John Sweeney Copyright 2013 by

Rochester Institute of John Sweeney

Technology No rights reserved
Rochester, New York Public Domain
[email protected]

Introductory Java 7 Computer Programming for Mathematicians

by

John Sweeney
Rochester Institute of Technology
 Java for Mathematicians/Page 2

Chapter One: First Java Program

Topic: Goal

I hope students of mathematics will find this book

as a helpful start to their use of the java computer
programming language.

Developing computer programming skills requires

extensive experience using a computer programming
language to solve problems. Being aimed at
mathematicians this book has problems that exclusively
focus on mathematics. Assuming a reader has completed
introductory courses in algebra, trigonometry, and
geometry this book contains references to mathematics
easily comprehended by novice mathematicians. The set
of problems do not focus on any area of mathematics,
nor do they cover all of mathematics. The problems
chosen for this book on java do not form an
introduction to mathematics, nor any area of
mathematics. Instead the chosen mathematical topics
 Java for Mathematicians/Page 3

and problems are intended to help the reader maintain

interest in completing the exercises and complete this
introduction to the java programming language. The
problems chosen for this book emphasize the
development of computer programming skills,
specifically the java computer programming language.

Since 1995 the java computer language has proven

itself as a good computer language for students
learning computer programming techniques built on
object-oriented concepts using a single processor.
Amongst students java is notorious for requiring all
computer programming variables to be defined before
being used. This contrasts with many popular
introductory computer programming languages such as
Visual Basic in which the computer allows variables to
hold any type of data and not even requiring their
definition before you begin using a variable. The
extensive structure and rules of java permit the
computer compiler to detect your programming errors
and provide a more appropriate error message. The
compiler does not understand the purpose of your
computer program, nor the student's approach to
solving a problem. The compiler can only enforce the
rules of the programming language, so all error
messages can be cryptic.

Developing a computer program requires an

extensive amount of time devoted to understanding the
problem. Next the programmer should develop
appropriate data sets to test the proposed computer
program. Only then should coding begin. Testing and
fixing errors in your computer program so much time
 Java for Mathematicians/Page 4

that it becomes the major portion of the entire

development effort. Although java programs do require
more typing the effort devoted to typing is a small
portion of the entire project. Studying java as their
first computer language has proven to assist students
in their learning of additional computer programming
languages.

Java has proven to be a difficult language to

teach as a first computer programming language. Most
books beg the reader to show patience since parts of
the their first java programs will not be discussed or
explained until a delay of several chapters. Every
effort has been made to explain, although sometimes
only in a cursory fashion, java constructs as soon as
they appear in example programs. A simple java program
involves a large number of concepts, so even a simple
introductory java program can overwhelm novice java
programmers.

This book emphasizes object-oriented programming

concepts. Java provides good support for your study of
those concepts. Example programs support the writing
of object-oriented solutions which are easily
comprehended by other computer programmers. Most
computer development involves large teams of
programmers and computer solutions to mathematical
problems should be reviewed by other mathematicians so
a clear programming style has importance to your
future work.

Studying a computer language, especially java,

requires a commitment of a large amount of your time
 Java for Mathematicians/Page 5

and energy. Carefully compare the effort to learn

java with its benefits to your future work. Learning
java simply to earn an academic grade will ensure that
you will quickly forget everything about java.
Completion of each chapter's exercises will aid in the
development of introductory object-oriented
programming skills using a single processor computer.
Employing java as a tool to solve problems is a
higher-level of learning. Such a large effort on your
part may lead to long term retention of the core
concepts of object-oriented programming using java.
 Java for Mathematicians/Page 6

Topic: Preparation

Besides this book you will need the java

development kit and a text editor or integrated
development environment(IDE). You can download the
java development kit from Oracle's Web site at
http://www.java.com. A quick method to locate the
appropriate file is to use a search engine, such as
Google, with the search string of “download JDK”. JDK
is an abbreviation of java development kit. SE is the
standard edition of java. Java SE 7 was used to
develop the examples in this book. Although the screen
image shown below includes the internet address of
that download, it may change in the future, so a Web
search for “download JDK 7” is preferable. You should
always
 Java for Mathematicians/Page 7

Topic: IDE

The examples in this book were written using the

jGRASP IDE available for download at
http://www.jgrasp.org. Many good quality IDEs are
available for your use, you should feel free to use
any of them. jGRASP was developed for novice
programmers and is used to develop small programming
projects.

You should enter the java program shown as

example 1 into your chosen IDE, then compile and run
your first java program. Consult your IDE's manual for
directions to complete that exercise.

The exercises in this book will benefit you best

 Java for Mathematicians/Page 8

if you write your own computer programs to further

your exploration of mathematics. Only with practice
can you make progress and develop expertise in writing
java programs.

Example 1: The Source code of your first java

computer language program

/**
* The FamousMathQuotes class
* This computer program prints a famous quote of
Carl Friedrich Gauss.
*/
class FamousMathQuotes {

public static void main(String[] args) {

System.out.println("Carl Friedrich Gauss
described mathematics as the Queen of the Sciences.");
// Display the string.
}

Exercise: Search the internet for other famous

sayings by mathematicians. Print a variety of famous
sayings by mathematicians.

You type the source code for your computer

program into the text area provided by your IDE.

Topic: Comments
 Java for Mathematicians/Page 9

Java programs have three ways to include

comments: block comments, inline comments and end of
line comments. This program uses a block comment. The
top four lines of the FamousMathQuotes program are a
block comment describing the purpose of the entire
program. /** marks the start of that comment and */
marks its end. Everything between those markers is a
comment. Notice that the beginning marker contains two
asterisks, not one. Customarily IDEs automatically
begin each new line of such comments with an asterisk
and a space, but they are not required. Java provides
two other methods for adding comments to your
program's source code.

Topic: Commenting Conventions

The java community of programmers has developed

extensive guidelines for the contents of comments.
Portions of those guidelines will be introduced in
later chapters. For now the beginning comment, which
is known as a doc comment, satisfies its goal of
describing the purpose of this java program. Doc
comments provide descriptions of the program or
method's goal, algorithms employed, and data
structures used. Doc comments should be separated by a
blank line from any preceding code. The first line
begins with /**. Following lines begin with an
optional asterisk, *, to reinforce the visual
perception of being a line of comment. The final line
begins with */. Thus /** begins the doc comment and */
ends that doc comment.
 Java for Mathematicians/Page 10

/**
* Program goal, algorithms, and
* data structures are described here.
*/

Oracle provides more information at

http://www.oracle.com/technetwork/java/codeconventions
-141999.html#385

The java compiler ignores all comments, which are

written solely to help us comprehend the computer
program. A second style of marker begins with /* and
ends with */. Programmers often use this style of
comment when multiple lines of description are
required to properly explain a portion of the computer
program. The third style of comment begins with // and
continues to the end of the line. Notice // Display
the string. near the end of the example 1 program.
Such comments are a popular method to include a short
description of the command on the same line. Because
the compiler will ignore all comments, you could
correctly assume comments are not required, thus the
same program could simple be written as shown in
example 1A.

