, the omission of this
clause will leave the error handling up to the Java compiler or the Java interpreter.
Lets look at an example of how the throws clause is used.

134 Chapter 5

In the following method declaration, the Java exception named

ArrayOutOfBoundsException is specified so that in the event an array range
error occurs, the method will know how to handle the error:
public class Array_check() {
String arr[5];
public static void main(void) throws ArrayOutOfBoundsException {
int i=0;
char ch;
// Specify which code should be tested
try {
while (i <= 5) ch = arr[i++];
}
// An error has occurreddisplay a message
catch {
System.out.println("Array out of bounds");
}
}
}

At some point main() will try to access a location outside the legal range of the
array arr[]. When this happens, an exception will be thrown and the catch
clause will handle it. Also notice the use of the try clause which is needed to
specify which code in the method should be tested. In our case, we want to
check each iteration of the while loop.

Method Body
All executable code for Java classes is contained in the body of the methods.
Unless a method is declared as abstract, native, or is declared in the body of an
interface, the code for the method is placed between a pair of curly braces. This
code can be any valid Java statements including variable declarations, assignment statements, method calls, control statements, and so on.
Heres an example of how a basic method is defined:
public int SimpleMethod(int Number) {
// The integer variable Number is assigned whatever is passed to it
int lowrange = 1;

// Local declarations for the method

Java Classes and Methods

135

int highrange = 10;

if (Number <= lowrange) return -1;
if (Number >= highrange) return 100
else return 50;
}

In this case, the methods name is SimpleMethod(). Because it is declared as

public, it can be used by any class in the package in which the method is defined. The return type for the method is int and it accepts one int parameter.
The method body contains a few local declarations and a set of if-then decisionmaking statements.
For a method declared as abstract, native, or one that is declared in an interface,
the body is left blank and the declaration is terminated with a semicolon. The
bodies are then defined elsewhere depending on how they are declared. Heres
an example:
abstract class Aparticles extends Quark {
abstract int abstract_method();

// Defined in the subclasses of the class

native void native_method ();

// Defined in an external process

public String normal_method() {

... // Definition of the method
}
}

Using the this and super Keywords

To access class variables and methods from within an object, you can reference
them by using the keywords this and super. When the Java compiler encounters
the this keyword in the body of a method, it knows that you are accessing other
methods and variables defined within the scope of the class the method is defined in. On the other hand, variables and methods that are available for accessing in the parent class (superclass) to the current class are referenced using the
super keyword. Heres an example of how each of these keywords can be used:
class Atom_ant extends Molecule {
int Number;
...
}

136 Chapter 5
class Quark extends Atom_ant {
int Proton;
int Neutron;
String Electon = "Negative attraction";
...
void Count() {
System.out.println(this.Proton + " is the number of Protons"); // Correct
System.out.println(Neutron + " is the number of Neutrons"); // Correct
System.out.println(super.Number + " is the number of Atoms"); // Correct
System.out.println(Atom_ant.Number + " is the number of Atoms");
// Correct
...
}
}

In this example, this.Proton references the local variable Proton defined in the
class Quark. But take a look at the second method call in the Count() method.
Here, the variable Neutron, which is also declared in Quark, is referenced without the use of the this keyword. What gives? Actually, since both of these variables are defined within Quark, the this keyword is not really needed.
As for the two following lines of code, they each reference the Number variable declared in the Atom_ant class, which serves as the parent to the Quark
class. Notice that the keyword super is placed in front of the variable Number
to allow it to be accessed. This is the same as using the superclass name in the
statement Atom_ant.Number to reference the value of Number. Superclass
names can be referenced further up the hierarchical tree but the super keyword can only be used to access class members that reside one level above the
current class. If the Molecule class contained a variable named M1, and we
wanted to reference it from the Quark class, a statement like this would be
required:
Proton = Molecule.M1;

Here the superclass named Molecule is included in the assignment statement. If

it was omitted or the super keyword was used instead,
Proton = super.M1;

the Java compiler would return an error because it would try to locate the M1
variable in the class that is directly above the Quark class.

Java Classes and Methods

137

Overloading and Overriding Methods

A method may be declared with multiple declarations, each specifying different
types and arguments that may be passed to the method. The context in which the
method is called will determine which actual method code is used. The techniques of
using a methods name more than once to define an operation in a class involves
overloading and overriding methods. As long as you can define each method having the same name so that it can be distinguished from the others sharing the same
name, the Java compiler will not give you an error. The technique for creating
overridden methods involves using different parameters (types and numbers) and
return types. Methods that are inherited from a superclass may be overridden but
the overriding method must provide at least the same access.
Lets look at some examples of how we can override methods:
class Atom_ant extends Molecule {
int Number;
protected void Count(String Astring, int Number) {
...
}
}
class Quark extends Atom_ant {
int Proton;
int Neutron;
String Electon = "Negative attraction";
...
public void Count(int Number, String Astring) { // Correct
...
}
protected void Count() {
...
}

// Correct

Here weve declared two classes: Atom_ant and Quark. Atom_ant serves as the
superclass. The method that is overridden is Count(). It is first introduced as a
protected method in the Atom_ant class. Notice that it is declared here as taking
two parameters: Astring and Number. Because Atom_ant is declared as a protected method, it is restricted from being called by other classes outside of the
package Atom_ant is declared in.

138 Chapter 5

The Quark class, which is derived from Atom_ant, provides two new variations
of the Count() method, each one being overridden from the base method defined in Atom_ant. Notice that each of the overridden methods uses different
parameters and/or return types than the original method.
To see how the different versions of the Count() method can be called, lets
expand the Quark class a little:
class Atom_ant extends Molecule {
int Number;
protected void Count(String Astring, int Number) {
...
}
}
class Quark extends Atom_ant {
int Proton;
...
public void Count(int Number, String Astring) { // Correct
...
}
void check() {
Atom_ant.Count("Hello", 5); //Correct refer to superclass method
super.Count("GoodBye", 5); //Correct same as previous
Molecule.Count("Hello World"); //Correct as long as it exists
Count(5, "World");
//Correct same as this.Count
}
}

The first two calls to the Count() method result in calling the Count() method
defined in Atom_ant. For the third call, we are making the assumption that the
class Molecule, which Atom_ant is derived from, contains a Count() method. If
it doesnt, a compiler error will occur. The last call to Count() accesses the method
of the same name defined in Quark.

ConstructorsThe Special Methods

Although constructors are identified as special methods, it is important to distinguish between the two. Methods define an objects behavior in terms of what operations the object can perform. A constructor, on the other hand, determines how an
object is initialized by creating a new instance of a class with specified parameters.

Java Classes and Methods

139

Methods and constructors actually differ in three ways. First, constructors do

not have their own unique names; they must be assigned the same name as their
class name. Second, constructors do not have a return typeJava assumes that
the return type for a constructor is void. And third, constructors are not inherited by subclasses, as are methods.
To understand how constructors work conceptually, lets return to the car analogy we introduced earlier in this chapter. Each car in our hierarchy represents an
object and the blueprint for each car is a class structure. Also recall that operations such as opening and closing car doors were considered to be our methods.
Now, imagine that we have a subclass, called BodyShop, which defines the
body style for a car. This class could be inserted under the general car class in
the class hierarchy. An object could be created from this class called
FrameCreation, which is responsible for making body frames for cars. The
process of building a frame could involve first calling a constructor to do the
dirty work of setting up the shop for building a particular car frame. The
manner in which the different classes are defined in the hierarchy will determine what frame a particular car gets at the BodyShop from the FrameCreation
team. (The FrameCreation team is responsible for initializing an object depending on the information passed to a constructor. )
Now lets assume we have three choices for making body frames:

4 Door(integer) Falcon(String)
3 Door(integer) Pinto(String)
2 Door(integer) Mustang(String), which is the default.

We could just say 2, 3, or 4 doors, but the FrameCreation team insists on a certain
format for each. The Falcon requires (integer Doors, String Name), the Pinto requires (String Name, integer Doors), and the Mustang doesnt require any values
(). When you pass these values, known as types to the FrameCreation team, they
immediately know which frame to create, or initialize, by the arrangement of the
information passed to them (data types and number of parameters). By passing the
information in a distinct pattern FrameCreation(Doors, Name), FrameCreation(Name,
Doors), or FrameCreation() to create an object, we are using a constructor.
A constructor is a special method that determines how an object is initalized
when created. The constructor is named the same as the class it follows. The
code for our example could be written like this:

140 Chapter 5
class FrameCreation extends BodyShop {
// ** Initializing the object newcar **
FrameCreation newcar = FrameCreation(4 , Falcon);
// ** The Beginning of the Constructor **
FrameCreation {
// ** An example of Overloading the Constructor **
FrameCreation(int, String) {
// Creates the FALCON
}
// ** An example of Overloading the Constructor **
FrameCreation(String, int) {
// Creates the Pinto
}
FrameCreation() {
// ** An example of Overloading the Constructor **
// Creates the Mustang
}
// ** The End of the Constructor **
}

FrameCreation is the constructor, which is declared multiple timeseach taking

different parameter configurations. When it is called with a configuration (a
number, a word), the constructor with the matching configuration is used.
In calling a constructor, you need to disregard the rules for calling methods.
Methods are called directly; constructors are called automatically by Java. When
you create a new instance of a class, Java will automatically initialize the objects
instance variables, and then call the classs constructors and methods. Defining
constructors in a class can do several things, including:

Setting initial values of the instance variables

Calling methods based on the initial variables
Calling methods from other objects
Calculating the initial properties of the object
Creating an object that has specific properties outlined in the new argument
through overloading

Components of a Constructor Declaration

The basic format for declaring a constructor is:

Java Classes and Methods

141

[ConstructorModifier] ConstructorIdentifier([ParameterList]) [Throws] {

ConstructorBody;
}

As with the other declarations weve introduced in previous sections, only the
identifier and body are necessary. Both the modifier and the throws clause are
optional. The identifier is the name of the constructor; however, it is important
to remember that the name of the constructor must be the same as the class
name it initializes. You may have many constructors (of the same name) in a
class, as long as each one takes a different set of parameters. (Because the different constructors in a class must have the same name, the type, number, and
order of the parameters being passed are used as the distinguishing factors.) For
example, all constructors for a class named Atom_ant, must be named Atom_ant,
and each one must have different set of parameters.
In addition to having a unique declaration from that of a method, a special
format is used for calling a constructor:
Typename([ParameterList]);

The only required element is Typename, which names the class containing
the constructor declaration. Heres an example of a constructor, with the
class Atom_ant and a constructor that uses the new operator to initialize
instance variables:
class Atom_ant {
String superhero;
int height;
Atom_ant(String s, int h) {
superhero = s;
height = h;
}

// Declare a constructor

void printatom_ant() {
System.out.print("Up and attam, " + superhero);
System.out.println("! The world's only " + height +
" inch Superhero!");
}
public static void main(String args[])
Atom_ant a;

142 Chapter 5
a = new Atom_ant("Atom Ant" , 1); // Call the constructor
a.printatom_ant();
System.out.println("------");
a = new Atom_ant("Grape Ape", 5000);
a.printatom_ant();
System.out.println("------");
}
}

The output for this program looks like this:

Up and attam, Atom Ant!
-----Up and attam, Grape Ape!
------

The world's only 1 inch Superhero!

The world's only 5000 inch Superhero!

Notice that each constructor call is combined with the new operator. This operator is responsible for making sure a new instance of a class is created and
assigned to the object variable a.

USING JAVAS DEFAULT CONSTRUCTOR

If you decide not to declare a constructor in a class, Java will automatically
provide a default constructor that takes no arguments. The default constructor
simply calls the superclass constructor super() with no arguments and initializes
the instance variable. If the superclass does not have a constructor that takes no
arguments, you will encounter a compile-time error. You can also set a classs
instance variables or call other methods so that an object can be initialized.
Here is an example of a Java class that does not use a constructor but instead
allows Java to initialize the class variables:
class Atom_ant2 {
String superhero;
int height;
Boolean villain;
void printatom_ant() {
System.out.print("Up and attam, " + superhero);
System.out.println("! The world's only " + height +
" inch Superhero!");
}
public static void main(String args[])
Atom_ant2 a;

Java Classes and Methods

143

a = new Atom_ant2();
a.printatom_ant();
System.out.println("------") ;
}
}

Because no constructor is defined for this example program, the Java compiler
will initialize the class variables by assigning them default values. The variable
superhero is set to null, height is initialized to zero, and villain is set to false.
The variable a, in the main() method, could have been initialized at the time the
constructor was called by substituting the code a = new Atom_ant2(); for
Atom_ant2 a = new Atom_ant2();. Either statement provides an acceptable
means of creating an instance of a classthe object a. Once this object is in
hand, the method printatom_ant() can be called.
The output for this program looks like this:
Up and attam, The world's only 0 inch Superhero!
------

CONSTRUCTOR MODIFIERS
Java provides three modifiers that can be used to define constructors:

public
protected
private

These modifiers have the same restrictions as the modifiers used to declare standard methods. Here is a summary of the guidelines for using modifiers with
constructor declarations:
A constructor that is declared without the use of one of the modifiers may
only be called by one of the classes defined in the same package as the constructor declaration.
A constructor that is declared as public may be called from any class that has
the ability to access the class containing the constructor declaration.
A constructor that is declared as protected may only be called by the subclasses of the class that contains the constructor declaration.
A constructor that is declared as private may only be called from within the
class it is declared in.

144 Chapter 5

Lets look at an example of how each of these modifiers can be used:

class Atom_ant2 {
String superhero;
int height;
String villain;
int numberofsuperheros;
Atom_ant2() {
this("Dudley Do Right", 60);
}
public Atom_ant2(String s, int h) {
superhero = s;
height = h;
}
protected Atom_ant2(int s, int h) {
numberofsuperheros = s;
height = h;
}
private Atom_ant2(String s, int h, String v) {
superhero = s;
height = h;
villain = v;
}
void printatom_ant() {
System.out.print("Up and attam, " + superhero);
System.out.println("! The world's only " + height +
" inch Superhero!");
}
public static void main(String args[]) {
Atom_ant2 a;
a = new Atom_ant2();
a.printatom_ant();
a = new Atom_ant2("Grape Ape", 5000);
a.printatom_ant();
}
}
class Molecule_mole extends Atom_ant2 {
String superhero;
int height;

Java Classes and Methods

145

public static void main(String args[]) {

Atom_ant2 a;
a = new Atom_ant2(); // Compile-time Error
a.printatom_ant();
a = new Atom_ant2("Atom Ant", 1);
a.printatom_ant();

// Correct

a = new Atom_ant2(5, 5); // Correct

a.printatom_ant();
// Compile-time Error
a = new Atom_ant2("Atom Ant", 1 , "Dudley Do Right");
a.printatom_ant();
}
}

In this example, the Atom_ant2 class uses constructors with all three of the
modifiers: public, protected, and private. In addition, a constructor is declared
that does not use a modifier. Notice how the constructors are called from the
Molecule_mole class. Each constructor type is both defined and called using a
different parameter configuration. (This is how the Java compiler knows which
constructor to use.)
The first constructor call, Atom_ant2(), produces a compiler error because of
Javas scoping rulesthe declaration of this constructor is outside of the range of
the Molecule_mole class, and the constructor was not declared as public or protected. Also notice that the call to the fourth constructor produces a compiler
error. In this case, the constructor was declared in the Atom_ant class as private,
which limits the constructor from being called by the class it is declared in.
As this example illustrates, you need to make sure you understand the restrictions that modifiers can place on method declarations. For example, here is an
example of a compile-time error you will encounter if you try to access a constructor from another class when its modifier has been declared as private:
Compiling...
E:\java\jm\Molecule_mole.java
E:\java\jm\Molecule_mole.java:8: No constructor matching _
Atom_ant2(java.lang.String, int, java.lang.String) found in class Atom_ant2.
a = new Atom_ant2("Atom ant",5,"Dudley");
^1 error
Compile Ended.

146 Chapter 5

Parameter List and Throws Clause

Both the parameter list and throws clause follow the same rules used for declaring and calling methods; after all, a constructor is just a special method. When
calling a constructor, different parameter configurations (type of parameters and
quantity) can be used as long as you have a matching declaration that uses the
same parameter configuration.

Constructor Body
The body of the constructor is essentially the same as the body of a method. The
only difference occurs in the first statement. If the constructor is going to call
itself (an alternate constructor for the same class having the same name) or call
the constructor of its superclass, it must do this in the first statement. To access its
own class, the this() statement is used as a placeholder for the classs identifier. To
refer to the classs superclass, the super() statement is used. Following each of the
clauses are parentheses containing the parameter list to be passed to the constructor, identified by the keyword. Here is an example of how both the this() and
super() statements are used within the constructors defined for Atom_ant2:
class Atom_ant2 extends Quark {
String superhero;
int height;
String villain;
int numberofsuperheros;
Atom_ant2() {
this("Atom Ant", 1);
}

// Call another Atom_ant2() constructor

public Atom_ant2(String s, int h) {

superhero = s;
height = h;
}
Atom_ant2(String s, int h, String v) {
super(s, h);
// Call the superclass's constructor
}
protected Atom_ant2(int s, int h) {
numberofsuperheros = s;
height = h;
}

Java Classes and Methods

147

synchronized void printatom_ant() {

System.out.print("Up and attam, " + superhero);
System.out.println("! The world's only " + height +
" inch Superhero!");
System.out.print("\n-----\n");
}
public static void main (String args[ ])
Atom_ant2 a;

a = new Atom_ant2();
a.printatom_ant();
System.out.println ("------") ;
}
}

When the program runs, the call to Atom_ant2() results in the first constructor
defined in the Atom_ant2 class being called. Then, the first constructor calls the
second constructor defined in the class. This process is illustrated in Figure 5.4.
In the first constructor, this() is used so that the constructor can directly call one
of Atom_ant2s other constructors. How does the compiler know which one to
use? It looks for a match between the parameters based on this(Atom Ant, 1)
and one of the other Atom_ant2(...) constructors. Since the this() statement
passes a string and an integer, the actual constructor that is called is the second
one defined in the Atom_ant2 class.
In the third constructor declaration, the super() statement performs a similar
operation except this time it searches the immediate superclasss constructor for
Class Atom_ant2
. . .

A
Atom_ant2( )

call first
constructor

A
Atom_ant2(s, h)
h)

a= new Atom_ant2(
)
A

. . .
}

Figure 5.4
The chain of constructor calls in the Atom_ant2 example.

call second
constructor

148 Chapter 5

a match. It is important to remember that when using either of these statements, you may not directly call instance variables of the object being created.
Furthermore, an instance variable cannot be dependent upon another variable
that has not yet been defined, or is defined after it.
Heres an example:
class Foo {
int variableNow = variableLater + 10;
int variableLater = 20;
}

As you can see, variableNow is trying to initialize itself before variableLater is

assigned a value.

Object Creation
There are two ways to create an instance of a class: use a literal, specific to the
String class or use the new operator. The new operator is placed in front of the
constructor. The parameter list of the constructor determines what constructor
is used to create an instance of an object.
...
public static void main(String args[])
Atom_ant2 a;

a = new Atom_ant2();
a.printatom_ant() ;
System.out.println ("------");
}
...

Here, the new operator initializes Atom_ant2 with an empty parameter list, initializes the variable to create an instance of the class Atom_ant2, and assigns it to a.

Variables for Classes

If youve read Chapter 3, you should already be familiar with the basics of declaring variables in Java. Lets refresh your memory: A variable is a named storage location that can hold various values, depending on the data type of the
variable. The basic format for declaring a variable is as follows:

Java Classes and Methods

149

VariableModifiers Type Indentifier = [VariableInitializer];

Only the Type and Identifier components are necessary. The modifiers are optional.
As with all the identifiers weve used throughout this chapter, the variable identifier simply names the variable. However, you can name any number of variables in the declaration by naming them in the identifier position and separating
them with commas. If you decide to declare multiple variables, also realize that
the modifiers and Type apply to all the variables that are named. For example, in
these declarations
int paul, david, kelly;
static String henry, diana;

the variables paul, david, and kelly are declared as integers, and the variables
henry and diana are declared as static strings.

VARIABLE MODIFIERS
Java provides seven different modifiers for declaring variables within classes. However, you can only use two of themone from each groupin a declaration.
Also, you cant use two modifiers that contradict each other in the same declaration. The two groups of modifiers are:

public, protected, private

static, final, transient, volatile

The public, protected, and private modifiers are discussed under the modifiers
sections of class, method, and constructors.

STATIC MODIFIERS
A static variable is also known as a class variable. This is because there is only one
variable of that name, no matter how many instances of the class are created.
Heres an example of how the static modifier can be used:
Atom_ant2() {
static int Doug = 9;
this("Atom Ant", 1);
}

150 Chapter 5
...
public static void main(String args[])
Atom_ant2 a, b, c, d;

a = new Atom_ant2();
b = new Atom_ant2();
c = new Atom_ant2();
d = new Atom_ant2();
a.printatom_ant() ;
System.out.println("------") ;
}
...

Here, no matter how many objects we create, there is exactly one variable Doug
for every instance of Atom_ant().

FINAL MODIFIER
When a variable is assigned final, it acts as a constant throughout the instance of
the class. They must be declared at time of initialization of the class or method.

TRANSIENT MODIFIER
This is a modifier that has been reserved by Java virtual machine language for
low level segments that do not pertain to the persistent state of an object. Other
implementations will follow for this modifier in future versions.

VOLATILE MODIFIER
These are modifiers that are processed through the multi-processor in an asynchronous manner. The variables are reloaded from and stored to memory every
time the variables are used.

The Art of Casting with Classes

When we introduced the fundamental data types in Chapter 3, we showed you
how to use casting techniques to convert the values assigned to variables of predefined data types to other data types. For example, in a set of statements like this
int i;
short s;

Java Classes and Methods

151

s = 10;
i = (int) s;

the contents of the variable soriginally defined to be of the short typeis

converted to an int type by using a cast in the assignment statement. When
casting variable types from one to another, no information will be lost as long as
the receiver is larger than the provider. Java also allows you to cast instances of a
class, known as objects, to instances of other classes. The declaration for an
explicit cast to a class is as follows:
(Classname)reference

The Classname is the name of the class you wish to cast to the receiving object.
The reference specifies the object that is to receive the cast. When applying a
narrowing effect to a class, as you will read about later, this type of cast is required by the Java compiler. Figure 5.5 illustrates this concept.
If a superclass attempts to cast an instance of itself to a subclass beneath it, a
runtime error will occur even though this type of cast will be accepted by the
Java compiler. The technique of passing object references down a class hierarchy
is referred to as widening. As a class is located at lower levels in a hierarchy it
becomes more specific and thus it contains more information than the classes
above it in the hierarchy. Superclasses, on the other hand, are usually more general than the classes beneath them. Conversions that occur when you pass the
references up the hierarchy are thus referred to a narrowing because not all the
information is passed along to the receiving object. Furthermore, all instance
variables of the same name in the receiving object are set to the class variables
that are being casted.