Topic: example java program 1A

class FamousMathQuotes {

public static void main(String[ ] args) {

System.out.println("Carl Friedrich Gauss
described mathematics as the Queen of the Sciences.");
 Java for Mathematicians/Page 11

Topic: Class

The computer program begins with "class." A class

describes an object, an example of such might be a
robot, which would have attributes and abilities. By
the standard naming convention we begin a class name
with a capital letter. The left curly bracket, or open
brace, symbol { begins the body of that class, whose
end is marked by the right curly bracket, or closing
brace, } symbol. Of course those symbols forms teams.
Forgetting to match brackets is an error and will
block the creation of the associated java program and
its execution. Most IDEs will automatically align
those symbols so you can easily notice the beginning
and end of a class's body. There are two different
standards for curly braces. The first standard puts
the left curly brace { at the end of the class line
with the closing right curly brace on a line by
itself. Then the closing curly brace, } aligns with
the class command, but the opening left curly brace
does not align.

class FamousMathQuotes {
...
}

Move your cursor over such a symbol and your IDE

might highlight the matching symbol. Attributes are
described early in the class. This class has no
 Java for Mathematicians/Page 12

attributes.
Within the class you will notice "public static
void main(String[] args)" which describes the method
named "main." In a world of objects the modifier
"public" means other objects can see and use your
method named "main." Classes are used to make objects.

The modifier static simply means the method stays

with the class and is not included with each object.
Recall you write methods to implement the object's
abilities, so our objects of type FamousMathQuotes
would have no abilities. Thus any object made from
class FamousMathQuotes has neither attributes nor
methods. They would be objects in name only.

The modifier void means your method does not

produce, or return, anything. All the action occurs
with the command: System.out.println("Carl Friedrich
Gauss described mathematics as the Queen of the
Sciences."); System is the name of a special class
containing attributes and methods of use generally to
all java programs. Out is the output print stream,
which is an attribute of the class System. Your IDE
automatically connects that output print stream with
the console of your IDE. For jGRASP users look at the
bottom of jGRASP's window where you will find it under
the tab labeled "Run I/O". "I/O" is an abbreviation of
"input / output" meaning your java program prints to
that area and there you can type replies to your
program's questions. All java methods are followed by
( and ). The string "Carl Friedrich Gauss described
mathematics as the Queen of the Sciences." must begin
and end with double quotation marks.
 Java for Mathematicians/Page 13

Topic: Example java program 2

/**
* The FamousMathQuotes class
* simply prints a famous quote of Carl Friedrich
Gauss to standard output.
*/
class FamousMathQuotesV2 {

/**
* Method go prints
* a famous quote of Carl Friedrich Gauss to
standard output.
*/
private void go()
{
System.out.println("Carl Friedrich Gauss
described mathematics as the Queen of the Sciences.");
// Display the string.
}

/**
* The main method is automatically executed.
*/
public static void main(String[] args) {
FamousMathQuotesV2 entity = new
FamousMathQuotesV2();
entity.go();
}

}
 Java for Mathematicians/Page 14

Study the second java program example. The class

has been renamed FamousMathQuotesV2 where V2 means the
second version. Each object has a method named "go"
which prints Gauss' famous saying. That method is
private, meaning only that object can use that method.
The FamousMathQuotes class had no attributes nor
methods. Our FamousMathQuotesV2 entity now has an
ability to print a famous saying.

Notice how the main method has changed. The print

statement has been moved to the go method. Now you
notice a command to create an object of type
FamousMathQuotesV2() by using the "new" command
followed by the name of the class. The new variable
named "entity" is of type FamousMathQuotesV2 so our
definition is the type then the variable's name thus
"FamousMathQuotesV2 entity." The equals command
stores the address of our new object in variable
"entity." The next line "entity.go();" calls the go
method of the object. Then that go method will print
our famous mathematician's saying.

Each method also has its own block comment. The

doc comment begins with /** and ends with */. Each
method and class should begin with a doc comment. A
doc comment begins with a brief description of the
method or class. Then a list of @param statements
describe each parameter. The @return statement
describes the value to be returned by a method.

Obviously our example 2 program produces the same

result as example 1. Example 2 demonstrates the role
of an object. You can use either approach with your
 Java for Mathematicians/Page 15

own java programs.

 Java for Mathematicians/Page 16

Chapter Two: java primitives

Topic: Using java primitives to represent

numbers

The followers of Pythagoras knew the first few

perfect numbers which equal the sum of their proper
positive divisors. For example 6 is the sum of 1, 2,
and 3 which are all its proper positive divisors.
Although 6 divides 6 it is not considered to be a
proper divisor. For more information concerning
perfect numbers you can consult Wikipedia at
http://en.wikipedia.org/wiki/Perfect_number.

Topic: Example java program 1

/**
* Print the sum of 6's proper positive divisors.
*/
public class SumPositiveDivisors {

/**
* This method prints the sum of 6's proper
positive divisors.
*/
private void go()
{
int divisor1 = 1;
int divisor2 = 2;
int divisor3 = 3;
int sum = divisor1 + divisor2 + divisor3;
System.out.println(sum); // Display the
answer.
}
 Java for Mathematicians/Page 17

/**
* The main method is automatically executed.
* Create an object of class SumTwoIntegers
* then call its method named go.
*/
public static void main(String[] args) {
SumPositiveDivisors entity = new
SumPositiveDivisors();
entity.go();
}

In addition to objects the java language includes

primitives, such as int to represent integers and
double to represent real numbers. All numbers in a
computer system are represented by binary numbers. An
int has a size of four bytes. In java the byte is
standardized at eight bits. An int contains 32 bits
thus providing for integers from -2,147,483,648 and a
maximum value of 2,147,483,647(inclusive). If your
work involves larger integers you can use the long
primitive providing double the width for a size of 8
bytes or 64 bits and integer values from
-9,223,372,036,854,775,808 and a maximum value of
9,223,372,036,854,775,807 (inclusive). For more
information consult Oracle's Web site on java
primitives at
http://docs.oracle.com/javase/tutorial/java/nutsandbol
ts/datatypes.html.

You can easily use int values in a command such

as sum = divisor1 + divisor2 + divisor3, but first
 Java for Mathematicians/Page 18

java demands that you define each variable. A

definition for divisor1 includes int, the variable's
type, and divisor1, the variable's name. You can also
set that variable to an initial value by using the =
operator. Thus int divisor1 = 1; defines a new int
value with a name of divisor1 and sets it to an
initial value of 1.

A definition consists of the following: type name

= initial value; We have seen this pattern in chapter
one in the command: FamousMathQuotesV2 entity = new
FamousMathQuotestV2(); This program has examples such
as int divisor1 = 1; and int divisor2 = 2;

Because the use of dynamic objects is considered

to be a better programming paradigm our future
examples will use that approach. Static methods and
classes will be limited to supporting roles.

Topic: Computer program error

The java compiler might produce the error message

“cannot find symbol” which confuses most novice
programmers.

The line number 13 helps programmers find the

source of the compiler error. The carat symbol ^
underneath divisor1 also helps identify the source of
 Java for Mathematicians/Page 19

the compiler's complaint. Most development

environments allow the programmer to display line
numbers. Also the erroneous line will be highlighted.

Compilers begin by reading each line from the top

of the class. The blanks neatly separate the line into
symbols. Either the compiler recognizes the symbol as
part of the java language or the compiler checks the
list of previously defined variables. When the symbol
cannot be found the compiler displays the message
error: cannot find symbol. Programmers should check
the spelling and check that all variables are defined
before being used.

Topic: Another computer programming error

 Java for Mathematicians/Page 20

Here is another example of a computer programming

error. The compiler cannot find the variables
divisor1, divisor2, and divisor2.

The computer program is shown below.

 Java for Mathematicians/Page 21

Those variables divisor1, divisor2, and divisor3

are defined above the go method, but those variables
are in the main method. During compilation the
compiler keeps a table of all variables and their
scope which means where the variable is known. Because
divisor1 is defined in method main its scope is method
main. Only commands within main can use that variable.
Thus the go method cannot use those variables.

Because variables divisor1, divisor2, and divisor3

are used only in the go method those variables should
be defined at the top of the go method.

Topic: Another programming error

 Java for Mathematicians/Page 22

This error message describes another way to

mistakenly type your java computer program.

Java identifiers are case sensitive. Thus Entity

is not same identifier as entity. Notice that error on
line 25.

Topic: java coding conventions

 Java for Mathematicians/Page 23

The java coding convention described at

http://www.oracle.com/technetwork/java/codeconv-
138413.html encourages programmers to begin the names
of classes with a single capital letter. Variables
should begin with lowercase letters. Thus entity
should begin with a lowercase letter, not a capital
letter.

Notice all three of the prior examples of program

errors produced the same error message cannot find
symbol. Only by understanding how the compiler works
and knowing all the rules of the java programming
language can you understand the cause of those errors.

Topic: Execution error

Find the error in the following java program's

source code. The program sums 1, 2, and 3 and prints
an answer of 5, not 6. No compiler error message is
displayed and the program executes seemingly in a
normal manner.

1 /**
2 * Print the sum of 6's proper positive divisors.
3 */
4 public class SumPositiveDivisorsV5 {
5
6 /**
7 * This method prints the sum of 6's proper
positive divisors.
8 */
9 private void go()
 Java for Mathematicians/Page 24

10 {
11 int divisor1 = 1;
12 int divisor2 = 2;
13 int divisor3 = 3;
14 int sum = divisor1 + divisor1 + divisor3;
15 System.out.println(sum); // Display the
answer.
16 }
17
18 /**
19 * The main method is automatically executed.
20 * Create an object of class SumTwoIntegers
21 * then call its method named go.
22 */
23 public static void main(String[] args) {
24 SumPositiveDivisorsV5 entity = new
SumPositiveDivisorsV5();
25 entity.go();
26 }
27
28 }

In this instance divisor1 is shown where divisor2

should be included in the list. Execution errors are
notoriously difficult to spot and fix. Apparently
people gloss over such slight errors and may even read
the second divisor1 as divisor2. We see what we
intended to type, not the reality of a typo.

To detect such errors programmers produce by hand

a list of the sample data, such as 1, 2, and 3, and
the expected results of 6. Only by carefully comparing
the computer program's result with our expectation can
 Java for Mathematicians/Page 25

we possibly detect an execution error.

Topic: approximations

Wikipedia lists historical approximations for pi

at http://en.wikipedia.org/wiki/Approximations_of_%CF
%80 where %CF%80 is the code for pi. The following
program calculates the values for approximations
credited to Ahmes, the Babylonians, and Archimedes.

/**
* Print three historical approximations of pi.
*/
public class ApproximatePi {

/**
* This method prints the sum of 6's proper positive
divisors.
*/
private void calc()
{
System.out.println("Math.pi: " + Math.PI);

double ahmesPi = 256.0/81.0;

System.out.println("Ahmes 256/81: \t" +
ahmesPi); // Display the answer.

double babylonianPi = 25.0/8.0;

System.out.println("Babylonian 25/8: " +
babylonianPi); // Display the answer.

double archimedesPi = 22.0/7.0;

System.out.println("Archimedes 22/7: " +
 Java for Mathematicians/Page 26

archimedesPi); // Display the answer.

}

/**
* The main method is automatically executed.
* Create an object of class ApproximatePi
* then call its method named calc.
*/
public static void main(String[] args) {
ApproximatePi object = new ApproximatePi();
object.calc();
}

The program displays the following results.

ÏÏÏMath.pi: 3.141592653589793
ÏÏÏAhmes 256/81: 3.1604938271604937
ÏÏÏBabylonian 25/8: 3.125
ÏÏÏArchimedes 22/7: 3.142857142857143

Notice how Amhes' aproximation is 256/81 but our

java computer program uses 256.0 / 81.0. The reason is
that 256 is an int value while 256.0 is a double
value. When java evaluates 256 / 56 both values 256
and 56 are int values, so the resulting answer will be
an int value of 3. Likewise the Babylonian
approximation of 25/8 would result in 3 and the
Archimedes approximation of 22/7 would produce and int
value of 3. Java would not produce any compiler errors
nor warnings. The results will simply be 3.
 Java for Mathematicians/Page 27

Both 256.0 and 81.0 are double values so the

result would be a double value. You might think that
defining the result variable as a double would change
the result, but it would not. The result would be a
double value of 3.0 not an int value of 3. Java
evaluates the expression to the right of the
assignment operator (=) and assigns that result to the
variable on the left of the assignment operator (=).
The result of 256 / 81 is evaluated first as an int
value of 3. Only later is that int value of 3 assigned
to the double value of ahmesPi as 3.0.

The following erroneous version of our pi

approximation program calculates 256/81.

/**
* Print three historical approximations of pi.
*/
public class ApproximatePi {

/**
* This method prints the sum of 6's proper positive
divisors.
*/
private void calc()
{
System.out.println("Math.pi: " + Math.PI);

double ahmesPi = 256/81;

System.out.println("Ahmes 256/81: \t" +
ahmesPi); // Display the answer.
 Java for Mathematicians/Page 28

double babylonianPi = 25.0/8.0;

System.out.println("Babylonian 25/8: " +
babylonianPi); // Display the answer.

double archimedesPi = 22.0/7.0;

System.out.println("Archimedes 22/7: " +
archimedesPi); // Display the answer.
}

/**
* The main method is automatically executed.
* Create an object of class ApproximatePi
* then call its method named calc.
*/
public static void main(String[] args) {
ApproximatePi object = new ApproximatePi();
object.calc();
}

The result of running the erroneous pi

approximation program is shown here. The Ahmes pi
variable contains 3.0.

Math.pi: 3.141592653589793
Ahmes 256/81: 3.0
Babylonian 25/8: 3.125
Archimedes 22/7: 3.142857142857143

When java must evaluate an expression composed of

mixed types such 256.0 / 81 which contains a double
value and an int value, the int value will first be
 Java for Mathematicians/Page 29

promoted to a double. Then the expression 256.0 / 81.0

will be evaluated producing a result of type double.
Such promotions occur automatically without any
compiler warnings.

When a value of type double must be assigned to a

variable of type int the compiler will produce an
error message because the fractional portion of that
double value would be truncated thus information would
be lost. If you wanted to know only the whole number
portion of a double value use the cast command (int).
For example Math.PI is a value of type double, so a
cast of type int is required for the following
command. The compiler will notice the cast and avoid
producing an error message.

int wholeIntPortion = (int)Math.PI;

When a number is too large to be stored for the

variable's data type, that assignment will cause an
overflow condition. Although the resulting value is
certainly incorrect java does not produce any warning
or errors.