Casting an Object vs. Creating an Object

When casting between instances of a class, an object only assumes reference of the class. A new instance of the class is not
created; the object merely points to the methods and variables
of the casting class. It is important not to confuse the process
of casting a object with the process of creating an object. Just
as you pass the value of a variable through different types (e.g,
int, float, double, and so on), you can pass an object through
different classes, as long as the class is in the current hierarchy.

152 Chapter 5
Not Allowed:
Objects

Classes

Widening

Parent

Child

Allowed:

Narrowing
Parent

Child

Figure 5.5
Widening and narrowing an instance of a class by using casts.

Here is an example of how you can cast references to objects between class types:
public class atom_ant {
String superhero = "Atom Ant";
int height = 10;

Java Classes and Methods

153

atom_ant() {
}
void print() {
System.out.print (superhero + " is " + height + "\n");
}
public static void main(String arg[]) {
atom_ant a1;
a1 = new atom_ant();
a1.print();
proton_pal p1, p2;
p1 = (proton_pal) a1; // Runtime error due to casting error
p1.print(); // Unable to execute because of the previous line
electron_enemy e1;
e1 = (electron_enemy) p2; // Compile-time error due to casting to a
// sibling class
e1.print(); // Unable to execute because of the previous line
atom_ant a2;
a2 = (atom_ant) p2;
a2.print();
}
}
class proton_pal extends atom_ant {
String superhero = "Proton Pal";
int height = 1;
proton_pal() {
}
void print() {
System.out.print (superhero + " is " + height + "\n");
}
}
class electron_enemy extends atom_ant{
String superhero = "Electron Enemy";
int height = -1;

154 Chapter 5
electron_enemy() {
}
void print() {
System.out.print (superhero + " is " + height + "\n");
}
}

Here weve modified our previous atom_ant class to illustrate the basics of casting. Notice that two of the casts used will produce a runtime and compile-time
error, respectively. (Thus, dont try to compile the code unless you remove the
two illegal casts.) The first cast used in the main() method, p1 = (proton_pal)
a1, produces a widening effect. Although this statement will compile, it produces a runtime error because the object a1 cannot be expected to grow to accommodate the new variables and methods it references in proton_pal. The
second casting statement used is a sibling cast: e1 = (electron_enemy) p2. It
generates a compile-time error because an illegal reference to a sibling class,
electron_enemy is used. This is due to the fact that the classes can have completely different variables and methods not related to each other. The last form
of casting that is addressed in the atom_ant class produces a narrowing effect. In
the statement, (a2 = (atom_ant) p2), the object p2 references variables that are
defined in the class, atom_ant, that is being casted. The reference is then past to
the variable a2.

Chapter

Interfaces and
Packages

Interfaces and
Packages
If youre ready to move beyond the stages of writing applets and simple standalone applications and
applets, youll find that Javas flexible interfaces
and packages provide a welcome relief.

After writing a few applets and applications, youll probably notice that the

directory your classes are written to will start to become obscenely large. This is
the downside of the way Java processes classes; but the good news is that Java
provides two key features called interfaces and packages to help you organize your
code. We put these two topics in a chapter by themselves instead of covering in
detail in the previous chapter to emphasize how important they are. (Many Java
books simply lump interfaces and packages in with classes, or they just skim
over themshameful!) As you start to work more with interfaces and packages,
youll discover a myriad of important program design issues that come into play
which youll need to master to use interfaces and packages effectively.
In this chapter youll learn about:

The basics of interfaces

Techniques for implementing interfaces
The hierarchical structure related to interfaces themselves
Techniques for using casts with interfaces
The basics of packages
Techniques for creating packages
Techniques for using Javas predefined packages

The underlying goal of this chapter is to help you transition from writing small
standalone Java applications and applets to creating classes that can be used over

157

158 Chapter 6

and over. As you start to adopt this style of programming, youll need the flexibility that interfaces and packages provide.

Understanding Interfaces
An interface is a collection of methods and variables that are declared as a unit
but they are not implemented until a later stage. Basically this means that the code
declarations placed in an interface serve as a shell so that you can create a truly
abstract class. The goal behind an abstract class is to provide a mechanism so that
you can define the protocols for a classhow a class should essentially communicate with other classesearly on in the development cycle. The upshot is that
when you create your interfaces or abstract classes, you dont have to specify all of
the details of how they will be implemented. This is saved for a later stage.
Before we jump in and start writing Java code for declaring interfaces, lets explore a few conceptual examples. The concept of abstract classes and interfaces
is tricky to grasp at first. In fact, many experienced object-oriented programmers will tell you that they didnt quite master the concepts until they had written a number of programs. Fortunately, we can help you understand and use the
techniques much quicker by providing the background information and conceptual models youll need to apply them.
The simplest form of an interface involves adding methods and/or variables that
are necessary to a particular class, but would disrupt the hierarchy of the class
structure you are currently building for an application. If you chose to actually
implement these elements in your class, they could limit how you planned to
use the class to derive other classes. To make your classes more flexible, you can
add interfaces to your classes in your hierarchy early on, so that the interfaces
can be used in multiple ways to help construct the behavior of other classes
that appear elsewhere in your class hierarchy. (If this discussion sounds like we
are talking in circleswelcome to the world of interfaces! Hopefully these fine
points will start to make sense to you in a moment when we look at a specific
example.)
Lets assume that we need to develop an application that processes information
about different forms of transportation. Figure 6.1 shows the hierarchy that
could be used along with the list of components that could be implemented as
interfaces.

Interfaces and Packages

159

Class Hierarchy
Transportation

Automobiles

Truck

Car
Solar

Boats

Sail

Powered

Airplanes

Gliders

Powered

Gas

Interfaces
Battery

battery_life

weight

Tires

Gasoline

liters

gas_type

diameter

brand_name

Figure 6.1
The hierarchy of classes for the transportation example.

As with typical class hierarchies, the classes shown in Figure 6.1 become more
specific as they appear further down in the hierarchy tree. The interface components are advantageous when you have operations that are to be performed in
one section of the hierarchy and not in the other areas. For example, the class
Car has two subclasses: Solar and Gas. Lets assume you need to calculate the
liters of gas that a gas car will use. You could include the methods and variables
for performing this operation in the Car superclass, or even better, up two levels
in the Transportation class, so that the Powered|Boats and Powered|Airplanes classes
could use this code also.
Unfortunately, when you consider the scope of the application and all of the
subclasses that inherit this useless information, youd probably agree that this
design approach is flawed. After all, the Solar|Car class would never calculate the
liters of gas used and neither would the Sail|Boats or Gliders|Airplanes classes. A
class that handles the gas calculating operation would be an incredible pain to

160 Chapter 6

incorporate at the Transportation level so that it could be designed into the hierarchy, and thus forcing all the subclasses to inherit all of its methods. If we were
creating a small application that only required a few classes, this approach could
be used. But if you are building an application that uses lots of classes from the
beginning or you are expecting to expand the application in the future, this
approach could quickly become a programmers nightmare because limitations
could arise from placing such restrictions early on.
In applications that have class hierarchies like our transportation example, interfaces become priceless because they allow us to mix-in classes into the application, adding them only where they become absolutely necessary. Another feature
that enhances the interfaces capabilities is the use of multiple implementations
of interfaces per class. For example, in our transportation application, theoretically the Car class would be interested in the Gasoline interface, but the Tire
interface could also be of use. An abstract class could incorporate both of these
interfaces (the methods and variables that define them) at the Transportation
level, but the Boat class would also be forced to inherit them. The Boat class
never would have any use for the Tires methods or variables.
Design Issues with Interfaces
Interfaces will usually fall into a class hierarchy without any problems when you are creating small scale applications. They also
help separate the design process from the implementation process because they keep you from having to combine the more
abstract design issues with implementation details in one component. They also allow you to derive classes without relying on
the more limited technique of single inheritance. If you recall
from Chapter 5, a single inheritance model requires you to create class hierarchy trees by deriving one class from a single
parent or superclass. Each class in the tree is created by using
only data and operations that were defined in the levels above
the current class.
Interfaces help you build class hierarchies that use more powerful object-oriented techniques like multiple inheritance. With interfaces, you can define classes that have multiple parents. You
can incorporate interfaces into a hierarchical class tree to include new methods and variables without having to worry about
disrupting your current implementation tree.

Interfaces and Packages 161

Declaring an Interface
Lets look at the basic declaration for an interface and then well show you the
syntax for implementing an interface. After that, well introduce some code to
illustrate how the transportation example we presented in the previous section
could be set up. The basic syntax for declaring an interface looks similar to the
syntax used for defining a Java class:
public interface InterfaceName {
StaticVariables;
AbstractMethods;
}

In this case, however, the class keyword is not used; the keyword interface takes its
place. The InterfaceName serves as the interface indentifier name and the rules for
specifying this name are the same as those used to name classes. The body of the
interface declaration simply consists of the declaration of static variables and the
names of one or more methods. Heres an example of an interface declaration:
public interface Gasoline {
// This variable is defined as a constant
public static final int Feet_in_Miles = 7245;
// A Method that is to be defined in a class
void gas_type(String Name);
// Another method to be defined later
void liters(int Amount);
}

Note that the variable Feet_in_Miles is declared as both static and final. This is
required because all variables in interfaces cannot be changed. This type of declaration essentially turns the variable into a constant. If you leave out the static and
final keywords, Java will force the variable to be declared as a constant. The two
methods listed include both the method name and the methods parameter list.
The actual code for the method will come when the interface is implemented.

Implementing an Interface
Declaring an interface is only half of the work. At some point, the interface
must be implemented. This is accomplished by using the interface definition (or
abstract class) to create a class. In a sense, a class can be derived using the
interface shell. The syntax for implementing an interface is:

162 Chapter 6
modifier class Identifier extends Superclass
implements InterfaceName [, InterfaceList ] {
ClassBody;
}

In implementing an interface, you are essentially defining a special type of class.

First, the class modifier is needed, followed by the class keyword. Then, the
name of the class is provided. Next, the extends keyword is used followed by a
superclass name to indicate that the class being defined is derived from a parent
class. The implements keyword followed by the name of one or more interfaces,
tells the Java compiler which interfaces will be used to implement the class. It is
important to keep in mind that a class can implement more than one interface.
The class body consists of all of the variables and method definitions for the
class. This is where all of the code must be placed for the methods that are listed
in the interface declarations that are used. Using the Gasoline interface we declared earlier, here is a class called Gas that implements the Gasoline interface:
public class Gas extends Car implements Gasoline {
int Miles;
// Variable declarations
...
void gas_type(String Name) {
... // Add code for this method
}
void liters(int Amount) {
... // Add code for this method
}
}

Notice that this class is derived from a superclass named Car.

Now that weve covered the basics of declaring and implementing an interface,
lets return to the transportation example we presented earlier. The first thing we
need to do is declare the interfaces for the ones listed in Figure 6.1Gasoline,
Batteries, and Tire:
public interface Gasoline {
// This variable is now a constant
public static final int Feet_in_Miles = 7245;
// A Method that is to be defined in a calling class
void gas_type(String Name);

Interfaces and Packages 163

// Another method to be defined later
void liters(int Amount);
}
public interface Batteries {
// A Method that is to be defined in a calling class
void battery_life(int Time);
// Another method to be defined later
void weight(int Amount);
}
public interface Tires {
// A Method that is to be defined in a calling class
void diameter(int Distance);
// Another method to be defined later
void brand_name(int Name);
}

With these interfaces in hand, were ready to create the two classesGas and
Poweredeach one will implement some of the interfaces in different ways. They
will also show you how multiple interfaces can be used in a class definition:
public class Gas extends Car implements Gasoline, Batteries, Tires {
int Feet_Traveled;
int Miles_Traveled = 20;
Feet_Traveled = Miles_Traveled * Feet_in_Miles;
public static gas_type(String Name) {
... // Any functions that are to be performed with gas_type
if(Name.equals("Diesel"))
System.out.println("Ah, good power");
if(Name.equals("Unleaded"))
System.out.println("ok power");
if(Name.equals("Leaded"))
System.out.println("eh, clogged injectors");
}
public static liters(int Amount) {
... // Any functions that are to be performed with liters
}
public static battery_life(int Time) {
... // Any functions that are to be performed with battery_life
}

164 Chapter 6
public static weight(int Amount) {
... // Any functions that are to be performed with weight
}
public static diameter(int Distance) {
... // Any functions that are to be performed with diameter
}
public static brand_name(int Name) {
... // Any functions that are to be performed with brand_name
}
}
public class Powered extends Boat implements Gasoline, Batteries {
int Feet_Traveled;
int Miles_Traveled = 20;
Feet_Traveled = Miles_Traveled * Feet_in_Miles;
public static gas_type(String Name) {
... // Any functions that are to be performed with gas_type
if(Name.equals("Diesel"))
System.out.println("Required");
if(Name.equals("Unleaded"))
System.out.println("Not applicable");
if(Name.equals("Leaded"))
System.out.println("Not applicable");
}
public static liters(int Amount) {
... // Any functions that are to be performed with liters
}
public static battery_life(int Time) {
... // Any functions that are to be performed with battery_life
}
public static weight(int Amount) {
... // Any functions that are to be preformed with weight
}
}

Notice that the Gas class is extended from the superclass Car and implements the
interfaces Gasoline, Batteries, and Tires. In the class body of Gas, the methods

Interfaces and Packages

165

declared for these interfaces are coded as well as other variables that the class
needs, such as Feet_Traveled and Miles_Traveled. The Boat class, on the other
hand, only implements two interfaces: Gasoline and Batteries. Notice that the
Boat class implementation for the gas_type() method (declared in the Gasoline
interface) differs from the version implemented in the Gas class.

Tips on Using Interfaces

The implements clause lists all of the interfaces that are to be
included in the class definition. By referencing the interface, the
class implementing it must restate the methods and their definitions in the body of the class. Constructorsthe special methods that initialize new objectsmay not be included in the
interface declaration because interfaces can not instantiate new
objects. Interfaces reference an object that is an instance of a
class. By doing this they state that the object being referenced
includes all the methods in the class that created the object.

The Art of Casting with Interfaces

If you recall from Chapter 5, where we covered casting between class types, we
discussed how a cast can be used to change a reference to an object and not the
actual object itself. We also showed you how instance variables are created and
initialized to reflect the current reference to an object. This occurs when the
names of the variable are the same in two classesthe one casting the object and
the object the variable references. (If this is beginning to sound Greek to you,
refer back to Chapter 5 for a refresher.) When we first introduced casting techniques for classes, this may have seemed to be a negative because a cast can relate
to different instances of the same variable name, but actually a cast works to our
advantage when used with interfaces.
Lets return to our Gas class example to see how we can use casts with interfaces.
This time around Gas will reference the interfaces Gasoline, Tires, and Batteries; and Gas will create objects that reference the interfaces in different ways.
Some of the references are correct and some of them will produce compile-time
errors. Weve included line numbers at the start of each line of code so that you
can easily refer to the example in the discussion that follows:
1 public class Gas extends Car implements Gasoline, Tires, Batteries {
2

166 Chapter 6
3 Gas
aCar
= makeGasCar();
4 Gasoline aGasCar = (Gasoline) makeGasCar();
// Use cast
5 Tires
aTireCar = (Tires) makeGasCar();
// Use cast
6
7
aGasCar.gas_type(Diesel);
// Valid
8
aGasCar.liters(5.8);
// Valid
9
10 aTireCar.diameter(6.9);
// Valid
11 aTireCar.gas_type(Unleaded);
// Not Valid
12
13 aCar.gas_type(Diesel);
// Valid
14 aCar.weight(12.7);
// Valid
15 aCar.diameter(6.9);
// Valid
16 aCar.brand_name(Bridgestone);
// Valid
17
18 . . .
// Any functions that you would perform on the Cars created
19}

Lets break down what is going on here so that you can better understand some
of the important and subtle Java programming techniques that are being used.
Our example is only missing one thing that is not shown in the codea method
named makeGasCar() that creates and returns an object. Line 3 shows that an
object is returned from the makeGasCar() method and is named aCar of type
Gas. By assigning the returned value of makeGasCar() to an object variable of
the type Gas, the object inheirits all the methods pertaining to the Gas class.
This means it acquires all the methods relating to the class, its superclass, and the
interfaces the class implements. In line 4, we acquire an object from the
makeGasCar() method, but this time we cast it as type Gasoline from the interface Gasoline. This means that the object, aGasCar, inheirits all the methods
that relate to the Gas class, its superclass, and only the methods and variables
declared in the interface Gasoline. As well see in a second, this means no methods or variables from the other interfaces are available for the object to reference.
The next line does the same as the previous line, but the Tires interface is used
in place of Gasoline.
Lines 7 and 8 both have the object aGasCar call the methods gas_type() and
liters(), which were originally declared in the Gasoline interface. These method
calls are valid because the correct parameters are used and the object aGasCar
has access to both of these methods because of the cast that was used. In line 10,
the aTireCar object references the diameter() method which is also valid

Interfaces and Packages 167

because this object was created using the (Tires) cast and the diameter() method
is declared within the Tires interface. But in line 11, the aTireCar object tries to
call a method that is declared in the Gasoline interface. This produces a compile-time error because the object does not implement the interface Gasoline.
Only the methods declared in the Tires interface are available from the object.
In the last section of the Gas class, lines 13 through 16, the object aCar may call
any of the methods available to the interfaces because this object is an instance
of the class Gas and is not casted to any particular class. This shows you the
versatility possible in creating objects using interfaces.

Tips on Implementing Interfaces

If you refer back to our ticker tape applet in Chapter 2, youll notice that it
implements an interface named Runnable for the explicit function of moving
(actually redrawing) text across the screen. When the applet is loaded into a
browser, the browser checks to see if the object ttapeThread, which is an instance of the class Thread from a package that is imported into our class
TickerTape, implements the Runnable interface. In this case, the browser detects the interface and uses the run() method declared in the class Thread during the operation of the applet:
// TickerTape Class
public class TickerTape extends Applet implements Runnable{
...
// Change coordinates and repaint
public void run(){
while(ttapeThread != null){
try {Thread.sleep(50);} catch (InterruptedException e){}
setcoord();
repaint();
}
}
...
}

This is a powerful feature for creating methods and variables in classes that can
be set up with interfaces for future use, as long as the interface explains how
information will be transferred to and from it. You dont need to allow others
access to your original classes.

168 Chapter 6

Using the instanceof Operator

To detect if an object implements an interface, you can use the
instanceof operator. This operator allows you to look at a group
of objects to pick out which ones can perform certain operations. Heres an example:
if (ttapeThread iinstanceof Runnable) {
((Runnable)ttapeThread).run(); // performs this function only
// if the object ttape implements
// the Runnable interface
}

In this case the if statement checks to see if the object

ttapeThread is an instance of the Runnable interface. If it is,
the run() method defined in the Runnable interface is called.

Creating and Using Packages

As you begin to design and code Java applications and applets that use multiple
classes and interfaces, youll need a way to organize your code so that you can
easily update and reuse your growing library of classes and interfaces. Because
the Java language is specifically designed to allow you to use classes and interfaces over and over, its likely that youll end up getting some of your class and
interface names mixed up.
Furthermore, another programmer may design an excellent class that performs
operations that you may want to use. Incorporating this class into one of your
applications that already uses a number of classes could become difficult, especially if the class name conflicts with the name of a class you are already using.
For example, you may have a custom print class named Print that performs
certain functions for printing to the screen. After youve developed the class,
another programmer might provide you with a class having the same name that
prints a certain format to a printer that you need to support. You could actually
use both of these classes even if they shared the name Print; however, they
must be packaged in different groups so that the Java compiler can easily determine which one you want to use.
To help us combine classes into unique groups, Java supports the concept of
packages. A package is essentially a device for grouping classes that you want to
be labeled as a unit. You can actually combine any classes that you want into a

Interfaces and Packages 169

single group. Usually, classes that share a common goal are combined in a class.
For example, if you were creating a set of classes to handle drawing-related functions for a design application, you might create a package called Draw and place
all of the related classes in this package.
You might have noticed back in Chapter 2 that some of the methods we implemented in the ticker tape applet were borrowed from classes or interfaces belonging to other packages. For example, one of the packages used was the Applet
packagea package that Java provides, which contains all the necessary classes
for creating an applet. A package is introduced to a class by using the import
keyword in the beginning of a source code file. This will be covered in more
detail later in the chapter. As you will see, classes and packages are segregated
according to the functions they perform. This reduces the risk of having methods that share the same name interfere with each other. Here is a simple example
of how you can implement methods that belong to different packages into a
common class:
// TickerTape Applet
import java.applet.*;
import java.awt.*;
// TickerTape Class
public class TickerTape extends Applet implements Runnable {
...
public void init(){
...
}
public void start(){
...
}
public void run(){
...
}
public void graphics() {
...
}
public void stop(){
...
}
...
} // End TickerTape

170 Chapter 6

This is the same applet that was used in Chapter 2. All of the methods declared
in this example come from somewhere other than the current class. They have
been overridden to perform a certain function specific to the operation of this
applet. For example, the methods init(), start(), and stop() are defined in the
Applet class that is contained in the java.applet package. The run() method is
defined in the Runnable interface contained in the package java.lang.

Naming and Referencing Packages

Besides the fact that you may want to repeat a simple class name over and over,
youll want to create packages so that you can distribute your classes to other
Java programmers. As with files on your computer, you list the directories in
which they are contained to reference them. This creates a path for the Java
compiler to follow so that it can locate designated classes and interfaces in your
packages. Figure 6.2 shows an example of the directory hierarchy used to reference the package java.awt.image.
By convention, the first level of the hierarchy has been reserved for the name of
the company that develops it. An example of this is sun.audio.AudioDataa
java

lang

io

net

util

awt

peer

image

Figure 6.2
A graphical image of the hierarchy of java.awt.image and a call to the (import)
java.awt.image on the other side.

Interfaces and Packages 171

package developed by Sun Microsystems. (Of course, as with every programming language, Java provides certain exceptionsone being the guideline for
naming and referencing packages. For example, java.io.File was developed by
Sun Microsystems, but this package is intended to be implemented by other
companies as a foundation for building additional I/O packages.) The sections
listed beneath the company name reference subdirectories that further specify
where the class is located. For example java.io.File is a subdirectory that contains
classes that relate to the input/output functions of Java. The extension .class has
been omitted from the reference to the File class because interfaces and classes
are the only format contained in a package and both end in the .class extension.

Uppercase vs. Lowercase Package Names

A specific format should be followed for naming packages and
the classes that are contained within them. All package names
and the directories that follow them should be specified using
lowercase letters. On the other hand, the class and interface
names you wish to reference within a package should be specified using an uppercase letter as the first character. This allows
other programmers who use your packages to easily determine
which components are directory names and which ones are
class and interface names.