/**
* Print an example overflow condition for an int.
*/
public class OverflowError {

/**
* This method prints the sum of 6's proper positive
divisors.
*/
 Java for Mathematicians/Page 30

private void calc()

{
System.out.println("maximum int: " +
2_147_483_647); // maximum int value 2,147,483,647

System.out.println("maximum int + 1: : " +

( 2_147_483_647 + 1) ); // overflow.
}

/**
* The main method is automatically executed.
* Create an object of class OverflowError
* then call its method named calc.
*/
public static void main(String[] args) {
OverflowError object = new OverflowError();
object.calc();
}

The execution of that program produces these

erroneous results. The computer programmer must
constantly consider the impact of potential overflow
conditions and test results.

Ïmaximum int: 2147483647

Ïmaximum int + 1: : -2147483648

Many areas of recreational mathematics such as

the search for prime numbers and square numbers have
already produced results exceeding the size of an int
 Java for Mathematicians/Page 31

value. A lack of compiler errors and a seemingly

successful execution of a computer program does not
guarantee correct values. Computer programmers must be
vigilant in checking the results of their computer
programs.

Although ancient Egyptians, Mesopotamians, and

Greeks used approximations to pi, the entire field of
approximate values is ignored in most books on
mathematics. The comic XKCD gives us their own
treatment of the topic at HTTP://XKCD.COM/1047/
 Java for Mathematicians/Page 32

Topic: Activity

Write your own java program to verify an

approximation from Wikipedia's history of mathematics
or XKCD.
 Java for Mathematicians/Page 33

Chapter Three: First Java Program

Topic: Classes

Java classes can represent mathematical objects

such as a square. For more information concerning
squares you can consult Wikipedia at
http://en.wikipedia.org/wiki/Square. All squares are
similar. They vary only in the length of the side.
Although many descriptions of a square are possible,
this chapter defines a class for a square solely by
the length of its side.

When a class describes an object, that class has

a name representing the object. A good name for such a
class would be “Square.” Thus the class's code begins
as follows. Remember a class's name begins with a
capital letter.

/**
* This class represents a square in two dimensions.
*/
public class Square {

Each class begins with a list of attributes for

the object it describes. Class Square needs only one
attribute named side which we represent with a Java
double primitive. A Java primitive is an item which is
not an object. So far in this book, you have worked
with the primitives int and double. Now the code for
class Square appears as follows.
 Java for Mathematicians/Page 34

/**
* This class represents a square in two dimensions.
*/
public class Square {

double side;

Let sq be the name of a square of class Square.

Your test program can easily access the contents of
sq's side by using the class name with the attribute
in a command such as System.out.println(sq.side);
Unfortunately a bad command such as sq.side = -5;
could also easily insert a bad value. Programmers
avoid such mistakes by making side a private
attribute. Then only methods within class Square can
access or change its contents. The accessor method
normally begins with get. Mutator methods begin with
set. Now class Square's code contains an accessor and
a mutator methods.

Topic: Example java program 1

/**
* This class represents a square in two dimension.
*/
public class Square {

private double side;

private double area;

/**
 Java for Mathematicians/Page 35

* This default constructor makes objects of class

Square.
*/
public Square() {
side = 0.0;
area = 0.0;
}

/**
* This accessor method returns the length of the
square's side.
*
* @return the length of the square's edge
*/
public double getSide() {
return side;
}

/**
* This mutator method sets the length of the
square's side.
*
* @param _side the new length for the square's edge
*/
public void setSide(double _side) {
side = _side;
area = side * side;
}

/**
* This accessor method returns the area of the
square.
*
 Java for Mathematicians/Page 36

* @return the area of the square

*/
public double getArea() {
return area;
}

Suppose you had made a square of class Square

with the name of square. System.out.println(“The
square has an edge of length “ + square.getSide());
would display the value of attribute side.
Square.setSide(5) would set the square to a size of 5
by 5. The list of arguments for a method is shown in
the parentheses. The argument for setSide method has
the new size of the square. A common name would be
side, but that name would cause confusion because the
class attribute is already named side. You might use a
different name such as newSizeValue or simply begin
with name with an underscore _size. Java recognizes
size and _size as different names representing
different values.

A different solution to avoid confusion uses the

special identifier this for the current object.
“this.side” means the object's attribute named side.
“side” automatically refers to the closest identifier
which is in the same method's list of arguments. The
setSide method uses this.side = side; as shown below.

Topic: Example java program 2

//**
 Java for Mathematicians/Page 37

* This class represents a square in two dimensions.

*/
public class Square {

private double side;

/**
* This accessor method returns the length of the
square's side.
*
* @return double the area of the square
*/
public double getSide() {
return side;
}

/**
* This mutator method sets the length of the
square's side.
*
* @param side the new length for the square's edge
*/
public void setSide(double side) {
this.side = side;
}

Notice java allows two different variables to

have the same name: side. The class Square has an
attribute side known throughout the entire class, but
not outside that class. Likewise the list of arguments
of method setSide has a variable also named side,
which is know throughout that method, but not outside
that method. Confusion can occur because the range of
 Java for Mathematicians/Page 38

the class Square includes the method setSide. Within

method setSide both the attribute side and the
argument side exist. There they share the same name,
side, but have different meanings. To avoid that
confusion java automatically uses the nearest
enclosing definition. Thus any reference to side
within method setSide automatically means the
argument, not the class attribute. Most programmers
avoid this concern by using different names. For
example the following class has been changed to use
the parameter name of _side.

/**
* This class represents a square in two dimension.
*/
public class Square {

private double side;

/**
* This default constructor makes objects of class
Square.
*/
public Square() {
side = 0.0;
}

/**
* This accessor method returns the length of the
square's side.
*
* @return the length of the square's edge
*/
 Java for Mathematicians/Page 39

public double getSide() {

return side;
}

/**
* This mutator method sets the length of the
square's side.
*
* @param _side the new length for the square's edge
*/
public void setSide(double _side) {
side = _side;
area = side * side;
}

The technical word, scope, means the part of the

class where the variable is known. _side is known only
in method setSide. Attribute side is known throughout
the entire class including method setSide.

Class square could also include an attribute for

the area of the square. The following example program
includes that new attribute named area.

/**
* This class represents a square in two dimension.
*/
public class Square {

private double side;

private double area;
 Java for Mathematicians/Page 40

/**
* This default constructor makes objects of class
Square.
*/
public Square() {
side = 0.0;
area = 0.0;
}

/**
* This accessor method returns the length of the
square's side.
*
* @return the length of the square's edge
*/
public double getSide() {
return side;
}

/**
* This mutator method sets the length of the
square's side.
*
* @param _side the new length for the square's edge
*/
public void setSide(double _side) {
side = _side;
area = side * side;
}

/**
* This accessor method returns the area of the
 Java for Mathematicians/Page 41

square.
*
* @return the area of the square
*/
public double getArea() {
return area;
}

Now the setSide method performs a double duty of

setting the length of the square's side and the
square's area. Normally methods produce only a single
result, so this is not the ideal design for this
class. A better approach would rewrite the getArea
method to access a pseudo-attribute. Instead of simply
returning the value of area, the getArea method will
calculate and return that result. The class Square now
appears to contain another attribute named area. Such
attributes are sometimes called pseudo-attributes.
Many IDEs will automatically generate the javadoc for
a class.
Java for Mathematicians/Page 42
 Java for Mathematicians/Page 43

When you create an object of class Square you use

command new. An example command would be Square obj =
new Square(); The command automatically calls the
class constructor. Constructors are special methods
usually written immediately after the class
attributes. Constructors must have the public
attribute and not return any value. Do not even use a
return type of void. Default constructors normally set
initial values for all the class attributes. Ints are
set to 0. Doubles to 0.0 and Strings to null or “”.
Thus class Square has the default constructor shown
below.
 Java for Mathematicians/Page 44

/**
* This default constructor makes objects of class
Square.
*/
public Square() {
side = 0.0;
}
 Java for Mathematicians/Page 45

Square obj = new Square(); makes a square with

name obj and its attribute side has the value 0.0.