Declaration for Creating Packages

To create a package, the following statement should be placed at the beginning
of a source file that contains a set of class definitions:
package PackageName;

Each class defined in the source file will automatically be added to the package
having the name specified by PackageName. The PackageName will be created
under the subdirectory you have defined in the CLASSPATH variable set in
your environment. (The instructions for setting this environment variable are
presented in the sidebar, Setting Your CLASSPATH Environment Variable.) As an
example, assume that you have a source file that contains a set of classes that
implement different types of airplanes. These classes could be combined into a
single package named airplanes by placing the package statement at the beginning of each source file that defines a public class:

172 Chapter 6
package airplanes;

// This statement must come first

// Provide source code for Glider class

public class Glider {
...
// Class definition
}
// The end of this source file

package airplanes;

// This statement must come first

// Provide source code for Single_engine class

public class Single_engine {
...
// Class definition
}
// The end of this source file

package airplanes;

// This statement must come first

// Provide source code for Twin_engine class

public class Twin_engine {
...
// Class definition
}
// The end of this source file

The actual PackageName is extended by the Java compiler by preceding it with

the CLASSPATH. (Each subdirectory included in the path name is separated
by a period.) The nice part is that you dont need to create the path for the
package you define yourself; it is generated by the compiler at compile-time
automatically.

Interfaces and Public Classes

If you recall from Chapter 5, only one public class may be
declared in any one source file. Only classes defined as public
may be referenced from outside the current package. Otherwise, the classes not defined as public are used to support the
public classes in the package.

Interfaces and Packages 173

In another example, if the package coriolis.books.programming.java

is declared, the directory structure will turn out like this:
c:\java\lib\coriolis\books\programming\java

Essentially, what the Java compiler does when it encounters

a statement like package coriolis.books.programming.java is create a new directory structure for coriolis.books.programming.java
using the directory path specified by the CLASSPATH environment variable. It then places all of the compiled class code defined
in the source file in the java directory. As the example above illustrates, the CLASSPATH would be:
c:\java\lib;

When the package is later referenced by a Java application, the

compiler will know exactly where to look for each class that is
referenced in the package.

Saving Java Source Code Files

It is wise to save your Java source code in the directories containing your compiled class. This will allow you to later edit your
source code if you wish, but more importantly, you wont have to
worry about your class definitions being overwritten with identical names in the default directory where you create and save your
source code (.java extension). Youll want to save the different versions of your source files because as you create more and more
classes, the chance for repeating a class name becomes more common. For example, assume you have a Spreadsheet class that contains two classes; one that prints a graph and the other that prints
a data sheet. Both classes perform very different operations, but
both of them could be assigned the name Print.class. In doing so,
you must take two steps in generating source code with identical
class names because the classes will share the same working directory in most instances. If you placed a statement like this in the
beginning of your source code
package acme.spreadsheet.graph;

the Java compiler would automatically place the Print.class in

the directory graph but the original source file (Print.java)

174 Chapter 6

would still be paced in the working directory. The next step

would be to place the source file in the same directory. This is
because the next Print.java source file created (for example,
the class responsible for printing the data sheet) will be saved
in the working directory, causing the old file to be overwritten. If you later need to modify the class file, you will still have
the original source code. The next source file should provide
the statement
package acme.spreadsheet.datasheet;

at the beginning. Remember, you are required to manually

move the source file to the appropriate directory.

Setting Your CLASSPATH Environment Variable

When your source code is compiled, the CLASSPATH environment variable specifies the default base directory for the packages you create. It also tells the compiler which directory path to
search for the classes that are predefined. The order of directories defined by CLASSPATH determines the order in which the
Java compiler will search for your classes. When a class is found
that meets the requirements of the calling class, the compiler
stops searching for a match. You should define the path of the
default package that accompanies the Java Development Kit (JDK)
and the temporary directory that you work from in this order.
Heres an example:
CLASSPATH = c:\java\lib;.

The period sets the current directory you are compiling from.
The first directory listed in the CLASSPATH also specifies where
your package structure will begin.

Using Packages
The one feature that makes the Java language very powerful is that it lets you use
the same code (classes) over and over countless times. This is accomplished by
referencing classes that are contained in packages. To use classes that have already been created by you or other Java programmers, you need to reference the
package(s) the classes are grouped in. You can do this in one of three ways:

Interfaces and Packages

175

Specify the full package reference each time a class is used that is defined in
an outside package. This approach is the most cumbersome and least often
used. Heres an example:
airplanes.Twin_engine twin = new airplanes.Twin_engine("Beach", 1100);

In this case, the object variable twin is declared and initialized as an instance of
a Twin_engine class which is included in the airplanes package. With this approach, each time a Twin_engine class is accessed, its corresponding package
name must also be included.
Import the actual class needed from the package it is defined in. As an example, we could rewrite the previous example by using this code:
import airplanes.Twin_engine;
...
Twin_engine twin = new Twin_engine("Beach", 1100);

Notice that once the desired class is imported, the name of the airplanes package is not needed to reference the Twin_engine class.
Import all of the classes defined in a package. The syntax for doing this is
illustrated with this statement:
import airplanes.*;

In this case, all of the public classes combined in the airplanes class, such as
Glider, Single_engine, and Twin_engine, would be included.

Importing Packages Is Like Including C/C++ Header Files

If you are an experienced C / C++ programmer, you can think
of the technique of importing a package as you would the technique of using an include file. Typically, you would use an
include file to specify the names of function prototypes you
wish to call that are defined in external files.
Every class defined in an external package that you want to reference by a class in
your Java application or applet must be called directly or with a wild card (*) in

176 Chapter 6

the immediate directory. For example, if you refer back to our ticker tape applet
presented in Chapter 2, we called an instance of the class FontMetrics that is
contained in the java.awt package (directory). The Applet class imports the
java.awt package with a wild card in the beginning of the code (e.g., import
java.awt.*;). The wild card tells the Java compiler to import all of the public
classes in the java.awt directory into the TickerTape class. The compiler wont,
however, import any of the classes that are contained in the peer or image directories beneath java.awt. To include the classes in those directories, you must
reference the directories directly (e.g., import java.awt.peer.*; or import
java.awt.image.*;).
// TickerTape Applet
import java.applet.*;
import java.awt.*;
// TickerTape Class
public class TickerTape extends Applet implements Runnable {
// Draw background and text on buffer image
public void paintText(Graphics g){
...
FontMetrics fmetrics = g.getFontMetrics();
...
}
}

Declaration for Importing Packages

When importing a package into a class, the declaration must appear before any
class declarations. The format for declaring a package is as follows:
import PackageName;

The PackageName represents the hierarchy tree separating the directories of the
package with decimals. The java.lang package is automatically imported into
every class that is created. If you look at the ticker tape applet presented in
Chapter 2, you will notice that it does not import the java.lang package but uses
many of the classes that are contained in the package. The classes String, Integer, and Thread are just a few of the classes that are called from this package.

Interfaces and Packages 177

// TickerTape Class
public class TickerTape extends Applet implements Runnable {
// Declare Variable
String inputText;
String animSpeedString;
int xpos;
int fontLength;
int fontHeight;
int animSpeed;
boolean suspended = false;

...
}

Standard Java Packages

Since we created our first applet in Chapter 2, we have been using packages
already defined by other developers including Sun Microsystems. These packages have been arranged by their category of usage. Table 6.1 shows the packages
currently being distributed with the JDK.

Hiding Classes Using the Wild Card

We mentioned before that the Java wild card (*) will only allow you to bring in
the public classes from an imported package. The benefit of this feature is that
you can hide the bulk of your classes that perform support operations for your
public classes. Users who use the public classes wont be able to look at the code
or directly access the internal support classes.
Table 6.1

Packages Distributed with the Java Development Kit

Package

Description

java.lang

Contains essential Java classes for performing basic functions. This package is automatically
imported into every class that is created in Java.

java.io

Contains classes used to perform input/output functions to different sources.

java.util

Contains utility classes for items such as tables and vectors.

java.net

Contains classes that aid in connecting over networks. These classes can be used in
conjunction with java.io to read/write information to files over a network.

java.awt

Contains classes that let you write platform-independent graphics applications. It includes
classes for creating buttons, panels, text boxes, and so on.

java.applet

Contains classes that let you create Java applets that will run within Java-enabled browsers.

Chapter

Java Exceptions

Java
Exceptions
Are you tired of writing applications that
mysteriously crash, leaving the user to give
up in frustration? If so, youll be glad to
learn that Java provides a powerful feature
called exceptions that automates the work of
catching and handling compile-time and
runtime errors.

One of the most difficult and time-consuming tasks of developing software

involves finding and fixing bugs. Fortunately, Java provides some built-in features that lend a hand in the debugging process. As errors occur in a Java program, and we all know they will, you can use Java exceptions to provide special
code for handling them.
Java programs can detect certain errors on their own and instruct the Java runtime system to take some predefined action. If you dont like the default operations that Java performs when it encounters certain errors, you can write your
own custom error handling routines.
In this chapter well start by explaining the basics of exceptions. Then, well
show you

Why exceptions are important

How to use try clauses to setup exceptions
How to use catch clauses to trap exceptions
When and how to use your own exceptions

181

182 Chapter 7

Understanding Exceptions
Exceptions catch your errors and handle them gracefully so that your programs
can continue to operate. In Java, this process is called throwing an error. This
type of error handling can greatly benefit both you and the user of your application. After all, nobody likes an application that just crashes out of the blue.
Unlike other languages, such as C, C++, and Pascal, where error detection and
reporting can often double and even triple the size of an application, Java provides the means to detect and handle errors and at the same time reduce the
overall size of your applications. The best part is that error handling in Java
replaces the multiple if this occurs then do this statements so often found in
programs written in languages like C.
Javas exceptions allow you to effectively code the main sections of your applications without you having to spend too much time writing code to detect and
handle potential errors. As youll learn in this chapter, exceptions create an object when an error occurs. The exception, which is a subclass of the Throwable
class, throws an object, which is passed up through the hierarchy of the calling
classes. The object will continue up through the classes until an exception handlera method that deals with the exceptioncatches the object. This process
is illustrated in Figure 7.1. If no exception handler is defined, a default exception handler is used to handle the error. This causes the error to be printed to the
command line and the program will cease running.

Using Javas Throwable Class

For a class to throw an error or catch one, it must be declared
as a subclass of the Java Throwable class. All of the classes in
the java package have incorporated the Throwable class in the
package. This is why you dont see the Throwable class imported at the beginning of Java source code files. Although if
you wish to refer to this class, you can directly import it into
an application by including the statement:
import java.lang.Throwable;

Having error checking and error handling features in Java is important because
Java programs, especially applets, run in multitasking environments. Often when
an applet is executed in a Web browser like Netscape 2, other applets will be
running at the same time. Each applet will have its own thread that the system

Java Exceptions 183

...

method
try{
call method( );
}
catch(Exception);

method throw

throw

method throw
method throws

throw

...

message

throw new Exception

Figure 7.1
The process of throwing and catching an error in Java.

will control. If one applet causes a fatal error, the system could crash. With
exceptions, on the other hand, critical errors are caught; and the Java runtime
environment will know how to handle each thread that it must manage.

Do You Really Need Exceptions?

Even the smallest program can quickly evolve into a programmers nightmare when
you are trying to locate and fix a troublesome error. Why should you waste your
time handling the possible errors that can occur? The fact is, you should only
handle errors that you can do something useful with. Grabbing every little error in
a program is useless if you do nothing intelligent with them. The best reason to
declare exceptions is to provide a means for you and others that follow to understand where your code is having problems during critical operations.

184 Chapter 7

For example, assume you have created a class to write data to a disk file. As your
program is running, a number of errors could occur such as your hard disk
being full, a file being corrupted, and so on. If you didnt have a way to catch
errors like these at some point, the program might crash, leaving the user with
nothing except a cryptic error message. Heres a Java program that performs a
critical file operation but doesnt provide any error handling:
// This program will not compile because an IOException handler is
// expected by the Java compiler
import java.io.*;
public class WriteAFile extends Object {
WriteAFile(String s) {
write(s);
}
// Writes to a file
public void write(String s) {
// I/O errors could occur here
FileOutputStream writeOut = null;
DataOutputStream dataWrite = null;
// Begin to Write file out
writeOut = new FileOutputStream(s);
dataWrite = new DataOutputStream(writeOut);
dataWrite.writeChars("This is a Test");
dataWrite.close();
}
// Where execution begins in a stand-alone executable
public static void main(String args[]) {
new WriteAFile(args[0]);
}
}

(Actually, this program wont compile just yet because the Java compiler expects to find an exception named IOException. Well explain this in a moment.) The part of the code that could get you into trouble is the write()
method. This method creates a new file output stream and attempts to write a
character string to the stream. If the operation fails for one reason or another,
a runtime error would occur, although an exception has not been setup to
handle such an error.

Java Exceptions 185

To catch potential I/O problems, Java provides a built-in exception called

IOException. This exception gets thrown whenever an I/O error occurs during a
transfer to a device, such as a printer, disk drive, and so on. In our sample program,
the Java compiler knows you must declare the exception because thewrite() method
calls other methods that have declared an IOException to be thrown to calling
methods. To remedy this, we could alter the write() method as shown here:
// Begin to Write file out
try {
writeOut = new FileOutputStream(s);
dataWrite = new DataOutputStream(writeOut);
dataWrite.writeChars("This is a Test");
dataWrite.close();
}
catch(IOException e) {
}

Notice that two changes have been made. First, the entire block of code has
been placed in a try { } clause. Essentially, this tells the Java environment to be
on the lookout for errors that might occur as each method call is executed.
The second change is the addition of the catch() method. This block of code
performs the job of handling an I/O error that could occur with any of the calls
contained in the try section. In our example, we are letting Java handle the work
of processing an I/O error on its own by using the built-in IOException, and
thats why no other code is provided with the catch statement.
These changes allow the code to compile and run. Unfortunately, they do not address any problems that could arise from actually writing a file to disk. In a perfect
world, this code would be sufficient for our needs, but we dont live in a perfect
world. For example, what if an error occurred while we were opening the file to be
written because the disk is full or not even present? And even if the file could be
opened, what would happen if an error occurred while we were writing the data to
the file. All of these conditions are valid exceptions to writing a file to a disk. Unfortunately, you or others who use your classes might not detect them until it is too late.
Remember, the advantage of using a language like Java or other flexible object-oriented languages is the ability to create robust code that can be reused by others.
Now, lets change the WriteAFile class once more to make it more robust. Dont
worry about the syntax right now, we will discuss the details of implementing
exceptions in the sections to follow.

186 Chapter 7
// Writes to a file
public void write(String s) {
FileOutputStream writeOut = null;
DataOutputStream dataWrite = null;
try {
writeOut = new FileOutputStream(s);
dataWrite = new DataOutputStream(writeOut);
}
catch (Throwable e) {
System.out.println("Error in opening file");
return;
}
try {
dataWrite.writeChars("This is a Test");
dataWrite.close();
}
catch(IOException e) {
System.out.println("Error in writing to file");
}
}

This time around, weve included two try clauses. The first one checks the methods
used to open the file, and the second one tests the methods used to write to the
file and close the file. Notice how each of the catch statements specifies the type
of object that it will catch or the exception that is thrown. Well show you how to
create custom error-handling routines later when we discuss the topic of catching. For now it is important to realize that we have separated the possible errors
we want to catch into two separate cases, opening and writing. By catching these
errors, we have prevented the program from crashing as a result of not being able
to open or write to a file. If errors like these are found, we could possibly ask the
user to change disks or try again instead of having the user loose his or her data.
In our case, we have simply written a message to the command-line telling the
user where the operation has failed if an error occurs.

Defining a Try Clause

The try statement is responsible for indicating which section of code in a Java
applet or application will most likely throw an exception. The syntax for using
this statement is as follows:
try {
statement;

Java Exceptions 187

statement;
}
catch (Throwable-subclass e) {
statement;
statement;
}

For every try section, you must include at least one catch block that follows the
try section. If an exception is thrown in the try section during the execution of
the code, control flow is transferred to the matching section defined in the catch
statement. If no match is found, the exception is passed up through the hierarchy of method calls. This allows each level to either handle the exception or pass
it on. Well cover this more when we present exception throws.

Using the catch Statement

If an exception is thrown during the execution of a try section, the flow of the
program is immediately transferred to the corresponding catch block. The object, which is a reference to an instance of the exception class being thrown, is
compared to the catchs parameter type, also known as an Exception Handler.
Here is the declaration for the catch block:
catch (ExceptionType ExceptionObject) {
statement;
statement;
}

The ExceptionObject reference parameter is a subclasses of the Throwable class.

In most code, this reference is declared as e to distinguish it as a reference to an
exception. In the event of an error, a subclass of the Throwable class is thrown,
which is triggered by a violation of one of the procedures. This violation creates
an object of the class type that the error originated from and is compared to the
Exception Handler listed in each of the catch blocks that immediately follow the
try section. The following code example illustrates how this process works:
import java.io.*;
// Reads from a file
public class ReadAFile extends Object {

188 Chapter 7
ReadAFile(String s) {
String line;
FileInputStream fileName = null;
BufferedInputStream bufferedInput = null;
DataInputStream dataIn = null;
try {
fileName = new FileInputStream(s);
bufferedInput = new BufferedInputStream(fileName);
dataIn = new DataInputStream(bufferedInput);
}
catch(FileNotFoundException e) {
System.out.println("File Not Found");
return;
}
catch(Throwable e) {
System.out.println("Error in opening file");
return;
}
try {
while ((line = dataIn.readLine()) != null) {
System.out.println(line + "\n");
}
fileName.close();
}
catch(IOException e) {
System.out.println("Error in reading file");
}
}
// Where execution begins in a stand-alone executable
public static void main(String args[]) {
new ReadAFile(args[0]);
}
}

Here, the try block instructs the code to watch for an exception to be thrown
from one of the methods contained in the block. The initializer that creates an
instance of the class type FileInputStream named fileName is capable of throwing an exception in the event of an error. More specifically, the method contained in the class FileInputStream declares that an exception is to be thrown to
the calling method. The topic of throwing exceptions will be covered later in the
chapter, but for now you just need to know that you are required to address all

Java Exceptions 189

exceptions thrown by handling them or passing them on. You handle the exception being thrown by placing exception handlers, declared in catch statements
that the errors are then compared to. In the event of an error, the code will break
from the normal flow of the code and immediately jump to the first exception
handler that matches the class type defined in the catch. In the ReadAFile()
method, the first catch identifies the FileNotFoundException class as a type
that may be thrown upon instance of an error. This is followed by another catch
identifying the Throwable class, which will act as a catch all for the exceptions
being thrown. This match occurs because all exception classes are derived from
the Throwable parent class.

When to Use the finally Statement

When an exception is thrown, the compiler does not necessarily return to the
exact spot it left off. The developers of Java realized that some procedures need
to perform additional routines after an exception is handled, so they defined a
finally statement. This statement instructs the Java Virtual Machine to return
and finish any code after handling the exception before moving on. Here is the
syntax required for using the finally statement:
try {
statement;
statement;
}
catch (Exception Handler) {
statement;
statement;
}
finally {
statement;
statement;
}

The finally statement is not necessary to handle an exception, but it can be

useful when you wish to handle an operation specific to a class. To see how it is
used, lets expand our WriteAFile class to incorporate a finally statement that
will create a backup file whether or not an exception occurs:
import java.io.*;
public class WriteAFile {

190 Chapter 7
WriteAFile(String s) {
write(s);
}
// Writes to a file
public void write(String s) {
FileOutputStream writeOut = null;
DataOutputStream dataWrite = null;
try {
writeOut = new FileOutputStream(s);
dataWrite = new DataOutputStream(writeOut);
dataWrite.writeChars("This is a Test");
dataWrite.close();
}
catch(IOException e) {
System.out.println("Error in writing to file");
}
catch(Throwable e) {
System.out.println("Error in writing to file");
}
finally {
System.out.println("\n\n.....creating a backup file.");
try {
writeOut = new FileOutputStream("MyBackup.sav");
dataWrite = new DataOutputStream(writeOut);
dataWrite.writeChars("This is a Test");
dataWrite.close();
}
catch (IOException e) {
System.out.println("Error in writing backup file");
}
}
}
// Where execution begins in a stand-alone executable
public static void main(String args[]) {
new WriteAFile(args[0]);
}
}

The Hierarchy of Exceptions

Like any other built-in Java classes you use in your applications, the standard
exceptions are designed around a class hierarchy. Every exception is derived from
the superclass Throwable as shown in Figure 7.2. The first subdivision is where
the class splits into two categories: Errors and Exceptions. The Exceptions cat-

Java Exceptions 191

Throwable

Errors

Exception

RunTimeException

Figure 7.2
The Hierarchy of the Throwable class.

egory consists of the more common exceptions that you will want to catch.
The Errors category, on the other hand, consists of the low level exceptions that
most programmers wont need to deal with.
The next major split occurs with the Run-Time Exception category, which is a
subclass of Exception. Sun has arranged the hierarchy like this because they realized that by separating commonly used exceptions from specific exceptions, programmers would not be forced to include tons of handlers in their code.

ERROR CLASS
The exceptions included in the Error class are problems such as Linkage Error,
ThreadDeaths, and other catastrophes that result in fatal errors. For the most part,
these exceptions will not be handled by most applications. They are reserved for
lower level programming tasks that require you to get into the internal workings of
the language. For that reason, it is not a good idea to derive your own exception
classes from Error unless you have a good working knowledge of Java. Table 7.1
provides a list of the key exceptions that are provided. All of the exceptions are
defined in the Java Language Package java.lang, expect for AWTError, which is
defined in the Advanced Windowing Toolkit Package, java.awt.

192 Chapter 7
Table 7.1

Exceptions Included in the Error Class

Exception

Description

AbstractMethodError

This exception is thrown when your code attempts to call an abstract method.

AWTError

This exception is thrown when an error occurs in the Advanced Windowing Toolkit.

ClassCircularityError

This exception is thrown when a class hierarchy tries to establish a circle by

linking itself to a parent class and the parent links itself to the child class or one
of the children classes beneath it.

ClassFormatError

This exception is thrown as a result of an invalid file format being implemented.

IllegalAccessError

This exception occurs when an illegal access has been triggered.

IncompatibleClassChangeError This exception is thrown when a class of incompatible types is changed.

InstantiationError

This exception occurs when a program attempts to instaniate an object from an

abstract class or interface.

InternalError

This exception is thrown when an internal error occurs in the Java Virtual Machine.

LinkageError

This exception is thrown when the current class is dependant on another class, but
the other class is not compatible with the current class.

NoClassDefFoundError

This exception is thrown when a class cannot be found by checking the path
specified by the CLASSPATH environment variable or the current directory.

NoSuchFieldError

This exception is thrown when a specific field cannot be found.

NoSuchMethodError

This exception is thrown when a particular method cannot be found in the current
class or one of its superclasses.