Classes also have a fully parametrized

constructor to properly set the initial values for all
attributes.

/**
* This fully parametrized default constructor makes
objects of class Square.
*/
public SquareV2(double _side) {
side = _side;
}

Square obj2 = new Square(5.2); makes a square with

name obj2 and its attribute side has the value 5.2.
Class Square now has the code shown below.

Topic: Example java program

//**
* This class represents a square in two dimensions.
*/
public class Square {

private double side;

/**
* This default constructor makes objects of class
Square.
*/
public Square() {
 Java for Mathematicians/Page 46

side = 0.0;
}

/**
* This fully parametrized default constructor makes
objects of class Square.
*/
public Square(double _side) {
side = _side;
}

/**
* This accessor method returns the length of the
square's side.
*/
public double getSide() {
return side;
}

/**
* This mutator method sets the length of the
square's side.
*/
public void setSide(double _side) {
side = _side;
}

Beside the length of the side squares have a list of

other properties. Such a list might include area,
parameter, diagonal, location of its center, the
surrounding circle, etc. You could write the Java code
to calculate each of those other properties in your
test class, or class Square could include methods to
return the results of those computations. Then those
 Java for Mathematicians/Page 47

new methods act as pseudo-attributes. For area you

might include an accessor, get, method as shown below.

/**
* This accessor method returns the area of the
square.
*/
public double getArea() {
// return side * side; or side raised to the
power of 2
return Math.pow(side,2);
}

The class Math is always available for your use.

Math contains many useful methods such as pow that
raises a double to a power. Searching the Java API for
the Math class you would find the pow method.

The method pow is static so our command begins

with the name of the class Math. Squaring a number
means to raise that number to the power 2, so our
command is Math.pow(side,2).

A square's perimeter is four times the length of

a side, so the following method will perform that
computation.

/**
* This accessor method returns the area of the
square.
*/
public double getPerimeter() {
final int NUMBER_SIDES = 4;
return side * NUMBER_SIDES;
 Java for Mathematicians/Page 48

Notice the final attribute at the start of the

declaration for NUMBER_SIDES. Final limits the program
to exactly one change to the value of that variable.
Because NUMBER_SIDES is set to 4, it cannot be changed
again. Such constants have names in all capitals with
the words separated by underscore characters. That
declaration is included in method getPerimeter because
NUMBER_SIDES is used only in that method. Because
NUMBER_SIDES is inside method getPerimeter the other
methods cannot use that constant. In technical
terminology the scope of NUMBER_SIDES is the method
getPerimeter.

Topic: Example java program

/**
* This class represents a square in two dimensions.
*/
public class Square {

private double side;

/**
* This default constructor makes objects of class
Square.
*/
public Square() {
side = 0.0;
}

/**
* This fully parametrized default constructor makes
objects of class Square.
 Java for Mathematicians/Page 49

*/
public Square(double _side) {
side = _side;
}

/**
* This accessor method returns the length of the
square's side.
*/
public double getSide() {
return side;
}

/**
* This mutator method sets the length of the
square's side.
*/
public void setSide(double _side) {
side = _side;
}

/**
* This accessor method returns the area of the
square.
*/
public double getArea() {
// return side * side;
return Math.pow(side,2);
}

/**
* This accessor method returns the area of the
square.
*/
public double getPerimeter() {
final int NUMBER_SIDES = 4;
return side * NUMBER_SIDES;
 Java for Mathematicians/Page 50

Topic: Exercise

Add a pseudo-attribute for the square's hypotenuse by

writing the getDiagonal method. Use the hypot method
of the Math class to calculate that diagonal's length.

Next, we examine the test class TestSquare that

will create an object of class Square and exercise its
methods. As in previous programs that test class will
create an object of its own class and execute its go
method as shown below.

/**
* This class tests class Square which represents a
square in two dimension.
*/
public class TestSquare {

/**
* This method creates an object of class Square
* and displays information concerning that square.
*/
private void go()
{
System.out.println("Create an object of class
Square.”);
}
 Java for Mathematicians/Page 51

/**
* The main method is automatically executed.
* Create an object of class TestSquare
* then call its method named go.
*/
public static void main(String[] args) {
TestSquare entity = new TestSquare();
entity.go();
}

The go method will create an object of class

Square and print information about the area and
perimeter of that square. Next the program creates a
square with an edge of length 5.2.

/**
* This class tests class Square which represents a
square in two dimension.
*/
public class TestSquare {

/**
* This method creates an object of class Square
* and displays information concerning that square.
*/
private void go()
{
Square sq = new Square();
System.out.println("Default square side: " +
sq.getSide() +
" area: " + sq.getArea() +
" perimeter: " + sq.getPerimeter()); //
 Java for Mathematicians/Page 52

Display the answer.

Square sq2 = new Square();
sq2.setSide(5.2);
System.out.println("Second square side: " +
sq2.getSide() +
" area: " + sq2.getArea() +
" perimeter: " + sq2.getPerimeter()); // Display
the answer.
}

/**
* The main method is automatically executed.
* Create an object of class TestSquare
* then call its method named go.
*/
public static void main(String[] args) {
TestSquare entity = new TestSquare();
entity.go();
}

In the previous program we did two steps to set

up a square with edge length of 5.2.

Square sq2 = new Square();

sq2.setSide(5.2);

By providing a second constructor we can reduce

that to one command: Square sq2 = new Square(5.2);
The new constructor accepts the size of the square as
an argument.

/**
* This fully parametrized constructor makes objects
of class Square.
 Java for Mathematicians/Page 53

*/
public SquareV2(double _side) {
side = _side;
}

In addition to objects the java language includes

primitives, such as int to represent integers and
double to represent real numbers. All numbers in a
computer system are represented by binary numbers. An
int has a size of four bytes. In java the byte is
standardized at eight bits, thus an int contains 32
bits providing for integers from -2,147,483,648 and a
maximum value of 2,147,483,647(inclusive). If your
work involves larger integers you can use the long
primitive providing double the width for a size of 8
bytes or 64 bits and integer values from
-9,223,372,036,854,775,808 and a maximum value of
9,223,372,036,854,775,807 (inclusive). For more
information consult Oracle's Web site on java
primitives at
http://docs.oracle.com/javase/tutorial/java/nutsandbol
ts/datatypes.html.

You can easily use int values in a command such

as sum = divisor1 + divisor2 + divisor3, but java
demands that you define each variable. A definition
for divisor1 includes int, the variable's type, and
divisor1, the variable's name. You can also set that
variable to an initial value by using the = operator.
Thus int divisor1 = 1; defines a new int value with a
name of divisor1 and sets it to an initial value of 1.
 Java for Mathematicians/Page 54

A definition consists of the following: type name

= initial value; We have seen this pattern in chapter
one in the command: FamousMathQuotesV2 entity = new
FamousMathQuotestV2(); This program has examples such
as int divisor1 = 1; and int divisor2 = 2;

Although Java automatically upgrades an int to a

double, it will not automatically downgrade a double to

an int. When you sum or multiply two int values the

result is an int value. Likewise when all the variables

are doubles the result will be a double. For a command
that mixes an int value with a double value, Java will
automatically upgrade the int value to a double before
computing the result. For example double x = 5 + 11.3;
the 5 will be upgraded to 5.0, a double value. Then
that 5.0 will be summed with 11.3 to produce a result
of 16.3 to be stored in variable x.