OutOfMemoryError

This exception is thrown in the event that no more memory can be allocated.

StackOverflowError

This exception signals that the stack has overflowed.

ThreadDeath

This exception is thrown in the thread that is being terminated. It should be

handled when additional procedures are needed to be carried out before the
stop() method has finished executing. If the exception is caught, it must be
rethrown to actually finish killing off the thread. Because the exception is not
required to be caught, it will not produce a command line message when it cycles
up to the base class.

UnknownError

Bearing a close relation to mystery meat, this exception is triggered when a

seriously unknown error occurs.

UnsatisfiedLinkError

This exception is thrown when a link to a library is unsuccessful.

VerifyError

This exception is thrown when the Java compiler is unable to verify if a linkage
between classes is valid.

VirtualMachineError

This exception is thrown when the Virtual Machine has depleted its resources.

Java Exceptions 193

EXCEPTION CLASS
The exceptions included in the Exception class represent the most common
errors that a programmer will want to deal with. For the most part, these excepTable 7.2

Exceptions Included in the Exception Class

Exception
ClassNotFoundException

Defined in the Language Package (java.lang)

Description
This exception is thrown when the compiler is unable to locate a class in the
current directory or the directory path specified by the environment variable
CLASSPATH.

CloneNotSupportedException This exception is thrown when an object attempts to clone an object that does not
want to be cloned.
IllegalAccessException

This exception is thrown when a method is called from a class that does not have
permission to do so. The access is determined by the modifiers used for the class
and methods, resulting in the compiler being able to see or not see the calling
method.

IllegalMonitorStateException This exception is thrown in the event that a monitor is accessed that you do not own.
InstantiationException

This exception is thrown because of an attempt to create an instance of an

abstract class or interface.

InterruptedException

This exception is thrown when a thread has interrupted the currently running
thread.

NoSuchMethodException

This exception is thrown when a method cant be found in the calling class.

Exception

Defined in the Utility Package (java.util)

Description

EmptyStackException

This exception is thrown in the event of an empty stack.

NoSuchElementException

This exception is thrown in the event that an enumeration is empty.

Exception

Defined in the Input/Output Package (java.io)

Description

EOFException

This exception is thrown when an EOF is reached unexpectedly during input.

FileNotFoundException

This exception is thrown when a file is not found.

IOException

This exception is thrown in the event of an I/O error.

InterruptedIOException

This exception is thrown when an I/O operation has been interrupted.

UTFDataFormatException

This exception is thrown when a malformed UTF-8 string has been read in a
DataInput stream.

continued

194 Chapter 7
Table 7.2

Exceptions Included in the Exception Class (Continued)

Exception

Defined in the Networking Package (java.net)

Description

MalformedURLException

This exception is thrown in the event of a bad URL.

ProtocolException

This exception is thrown when connect receives an EPROTO. This exception is

specifically caught in the Socket class.

SocketException

This exception is thrown when an error occurs during the use of a socket.

UnknownHostException

This exception is thrown when there is an error in the connection to server from
the client.

UnknownServiceException

This exception is thrown when a service is not identified.

Exception

Defined in the Advanced Windowing Toolkit Package (java.awt)

Description

AWTException

This exception is thrown in the event of an error with the Advanced Windowing
Toolkit.

tions can effectively be handled in the average program, to address problems

between the user and the program. This class makes an obvious choice to derive
your own personal classes from. Table 7.2 provides a list of the key exceptions
that are provided in the Exception class.

RUNTIME CLASS
The exceptions included in the Runtime class are thrown during the execution
of Java code. All of these exceptions are exempt from the restrictions of handling
the exception at compile time. These exceptions are optional because of the
need to keep Java code compact and easy to read. Table 7.3 provides a list of the
key exceptions that are provided in the Runtime class. These exceptions are
defined in the Language Package (java.lang).

Declaring a Method Capable

of Throwing Exceptions
All methods capable of throwing an exception to a calling method must declare
the exception in the method declaration. The type of exception being thrown to
the calling method must be declared so that it understands how to handle the
object it is receiving. The format for a method capable of throwing an exception
is as follows:

Java Exceptions 195

Table 7.3

Exceptions Included in the Runtime Class

Exception

Description

ArithmeticException

This exception is thrown when an integer is divided by zero.

ArrayIndexOutOfBoundsException This exception is thrown when an array is referenced outside the legal
range.
ArrayStoreException
`

This exception is thrown when you attempt to store an incompatible class or

type in an array.

ClassCastException

This exception occurs when you attempt to cast an instance of a class to a

subclass or a sibling class.

IllegalArgumentException

This exception is thrown when an invalid parameter is passed to a method

that is outside the legal range or value.

IllegalThreadStateException

This exception occurs when a thread state is changed to an invalid state or

one that the thread is currently in.

IndexOutOfBoundsException

This exception is thrown when an index to an array is outside the legal

range.

NegativeArraySizeException

This exception occurs when an array of negative size is to allocated.

NullPointerException

This exception is thrown in the event that an object contains a null

reference.

NumberFormatException

This exception is thrown when an invalid string to a number or a number to

a string is encountered.

SecurityException

This exception is thrown when an applet attempts to breach the security

defined by the browser.

StringIndexOutOfBoundsException This exception occurs when a string is accessed outside the legal length of a
string.
[Modifier] Return-type Identifier ([Parameter List]) [throws
ExceptionName]
{
Body;
}

The throws clause may list as many exceptions as will be thrown to it by separating each of them with a comma. For an example, lets take our ReadAFileAFile
class to the next level and introduce a throws method:
import java.io.*;
public class wordProcessor extends Object {

196 Chapter 7
String fileName;
void save(String fileName) {
System.out.print ("Saving File Procedure\n");
try {
System.out.print ("Saving File " + fileName + "\n");
ReadAFile aFile = new ReadAFile(fileName );
}
catch(FileNotFoundException e) {
System.out.print ("Procedure to get another name and try again\n");
// Procedure to get another name and try again
}
catch(IOException e) {
System.out.print ("Procedure to try again\n");
// Procedure to try again
}
finally {
System.out.print ("Perform any cleanup\n" );
// Perform any cleanup
}
}
// Where execution begins in a stand-alone executable
public static void main(String args[]) {
wordProcessor myProgram = new wordProcessor();
myProgram.save(args[0]);
}
}
// Reads from a file
class ReadAFile extends wordProcessor {
ReadAFile(String s) throws FileNotFoundException, IOException {
String line;
FileInputStream fileName = null;
BufferedInputStream bufferedInput = null;
DataInputStream dataIn = null;
try {
fileName = new FileInputStream(s);
bufferedInput = new BufferedInputStream(fileName);
dataIn = new DataInputStream(bufferedInput);
}
catch(FileNotFoundException e) {
System.out.println("File Not Found");
throw e;
}

Java Exceptions 197

catch(Throwable e) {
System.out.println("Error in opening file");
}
try {
while ((line = dataIn.readLine()) != null) {
System.out.println(line + "\n");
}
fileName .close();
}
catch(IOException e) {
System.out.println("Error in reading file");
throw e;
}
}
}

Notice that we didnt need to make many changes to the ReadAFile class used in
this application. This class can quickly be made to pass exceptions as well as
handle the ones that apply specifically to the class. The objectmyProgram, which
is an instance of the class wordProcessor, calls the method save(). This method
then calls the ReadAFile() method which declares that it will pass an exception
to the calling method in the event of an error. Because the ReadAFile() method
declares that it throws an exception, save() is required to address the exception
that is being passed to it. If the method will not handle the exception, it must
declare that it passes the particular exception on to the method that derived it:
ReadAFile(String s) throws
...

FileNotFoundException, IOException {

In our example, this line of code tells the method, ReaAFile(), that two exceptions, FileNotFoundException and IOException, can be thrown from the try
block. This requires the save() method to handle them or declare the exceptions
to be passed on to the method main() to deal with them.

Throwing Exceptions
The throw operator declares a particular exception may be thrown from the
current method on to the calling method. This effectively passes the exception
to the next method for it to deal with. In our previous example, the ReadAFile
class declared that the method save() would pass two exceptions. In the code
that follows, the example identifies which exceptions will be thrown.

198 Chapter 7
try {
fileName = new FileInputStream(s);
bufferedInput = new BufferedInputStream(fileName);
dataIn = new DataInputStream(bufferedInput);
}
catch(FileNotFoundException e) {
System.out.println("File Not Found");
throw e;
}
catch(Throwable e) {
System.out.println("Error in opening file");
}
try {
while ((line = dataIn.readLine()) != null) {
System.out.println(line + "\n");
}
fileName .close();
}
catch(IOException e) {
System.out.println("Error in reading file");
throw e;
}

The statement throw e specifies that the exception will be passed on for the
calling method to deal with. Furthermore, much like error codes in other languages, messages can be passed along with the object to identify particular details to help process the exception. The following line of code shows how to
throw an exception with a message attached:
throw new FileNotFoundException("MyFile.txt");

To reference the message in the calling method, you could simply call a
getMessage() method to read the message attached to the file. The following
code presents an example of this method:
catch(FileNotFoundException e) {
System.out.println("The file " + e.getMessage +
" was unable to be located.");
}

When to Catch and When to Throw

The issue of knowing when to catch an exception versus when to throw it to the
calling method is typically a factor of what the exception does. If you refer back

Java Exceptions 199

to our WriteAFile example, youll see that we deal with a couple of exceptions.
One of them caught an error that occurs in the event of an IOException by
printing a message to the command line. This notifies the user of an error when
writing to a file; but suppose WriteAFile class was a subclass in the hierarchy of
the class wordProcessor. Here is a new version of our example that has been
expanded to handle this:
import java.io.*;
public class WriteAFile extends wordProcessor{
WriteAFile(String s) throws IOException {
write(s);
}
// Writes to a file
public void write(String s) throws IOException {
FileOutputStream writeOut = null;
DataOutputStream dataWrite = null;
try {
writeOut = new FileOutputStream(s);
dataWrite = new DataOutputStream(writeOut);
}
catch (Throwable e) {
System.out.println("Error in opening file");
return;
}
try {
dataWrite.writeChars("This is a Test");
dataWrite.close();
}
catch(IOException e) {
System.out.println("Error in writing to file");
throw e;
}
}
}
import java.io.*;
public class wordProcessor extends Object {
wordProcessor(String s) {
new WriteAFile(s);
}

200 Chapter 7
wordProcessor() {
System.out.println("Create a backup file");
}
// Where execution begins in a stand-alone executable
public static void main(String args[]) throws IOException {
new wordProcessor(args[0]);
}
}

Now, lets suppose we pass a filename to the write() method and it triggers the
IOException. The IOException again writes a message to the command line,
but notice it re-throws the exception to the calling method wordProcessor().
This method then allows for an additional message to be printed, in this case
Create a backup file. In place of the message, we could write an additional file
to another location or do some other operation specific to the classwordProcessor.
In addition, any other class could call the method and use it to fit its needs
without being forced to perform an operation specific to wordProcessor.

Knowing When to Create Your

Own Exceptions
The process of creating your own exceptions in Java is similar to creating other
types of classes. Knowing when to create an exception is sometimes trickier than
writing the exception itself. Here are some guidelines to help you create your
own exceptions:

Make sure you derive your new exception from the correct class. For example, if you create an exception that detects the corruption of a file, youd
want to subclass it beneath an IOException. Deriving the new exception
from an exception like ArrayOutOfBoundsException would be pointless.
If your code generates an error condition, you should handle it unless there
is an obvious exception already created. For example, in the ReadAFile class
we coded in this chapter, we used an exception to detect if a file cannot be
found. On the other hand, if you created a class that determines whether a
file has a virus or not, an IOException wouldnt necessarily be a wise choice.
This would be a good place to subclass, however.
Exceptions created in most applications should be derived from the Exceptions class. Only specific (lower-level) situations should require exceptions
that need to be derived from the Errors or RunTime classes.

Java Exceptions 201

To create and use your exception classes, follow the same rules as standard classes.
If you need a refresher, refer to Chapter 5. Here is a basic example of an exception:
public class AVirusDetectedException extends Exception {
AVirusDetectedException(String fileName) {
//perform some actions like read in libraries of virus types
while(viruslibrary != null) {
if (virus(fileName)) {
throw new AVirusDetected(NameofVirus);
//code after the throw operator is never executed.
}
}
//this code is only executed if no virus is found
}
int virus(String fileName) {
//perform some actions like read in libraries of virus types
//test the byte code against patterns associated to viruses
if (fileName = viruspattern) {
return 1;
}
return 0;
}
}

Trying to compile the source code will only result in an error. We subclassed the
AVirusDetectedException from the Exception class because it will be triggered in the
event of an I/O operation, but it does not fall under one of the predefined exceptions. This is used to demonstrate how an exception would look if it were created by
you. To call this exception in your code, place the following code in your program:
try {
if (file is questioned) {
throw new AVirusDetectedException(fileName);
}
} catch (AVirusDetectedException e) {
System.out.println(e.getMessage + " has been found in " + fileName);
}

This tests whether a file was read from a disk drive, downloaded, and so on. An
exception is then thrown in the event of a virus, as declared in the exception
code above.

Chapter

Threads

Threads
To create Java applets and applications that
wont turn into system resource hogs, youll
need to arrange your programs into separate
processes, which are called threads.

I magine what our lives would be like if we could only do one thing at a time.

You wouldnt be able to listen to music and program at the same time; and you
definitely couldnt cook dinner, watch TV, and carry on a conversation with a
friend. Although programming languages dont need to perform tasks like these,
newer operating systems and environments like the Web are placing greater demands on programs, requiring them to handle multiple processes at the same time.
Java offers an advantage over most other languages because it was designed from
the ground up to support multiple processes. When a Java applet runs in an
environment like a Web browser, the browser can determine which parts of the
program are separate processes and manage them to keep the program from
draining the available system resources. As you gain more experience writing
Java programs, youll learn how to structure your programs to take advantage of
the flexibility that multiple processes provide.
In this chapter well examine how threads are used to create multiple processes
in Java programs. Youll learn how to create threads using either the pre-defined
Thread class or the Runnable interface.

What Is a Thread?
One of the key jobs performed by the Java runtime system is to be able to handle
programs that contain multiple processes called threads. If youve done any programming for an operating system such as Windows 95 or Windows NT, youve

205

206 Chapter 8

probably come across the concept called multithreading. The idea is to create
applications that can handle different tasks at the same time, or at least be able
to convince the user that multiple tasks are being performed. For example, a
multithreaded version of an application that monitors the stock market would
be able to download data from a central computer network, perform calculations in the background, and accept input from the user. Although only one
thread can be executed at a time, the operating system that runs the program
expertly divides up the different processes or threads, runs pieces of them, and
jumps from thread to thread.
If you have ever loaded a Web page that contains multiple applets, you can see the
process of multithreading at work. Assuming each applet is coded properly, your
Web browser will make it look like each one is running at the same time. Of
course, since most computers have only a single processor, the Web browser must
be able to juggle each process so that each one gets its share of processing time.
To better understand how threads are coded, lets start with a simple example
that contains a single process. Then well add to it so that you can see the effect
that using threads has on the execution of the program. In the following
Hi3Person class, the code executes in a linear fashion until the last line is reached,
and the process ends:
public class Hi3Person {
public static void main(String args[]) {
Hi3Person people = new Hi3Person();
people.hi("Person");
people.hi("Person 2");
people.hi("Person 3");
System.out.println("Hello Everyone");
}
void hi(String who) {
System.out.println("Hi " + who);
}
}

Code execution begins by creating an instance of the class Hi3Person. Next, the
three hi() methods are called. Each of these is executed one at a time, returning

Threads

207

control back to the main body of the code. The final statement in main() writes
the text Hello Everyone before the program ends.
As we introduce the concept of threads to Hi3Person, the linear path of execution will be disrupted. The program will be split into multiple processes, each
responsible for writing to the screen. Lets look at the new version of our code to
see what is going on behind the scenes:
public class Hi3People implements Runnable {
public static void main(String args[]) throws InterruptedException {
int i = 0;
Hi3People person = new Hi3People();
// Create thread #1
Thread aThread = new Thread(person, "Person 1");
// Create thread #2
Thread anotherThread = new Thread(person, "Person 2");
aThread.start();
anotherThread.start();

// Start the first thread

// Start the second thread

// Body of main program

while ((aThread.isAlive()) || (anotherThread.isAlive())) {
i++;
}
// Executes after both threads have finished
System.out.println(i + "\n");
System.out.println("Hello Everyone");
aThread.stop();
anotherThread.stop();

// Stop the first thread

// Stop the second thread

}
public void run() {
System.out.println("Hi " + Thread.currentThread().getName());
}
}

(For now, dont worry about the syntax used to create the threads that are used.
Well explain the techniques for implementing threads a little later in this chapter.) Notice that the Hi3People class initiates two threads that run concurrently
as our application continues on. After each thread has been created, the start()

208 Chapter 8

method of the thread is called, which tells the Java interpreter to begin processing this thread. The main() method is responsible for setting up each thread and
determining when the threads are finished. This is necessary because our program needs to know when it is safe to execute the code starting with the line:
System.out.println(i + "\n");

Otherwise, the program will end before the threads have finished and it will
hang. In this case, we have placed a while loop to count continuously during the
execution of the threads:
while ((aThread.isAlive()) || (anotherThread.isAlive())) {
i++;
}

If you compile and run this example, you will notice that the value stored in the
variable i will change after each execution of the code. This variable stores the
number of times the while loop repeats during the life of both threads. The fact
that this value changes illustrates the control that Java can have as it executes
programs that are divided up into separate processes. Running a single- or multithreaded program is not the only task that the Java runtime system must perform. Java also has its own internal threads that it must perform to manage tasks
such as garbage collection.

To Thread or Not to Thread

The trick to programming with threads is knowing when you
need to use them. Threads have the ability to hinder as well as
help the execution of programs. You should only use threads
when two or more processes need to exist at the same time.
For example, in a windows environment, multiple windows
can be opened at once to give the impression that two operations are occurring at the same time. These are examples of
threads being implemented into real world application. Most
Java programmers implement threads for building interface
components by using the AWT package, which well explore
in Chapter 9.

Threads

209

Keep in mind that when an application runs, multiple processes dont actually run at the same time. Its the operating
systems job to give the user the impression that everything
happens at once. Even in a multithreading environment like
Unix, processes do not occur at the same time. As Figure 8.1
shows, the illusion of processes running concurrently is created by carefully and quickly cycling instructions through a
channel. The Java Virtual Machine handles its own type of
processor management by determining what executions will
occur and in what order.

Process

When you add multiple threads to your program, you effectively set up an events manager. You must manage how the
instructions from your application are handled. You determine
what process receives more or less time and when to change
the focus of the program. As you will see later in this chapter,
you can make your application appear transparent or painfully slow just by the way you schedule your threads.

Process

Process

Process

Processor

Figure 8.1
The technique of managing multiple processes.

210 Chapter 8

Creating a Thread
Before you can create a thread, you must set up a class to handle the thread. This
is done in either of two ways: extending a class (subclassing the Thread class) or
implementing an interface. As youll see in the next two sections, the approach
you use will depend on your own needs.

Subclassing the Thread Class

The most obvious way to create a thread is to subclass the Thread class that Java
provides. This approach allows you to override the methods in the Thread class
to perform the functions you define for your program. Here is the syntax for
creating a class using the Thread class.
[Modifier] class ClassName extends Thread {
ClassBody;
}

Heres an example Java application that utilizes the Thread class as a superclass:
public class Hiagain extends Thread {
public static void main(String args[]) {
int i =0;
Hiagain tony = new Hiagain(); // Create a new object
Thread t1 = new Thread(tony); // Create each thread of the object type
Thread t2 = new Thread(tony);
t1.start();
t2.start();

// Start each thread

while ((t1.isAlive()) || (t2.isAlive())) {

i++;
}
System.out.println("Hello Everyone");
t1.stop();
// End the threads
t2.stop();
}
public void run() {
System.out.println("Hi");
}
}

Threads

211

The class Hiagain subclasses the Thread class and overrides the run() method
defined by the Thread class. Because Hiagain is derived from Thread, it inherits all of the methods defined in Thread including start(), run(), and stop().
The original versions of these methods defined in Thread are used except in the
case of the run() method, which has been overridden. Because this method tells
the thread which operations to perform after the thread has been started, it
typically needs to be overridden. The key methods that are defined in theThread
class will be presented later in this chapter.

Implementing the Runnable Interface

When designing class hierarchies for your Java applications or applets, situations arise where subclassing from the Threads class is just not possible. In such
cases, you can implement the Runnable interface to avoid a conflict. To declare
a class Runnable, follow the template shown here:
[Modifier] class ClassName extends SuperClass implements Runnable {
ClassBody;
}

The advantage of this approach is that you can create a new class that is both
derived from another class and uses the methods defined by the Runnable interface. Of course, you will then be required to go along with the implementation
created by the designers of this interface. Lets revisit our ticker tape applet introduced in Chapter 2 because it provides a good example of an implementation
of the Runnable interface:
// TickerTape Class
public class TickerTape extends Applet implements Runnable{
....
// Initialize Applet
public void init(){
....
}
....
// Start Applet as thread
public void start(){
if(ttapeThread == null){
ttapeThread = new Thread(this);

212 Chapter 8
ttapeThread.start();
}
}
...
// Change coordinates and repaint
public void run(){
while(ttapeThread != null){
try {Thread.sleep(50);} catch (InterruptedException e){}
setcoord();
repaint();
}
}
....
// Stop thread then clean up before close
public void stop(){
if(ttapeThread != null)
ttapeThread.stop();
ttapeThread = null;
}
} // End TickerTape

As with all applets, you must use the Runnable interface to implement threads.
(You are forced to subclass the Applet class to perform the basic operations of
initializing and starting your applet.) The reason you would want
to implement threads in an applet is to reduce the processor load for performing operations that occur over and over. One example would be the graphics
routine in our ticker tape applet that updates the screen to make the text appear
as if it floats across the screen. The run() method is coded to redraw the screen
and then set the thread to sleep for a specified amount of time:
// Change coordinates and repaint
public void run() {
while(ttapeThread != null) {
try { Thread.sleep(50); } catch (InterruptedException e) {}
setcoord();
repaint();
}
}

The reason for putting the thread to sleep is covered in more detail later in the
chapter. The two methods responsible for moving the text are setcoord() and
repaint(). They are executed as long as the thread exists.

Threads

213

Initializing a Thread
Before you can use a thread, you must initialize it by creating an instance of the
Thread class. The best way to do this is to use the constructor for the Thread
class. The simple form of this constructor is:
Thread Identifier = new Thread();

A few other variations of this constructor are also supported:

Thread(ObjectReference);
Thread(StringName);
Thread(ObjectReference, StringName);

In the first example, a parameter that references the object to be used to create
the thread for is provided. We actually used this type of constructor in theHiagain
class example presented earlier:
Hiagain tony = new Hiagain();
Thread t1 = new Thread(tony);

// Create a new object

// Create a thread of the object type

The next two constructors allow you to pass a string to create references to
individual threads, which can then be used as symbolic references. Well show
you how to use this feature later in this chapter to keep track of multiple threads
created from the same class instance.
If you return to the ticker tape applet outlined above, youll see that the thread is
created in the start() method for the applet:
// Start Applet as thread
public void start() {
if(ttapeThread == null) {
ttapeThread = new Thread(this);
ttapeThread.start();
}
}

In this case, notice the Thread() constructor is passed the this parameter. Using
this allows us to tell the constructor the name of the class that implements the
Runnable interface. The new thread object that is created is assigned to the

214 Chapter 8

variable ttapeThread. Once the thread has been initialized, the start() method
for the thread can be called by using the statement ttapeThread.start().

Who Goes First; Who Finishes Last?

Although only one thread can be started at a time, dont expect the first one
called to always be the one that finishes first. In fact, the order in which threads
are called wont necessarily determine the order in which they finish. In most
cases, it is impossible to determine which thread will finish first. Lets return to
our simple Hi3People class introduced in the beginning of this chapter, to see
how the execution of threads can be hard to predict:
public class Hi3People implements Runnable {
public static void main(String args[]) throws InterruptedException {
int i1 = 0;
int i2 = 0;
Hi3People person = new Hi3People();
Thread aThread = new Thread(person, "Person 1");
Thread anotherThread = new Thread(person, "Person 2");
aThread.start();
anotherThread.start();
while ((aThread.isAlive()) || (anotherThread.isAlive())) {
if (aThread.isAlive()) { ++i1;} // Counter for the first thread
if (anotherThread.isAlive()) { ++i2;} // Counter for the second thread
}
System.out.println("The time for Person1 is " + i1 + "\n");
System.out.println("The time for Person2 is " + i2 + "\n");
aThread.stop();
anotherThread.stop();
}
public void run() {
System.out.println("Hi " + Thread.currentThread().getName());
}
}

First, notice the types of constructors that are used:

Hi3People person = new Hi3People();
Thread aThread = new Thread(person, "Person 1");
Thread anotherThread = new Thread(person, "Person 2");

Threads

215

Both the object name (person) and a unique string is passed to each call to
Thread(). Since both threads are created using the same object, the string is
passed to assign each thread its own unique name. In the run() method of the
program, the getName() method is used to display the name of the current
thread. A companion method named setName() is provided in the Thread class
for setting the name of a thread.
Next, by changing a few lines of code, we converted our while loop to count the
time it takes to process each thread instead of counting the time it takes to
process the two together. You would need to run this code about 10 to 15 times
before running across an instance where the first person beats the second one.
This is due to the fact that Javas scheduler is still in a beta version. (Hopefully,
Sun will consider implementing a method for determining the order in which
threads are processed.) The scheduler is responsible for determining what threads
may run and which ones must wait in the queue. This process is determined in
either one of two methods: priority or first in first out (FIFO).

Priority versus FIFO Scheduling

When a thread is processed, it is automatically thrown into a queue. Then, the
thread must wait until its turn is up. This process of waiting is called blocking.
The implementation of the thread that is ahead of the one waiting will determine if a thread will wait until the current thread has completed. This method
of waiting is referred to as First in First Out (FIFO). Like everything in this
world, there are exceptions to the rules. If a thread has priority over another
thread, it switches places with the thread. This process continues up the queue
until a thread reaches an equal or greater priority or it is executed. The threads
that were pushed aside may not continue their operation until all the higher
priority threads either step aside or finish. The most common case of this is
the Garbage Collector, the thread that runs in the background and has the
lowest priority.
To control how the priority is set for a thread, the Thread class provides the
following three variables:

MAX_PRIORITY
NORM_PRIORITY
MIN_PRIORITY

216 Chapter 8

Each of these variables holds integer values that specify a threads priority level.
For example, MAX_PRIORITY stores a number that indicates the maximum
allowable value for a threads priority. To set or retrieve the current priority setting of a thread, the Thread class provides the setPriority() and getPriority()
methods. In setting priorities, you can use one of the three priority instance
variables. For example, the following method call would set a threads priority to
the value contained in the NORM_PRIORITY variable:
Thread aThread = new Thread(person, "Person 1");
aThread.setPriority(aThread.NORM_PRIORITY);

Controlling the Life of a Thread

A thread is like a human or plant life form; it has a beginning, middle, and an
ending. During these stages, a thread will take many different paths depending
on the objective to be reached. Figure 8.2, shows the cycle of life that a thread
can possibly take.
The stages of life of a thread are determined by a set of pre-defined methods that
can be overridden to perform certain tasks.

The start() Method

This method starts the birthing process for a thread. By default, it sets up a
few initializations and calls the start() method. You can override it to initialize
your own variables and then call the start() method:
// Start Applet as a thread
public void start() {
if(ttapeThread == null) {
ttapeThread = new Thread(this);
ttapeThread.start();
}
}

In this example the start() method checks to see if the thread ttapeThread exists. If it doesnt, it creates an instance of the Thread class and assigns the variable ttapeThread to it.

Threads

217

NEW BORN
Runnable

Blocked

Running

DEAD
Figure 8.2
The cycle of life pertaining to a thread.

The run() Method

This method defines the actions to take place during the life of a thread. Heres
an example:
// Change coordinates and repaint
public void run() {
while(ttapeThread != null) {
try {Thread.sleep(50);} catch (InterruptedException e) {}
setcoord();
repaint();
}
}

In this example, the run() method makes the process sleep for 50 milliseconds
while the instance of the class named ttapeThread is not equal to null. Then,
the setcoord() method is called followed by the repaint() method.

218 Chapter 8

The sleep() Method

This method releases control of a thread from the processor for a specified amount
of time. The syntax for calling this method is:
Thread.sleep(Miliseconds);

In the run() method we just presented, the sleep() method is called to allow other
threads to be processed while the ttapeThread is put on hold. This allows the browser
to accept other input and not redraw the screen every instance of the thread.

The suspend() Method

This method suspends execution of a thread. As the following example shows, it
requires no parameters:
// Handle mouse clicks
public boolean handleEvent(Event evt) {
if (evt.id == Event.MOUSE_DOWN) {
if (suspended) {
ttapeThread.resume();
} else {
ttapeThread.suspend();
}
suspended = !suspended;
}
return true;
}

This line of code states that in the event of a mouse click and the thread is
running, the thread will be suspended from operation. This allows other threads
in the queue to be processed; but as soon as the thread is reactivated, it will
resume its place in the queue as long as a higher priority thread is not executing.

The resume() Method

This method resumes a suspended thread. Heres an example of how it is used:
// Handle mouse clicks
public boolean handleEvent(Event evt) {
if (evt.id == Event.MOUSE_DOWN) {
if (suspended) {
ttapeThread.resume();
} else {
ttapeThread.suspend();

Threads

219

}
suspended = !suspended;
}
return true;
}

The resume() method is responsible for reactivating a method that is asleep.

This allows for the thread to re-enter the queue.

The yield() Method

This method causes the current thread to move to the end of the queue and lets the
next thread take control of the processor. Heres an example of how it can be used:
// Change coordinates and repaint
public void run() {
while(ttapeThread != null) {
setcoord();
repaint();
yield();
}
}

If the thread exists, the setcoord() method is executed followed by the repaint()
method. Then the yield() method is called to permit the next thread in line to
execute. Unfortunately, this is not wise if we are to depend on a scheduled repaint(). We could fall victim to the mercy of the threads that will be placed
before the current thread that is moved to the end of the queue.

The stop() Method

This method ceases the life of a thread and performs the required cleanup operations:
// Stop thread then clean up before close
public void stop(){
if(ttapeThread != null)
ttapeThread.stop();
ttapeThread = null;
}

In the event that the end of the process is reached, this method is called to clean
up after the thread and perform any final procedures before closing out.

220 Chapter 8

The destroy() Method

This method causes a thread to die without any cleanup operations being performed:
// Stop thread then clean up before close
public void stop(){
if(ttapeThread != null)
ttapeThread.destroy();
ttapeThread = null;
}

In the event that the stop() method of the applet is called, the thread ttapeThread
will be destroyed and no further lines of code for that object will be executed.

Multiple Objects Interacting

with One Source
When you have multiple threads in your application running all at once, the
need for limiting access to devices that write data and perform other critical
tasks becomes absolutely necessary. After all, there is no way of telling when
objects may try to update a storage device like a file at the same time. As a result,
data may become corrupt or false information may be extracted from a device.

Synchronizing Revisited
If you recall from Chapter 5, we showed you how to declare a synchronized
method. If you dont remember, here is the syntax.
synchronized ReturnType Identifier([ParameterList]) [Throws]
{
MethodBody;
}

The synchronized modifier is used to declare a method of which only one object can execute at any one time. This is accomplished by the Java Virtual Machine setting up an object monitor for every portion of code that declares itself
as synchronized. In order for the code to run, the object must attain this monitor. In the event that a thread wants to run a synchronized section of code, it is

Threads

221

blocked until the thread ahead of it finishes executing the particular section of
code. Lets look at an example of how the synchronized techniques works:
import java.awt.*;
import java.lang.*;
public class MyApp2 extends Frame implements Runnable {
static TextArea t1;
static TextArea t2;
MyApp2() {
// Calls the parent constructor Frame(string title)
// Same as setTitle("Duck Duck Goose");
super("Counting example");
// A new panel to the south that 4 buttons and 1 choice
t1 = new TextArea();
t2 = new TextArea();
add("East", t1);
add("West", t2);
pack();
show();
}
public static void main(String args[]) {
int i = 0;
MyApp2 game = new MyApp2();
Thread person1 = new Thread(game, "duck");
Thread person2= new Thread(game, "goose");
person1.start();
person2.start();
while ((person1.isAlive()) || (person2.isAlive())) {
++i;
t2.setText(Integer.toString(i));
}
t2.appendText("\n Time through the loop \n\nYour It");
person1.stop();
person2.stop();
}

222 Chapter 8
public synchronized void run() {
int d = 0;
int change = 0;
while(d < 100) {
t1.appendText("\n
" " + d );
++d;
}

" + Thread.currentThread().getName() +

}
public boolean handleEvent(Event evt) {
switch(evt.id) {
case Event.WINDOW_DESTROY: {
System.exit(0);
return true;
}
default:
return false;
}
}
}

The above code initiates two threads that cycle through the synchronized run()
method. When you compile this program, you will see the first thread, person1, count up to 99, followed by the next thread, person2, count up to 99
and end. The thread person2 must wait until the monitor is released by the
previous thread before executing. While this process is occurring, notice that
the counter timing the execution of the synchronized threading event is running alongside.

Wait() a Second... Notify() Me When...

Running a thread through a synchronized method is perfectly fine if you dont
need any additional information from the outside. But lets suppose you wish to
execute a thread within a synchronized method, and half way through you need
to collect information from an additional thread. The first thread establishes a
foundation, perhaps opening a file. Then, the following thread will enter the
method and write the information and leave. Finally, the original thread will
perform the cleanup necessary for closing the file. Well, this is all great, but
remember we synchronized the method to permit only one thread to execute at

Threads

223

any one time. This is easily remedied by using the wait() method, which causes
the currently executing method to release the monitor to the next thread. The
thread that released the monitor can then reacquire the monitor when the notify() method is called from within the same method. The thread waiting then
picks up right from the point where it began waiting. Lets modify our previous
example MyApp2 to utilize the wait() and notify() methods:
public synchronized void run() {
int d = 0;
int change = 0;

while(d < 100) {

t1.appendText("\n " + Thread.currentThread().getName() + " " +
Integer.toString(d) );
++d;
if( d == 50) {
try {
if (Thread.currentThread().getName().equals("duck")) {
this.wait();
}
}
catch(InterruptedException e) {
}
}
}
if (Thread.currentThread().getName().equals("goose")) {
this.notify();
}
}

After compiling the class again and running it, you will notice that the first
thread counts to 50. The thread person1 then releases the monitor to the next
thread, person2. The thread then counts up to 99 and notifies the previous
thread to begin executing from where it left off.

Grouping Your Threads

Once you have created your threads, they are like children who run loose in the
mall. No matter how often you call them, they will go off into their own little
world. Just like children, threads themselves can have parents. These parents are
established by assigning them to a ThreadGroup. The ThreadGroup is actually

224 Chapter 8

a class just like the one the threads are derived from. You can create a
ThreadGroup the same way you initialize any class in Java:
ThreadGroup parentAuthor = new ThreadGroup( " The Potts ");

This statement sets up a ThreadGroup named The Potts with a reference to

the object parentAuthor. From here we can assign threads to this group by passing the name of the ThreadGoup with the initialization of the thread. For example, if we wish to add to threads to the ThreadGroup parentAuthor, we
would enter the following:
Thread child1 = new Thread( parentAuthor, "Angela");
Thread child2 = new Thread( parentAuthor, "Anthony");

Creating hierarchies of these groups is just as easy as assigning threads to the

group. For example, lets suppose that the parentAuthor ThreadGroup also wants
to have a subgroup underneath it named petsAuthor. To accomplish this we
would simply use the following code:
ThreadGroup petsAuthor = new ThreadGroup( parentAuthor, "Our Pets");

This allows for quick subgrouping of like threads. There are three main advantages to subgrouping threads:

Controlling the states of all the threads contained within the group without
having to individually set each one.
Retrieving all the threads in the group easily so that you can identify a thread
quickly.
Setting the priority of all the threads within the group with one command
to the group.
Note: Setting the priority of the ThreadGroup only effects the threads of less
priority than the calling method. If a thread is currently set at a high priority, it will continue at this level until it dies.

Chapter

The Java AWT

The Java
AWT
If youre wondering where to look for information on creating interface components for
Java programs, youve come to the right place.
The AWT provides a treasure chest of powerful interface classes.

No one would use a programming language these days if it did not have built-

in support for common user interface objects like windows, menus, dialogs, and
buttons. Fortunately, the designers of Java did not overlook this. They created a
package called the Abstract Window Toolkit or AWT, which allows Java programmers to build GUIs very easily. Although AWT is very flexible and powerful, its
shear size will overwhelm you until you understand how it is organized and the
basics for using it.
To help you get more out of the AWT, well look at how the AWT is arranged.
Then well present each of the key AWT classes. Well show you how to use the
layout manager, and well present the basics for creating interface components
such as menus. If we tried to cover the AWT in a lot of detail, we could easily
turn this chapter into an entire book. However, since Java is still a very young
language and much of the AWT is still being solidified, we will only cover enough
of this library to get you started with the AWT so that you can use it effectively.
Youll want to keep your browser tuned to Suns Java site for the latest information on this powerful package.

Introducing the AWT

When you first start to work with the AWT, you may notice that it lacks many
of the features that you would find in other graphical user interface (GUI) li227

228 Chapter 9

braries. This is because Java is a cross-platform language, and the tools provided
with any Java library must be designed so that they can work with all systems.
As you build more complex applets and applications, you will find it difficult to
not use the AWT because of its extreme flexibility. If a component such as a
window or menu doesnt do what you need, you can simply subclass it and add
your own custom features.
To use the AWT in a Java applet or program, you must first import the AWT
package by using the import statement as shown here:
import java.awt.*;

// Include the AWT

The asterisk (*) is used with the import statement to tell the Java compiler to
include all classes in the immediate subdirectory. Once you include this package, you can use any of the AWT controls or packages to derive your own. With
just a little bit of programming effort, youll be amazed at the types of interface
components you can create for your Java applicationseverything from scrollable
windows to fully functional pop-up menus.
Heres an example of a simple Java program that displays a window that contains
a text message:
import java.awt.*;

// Include the AWT

public class testWin extends Frame {

public testWin(){}

// Use the Frame class

// Constructor

public static void main(String args[]) {

testWin Test = new testWin();
// Display a line of text
Test.setText("This text will be displayed in the window");
// Add a second line
Test.appendText(" Add this text to the next line in the window");
Test.show();
// Display the frame
}
}

Introducing the Layout Manager

Creating applications with a visual programming language like Visual Basic
and Delphi can simply involve choosing from a selection of custom compo-

The Java AWT 229

nents written by other programmers and then dragging a component onto a

form. Visual programmers like to refer to this practice as drop-and-drag
programming. Unfortunately, Java programming is not quite this easy (although
Java development is headed in this direction). When you place controls on
forms to build applications with visual languages, you usually specify absolute
positions. In other words, you tell your development environment exactly where
you want to place a controlright down to the pixel. The problem with this
approach is that different operating systems use different methods to display
graphical components. So, a form that looks good on a PC may not look right
on a Mac screen.
If you have only programmed for an environment like Windows, many of these
problems are not as apparent because Windows takes care of specific system and
interface related details for you. To provide this type of flexibility, Java provides
a development tool called the layout manager, which works in conjunction with
the AWT.
Javas layout manager helps you control where items will appear on the screen.
The layout manager is actually an abstract class itself that you can use to create
custom layouts. There are several custom layout methods that come with the
Java Development Kit. As we present the AWT in this chapter, youll learn more
about the layout manager.

What About Menus?

You cant create a good user interface without implementing some form of menus.
The Macintosh and Windows environments are filled with menus. Computer
users expect your applications to present them with some sort of menuing system so that they can access the available features. When creating applets, menus
are not as important and can be confusing because your Web browser already
has its own menu system. But when you are creating Java applications, youll
more than likely want to add menus to your programsand thats where the
AWT comes in. The AWT can help you create menus very easily. However, its
up to you to make them functional and intuitive.

The AWT Hierarchy

The AWT package consists of many classes and subclasses. Most of the controls,
like buttons and text fields, are derived from the Component class. Since all of the

230 Chapter 9

AWT controls are descendants from the Component class, they all share some of
the same key methods such as createImage(), disable(), and hide(). Figure 9.1
presents a tree that illustrates the class hierarchy for the controls of the AWT and
Table 9.1 lists the complete set of classes.
If you have done any graphics or interface programming before, some of these
class names should be somewhat familiar to you. Instead of reinventing the wheel,
the developers of Java used traditional user interface componentswindows,
dialogs, scrollbars, and so onto build up the AWT. Of course, youll find that
the AWT heavily embraces object-oriented programming techniques.
Component
Button
Canvas
Checkbox
Choice
Container
Panel
Window

Dialog

Frame

Label
List
Scrollbar
TextComponent
TextArea
TextField
Figure 9.1
The class hierarchy of the AWT.

The Java AWT 231

The Component Class

Since all of the controls and GUI elements that well be using are subclassed
from the Component class, lets start with it. The Component class is rarely (if
ever) used as is. It is almost always subclassed to create new objects that inherit
all its functionality. Component has a tremendous number of methods built
into it that all of the classes that subclass it share. You will be using these methods quite often to perform tasks such as making the control visible, enabling or
disabling it, or resizing it.

Key Component Class Methods

The declarations and descriptions for the Component class would fill up a hundred pages. Fortunately, they are available online at Suns Java site. You can download the API reference and load up the Component class to examine the methods
Table 9.1
BorderLayout

The Complete List of Classes in AWT

FlowLayout

MenuComponent

Button

Font

MenuItem

Canvas

FontMetrics

Panel

CardLayout

Frame

Point

Checkbox

Graphics

Polygon

CheckboxGroup

GridBagConstraints

Rectangle

CheckboxMenuItem

GridBagLayout

Scrollbar

Choice

GridLayout

TextArea

Color

Image

TextComponent

Component

Insets

TextField

Container

Label

Toolkit

Dialog

List

Window

Dimension

MediaTracker

Event

Menu

FileDialog

MenuBar

232 Chapter 9

available in this class. However, there are a few methods that are important, and
you will probably use them for all the controls that subclass the Component
class. Well introduce these methods next, and well examine some of the key
event handling methods in Chapter 10.
BOUNDS()

This method returns the current rectangular bounds of the component.

DISABLE()

This method disables a component.

ENABLE([BOOLEAN])

This method enables a component. You can pass zero arguments, or a Boolean
argument to enable or disable a control. Heres a few examples of how this method
can be called:
myComponent.enable();
myComponent.enable(x==1);

GETFONTMETRICS()

This method gets the font metrics for the component. It returns null if the
component is currently not on the screen.
GETGRAPHICS()

This method gets a graphics context for the component. This method returns
null if the component is currently not on the screen. This method is an absolute
necessity for working with graphics.
GETPARENT()

This method gets the parent of the component.

HANDLEEVENT(EVENT EVT)

This method handles all window events. It returns true if the event is handled
and should not be passed to the parent of the component. The default event
handler calls some helper methods to make life easier on the programmer. This
method is used to handle messages from the operating system.

The Java AWT 233

HIDE()

This method hides the component. It performs the opposite operation of show().
INSIDE(INT X, INT Y)

This method checks to see if a specified x,y location is inside the component.
By default, x and y are inside a component if they fall within the bounding box
of that component.
ISE NABLED()

This method checks to see if the component is enabled. Components are initially enabled.
ISSHOWING()

This method checks to see if the component is showing on screen. This means
that the component must be visible, and it must be in a container that is visible
and showing.
ISVISIBLE()

This method checks to see if the component is visible. Components are initially
visible (with the exception of top level components such as Frame).
LOCATE(INT X, INT Y)

This method returns the component or subcomponent that contains the x,y
location. It is very useful for checking for mouse movement or mouse clicks.
LOCATION()

This method returns the current location of the component. The location will
be specified in the parents coordinate space.
MOVE(INT X, INT Y)

This method moves the component to a new location. The x and y coordinates
are in the parents coordinate space.
REPAINT([TIME])

This method repaints the component. This will result in a call to the update()
method as soon as possible. You can specify a time argument so that Java knows that
you want the component repainted within a specified number of milliseconds. You

234 Chapter 9

can also update just a certain portion of a control by sending the x and y coordinates that specify where you want to start the update and a width and height
that specify how much to update. Here are some examples:
// Regular update
myComponent.update();
// Update within 250 milliseconds
myComponent.update(250);
// Update rectangle
myComponent.update(50, 50, 200, 200);
// Update same rectangle within 250 milliseconds
myComponent.update(250, 50, 50, 200, 200);

RESIZE(INT

WIDTH,

INT

HEIGHT)

This method resizes the component to the specified width and height. You
could also pass it a dimension object instead of the two integers. Here are a
few examples:
myComponent.resize(300, 200);
myComponent.show(new dim(300, 200));

SETFONT(FONT)

This method sets the font of the component. The argument passed is a Font
object. For example, this method call
myComponent.setFont(new Font("Helvetica", Font.PLAIN, 12);

would change the font of myComponent to a Helvetica type face with no bolding
or italics or underline and a point size of 12.
SHOW([BOOLEAN])

This method shows the component. By calling this method you make a
control visible or not. It can also be passed a conditional statement. Here are a
few examples:
myComponent.show();
myComponent.show(x==1);

The Java AWT

235

SIZE()

This method returns the current size of the component. The size is returned in
dimensions of width and height.