If you want only the integer portion of the

result you can cast the answer to an int value by
using the cast operator (int). For example int rating
= (int)(5 / 3.9); will upgrade the 5 to 5.0, then
divide 5.0 by 3.9 producing an answer of
1.282051282051282 which the cast operator (int) will
truncate to 1. Be aware that the (int) cast will
truncate, not round, the result. If you want to round
the result then use the round method of the Math class.

static rint(double a)
 Java for Mathematicians/Page 55

Returns the double value that is closest in value

double
to the argument and is equal to a mathematical integer.
round(double a)
static
Returns the closest long to the argument, with ties
long
rounding up.

Those methods, rint and round, do not return an

int value, so a cast of (int) would still be required.

Topic: Inheritance

Use extend to expand upon an existing class. The

new class will have its own method plus access to all
the methods and attributes of its super class. For
example class ColorSquare can expand upon class Square
by including an attribute for color.

import java.awt.Color;

/**
* This class represents a square in two dimension.
*/
public class ColorSquare extends Square {

private Color color;

/**
* This default constructor makes objects of class
Square.
*/
public ColorSquare() {
color = Color.RED;
}
 Java for Mathematicians/Page 56

/**
* This accessor method returns the length of the
square's side.
*/
public Color getColor() {
return color;
}

/**
* This mutator method sets the length of the
square's side.
*/
public void setColor(Color _color) {
color = _color;
}

Color is a class within java, so the program's

code begins with an import statement to allow java
access to the Color class store in the awt directory.
Awt is an abbreviation of abstract window toolkit and
contains many utility classes. Often programmers use
the asterisk operand to allow access to all classes
within the awt directory. Then the import command is
import java.awt.*; This broader directive will not
make your java program bigger; java will import only
the classes needed for your program.

Color contains int values from 0 to 255 for each

of three primary colors: red, green, and blue
according to the additive color theory established by
 Java for Mathematicians/Page 57

the computer industry and known as sRGB. For further

information go to
http://en.wikipedia.org/wiki/SRGB_color_space. The
Color class has static attributes for common colors
such as Color.RED. All objects are extensions of one
object name Object, which has a toString method which
returns a string containing the object's name and
address in memory. The Color class defines its own
toString method which displays the three values for
red, green, and blue. For Color.RED toString displays
java.awt.Color[r=255,g=0,b=0], so red has a maximum
value of 255 and both blue and green have values of
zero, so they are not displayed.

The test class includes commands to set the

square's color and to display that color.

import java.awt.Color;

/**
* This class represents a triangle in two dimension.
*/
public class TestColorSquare {

/**
* This method creates an object of class Square
* and displays information concerning that square.
*/
private void go()
{
ColorSquare sq = new ColorSquare();
System.out.println("Default square side: " +
sq.getSide() +
 Java for Mathematicians/Page 58

" area: " + sq.getArea() +

" perimeter: " + sq.getPerimeter() +
" color: " + sq.getColor()); // Display the answer.
ColorSquare sq2 = new ColorSquare();
sq2.setSide(5.2);
sq2.setColor(Color.RED);
System.out.println("Second square side: " +
sq2.getSide() +
" area: " + sq2.getArea() +
" perimeter: " + sq2.getPerimeter() +
" color: " + sq2.getColor()); // Display the answer.
}

/**
* The main method is automatically executed.
* Create an object of class SumTwoIntegers
* then call its method named go.
*/
public static void main(String[] args) {
TestColorSquare entity = new TestColorSquare();
entity.go();
}

Program execution displays the color as shown

below.

Default square side: 0.0 area: 0.0 perimeter: 0.0

color: java.awt.Color[r=255,g=0,b=0]
Second square side: 5.2 area: 27.040000000000003
perimeter: 20.8 color: java.awt.Color[r=255,g=0,b=0]
 Java for Mathematicians/Page 59

Practice using other named colors. The javadocs for

the Color class contains a list of the predefined
colors such as Color.BLUE and Color.WHITE.

Topic: ArrayList

An ArrayList is a Java language object for

storing other objects in a list. The javadocs page for
ArrayList has the Internet address of
http://docs.oracle.com/javase/7/docs/api/java/util/Arr
ayList.html.
The name of the ArrayList object is squares, so a
declaration is ArrayList squares; To declare and
create that object named squares in one Java command
is possible as shown here.

ArrayList squares = new ArrayList();

The newly created ArrayList has no content. The

variable named squares is a pointer to an empty list.
Our next commands create objects of class Square and
add them to ArrayList squares.

Square sq1 = new Square(1.2);

squares.add(sq1);
Square sq2 = new Square(0.5);
squares.add(sq2);
Square sq3 = new Square(2.3);
squares.add(sq3);

Topic: for each loop

 Java for Mathematicians/Page 60

Next our program defines a new variable of type

double and name of sum to hold the answer as we sum
the areas of the three squares.
double sum = 0.0;
The “for each” loop will process each of those three
squares in ArrayList squares. The “for each” loop
begins with the type and name of a variable to hold an
element of the array. Next a colon separates that
variable from the name of the array.

for (Square sq: squares)

{
sum = sum + sq.getArea();
}

The general format is shown below.

for (type element: collection)

{
statement
}

A collection is a java class for holding items.

Programs use an extension of that collection class
such as ArrayList which maintains a list of items.
ArrayList is a type of Java collection, so the general
format used “collection”, not array.

The value of sum is replaced with the sum's

current total plus the area of the next array element.
That command can be abbreviated to sum +=
sq.getArea(); In general all Java statements of the
form variable = variable + number; can be abbreviated
using the += operand. That abbreviation also works
 Java for Mathematicians/Page 61

with subtraction, multiplication, and division.

ArrayList objects can contain any objects.

Because this program only stores objects of class
Square the array declaration should be limited to
holding only objects of class Square as shown below.

ArrayList<Square> squares = new

ArrayList<Square>();

The Java program begins with an import statement.

import java.util.ArrayList; So, the java compiler can
obtain the code for an arrayList from that location.

The entire program is shown here.

import java.util.ArrayList;

/**
* This class sums the area of several objects of class
Square.
*/
public class SumSquares {

/**
* This method creates several objects of class
Square
* and displays information concerning those
squares.
*/
private void go()
{
ArrayList<Square> squares = new
ArrayList<Square>();
 Java for Mathematicians/Page 62

Square sq1 = new Square(1.2);

squares.add(sq1);
Square sq2 = new Square(0.5);
squares.add(sq2);
Square sq3 = new Square(2.3);
squares.add(sq3);
double sum = 0.0;
for (Square sq: squares)
{
sum = sum + sq.getArea();
}
System.out.println("Total area: " + sum);
}

/**
* The main method is automatically executed.
* Create an object of class SumSquares
* then call its method named go.
*/
public static void main(String[] args) {
SumSquares entity = new SumSquares();
entity.go();
}

The program produces this result: Total area:

6.979999999999999

Because ColorSquare extends Square objects of

class ColorSquare are also objects of class Square.
Our test program would also work with sq3 of class
ColorSquare.