The Frame Class

The Frame class is used to create standard application windows. It is a direct
descendant of the Window class and is used to create windows that automatically include the menu bar, title bar, control elements, and so on. Frame looks
like a standard window that you would expect to see running under most operating systems like Windows or Macintosh OS.
You will probably use the Frame class to create window interface components
for the majority of your applications. However, when you need to create custom
windows for special situations, you may need to use the Window class instead.

Hierarchy for Frame

java.lang.Object
java.awt.Component
java.awt.Container
java.awt.Window
java.awt.Frame

Declaration for Frame

To create a frame, you need to use a constructor to initialize your class. The
constructor method does not even need to have any code in it; the method just
needs to be provided so that you can instantiate your object. The following
listing shows a complete Java application that uses a frame:
import java.awt.*;

// Include the AWT

public class testWin extends Frame {

public testWin(){}

// Use the Frame class

// Constructor

public static void main(String args[]) {

testWin Test = new testWin();
Test.show();
// Display a window
}
}

236 Chapter 9

Figure 9.2
A simple windowed Java application.

In the main() method we use our classs constructor to create our object named
Test. Then we call the objects show() method to make the window frame visible
(frames are invisible by default). If you were running this Java program in Windows 95, youd see a window that looks like the one shown in Figure 9.2.
As you can see, the window is quite simple. You need to use several of the Frame
classs methods to make the frame useful. The other thing you may notice is that
when you try and close the window and terminate the program, nothing happens! Thats because you have not told Java to do it. You need to add an event
handling method to catch windows messages. In this case, we are looking for the
WINDOW_DESTROY call from the operating system. Here is the extended
code that sets the sizes of the frame, gives it a title, and catches messages:
import java.awt.*;
public class winTest1 extends Frame {
public winTest1() {}
public synchronized boolean handleEvent(Event e) {
if (e.id == Event.WINDOW_DESTROY) {
// Has window been destroyed?
System.exit(0);
return true;
}
return super.handleEvent(e);
}
public static void main(String args[]) {
winTest Test = new winTest();
Test.setTitle("Test Window");
Test.resize(300 ,200);
Test.show();
}
}

Figure 9.3 shows what the new application looks like.

The Java AWT 237

Figure 9.3
A fully functioning application using a frame.

We are gong to discuss event handling in more detail in the next chapter, so
dont get worried if you do not understand that part of the above code.
What you should notice is the two new calls to two of the frames methods:
setTitle() and resize(). These methods perform the function they are named
after; they set the title of our application and resize it respectively.
Lets look at the methods that are specific to the Frame class.

Methods for the Frame Class

DISPOSE()

This method removes the frame. It must be called to release the resources that
are used to create the frame. This method should be called when you exit a
window in an applet. In an application, you would usually use the System.exit()
method to terminate the program and release all memory used by the application. This method overrides the dispose() method from the Window class.
GETICONIMAGE()

This method returns the icon image for the frame.

GETMENUBAR()

This method gets the menu bar for the frame.

GETTITLE()

This method gets the title of the frame.

238 Chapter 9
ISRESIZABLE()

This method returns true if the user can resize the frame.
REMOVE(MENUCOMPONENT M)

This method removes the specified menu bar from the frame.
SETCURSOR(IMAGE IMG )

This method sets the current cursor image that will be used while the cursor is
within the frame.
SETICONIMAGE(IMAGE IMG)

This method sets the image to display when the frame is iconized. Note that not
all platforms support the concept of iconizing a window. The icon will also be
displayed in the title window in Windows 95.
SETMENUBAR(MENUBAR MB)

This method sets the menu bar for the frame to the specified MenuBar.
SETRESIZABLE(BOOLEAN BOOL)

This method sets the resizable flag.

SETTITLE(STRING TITLE)

This method sets the title for the frame to the specified title.

The Panel Class

Panels are probably one of the most useful components that no one will ever see.
Most of the time, panels are just used as storage containers for other components.
You could also use a panel for drawing or to hold images, but panels are not
usually used for these tasks.
Panels are important because they offer you a way to gain very strict control over
the layout of the other interface controls in your program. Unless you have a
very simple application, you will need multiple controls on the screen at once.
Usually these controls will not all fit where you want them using one layout class
or another. Panels provide a mechanism so that you can mix and match.

The Java AWT 239

Lets look at an example. Assume you have a frame that you want to fill with a
text field in the upper part of the frame, and three buttons lined up along the
bottom. If you only used a single layout class for the entire form, you would not
have enough control to do this. Figure 9.4 illustrates what we want the frame to
look like. Figure 9.5 and 9.6 shows the best you can achieve using a single layout
class. This is not bad, but if you resize the frame the layout gets pretty ugly.
What we need to be able to do is use one type of layout class for the upper part
of the frame, and another for the bottom. We can do this by using a pair of
panels, one for the top using a border style layout and another panel for the
bottom using a flow style layout. Now, when we add our controls, we add them
to their respective panels instead of the frame, and everything is taken care of for

Figure 9.4
Creating a window using the Panel class.

Figure 9.5
A close approximation with a single layout class.

Figure 9.6
Resizing the Panel.

240 Chapter 9

us. The user can resize the control all they want and our controls will stay where
we placed them originally. Here is the code that performs this new configuration. The output is shown in Figure 9.7.
import java.awt.*;
public class mixLayout extends Frame {
public mixLayout() {
super("Mixed Layout Demo");
setLayout(new BorderLayout());
Panel top = new Panel();
Panel bottom = new Panel();
top.setLayout(new BorderLayout());
top.add("Center", new TextArea("HelloWorld", 15, 5));
bottom.setLayout(new FlowLayout());
bottom.add(new Button("Load"));
bottom.add(new Button("Save"));
bottom.add(new Button("Quit"));
add("Center", top);
add("South", bottom);
resize(300, 200);
show();
}
public static void main(String args[]) {
mixLayout test = new mixLayout();
}
}

Hierarchy for Panel

java.lang.Object
java.awt.Component
java.awt.Container
java.awt.Panel

Declaration for Panel

Panels have a very straight-forward declaration because they cannot accept any
arguments whatsoever.
Panel myPanel = new Panel();

The Java AWT 241

Figure 9.7
The new two-panel program with different layout methods.

Methods for Panel

Panels have very few new methods ouside of the ones inherited from the Container class. The key new method that is introduced is setlayout().
SETLAYOUT(LAYOUT MANAGER)

As you have already seen, the setlayout() method is used to define which layout
manager will be used to place controls on the panel. If you do not set a layout
manager, the panel control defaults to flowLayout().

The Label Class

The Label class defines a very simple control that is simply used to display a text
string. This text string is usually used to indicate what tasks are controlled by
another user interface control. For example, if you had a group of radio buttons,
a label control might be used to tell the user what the group is about.

Hierarchy for Label

java.lang.Object
java.awt.Component
java.awt.Label

Declaration for Label

The most common declaration for a label is to assign the text string as the control is being declared. Heres an example:
new Label("Fruits of the world:");

242 Chapter 9

We can also assign the text string to an object variable like this:
Label fruitLabel = new Label("Fruits of the world:");

Table 9.2 shows the three ways you can declare a Label class.

Methods for Label

Labels have very little functionality of their ownthey cant even process mouse
clicks. They are not intended to do much more than sit there and display their
string. There are, however, a few methods you should know about.
GETALIGNMENT()

This method returns the current alignment of the label.

GETTEXT()

This method does what it sounds likeit returns the text the label is displaying.
SETALIGNMENT(INT)

This method changes the alignment of the label. The argument is the same as
the one used with Label.LEFT, Label. CENTER, and Label.RIGHT (see Table
9.2 for more information).
Figure 9.8 shows a few labels with different alignments. Here is the code used to
produce them:
add(new Label("Left")); // no need to specify alignment because it
// defaults to left
add(new Label("Center", Label.CENTER));
add(new Label("Right", Label.RIGHT));

Table 9.2 Options for Declaring a Label Class

Declaration

Description

Label()

This constructor will create an empty label.

Label(String)

This constructor will create a label with the given string.

Label(String, int)

This constructor will create a label with the given string as well as define the
alignment of the string. The int part of the declaration is represented by setting
it to Label.LEFT, Label.CENTER, or Label.RIGHT. (These labels should be
pretty self-explanatory.) Remember though, that if you do not assign an
alignment, the label control will default to left-justified.

The Java AWT 243

Figure 9.8
The Label component.
SETTEXT(STRING)

This method sets or changes the displayed text.

Button Class
The Button class is one of the most often used classes, for obvious reasons.
How often do you see Windows or Mac programs without some sort of
button somewhere?
Buttons are UI components that trigger actions when they are pressed. They
have multiple states to provide feedback to the user. The neat thing about buttons (and all the UI elements for that matter) is that a button you create for a
Java program will change its appearance depending on the operating system.
Thats one of the benefits of cross-platform. A button on your application or
applet will look like a standard Windows button on a PC. Or, it will look like a
standard Mac button on a Mac. The disadvantage here is that if you create other
elements that are dependent on the size and/or shape of the graphics for your
button, then you will run into trouble.
Figure 9.9 illustrates a few different buttons. Notice that the size of the buttons
depends on the length of the caption.

Hierarchy for Button

java.lang.Object
java.awt.Component
java.awt.Button

Declaration for Button

Buttons have two different constructor options. The first one simply creates
a blank button. The second option creates a button that you assign a string
to display.

244 Chapter 9

Figure 9.9
A few Button components.
new Button(); // Empty button
new Button(String); // Button with "String" as the caption

Methods for Button

There are only two methods specific to the button component.
GETLABEL()

This method returns the current caption of the button.

SETLABEL()

This method changes the caption of the button.

The real power behind buttons is realized when you handle the events a button
triggers. This is usually handled in the handleEvent() method that we will be
showing you how to use in the next chapter.

The Canvas Class

The Canvas class is a very simple component. It is designed to be subclassed and
used as a container for graphics methods, much like an artists canvas.

Hierarchy for Canvas

java.lang.Object
java.awt.Component
java.awt.Canvas

The Java AWT

245

Declaration for Canvas

The Canvas component has only a single constructor:
new Canvas();

Methods for Canvas

The only method that is used with the Canvas component is the paint() method
which is called to update the contents. If you do not override this method, the
Canvas component is pretty much worthless because it has very little functionality of its own.
When you subclass a Canvas component, you can really have some fun. Override the paint() method and add in calls to painting and geometric methods.
Or, you can create a Canvas object and make up your own calls to it that react to
button clicks, mouse movements, and so on.

The Checkbox Class

The Checkbox class is actually two controls in one. Along with checkboxes you
also can implement the functionality of radio buttons by grouping multiple
checkbox controls. Checkboxes are great for giving yes/no options. Radio buttons are good for multiple choice questions.
Figure 9.10 shows a couple of individual checkboxes on the left and a single
group of checkboxes on the right. Here is the code that produced them:
import java.awt.*;
public class testMe extends Frame {
public testMe() {
super("Checkbox Demo");
Panel P1 = new Panel();
Panel P2 = new Panel();
CheckboxGroup G1 = new CheckboxGroup();
setLayout(new GridLayout(1,2));
add(P1);
add(P2);
P1.setLayout(new FlowLayout());
P1.add(new Checkbox("E-Mail Tom"));
P1.add(new Checkbox("E-Mail Jack"));
P2.setLayout(new FlowLayout());

246 Chapter 9
P2.add(new Checkbox("Me", G1, true));
P2.add(new Checkbox("You", G1, false));
P2.add(new Checkbox("Them", G1, false));
resize(300, 200);
show();
}
public static void main(String args[]) {
testMe test = new testMe();
}
}

Hierarchy for Checkbox

java.lang.Object
java.awt.Component
java.awt.Checkbox

DECLARATION

FOR

CHECKBOX

A number of options are available for calling a Checkbox constructor. Here is

each option with a summary description:
new Checkbox();

Constructs a checkbox with no label, no checkbox group, and initializes it to a

false state.
new Checkbox(String);

Constructs a checkbox with the specified label, no checkbox group, and initializes it to a false state.

Figure 9.10
A few iterations of the Checkbox component.

The Java AWT 247

new Checkbox(String, boolean);

Constructs a checkbox with the specified label, no checkbox group, and initializes it to a specified boolean state.
new Checkbox(String, CheckboxGroup, boolean);

Constructs a checkbox with the specified label, specified checkbox group, and
initializes it to a specified boolean state.

Methods for Checkbox

Although checkboxes are simple UI components, they offer us a lot of options
and therefore provide a few custom methods we can use to control them.
GETCHECKBOXGROUP()

This method returns the checkbox group that this checkbox belongs to.
GETLABEL()

This method gets the label of the button.

GETSTATE()

This method returns the boolean state of the checkbox.

SETCHECKBOXGROUP(CHECKBOXGROUP)

This method sets the CheckboxGroup of the check box.

SETLABEL(STRING)

This method changes the label of the checkbox.

SETSTATE(BOOLEAN)

This method sets the checkbox to the specified boolean state.

The Choice Class

The Choice class implements a pop-up menu that shows the currently chosen
option normally; but when it has the focus, it opens up and displays all the
options a user can select from. Figure 9.11 shows a Choice component at rest
and Figure 9.12 shows the same component in action.

248 Chapter 9

Here is the code that creates the Choice component shown in Figures 9.11
and 9.12:
import java.awt.*;
public class testMe extends Frame {
public testMe() {
super("Choice Demo");
Choice C1 = new Choice();
setLayout(new FlowLayout());
add(C1);
C1.addItem("You");
C1.addItem("Me");
C1.addItem("Them");
C1.addItem("Us");
C1.addItem("Everyone");
resize(300, 200);
show();
}
public static void main(String args[]) {
testMe test = new testMe();
}
}

Hierarchy for Choice

java.lang.Object
java.awt.Component
java.awt.Choice

Declaration for Choice

When you are creating a Choice component, you cannot set any options. All the
options must be set after the object is created. The call for initializing Choice is
simply:
new Choice();

Methods for Choice

The set of special methods available for the Choice class include:

The Java AWT 249

Figure 9.11
The Choice component without the focus.

Figure 9.12
The Choice component with the focus.
ADDITEM(STRING )

This method adds an item to the list of choices.

COUNTITEMS ()

This method returns the number of items.

GETITEM(INT)

This method returns the item at the specified index.

GETSELECTEDINDEX()

This method returns the index of the currently selected item.

GETSELECTEDITEM()

This method returns a string representation of the current choice.

SELECT(INT)

This method selects the item with the specified index position.

250 Chapter 9
SELECT(STRING)

This method selects the item with the specified String if present.

The List Class

The List class is actually very similar to Choice except that it allows you to
display multiple items and scroll through the ones not displayed. It also gives
you the ability to select multiple items.
Figure 9.13 shows a form that contains a simple list with multiple elements.
Here is the code that created these lists:
import java.awt.*;
public class testMe extends Frame {
public testMe() {
super("List Demo");
List L1 = new List();
setLayout(new FlowLayout());
add(L1);
L1.addItem("You");
L1.addItem("Me");
L1.addItem("Them");
L1.addItem("Us");
L1.addItem("Everyone");
resize(300, 200);
show();
}
public static void main(String args[]) {
testMe test = new testMe();
}
}

Hierarchy for List

java.lang.Object
java.awt.Component
java.awt.List

The Java AWT

251

Figure 9.13
Create lists with the List class.

Declaration for List

Here are the two constructors you can call to initialize a List class:
new List();

Creates a scrolling list initialized with no visible lines and multiple selections
disabled.
new List(int, boolean);

Creates a new scrolling list initialized with the specified number of visible lines
and a boolean stating if multiple selections are allowed or not.

Methods for List

Youll find a number of methods for this class because of all the different options
it offers.
ADDITEM(STRING )

This method adds the specified item to the end of list.

ALLOWSMULTIPLESELECTIONS()

This method returns true if the list allows multiple selections.

CLEAR()

This method clears the list.

252 Chapter 9
COUNTITEMS ()

This method returns the number of items in the list.

DELITEM(INT)

This method deletes an item from the list at the specified index.
DELITEMS (INT, INT)

This method deletes multiple items from the list. Items are deleted from the
index position specified by the first parameter to the index position specified by
the second parameter.
DESELECT(INT)

This method deselects the item at the specified index.

GETITEM(INT)

This method gets the item associated with the specified index.
GETROWS()

This method returns the number of visible lines in the list.

GETSELECTEDINDEX()

This method returns the selected item in the list or -1 if no item is selected.
GETSELECTEDINDEXES()

This method returns the selected indexes in the list in the form of an array.
GETSELECTEDITEM()

This method returns the selected item in the list or null if no item is selected, or
it returns the first selected item if multiple items are selected.
GETSELECTEDITEMS()

This method returns the selected items in the list into an array of strings.
GETVISIBLEINDEX()

This method gets the index of the item that was last made visible by the method
makeVisible().

The Java AWT

253

ISSELECTED(INT)

This method returns true if the item at the specified index has been selected;
false otherwise.
MAKEVISIBLE(INT)

This method forces the item at the specified index to be visible. The method
automatically scrolls the list to display the specified index.
MINIMUMSIZE()

This method returns the minimum dimensions needed for the list.
MINIMUMSIZE(INT)

This method returns the minimum dimensions needed for the number of rows
in the list.
PREFERREDSIZE()

This method returns the preferred dimensions needed for the list.
PREFERREDSIZE(INT)

This method returns the preferred dimensions needed for the list with the specified amount of rows.
SELECT(INT)

This method selects the item at the specified index.

SETMULTIPLESELECTIONS(BOOLEAN)

This method sets up a list to allow for multiple selections or not.

TextField and TextArea Classes

The TextField and TextArea classes implement other controls that you will use
often. They are often accompanied with Label controls that tell the user what a
text entry box is used for. Text fields can only have a single line of text, while text
areas can have multiple lines like a word processor. One of the nice things about
the TextField and TextArea components is that they interact with both the
keyboard and the mouse. You can enter text by typing it on the keyboard. Then,
you can use the mouse to place the cursor within that text. If you are a Windows

254 Chapter 9

95 user, the components even support the right mouse button. If you right-click
within a text field or text area, a pop-up menu will be displayed like the one
shown in Figure 9.14.

Hierarchy for TextField and TextArea

java.lang.Object
java.awt.Component
java.awt.TextComponent
java.awt.TextField
java.awt.TextArea

Declaration for TextField and TextArea

A number of constructors are available for initializing TextField and TextArea classes:
new TextField();

Constructs a new TextField.

new TextField(int);

Creates a new TextField initialized with the specified number of columns (1

column = 1 character).
TextField(String)

Creates a new TextField initialized with the specified text.

Figure 9.14
Right-clicking on a TextField component to display a pop-up menu.

The Java AWT

255

new TextField(String, int);

Creates a new TextField initialized with the specified text and number of columns.
new TextArea();

Creates a new TextArea.

new TextArea(int, int);

Creates a new TextArea with the specified number of rows and columns.
new TextArea(String);

Constructs a new TextArea with the specified text displayed.

new TextArea(String, int, int);

Creates a new TextArea with the specified text and the specified number of rows
and columns.

Methods for TextField and TextArea

Many of the methods used with these two classes are actually methods of the
TextComponent class which both of these classes subclass. Lets look at these
methods first.
GETSELECTIONEND()

This method returns the selected texts end position.

GETSELECTEDTEXT()

This method returns the selected text contained in the text component.
GETSELECTIONSTART()

This method returns the start position of the selected text. If no text is selected,
the method returns -1.
GETTEXT()

This method returns the text contained in the text component.

256 Chapter 9
ISE DITABLE()

This method returns a boolean value that tells us if the text component is editable
or not. Text components are editable by default.
SELECT(INT, INT)

This method selects the text found between the specified start and end locations.
SELECTALL()

This method causes all of the text in the text component to be selected.
SETEDITABLE(BOOLEAN)

This method sets whether or not this text component can be edited.
SETTEXT(STRING)

This method changes the text of the text component to the specified text.
Now, lets look at a few methods that are specific to the TextField component:
ECHOCHARISSET()

This method returns true if the TextField has a character set for echoing. Echoing is used for situations where you do not want the text people are typing to be
displayed.
GETCOLUMNS()

This method returns the number of columns in the TextField.

GETECHOCHAR()

This method returns the character to be used for echoing. The character is not
returned in a String format, just a simple char.
MINIMUMSIZE()

This method returns the minimum size dimensions needed for the TextField in
columns.
MINIMUMSIZE(INT)

This method is used to request a minimum size for the text box. The parameter
specifies the number of columns for the text box. The method returns the mini-

The Java AWT

257

mum size dimensions needed for the TextField with the specified amount of
columns.
PREFERREDSIZE()

This method returns the preferred size dimensions needed for the TextField class.
PREFERREDSIZE(INT)

This method returns the preferred size dimensions needed for the TextField
with the specified amount of columns being passed to it.
SETECHOCHARACTER(CHAR)

This method sets the echo character for the TextField. Most often youll want to
set this character to the asterisk symbol, especially if you are working with password boxes.
Now, we need to look at the methods specific to the TextArea class:
GETCOLUMNS()

This method returns the number of columns in the TextArea.

GETROWS()

This method returns the number of rows in the TextArea.

INSERTTEXT(STRING, INT)

This method inserts the specified text at the specified position. The position tells
Java the number of characters it needs to move over before it inserts the string.
PREFERREDSIZE()

This method returns the preferred size dimensions of the TextArea.

PREFERREDSIZE(INT, INT)

This method returns the row and column dimensions of the TextArea.
MINIMUMSIZE()

This method returns the minimum size dimensions of the TextArea.

MINIMUMSIZE(INT, INT)

This method returns the specified minimum size dimensions of the TextArea.

258 Chapter 9
REPLACETEXT(STRING, INT, INT)

This method replaces text from the indicated start to end positions with the
specified new text.

The Scrollbar Class

Several of the Java components weve been discussing in this chapter automatically use scrollbars when needed. For example, the TextArea control we just
described will automatically create scrollbars when the text runs off the edge or
the bottom. Scrollbars can also be created as standalone components. Scrollbars
are useful when you need a user to pick a value. They offer a graphical interface
rather than a simple text box where a user would usually have to type in a value.
Scrollbars are implemented using the concepts of a minimum, maximum, and
current value. A scrollbar can be moved in three ways: small increments, large
increments, or to an absolute position. The small increment occurs when the
user clicks on the arrows at the top or bottom of the scrollbar. The large changes
occur when the user clicks on the space between the current position and either
arrow. Finally, the absolute change occurs when the user drags the current location indicator (sometimes called an elevator) and drags it to a specific spot within
the scrollbar. Figure 9.15 shows a standard scrollbar and labels each part.
Lets consider a very typical use of a scrollbar. In this case, we have two components, a Textfield control and a Scrollbar control. We want to display the current value of the Scrollbar in the text box. Also, when the user changes the value
in the text box, the scrollbar should match the change. Figure 9.16 shows what
this simple application looks like.