The second version of our test program has sq3 as

an object of class ColorSquare. Notice that this
 Java for Mathematicians/Page 63

version also includes an import statement for the

Color class.

import java.awt.Color;
import java.util.ArrayList;

/**
* This class sums the area of several objects of class
Square.
*/
public class SumSquaresV2 {

/**
* This method creates several objects of class
Square
* and displays information concerning those squares.
*/
private void go()
{
ArrayList<Square> squares = new
ArrayList<Square>();
Square sq1 = new Square(1.2);
squares.add(sq1);
Square sq2 = new Square(0.5);
squares.add(sq2);
ColorSquare sq3 = new ColorSquare();
sq3.setSide(5.2);
sq3.setColor(Color.BLUE);
squares.add(sq3);
double sum = 0.0;
for (Square sq: squares)
{
sum = sum + sq.getArea();
 Java for Mathematicians/Page 64

}
System.out.println("Total area: " + sum);
}

/**
* The main method is automatically executed.
* Create an object of class SumSquares
* then call its method named go.
*/
public static void main(String[] args) {
SumSquaresV2 entity = new SumSquaresV2();
entity.go();
}

The second version of our test program produces

this result: Total area: 28.730000000000004

Topic: IF statement

The next challenge is to find the size of the

largest square in our collection named squares. The
program begins with a declaration for the variable to
hold the size of the largest square. That variable
always begins with the smallest possible value, which
for area is zero.

double largest = 0;

In the “for each” loop the if statement compares

the current element's area with the value in variable
largest. If it is bigger then the program stores the
 Java for Mathematicians/Page 65

new value in variable largest.

double largest = 0;
for (Square sq: squares)
{
if (sq.getArea() > largest)
{
largest = sq.getArea();
}
}
System.out.println("Area of largest square: " +
largest);

Our program prints the answer Area of largest

square: 5.289999999999999

Next our program searches for the smallest area

of the squares in our collection. The declaration for
our variable to hold that answer is double smallest;
which is set to Double.MAX_VALUE because the program
is searching for the smallest value for area. Read the
documentation for the class named Double. The static
variable MAX_VALUE has the largest positive value for
a double.

double smallest = Double.MAX_VALUE;

for (SquareV2 sq: squares)
 Java for Mathematicians/Page 66

{
if (sq.getArea() < smallest)
{
smallest = sq.getArea();
}
}
System.out.println("Area of smallest square: " +
smallest);

The program prints Area of smallest square: 0.25

Topic: Exercise

Construct a class for rectangles with attributes

of length and width. Use that class to implement the
12 orthogons of Wersin as described in Wikipedia's
article on Dynamic Rectangles at
http://en.wikipedia.org/wiki/Dynamic_rectangle. Show
that those rectangles taken together can indeed have
an area equal to a double square as described in that
article.

“The 12 orthogons of Wersin

According to Wolfgang von Wersin's The Book of Rectangles,
Spatial Law and Gestures of The Orthogons Described (1956), a set of 12
special orthogons (from the Gr. ορθος, orthos,
"straight"[9] and γονια, gonia, "angle"; "a right angled figure", which, as a
consequence, is rectangular and tetragonal[10]) has been used
historically by artists, architects and calligraphers to guide the placement
and interaction of elements in a design.[3][11] These orthogons are:
[12]
 Java for Mathematicians/Page 67

•Square (1:1 or 1:√1)

•Diagon (1:√2)
•Hecton or sixton (1:√3)
•Doppelquadrat (1:2 or 1:√4)
•Hemiolion (2:3)
•Auron (the golden rectangle, 1:φ)
•Hemidiagon (1:½√5)
•Penton (1:√φ)
•Trion (1:⅔√3)
•Quadriagon
•Biauron (1:2φ)
•Bipenton
Wolfgang Von Wersin's book includes an extraordinary copy of text
from the year 1558 (Renaissance), with diagrams of seven of the 12
orthogons and an invitation from the passage to pay careful attention as
the "ancient" architects believed "nothing excels these proportions" as "a
thing of the purest abstraction."[13]
All 12 orthogons, when formed together, create an entire unit: a
square that is developed into a double square.[14]
Perhaps the most popular among the ortogons is
the auron or golden rectangle, which is produced by projecting the
diagonal that goes from the middle point of a side of a square to one of the
opposite vertexes, until it is aligned with the middle point.
Four of these orthogons are harmonic rectangles:
the diagon or root-2 rectangle is produced by projecting the diagonal of a
square; the sixton, hecton or root-3 rectangle is produced by projecting the
diagonal of a diagon; the double square or root-4 rectangle is produced by
projecting the diagonal of an hecton; the root-5 rectangle is produced by
projecting the diagonal of a double square (or by projecting 180° both
diagonals that go from the middle point of a side of a square to the
 Java for Mathematicians/Page 68

opposite vertexes).
Two of the most complicated of these figures are; the penton, with
proportions 1:√φ is related to the section of the golden pyramid,
the bipenton's longer side is equal to the shorter multiplied by two thirds of
the square root of three, longer side of the biauron is √5 - 1 or 2τ times the
shorter.
The quadriagon is related to the diagon in the
sense that its longer side is produced by projecting
the diagonal of a quarter of a square. The trion has
the height of an equilateral triangle and the width of
the side. The hemidiagon (1:½√5) longer side is half
the one of the root-5 rectangle and is produced by
projecting the diagonal of half a square until it is
perpendicular with the origin.
Besides the square and the double square, the
only other static rectangle included in the list is
the hemiolion, which is produced by projecting 90° or
180° half the side of a square.” Source: Wikipedia
Dynamic Rectangle.

Because the use of dynamic objects is considered

to be a better programming paradigm our future
examples will use that approach. Static methods and
classes will be limited to supporting roles.
 Java for Mathematicians/Page 69

Chapter Four: java interface

Topic: Polygonal Numbers

Triangular numbers 0, 1, 3, and 6 can be represented

by 0, 1, 3, and 6 dots arranged in triangles.
Wikipedia provides more detailed information at
http://en.wikipedia.org/wiki/Triangular_numbers.

http://en.wikipedia.org/wiki/Figurate_number gives an
equation for calculating the size of each triangular
number of number n as n(n+1)/2.

Class TriangleNumber represents a triangular number.

Each triangular number has a sequential ID as shown in
the diagram above. Mathematicians commonly name each
triangular number T followed by its ID. So the names
are T0, T1, etc. Those ID numbers are positive
integers beginning with zero. T0 is the empty set,
thus not shown in most drawings. The ID is an
attribute of class TriangleNumber, so it is private,
is set to zero by the default constructor, set to a
positive integer by the fully parametrized
constructor, and has an public accessor. Because a
triangular number's ID never changes, attribute ID
does not have an associated mutator method. This
change from the norm of an accessor and a mutator for
each attribute justifies the inclusion of a comment
within the program's source code.