Figure 9.15
A typical Java scrollbar.

The Java AWT

259

Figure 9.16
A Scrollbar control and a Textfield control with linked values.

And, here is the code for the entire application. Type it in and give it a try:
import java.awt.*;
public class ScrollTest extends Frame {
TextField text;
Scrollbar scroll;
public ScrollTest() {
super("Scroll-Text Test");
text = new TextField(5);
scroll = new Scrollbar(Scrollbar.VERTICAL, 0, 10, 0, 100);
setLayout(new FlowLayout());
add(text);
add(scroll);
resize(300, 200);
show();
}
public boolean handleEvent(Event evt) {
if (evt.target.equals(scroll)) {
text.setText("" + scroll.getValue());
} else if (evt.target.equals(text))
{scroll.setValue(Integer.parseInt(text.getText()));
} else if (evt.id == Event.WINDOW_DESTROY) {
System.exit(0);
return true;
}
return super.handleEvent(evt);
}
public static void main(String args[]) {
ScrollTest test = new ScrollTest();
}
}

260 Chapter 9

Keep in mind that the scrollbar, like all the other controls, will change its appearance to match the operating system. Obviously, you would not want Windows 95 style scrollbars being displayed on a Macintosh. That would sure cause
a commotion!

Hierarchy for Scrollbar

java.lang.Object
java.awt.Component
java.awt.Scrollbar

Declaration for Scrollbar

The three constructors provided for initializing scrollbars include:
new Scrollbar();

Constructs a new vertical Scrollbar.

new Scrollbar(int);

Constructs a new Scrollbar with the specified orientation. You can specify
Scrollbar.HORIZONTAL or Scrollbar.VERTICAL. Scrollbars are vertical
by default.
new Scrollbar(int, int, int, int, int);

Creates a new Scrollbar with the specified orientation, current value, large change
size, minimum value, and maximum value.

Methods for Scrollbar

Here are the set of specialized methods for the Scrollbar class:
GETMAXIMUM()

This method returns an integer representing the maximum value of the scrollbar.
GETMINIMUM()

This method returns an integer representing the minimum value of the scrollbar.

The Java AWT 261

GETORIENTATION()

This method returns an integer that gives us the orientation for the scrollbar.
You can check the returned integer against Scrollbar.HORIZONTAL and
Scrollbar.VERTICAL to determine the scrollbars orientation.
GETVALUE()

This method returns an integer representing the current value of the scrollbar.
GETVISIBLE()

This method returns the visible amount of the scrollbar.

SETVALUE(INT)

This method sets the value of the scrollbar to the specified value. If you try to set
the current value to a number that is greater than the maximum or less than the
minimum, the number becomes the new maximum or minimum, respectively.
SETVALUES(INT, INT, INT, INT)

This method changes the values for the scrollbar. The arguments in order of
appearance include:

value - the position in the current window

large change - the value that will be moved each time the user clicks in the
area between the elevator and the arrows. This value is also called the amount
visible per page.
minimum - the minimum value of the scrollbar
maximum - the maximum value of the scrollbar

Building Menus
Menus in Java are as easy to build and manage as the other visual components.
Every part of a menu, from the menu bar to individual items is represented as a
separate class, each having specialized properties and methods. Lets start our
investigation of menus with the main componentthe menu bar. Then, well
work our way down into the other more specialized components, such as a menu
itself and menu items.

262 Chapter 9

The MenuBar Class

Each window or frame you create in a Java application or applet can have its
own menu bar. The standard menu bar is always displayed along the top of the
window. You are not allowed to have multiple menu bars within a single window unless you create the entire system yourself (but why would you want to!).
The MenuBar class itself provides little functionality. By itself it looks like a
blank bar across the top of your window. To give your menuing system functionality, you need to add Menu and MenuItem components, which we will
discuss later.

Hierarchy for MenuBar

java.lang.Object
java.awt.MenuComponent
java.awt.MenuBar

Declaration for MenuBar

Only a single constructor is provided for initializing a MenuBar class. This constructor creates a menu object but does not display it. You must first assign the
MenuBar to a frame for it to be displayed:
new MenuBar();

Methods for MenuBar

ADD(MENU)

This method adds the specified menu to the menu bar.

COUNTMENUS()

This menu returns an integer representing the number of menus on the

menu bar.
GETHELPMENU()

This method returns the name of the menu component on the current menu
bar that has been designated as the Help menu. Help menus are discussed in
the next section in more detail.

The Java AWT 263

GETMENU(INT)

This menu gets the specified menu. Input is an integer representing the index
position of a menu and it returns a Menu object.
REMOVE(INT)

This method removes the menu located at the specified index from the menu bar.
REMOVE(MENU)

This method removes the specified menu from the menu bar.
SETHELPM ENU(MENU)

This method sets the current help menu to the specified menu on the menu bar.

The Menu Class

The Menu class is used to implement the selections for the section headings for
each type of menu. Typical Menu components will be labeled File, Options,
and Help. Figure 9.17 shows a few menus.
Menu classes are always children of a single MenuBar class. They usually have
MenuItem classes as children. However, a Menu class does not need to have menu
items under it; it can react to events on its own. In Figure 9.17, you could click on
the Menu component labeled File to expose its MenuItem children. However, if
you were to click on the Help Menu item, it would not display any child
MenuItem components because it does not have any. It acts on its own.

Hierarchy for Menu

java.lang.Object
java.awt.MenuComponent
java.awt.MenuItem
java.awt.Menu

Figure 9.17
Examples of menus created with the Menu class.

264 Chapter 9

Declaration for Menu

Here are the two constructors provided for the Menu class:
new Menu(String);

Constructs a new Menu with the specified string as the label. This menu will
not be able to be torn off. Tear-off menus are menus that will still appear on
the screen after the mouse button has been released.
new Menu(String, boolean);

Constructs a new Menu with the specified label. The menu will be able to be
torn off if the boolean value is set to true.

Methods for Menu

The specialized methods for the Menu class include:
ADD(MENUITEM )

This method adds the specified item to the menu.

ADD(STRING)

This method adds an item with the specified label to the menu.
ADDSEPARATOR()

This method adds a separator line to the menu at the current position.
COUNTITEMS ()

This method returns the number of elements in the menu as an integer.

GETITEM(INT)

This method returns the menu item located at the specified index of the menu.
ISTEAROFF()

This method returns true if the menu is a tear-off menu.

REMOVE(INT)

This method deletes the item at the specified index from the menu.

The Java AWT

265

REMOVE(MENUITEM)

This method deletes the specified item from the menu.

The MenuItem Class

The MenuItem class is the last in the line of children of the three menu classes.
We should mention, however, that Menu classes can have other Menu classes as
children. These child Menu components look like menu items, but another
menu would be under each menu to display more menu items. This is technique is used to create cascading menus as shown in Figure 9.18.
Here is the Java code that implements the menu system shown in Figure 9.18:
import java.awt.*;
public class testMe extends Frame {
public testMe() {
super("Menu Demo");
MenuBar MB = new MenuBar();
Menu M1 = new Menu("File");
Menu M2 = new Menu("Options");
Menu M3 = new Menu("More");
MB.add(M1);
MB.add(M2);
MB.add(new Menu("Help"));
M1.add(new MenuItem("Open"));
M1.add(new MenuItem("Close"));
M1.add(new MenuItem("Save"));
M2.add(new MenuItem("General"));
M2.add(M3);
M3.add(new MenuItem("Screen"));
M3.add(new MenuItem("Font"));
setMenuBar(MB);
resize(300, 200);
show();
}
public static void main(String args[]) {
testMe test = new testMe();
}
}

The AWT also provides a subclass of the MenuItem class called

CheckboxMenuItem. This class is identical to the standard MenuItem class

266 Chapter 9

Figure 9.18
Creating cascading menus using Menus as subclasses.

except that it provides the ability to be checked or unchecked when the user
clicks on a menu item. Figure 9.19 shows an example of this component.

Hierarchy for MenuItem

java.lang.Object
java.awt.MenuComponent
java.awt.MenuItem

Declaration for MenuItem

The MenuItem class provides a single constructor that takes one argument:
new MenuItem(String);

Figure 9.19
Using the CheckboxMenuItem to check and uncheck menu items.

The Java AWT 267

This class constructs a new MenuItem with the specified String displayed as the
menu components label. Note that the hyphen symbol (-) is reserved to mean a
separator between menu items. Separators should do nothing except delineate
different sections of a menu.

Methods for MenuItem

The specialized methods for the MenuItem class include:
DISABLE()

This method makes the menu item unselectable by the user and grays it out.
ENABLE()

This method makes the menu item selectable by the user. The user is given a
visual cue when the menu is disabled because it is grayed out.
ENABLE(BOOLEAN)

This method conditionally enables a component.

GETLABEL()

This method gets the label for the menu item. The value is returned as a string.
ISE NABLED()

This method checks to see if the menu item is enabled. The return value is a
boolean value.
SETLABEL()

This method sets the label to be the specified string.

The following two methods are used only with the CheckboxMenuItem component.
GETSTATE()

This method returns a boolean value that represents the state of the menu item.
SETSTATE(BOOLEAN)

This method sets the state of the menu item.

268 Chapter 9

Creating a Sample Menu Application

Now that weve introduced you to each of the three key classes for creating
menus (MenuBar, Menu, and MenuItem), youre probably anxious to write a
Java program that puts them to work. Lets build a sample application that incorporates all of them to create a practical menuing system. The application will
not respond to any of the menu items being chosen, but it will show you the
basic techniques for constructing a menuing system.
We need to start by planning our menu. Lets use three Menu components with
the labels File, Options, and Help. Under the File menu we will have
seven MenuItems including two separators. The Options Menu component
will have two CheckboxMenuItem components, and the Help menu will not
have any menu items associated with it.
Figures 9.20 through 9.22 show several different views of our test program with
different menu options being chosen. The code that creates this menuing system is as follows:
import java.awt.*;
class TestFrame extends Frame {
TestFrame() {
super("Menu Test");
MenuBar mb = new MenuBar();
Menu fileMenu = new Menu("File");
Menu optionMenu = new Menu("Option");
Menu helpMenu = new Menu("Help");
fileMenu.add(new MenuItem("New"));
fileMenu.add(new MenuItem("-"));
fileMenu.add(new MenuItem("Open"));
fileMenu.add(new MenuItem("Close"));
fileMenu.add(new MenuItem("Save"));
fileMenu.addSeparator();
fileMenu.add(new MenuItem("Exit"));
optionMenu.add(new CheckboxMenuItem("Large Fonts"));
optionMenu.add(new CheckboxMenuItem("Save Settings on Exit"));
mb.setHelpMenu(helpMenu);
mb.add(fileMenu);
mb.add(optionMenu);
mb.add(helpMenu);
setMenuBar(mb);
resize(300,200);
show();
}

The Java AWT 269

public boolean action(Event evt, Object obj) {
String label = (String)obj;
if (evt.target instanceof MenuItem) {
if (label.equals("Exit")) {
System.exit(0);
return true;
}
}
return true;
}
public boolean handleEvent(Event evt) {
if (evt.id == Event.WINDOW_DESTROY) {
System.exit(0);
return true;
}
return super.handleEvent(evt);
}
public static void main(String args[]) {
TestFrame tf = new TestFrame();
}
}

You my notice that we used two different methods for creating the two separators in the File menu. The addSeparator() method is probably easier. However, if using the standard add() method with the hyphen character, you can
create a full-fledged menu item that you can then change options for and set up
so that it can respond to mouse clicks. Now that youve seen the basics for
creating menus and GUI objects, youll need a way to position them within the
frames you build. And thats where the layout manager comes in.

Figure 9.20
Menu test app view #1.

270 Chapter 9

Figure 9.21
Menu test app view #2.

Figure 9.22
Menu test app view #3.

Working with the Layout Manager

To most programmers, the layout manager is a pretty strange concept. Instead
of telling Java exactly where you want components to be located within a window, you use the layout manager to actively place interface components depending on the container size, shape, and on a few variables you set.
All controls you create for a Java application must always use the layout manager. Even if you do not implicitly tell Java what to do, it defaults to using a
particular version of the layout manager. The layout manager comes in different
flavors for different situations. At this point in the evolution of Java, five different layout managers are provided:

The Java AWT 271

FlowLayout
BorderLayout
GridLayout
GridBagLayout
CardLayout

The layout manager is actually just an Interface. You then create classes that
implement the LayoutManager interface. These classes are used by the AWT
and your program to get the look and feel you want for the user. Lets look at
each of the layout manager classes in detail.
Any component that is a container in Java can use a different layout. So, you
could have a Frame that uses one class to lay out two panels which each have
their own layouts, and so on.

The FlowLayout Class

The FlowLayout class is the default layout for all containers in Java. This class
tries to lay out components in a very orderly fashion that is much like a word
processor that wraps your words and centers them. The order in which you add
components to a container using FlowLayout is vital. The first component you
add is the first in line and the last will be placed at the end. It is possible to go
back and insert and remove components, but it is much easier to do it all correctly at the beginning.
Figure 9.23 shows several buttons on a panel. As you can see, they are centered
as a group within the panel. If we resize the panel, the buttons automatically
align themselves again to be centered. If we continue to shrink the panel so that
all the buttons no longer fit across the panel, the last button will be wrapped
to the next row as shown in Figure 9.24.

Declaration for FlowLayout

To set the layout for a container, you use the setLayout() method. This method
has a single argument which is an instance of one of the layout classes. Lets look
at an example. The code shown here is used to create the arrangements shown in
Figures 9.23 and 9.24:

272 Chapter 9

Figure 9.23
Using the FlowLayout class to lay out some buttons.

Figure 9.24
The same panel resized so that the buttons wrap to the next row.
import java.awt.*;
class LayoutFrame extends Frame {
LayoutFrame() {
super("Layout Test");
setLayout(new FlowLayout());
add(new Button("Button 1"));
add(new Button("Button 2"));
add(new Button("Button 3"));
add(new Button("Button 4"));
add(new Button("Button 5"));
resize(300,200);
show();
}

The Java AWT 273

public static void main(String args[]) {
LayoutFrame lf = new LayoutFrame();
}
}

Here we call the setLayout() method of the Frame we have extended, sending it
a new instance of the FlowLayout class. The frame will then query the FlowLayout
object where to position each component. This query happens whenever the
window needs to be refreshed, such as when it is resized or uncovered.

Methods for FlowLayout

The specialized methods for FlowLayout include:
LAYOUTCONTAINER(CONTAINER)

This method lays out the container. It will actually reshape the components in
the target container to satisfy the constraints of the FlowLayout object.
MINIMUMLAYOUTSIZE

(CONTAINER)

This method returns the minimum dimensions needed to lay out the components contained in the specified target container. These dimensions are extremely
useful because they can help you ensure that the user will not shrink the container to such a small size that it forces some of the UI components to slip off the
visible screen.
PREFERREDLAYOUTSIZE(CONTAINER)

This method returns the preferred dimensions for this layout given the components in the specified target container. The return value is a Dimension variable
consisting of two values representing the width and height of the layout.
TOSTRING()

This method returns a string representation of this FlowLayouts values, including: (in this order) X position, Y position, container dimensions, layout class
being used, whether or not the container can be resized, and the title of
the container.

274 Chapter 9

The BorderLayout Class

The BorderLayout class creates a unique layout scheme that is useful for working with components that need to maintain their position relative to an edge of
your container. Border layouts use a compass analogy to allow you to specify
which side of the container to attach your control to.
The controls you use for border layouts are automatically sized to take up as
much space as possible. Figure 9.25 shows a sample container with five panels,
one in each area of the container that the border layout allows.
To create this sample, we used the same code from the previous figure and made
a few very minor changes. In fact, all the changes take place in the constructor
for the class, so we will only show you that:
LayoutFrame() {
super("Layout Test");
setLayout(new BorderLayout());
add("North", new Button("North"));
add("East", new Button("East"));
add("South", new Button("South"));
add("West", new Button("West"));
add("Center", new Button("Center"));
resize(300,200);
show();
}

The first change is obviously the switch to specifying the BordeLayout() class as
our layout scheme. The other changes occur in the add() method. What are

Figure 9.25
The BorderLayout() class in action.

The Java AWT

275

those extra strings doing there? They specify which side of the container to place
the new control. In this case, we are using buttons with labels that match the
position names (North, East, South, West, and Center). We used buttons here
for clarity sake, but you would probably not use them for any real projects.
Panel components are probably the best for this type of layout. You would specify
a few panels using a border style layout and then use a different layout scheme
for the individual panels. Since panels are for design purposes mostlythey do
not show up since they default to the same color as the background of the frame
they blend in perfectly and bring the whole thing together.

Declaration for BorderLayout

There are two ways to declare a border layout. You can use the simple constructor we
used in the sample application or you can also specify a vertical and horizontal space.
new BorderLayout();

The simple way. Constructs a new BorderLayout.

new BorderLayout(int, int);

Constructs a BorderLayout with the specified gaps. The first integer represents
the horizontal gap to be placed between components and the second integer
represents the vertical gap to be used.

Methods for BorderLayout

The specialized methods for BorderLayout include:
ADDLAYOUTCOMPONENT(STRING, COMPONENT)

This method adds the specified named component to the layout. The String
argument gives us a name to refer to the component within the layout. The
component can be any type of interface component we want to add.
LAYOUTCONTAINER(CONTAINER)

This method lays out the specified container. It will actually reshape the components in the specified target container to satisfy the constraints of the
BorderLayout object.

276 Chapter 9
MINIMUMLAYOUTSIZE(C ONTAINER)

This method returns the minimum dimensions needed to lay out the components contained in the specified target container. The return value is a dimension variable.
PREFERREDLAYOUTSIZE(CONTAINER)

This method returns the preferred dimensions for the layout given the components in the specified target container. The method also returns a dimension
style variable.
REMOVELAYOUTCOMPONENT(COMPONENT)

This method removes the specified component from the layout.

TOSTRING()

This method returns the string representation of the BorderLayouts values. At

this point in the development of Java, this method only returns the size of the
horizontal and vertical gaps.

The GridLayout Class

The grid style layout works just like it sounds; it creates a grid pattern for laying
out components. You can control the number of rows and columns as well as the
space between the components and the space between the components and the
edge of the container. Figure 9.26 shows a sample frame using a grid layout.

Figure 9.26
Using the GridLayout() class to create a sample frame.

The Java AWT 277

Declaration for GridLayout

Like the BorderLayout class, GridLayout gives you a few constructor options. If
the number of rows or columns is invalid, you will get an error (i.e., no negative
numbers). Also, if you specify more columns than you need, the columns will
not be used and your grid will appear as if youve specified fewer columns. However, if you specify more rows than you need, they will show up as blank space
with enough space that matches the amount of space taken up by a component
in this layout. We think this is a bug in Java, so you should experiment with this
phenomenon if you plan on using this layout style.
new GridLayout(int, int);

Creates a grid layout with the specified rows and columns.

new GridLayout (int, int, int ,int);

Creates a grid layout with the specified rows, columns, horizontal gap, and vertical gap.

Methods for GridLayout

The specialized methods for GridLayout include:
ADDLAYOUTCOMPONENT(STRING, COMPONENT)

This method adds the specified named component to the layout. The String
argument gives us a name to refer to the component within terms of the layout.
The component can be any interface component you want to add.
LAYOUTCONTAINER(CONTAINER)

This method lays out the specified container. It will actually reshape the components in the specified target container to satisfy the constraints of theGridLayout
object.
MINIMUMLAYOUTSIZE(CONTAINER)

This method returns the minimum dimensions needed to lay out the components
contained in the specified target container. The return value is a dimension variable.

278 Chapter 9
PREFERREDLAYOUTSIZE(CONTAINER)

This method returns the preferred dimensions for this layout given the components in the specified target container. It also returns a dimension style variable.
REMOVELAYOUTCOMPONENT(COMPONENT)

This method removes the specified component from the layout.

TOSTRING()

This method returns the string representation of the GridLayouts values. At

this point in the development of Java, this method only returns the size of the
horizontal and vertical gaps.

The GridBagLayout Class

The GridBagLayout class is one of the most versatile of all the layout classes. It
allows you to create a grid of an arbitrary size and use it to create components
within the grid that are of variable size.
You use this layout method to align components vertically and horizontally,
without requiring that the components be the same size or without them being
sized for you in ways you may not want. Each GridBagLayout uses a rectangular
grid of cells, with each component occupying one or more cells.
Each component that resides in a container that is using a GridBagLayout has
an associated GridBagConstraints instance that specifies how the component is
laid out within the grid. How a GridBagLayout places a set of components
depends on each components GridBagConstraints, minimum size, and preferred size.
To use a GridBagLayout effectively, you must customize one or more of its
components GridBagConstraints. Here are some of the variables you need to
customize to create a layout:

gridx, gridy Specifies the cell in the grid at the upper left of the component. The upper-left-most cell of a container has address gridx=0, gridy=0.
gridwidth, gridheight Specifies the width and height of our component in grid
space terms. You can set either of these to GridBagConstraints.REMAINDER to
make the component be the last one in its row or column.

The Java AWT 279

fill Used when the components display area is larger than the components
requested size to determine whether (and how) to resize the component.
ipadx, ipady Specifies the internal padding. Padding represents the amount
of space to add to the minimum size of the component. The width of the
component will be at least its minimum width plus ipadx*2 pixels (since the
padding applies to both sides of the component). Similarly, the height of the
component will be at least the minimum height plus ipady*2 pixels.
insets Sets the external padding of the componentthe minimum amount
of space between the component and the edges of its display area.
anchor Used when the component is smaller than its display area to determine where to place the component. Valid values are:
GridBagConstraints.CENTER (the default)
GridBagConstraints.NORTH
GridBagConstraints.NORTHEAST
GridBagConstraints.EAST
GridBagConstraints.SOUTHEAST
GridBagConstraints.SOUTH
GridBagConstraints.SOUTHWEST
GridBagConstraints.WEST
GridBagConstraints.NORTHWEST

weightx, weighty Used to determine how to distribute space; this is important for specifying resizing behavior. Unless you specify a weight for at least
one component in a row and column, all the components clump together in
the center of their container. This is because when the weight is zero (the
default), the GridBagLayout puts any extra space between its grid of cells
and the edges of the container.

Figure 9.27
Sample program using the GridBagLayout class.

280 Chapter 9

It is probably easiest to give you an example. Figure 9.27 shows the layout we
wish to end up with. Following it is the code that we used to create it:
import java.awt.*;
import java.util.*;
public class GridBagTest extends Frame {
GridBagTest() {
super("GridBag Test");
GridBagLayout gridbag = new GridBagLayout();
GridBagConstraints c = new GridBagConstraints();
setFont(new Font("Helvetica", Font.PLAIN, 14));
setLayout(gridbag);
c.fill = GridBagConstraints.BOTH;
c.weightx = 1.0;
makebutton("Button1", gridbag, c);
makebutton("Button2", gridbag, c);
makebutton("Button3", gridbag, c);
c.gridwidth = GridBagConstraints.REMAINDER; //end row
makebutton("Button4", gridbag, c);
c.weightx = 0.0;
//reset to the default
makebutton("Button5", gridbag, c); //another row
c.gridwidth = GridBagConstraints.RELATIVE; //next-to-last in row
makebutton("Button6", gridbag, c);
c.gridwidth = GridBagConstraints.REMAINDER; //end row
makebutton("Button7", gridbag, c);
c.gridwidth = 1;
//reset to the default
c.gridheight = 2;
c.weighty = 1.0;
makebutton("Button8", gridbag, c);
c.weighty = 0.0;
//reset to the default
c.gridwidth = GridBagConstraints.REMAINDER; //end row
c.gridheight = 1;
//reset to the default
makebutton("Button9", gridbag, c);
makebutton("Button10", gridbag, c);
resize(300, 100);
show();
}
protected void makebutton(String name, GridBagLayout gridbag,
GridBagConstraints c) {
Button button = new Button(name);
gridbag.setConstraints(button, c);
add(button);
}

The Java AWT 281

public static void main(String args[]) {
GridBagTest test = new GridBagTest();
}
}

Declaration for GridBagLayout

Since most of the work of setting up a GridBagLayout is achieved using a
GridBagConstraints class, the constructor for GridBagLayout is very simple, in
fact, it requires no arguments at all:
new GridBagLayout();

Methods for GridBagLayout

The specialized methods for GridBagLayout include:

DUMPCONSTRAINTS(GRIDBAGCONSTRAINTS)
This method prints the layout constraints to the System object. It is useful for
debugging.
GETCONSTRAINTS(C OMPONENT)

This method returns a copy of the GridBagConstraints object for the specified
component.
LAYOUTCONTAINER(CONTAINER)

This method lays out the specified container. This method will actually reshape
the components in the specified target container to satisfy the constraints of the
GridBagLayout object.
LOOKUPCONSTRAINTS(COMPONENT)

This method retrieves the constraints for the specified component. The return
value is not a copy, but is the actual constraints class used by the layout mechanism. The object returned by this method can then be altered to affect the looks
of the component.
MINIMUMLAYOUTSIZE(CONTAINER)

This method returns the minimum dimensions needed to lay out the components contained in the specified target container. The return value is a dimension variable.

282 Chapter 9
PREFERREDLAYOUTSIZE(CONTAINER)

This method returns the preferred dimensions for this layout given the components in the specified target container. It also returns a dimension style variable.
SETCONSTRAINTS(COMPONENT,

GRIDBAGCONSTRAINTS)

This method sets the constraints for the specified component.

TOSTRING()

This method returns the string representation of the GridBagLayouts values.

At this point in the development of Java, this method only returns the size of the
horizontal and vertical gaps.

The CardLayout Class

The card layout style is much different than the previous four. Instead of using
this class to lay out your controls, you use it like a layering system to specify
which layer certain controls appear on. The most common use for this class is to
simulate tabbed dialogs. You can create a card layout containing several panels
that each use their own layout method and controls. Then, you can make a call
to the class to specify which card to display and therefore which panel and respective components to show.

Declaration for CardLayout

Two versions of the CardLayout() constructor are availableone requires no
parameters. To add extra cards, you simply use the add() method and the layout
manager automatically handles the extra cards. Optionally, you can specify the
horizontal and vertical gaps that surround your cards.
new CardLayout();
// Creates a new card layout.
new CardLayout(int, int);

Creates a card layout with the specified horizontal and vertical gaps.

Methods for CardLayout

The specialized methods for CardLayout include:

The Java AWT 283

ADDLAYOUTCOMPONENT(STRING, COMPONENT)

This method adds the specified named component to the layout. The String
argument gives us a name to refer to the component within terms of the layout.
The component can be any interface component you want to add.
FIRST(CONTAINER)

This method flips to the first card. The argument is the parent container that
you assigned the layout style to.
LAST

(CONTAINER)

This method flips to the last card of the specified container.

LAYOUTCONTAINER(CONTAINER)

This method lays out the specified container. This method will actually reshape
the components in the specified target container to satisfy the constraints of the
CardLayout object.
MINIMUMLAYOUTSIZE(CONTAINER)

This method returns the minimum dimensions needed to layout the components contained in the specified target container. The return value is a dimension variable.
NEXT(CONTAINER)

This method flips to the next card in the stack.

PREFERREDLAYOUTSIZE(CONTAINER)

This method returns the preferred dimensions for the layout given the components in the specified target container. This also returns a dimension style
variable.
PREVIOUS(CONTAINER)

This method flips to the previous card.

REMOVELAYOUTCOMPONENT(COMPONENT)

This method removes the specified component from the layout.

284 Chapter 9
SHOW(CONTAINER,

STRING)

This method flips to the specified component name in the specified container.
This method is best used when you cannot use any of the previous four methods
and/or you want to switch directly to a specified card. The Container argument
specifies the owner of the card layout and the string is the name of the component you wish to switch to.
TOSTRING()

This method returns the string representation of the CardLayouts values. At

this point in the development of Java, this method only returns the size of the
horizontal and vertical gaps.

Index
A
Abstract classes, 33, 121, 158
Abstract methods, 132
Abstract Window Toolkit, 32
Action( ) method, 294
Add method, 274
Addition, 98
Addressing
Internet, 353
Animation
buffering, 40
speed, 26
API documentation, 64
Applet class
hierarchy, 288
methods available, 288
methods derived, 290
Applets, 5, 6
browser interaction, 294
class, 287
closing, 51
compiling, 53
defined, 21
drawbacks, 291
file access, 292
file execution, 292
fonts, 43
images, 298
navigation sample, 293
network communication, 292
package, 31

parameters, 39
passing information, 367
sounds, 297
tags, 39
threading, 51
ticker tape sample, 22
vs. applications, 21
Appletviewer, 7
Applications
command line arguments, 86
defined, 21
networking, 353
sample menu, 268
vs. applets, 21
Architecture natural, 5
Arguments
accessing, 88
indexing, 89
numeric, 89
passing, 87
reading, 87
Arrays, 16, 82
accessing data, 86
declaring, 82
elements, 83
indexing, 85
multidimensional, 85
sizing, 83
Assignment boolean operators, 102
Assignment operators, 93, 95
Audio clips, 297
AWT, 31, 227
AWTError, 191

403

404 Index
class hierarchy, 230
components, 229
defined, 32
importing, 228
layout manager, 228
menus, 229

B
Bandwidth considerations, 360
Binary, 97
Binary integer operators, 99
Binary operators, 98
Bitwise complement, 97
Bitwise operators, 99
Blocking, 217
Body (class), 128
Boolean data type, 78
Boolean operators
assignment, 102
evaluation, 101
logical, 100
negation, 100
ternary, 102
BorderLayout class, 274
declaration, 275
methods, 275
Break statement, 110
Browsers
HotJava, 6
Netscape, 26
BufferedInput Stream class, 327
BufferedOutputStream class, 327
Buffering, 40, 45
Button class
declaration, 243
getLabel( ), 244
hierarchy, 243
setLabel( ), 244
Buttons, 32
Byte streams, 321
Byte type, 76
ByteArrayInputStream class, 328

ByteArrayOutputStream class, 328

Bytecodes, 5, 53

C
Canvas class
declaration, 245
hierarchy, 244
paint( ), 245
CardLayout class, 282
declaration, 282
methods, 282
Case-sensitivity
declarations, 36
package names, 171
parameters, 40
variable names, 36
Casting
interfaces, 165
vs. creating, 151
Casts, 103
Catch statements, 187
Catch( ) method, 185
Catching errors, 186
CGI. See Common Gateway
Interface
Char data type, 79
Character arrays, 16
Character literals, 73
Checkbox class, 245
declaration, 246
getCheckboxGroup( ), 247
getLabel( ), 247
getState( ), 247
hierarchy, 246
setCheckboxGroup( ), 247
setLabel( ), 247
setState( ), 247
Choice class, 247
addItem( ), 249
countItems( ), 249
declaration, 248, 251
getItem( ), 249

Index
getSelectedIndex( ), 249
getSelectedItem( ), 249
hierarchy, 248, 250
methods, 251
select( ), 250
Classes, 5
abstract, 33, 121, 158
advantages, 116
applet, 287
body, 128
bufferedInputStream, 328
bufferedOutputStream, 327
button, 243
byteArrayInputStream, 328
byteArrayOutputStream, 328
canvas, 244
casting, 150
checkbox, 245
choice, 247
component, 290
container, 290
dataInputStream, 330
dataOutputStream, 330
declaring, 116
defined, 32
documenting, 63
error, 191
event, 304
exception, 193
extending, 124
fileInputStream, 333
fileOutputStream, 333
filterInputStream, 335
filterOutputStream, 335
final, 33, 123
flowLayout, 271
frame, 235
fully qualified name, 118
hiding, 177
identifiers, 124
importing packages, 176
InetAddress, 354
inputStream, 325
label, 241

lineNumberInputStream, 337
list, 250
menuItem, 265
modifiers, 33, 119
name space, 34, 129
naming, 124
networking, 353
object, 34
outputStream, 325
panel, 238
pipedInputStream, 339
pipedOutputStream, 339
printStream, 340
private, 33
protocols, 158
public, 33, 120
pushbackInputStream, 342
runtime, 194
scrollbar, 258
sequenceInputStream, 342
socket, 355
stringBufferInputStream, 343
super( ), 142
superclass, 34
System, 321
textArea, 253
textField, 253
throwable, 182
URL, 364
variables, 148
WriteAFile, 185
CLASSPATH, 171, 173, 174
Client, 350
Client/server technology, 350
Code parameter, 27
Color method, 37
Command line arguments, 86
indexing, 89
numeric, 89
passing arguments, 87
reading, 87
Comments, 30
styles, 59
tags, 67

405

406 Index
Common Gateway Interface, 10
Compilers, 7, 53
Component class
bounds( ), 232
disable( ), 232
enable([Boolean]), 232
getFontMetrics( ), 232
getGraphics( ), 232
getParent, 232
handleEvent(Event evt), 232
hide( ), 233
inside(int x, int y), 233
isEnabled( ), 233
isShowing( ), 233
isVisible( ), 233
locate(int x, int y), 233
location( ), 233
move(int x, int y), 233
repaint( ), 233
resize( ), 234
setFont( ), 234
show([Boolean]), 234
size( ), 235
Components, 60
Compound expressions, 104
Compound statements, 106
Constructors, 37, 138
body, 146
calling, 140
declaring, 140
FontMetrics, 48
Java default, 142
modifiers, 143
object creation, 148
protected, 143
public, 143
return type, 139
Container class, 290
Control flow, 106
Control structures
do...while, 108
for, 110
if...else, 106
labels, 111

list of, 107

switch, 109
while, 108
Controls, 229
buttons, 243
canvas, 244
checkbox, 245
components, 231
frame, 235
label, 241
layout manager, 270
lists, 250
menus, 229, 263
panel, 238
pop-up menus, 247
scrollbar, 258
text areas, 253
text fields, 253
Converting values
casting, 150

D
Data types, 35
boolean, 78
byte, 76
casting, 103
char, 79
double, 78
float, 78
int, 71, 77
long, 71, 77
separators, 75
short, 76
string, 79
variables, 76
DataInputStream class, 330
DataOutputStream class, 330
Debugging, 181
Decrement operator, 98
Destroy( ) method, 222
Developers Kit, 17
Directories

Index
search path, 174
Disassembler program, 17
Distributed programming, 6
Distributed software, 10
Division, 98
Do...while, 108
Doc comment clauses, 119
Documenting classes, 63
Double buffering, 45
Double data type, 78

E
Encapsulation, 43
Equal to operators, 102
Error handling, 181
Errors
catching, 186
checking, 323
file not found, 189
input/output, 185
throws, 133
try clauses, 186
Evaluation operators, 101
Event class, 304
methods, 306
variables, 305
Event handling, 53
Events
hierarchy, 313
processing problems, 318
system, 315
types, 304
Exceptions, 15, 181, 182
class, 193
creating, 200
error, 191
file not found, 189
finally statements, 189
handler, 182
IOException, 185
try clauses, 186
URL, 364

Executable content, 10
Export statement, 228
Expressions
assignment, 105
writing, 104
Extending classes, 124
Extends keyword, 34

F
Fatal errors, 191
File
access, 292
execution, 292
input/output, 321
saving, 173
File Transfer Protocol. See FTP
FileInputStream class, 333
FileOutputStream class, 333
FilterInputStream, 335
FilterOutputStream class, 335
Final classes, 33
Final methods, 132
Finally statement, 189
Finger protocol, 349
Float data type, 78
Floating-point, 72
operators, 102
FlowLayout class, 271
declaration, 271
methods, 273
Font metrics, 48
Fonts, 43
For loops, 110
Frame class, 235
declaration, 235
dispose( ), 237
getIconImage( ), 237
getMenuBar, 237
getTitle( ), 237
hierarchy, 235
isResizable( ), 238
remove( ), 238

407

408 Index
setCursor( ), 238
setIconImage( ), 238
setMenuBar( ), 238
setResizeable( ), 238
setTitle( ), 238
FTP, 349

G
Garbage collection, 6, 15, 37
Gateways, 355
Graphical User Interface
button class, 243
canvas class, 244
checkbox class, 245
choice class, 247
component class, 231
frame class, 235
label class, 241
lists, 250
menu class, 263
menu items, 265
menuBar class, 261
panel class, 238
scrollbar class, 258
text areas, 253
text fields, 253
Graphics methods, 46
GridBagLayout class, 278
declaration, 281
methods, 281
variables to customize, 278
GridLayout class, 276
declaration, 277
methods, 277

H
Header files, 13
Height parameter, 27
Helper programs, 17
Hexadecimal format, 71

History of Java, 8
HotJava, 6, 10
HTML. See Hyper Text Markup
Language
applet tags, 39
HTTP, 349
Hyper Text Markup Language, 25
Hyper Text Transfer Protocol. See HTTP

I
Identifiers, 65
class, 118
classes, 124
errors, 67
If...else, 106
Image buffer, 41
Images, 298
Implements clause, 126
Implements keywords, 34
Import statements, 31, 228
Including packages, 31
Increment operator, 98
Index, 84
InetAddress class, 354
Init( ), 130
Input streams, 321, 324
InputStream class, 325
methods, 325
Instanceof operator, 17, 168
Int data type, 77
Integers, 71
literals, 72
operators, 93, 97
Interfaces, 34, 158
casting, 165
class, 126
declaring, 161
design issues, 160
implementation tips, 167
implementing, 161
implements clauses, 126
keyword, 161

Index
layout manager, 271
runnable, 34
tips on using, 165
Internet
addressing, 353
java.net package, 352
Request for Comments, 351
IOException, 324

J
Java language
advantages, 4
benefits, 11
compared to C++, 9
developers kit, 7, 17
history, 8
interfaces, 158
jargon, 5
tools, 8
virtual machine, 6
JAVAC, 7, 53
JAVADOC.EXE, 63
Java-enabled, 7
JAVAP, 17
JavaScript, 7
Just-in-Time compiler, 7

K
Keyboard events, 311
keyDown( ), 311
keyUp( ), 311
Keywords, 68
class, 124
extends, 34, 124
implements, 34, 162
interface, 161
list of keywords, 69
super, 135
this, 50, 135

L
Label class, 241
declaration, 241
getAlignment( ), 242
getText( ), 242
hierarchy, 241
setAlignment( ), 242
setText( ), 243
Labels, 111
Layout manager, 228, 270
borderLayout class, 274
cardLayout class, 282
flowLayout class, 271
gridBagLayout class, 278
gridLayout class, 276
Lexical structures, 58
comments, 59
identifiers, 65
keywords, 68
separators, 75
LineNumberInputStream class, 337
List class, 250
Literals, 71
character, 73
numeric, 71
Logical operators, 100
Long data type, 77
Long integers, 71

M
Main programs, 27
Menu class, 263
declaration, 264
hierarchy, 263
methods, 264
MenuBar class, 262
declaration, 262
hierarchy, 262
methods, 262
MenuItem class, 265

409

410 Index
declaration, 266
hierarchy, 266
methods, 267
Menus, 32
creating, 229
Methods, 7, 38, 130
abstract, 132
action( ), 294
add( ), 274
applet class, 288
body, 134
catch( ), 185
color, 37
constructors, 138
createImage( ), 230
declaring, 130
defined, 28
destroy( ), 222
disable( ), 230
documenting, 63
drawString( ), 48
final, 132
getGraphics( ), 41
getMessage, 198
getParameter( ), 40
graphics, 46
handleEvent( ), 53
hide( ), 230
init( ), 28, 130
main( ), 87
modifiers, 131
native, 132, 292
overloading, 137
overriding, 43, 137, 170
paint( ), 29, 44
parameter lists, 133
parse( ), 89
private, 132
protected, 131
public, 131
resume( ), 220
return type, 133
Run( ), 29, 214

sleep( ), 50
start( ), 29
static, 132
stop( ), 51, 221
suspend( ), 220
synchronized, 132
throwing an exception, 194
throws, 133
valueOf( ), 89
write( ), 184
yield, 221
Modifiers
abstract, 121
constructor, 143
final, 123, 150
method, 131
modifiers, 33, 119
public, 120
transient, 150
volatile, 150
Modulus operator, 99
Mouse events, 304, 307
mouseDown( ), 307
mouseDrag( ), 309
mouseEnter( ), 310
mouseExit( ), 310
mouseMove( ), 309
mouseUp( ), 308
Multidimensional arrays, 85
Multiple inheritance, 14
Multiplication, 98
Multithreading, 7, 208
grouping, 226
synchronizing, 222

N
Name space, 129
Native methods, 132, 292
Negation operator, 97
Netscape
applet, 294
Network communication, 292

Index
Network News Transfer Protocol.
See NNTP
Networking, 347
between applets, 367
classes, 353
client/server, 350
concerns, 360
java.net, 352
ports, 350
protocols, 348
sockets, 355
URLs, 364
New lines, 356
NNTP, 349
Not equal to operators, 102
Numeric literals, 71

O
Object-oriented programming, 12
Objects
arrays, 82
class, 34
creation, 148
declaring, 118
Octal integers, 71
Operators, 74, 93
addition, 98
assignment, 95, 102
binary, 98
binary integer, 99
bitwise, 99
bitwise complement, 97
boolean negation, 100
compound expressions, 104
decrement, 98
division, 98
equal to, 102
evaluation, 101
floating-point, 102
increment, 98
instanceof, 17, 168
integer, 97

logical AND, 100

modulus, 99
multiplication, 98
negation, 97
not equal, 102
precedence, 93
subtraction, 98
ternary, 102
Output streams, 324
class, 325
Overloading methods, 137
Overriding methods, 137

P
Packages, 30
applet, 31
awt, 31
case sensitivity, 171
classes, 30
creating, 168
documenting, 63
import keyword, 169
java.io, 322
java.lang, 30
java.net, 352
naming, 170
public classes, 172
standard Java, 177
Paint( ) method, 44
Panel class, 238
declaration, 240
hierarchy, 240
setlayout( ), 241
Parameter lists
constructor, 146
Parameters, 39
code, 27
height, 27
speed, 26
values, 27
width, 27
Parsing, 89

411

412 Index
Performance issues
threading, 51
PipedInputStream class, 339
PipedOutputStream class, 339
Pointers, 13
Ports, 350
Internet, 351
numbers, 350
Precedence (operators), 93
PrintStream class, 340
Private
constructors, 143
methods, 132
Processing parameters, 39
Protected
constructors, 143
methods, 131
Protocols
class, 158
finger, 349
FTP, 349
Internet, 351
NNTP, 349
Request for Comments, 351
SMTP, 348
TCP/IP, 348
WhoIs, 349
Public
classes, 33, 120
constructors, 143
keyword, 162
method, 131
PushbackInputStream class, 342

R
Request for Comments. See Request for
Comments
Resizing, 239
Resource allocation, 37
Resume( ) method, 220
Return type, 133
Returns, 356

RFCs. See Request for Comments

Run method, 214
Runnable interface, 213
Runtime class, 194

S
Savings files, 173
Scripting language, 7
Scrollbar class, 258
hierarchy, 260
methods, 260
Security, 12, 15, 292
Seprators, 75
SequenceInputStream class, 342
Servers, 350
sample, 361
setting up, 360
ServerSocket class, 360
Shadowing, 129
Short type, 76
Simple Mail Transfer Protocol.
See SMTP
Simple statements, 105
Single inheritance, 121
Sleep( ) method, 50, 219
Socket class, 360
Sockets, 355
Sounds, 297
Source code
saving, 173
Statements, 105
catch, 187
compound, 106
control flow, 106
finally, 189
simple, 105
switch, 109
using semi-colons, 106
writing, 104
Static methods, 132
Status bar, 296
Stop( ) method, 221

Index
Streams, 321
inputStream, 324
outputStream, 324
String arrays, 16
String type, 79
StringBufferInputStream class, 343
Subclasses, 44
Subtraction, 98
Super classes, 16
Super keyword, 135
Super( ), 142
Suspend( ) method
suspending execution, 220
Switch, 109
Synchronized methods, 132
System class, 321
system.in, 322
System events, 315
action( ), 317
handleEvent( ), 317

T
Tags, 67
TCP/IP, 348
Ternary operators, 102
TextArea class, 253
declaration, 254
hierarchy, 254
methods, 255
TextField class, 253
declaration, 254
hierarchy, 254
methods, 255
This keyword, 50, 135
ThreadGroup, 226
Threads, 29, 49, 182, 207, 212
blocking, 217
creating, 211
destroy( ) method, 222
first in first out, 217
grouping, 226
initializing, 215

413

life cycle, 218

priority, 217
resuming, 220
run( ) method, 219
runnable interface, 213
sleep( ) method, 219
start( ) method, 219
stop( ) method, 221
subclassing, 212
suspending execution, 220
synchronizing, 222
when to use, 210
while loops, 210
yield( ) method, 221
Throws, 133
constructor, 146
exceptions, 194
Transient modifiers, 150
Transmission Control Protocol.
See TCP/IP
Try clauses, 186
Types, 76

U
Unary, 97
Unicode, 73
Uniform Resource Locator. See URLs
URLs, 364
User input, 52
User interface
component class, 231
layout manager, 271
menus, 229

V
Variable declarations, 35
Variables
constructors, 37
declarations, 79
modifiers, 149
naming, 36

414 Index
static, 149
variables, 148
vs. types, 76
Virtual machine, 6, 210
Volatile modifiers, 150

W
Web sites
Coriolis, 25
Javasoft, 54
While, 108

While loops, 210

WhoIs protocol, 349
Widening, 151
Width parameter, 27
Wild cards
hiding classes, 177
Windows, 32

Y
Yield( ) method, 221