/**
* Triangular numbers can be represented by a group of
 Java for Mathematicians/Page 70

dots
* organized in an equilateral triangle.
* The first few triangular numbers are of size 0, 1,
3,
* 6, and 10. Those are the sizes of
* triangles with edge length of 0, 1, 2, 3, and 4.
* The edge lengths are the IDs for those
* triangular numbers. Only one instance should be
made for each ID.
* Class TriangleNumber represents a single triangular
number.
*/

public class TriangleNumber

{

private final int tn; // which triangle number

/**
* This default constructor makes instance zero of
class TriangleNumber.
*/
TriangleNumber()
{
tn = 0;
}

/**
* This fully parametrized constructor makes
instance n of class TriangleNumber.
*
* @param n the ID for an instance of
TriangleNumber of size n
 Java for Mathematicians/Page 71

*/
public TriangleNumber(int n)
{
tn = n;
}

/**
* Returns the ID for this instance of
TriangleNumber.
*
* @return returns the ID of this instance of
TriangleNumber
*/
public int getID()
{
return tn;
}

/**
* This pseudo accessor calculates the size of this
instance of
* class TriangleNumber according to the equation
found in
* http://en.wikipedia.org/wiki/Figurate_number.
*
* @return the value or size of the triangle
*/
public int getValue()
{
return (tn * (tn + 1)) / 2;
}

/**
 Java for Mathematicians/Page 72

* The gnomon is how much must be added to produce

the
* next triangular number. This is a pseudo-
accessor.
* @return gnomon
*/
public int getGnomon()
{
return tn + 1;
}

/**
* Overrides toString method to display tn: value
*
* @return returns String tn: size
*/
public String toString()
{
return "t" + getID() + ": " + getValue();
}

The test class creates one instance of a triangular

number and displays its size.

class TestOneTriangleNumber
{

public static void main(String[] args)

{
TriangleNumber t = new TriangleNumber(6);
System.out.println("t6: " + t.getValue());
 Java for Mathematicians/Page 73

The resulting output is t6: 21.

Using the instance's toString method produces a

shorter program that is easier to understand. The print
and println commands automatically call the toString
method of an object. If the instance of an object is
named t then the print command would be println(t);.
Using toString methods allow your computer program to
be briefer and easier to comprehend.

class TestOneTriangleNumber
{

public static void main(String[] args)

{
TriangleNumber t = new TriangleNumber(6);
// System.out.println("t6: " + t.getValue());
System.out.println(t);
}
}

The resulting output continues to be t6: 21.

Topic: JOptionPane's static method

showInputDialog

import javax.swing.JOptionPane;
 Java for Mathematicians/Page 74

The showInputDialog static method of class JOptionPane

displays a message in a dialog box and accepts an
input string in response. The showInputdialog method
returns the user's input as an instance of String.
Next the program converts that string to an int for
use in its getValue method. Again the program uses a
static method of a class. The static method parseInt
of class Integer returns an int value matching the
contents of a string. Thus Integer.parseInt(“36”)
returns the int value of 36 as expected.

import javax.swing.JOptionPane;

class TestOneTriangleNumberV2
{

public static void main(String[] args)

{
String triangleID =
JOptionPane.showInputDialog("Enter
triangle number ID: ");
System.out.println("input: " + triangleID);
int tn = Integer.parseInt(triangleID);
TriangleNumber t = new TriangleNumber(tn);
System.out.println(t);
}

}
 Java for Mathematicians/Page 75

Version 2 of our test program produced this result.

input: 4
ÏÏ t4: 10

Topic: for loop

The next challenge is to list the first sixteen

triangular numbers: 0, 1, 3, 6, 10, 15, 21, 28, 36,
45, 55, 66, 78, 91, 105, and 120. The program uses a
for loop to iterate from zero to fifteen, not sixteen.
The stop value is called a sentinel value. For more
information go to
http://en.wikipedia.org/wiki/Sentinel_value.

final int STOP = 16;

for (int i = 0; i < STOP; i++)
{
… statements to be executed repeatedly
}

The for loop's control section has three portions:

initialize, check for termination, and increment.

for (initialize; terminate; increment)

{
 Java for Mathematicians/Page 76

… statements to be executed repeatedly

}

For initialization this for loop defines a control

variable named i and initializes its value to 0. The
increment portion uses the command i++ which is
equivalent to i = i + 1;. That is not a mathematical
statement. The = symbol assigns the value from the
right side to the variable on the left. Thus i + 1 is
evaluated first. For example i begins with a value of
zero, then i + 1 has the value of 1. That value of 1
is assigned to the variable on the left side of the =
operator. That results in the variable i having a new
value of 1. The termination portion compares the
variable i with the stop value of 16. Previously the
program initialized an int variable named STOP to 16.
Because that value will never change that variable was
made final, thus the name is in all capital letters as
STOP. Such variables are known as sentinel variables.

For each iteration of the loop this program makes an

instance of the next triangular number and prints its
contents.

class ListTriangleNumbers
{

/*
OEIS: On-Line Encyclopedia of Integer Sequences
The sequence of triangular numbers (sequence
A000217 in OEIS),
starting at the 0th triangular number, is:
 Java for Mathematicians/Page 77

0, 1, 3, 6, 10, 15, 21, 28, 36, 45, 55, 66, 78,

91, 105, 120
` */
public static void main(String[] args)
{
final int STOP = 16;
for (int i = 0; i < STOP; i++)
{
TriangleNumber t = new
TriangleNumber(i);
System.out.print(t.getValue() + " ");
}
System.out.println();
}
}

The program prints this result: 0 1 3 6 10 15 21 28 36

45 55 66 78 91 105 120 on one line because the print
statement used the print method. Only after the for
loop was completed did the program execute a println
for a new line.

Java classes can represent mathematical objects such

as a square. For more information concerning squares
you can consult Wikipedia at
http://en.wikipedia.org/wiki/Square. All squares are similar.
They vary only in the length of the side. Thus the
class for a square retains only the length of the
side.

When a class describes an object, that class has a

 Java for Mathematicians/Page 78

name representing the object. A good name for such a

class would be “Square.” Thus the class's code begins
as follows. Remember a class's name begins with a
capital letter.

/**
* Each instance of this class represents a square
number.
* They are 0, 4, 9, 16, etc.
*/
public class SquareNumber
{

private final int sqN; // square's ID

/**
* This default constructor creates an instance of
SquareNumber with ID of 0.
*/
public SquareNumber()
{
sqN = 0;
}

/**
* This fully parametrized constructor creates an
instance of
* SquareNumber with ID of n.
*/
public SquareNumber(int n)
{
sqN = n;
}
 Java for Mathematicians/Page 79

/**
* This accessor returns the instance's ID.
*
* @return the square number's ID
*/
public int getID()
{
return sqN;
}

/**
* This virtual accessor returns the size of the
square number.
*
* @return the size of the square number
*/
public int getValue()
{
return sqN * sqN;
}

/**
* This virtual accessor returns the additional
amount needed
* to increase to the next square number. This is
the edge across
* the top and on the right side of the current
square. Thus
* the gnomon is 2*n + 1.
*
* @return gnomon for this square
*/
 Java for Mathematicians/Page 80

public int getGnomon()

{
return (2 * sqN) + 1;
}

Class SquareNumber is similar to class TriangleNumber.

Likewise the program to list the first sixteen square
numbers beginning with zero is similar to the program
to list the first sixteen triangular numbers.

class ListSquareNumbers
{

/*
OEIS: On-Line Encyclopedia of Integer Sequences
The squares (sequence A000290 in OEIS) smaller
than 602 are:

0**2 = 0
1**2 = 1
2**2 = 4
3**2 = 9
4**2 = 16
5**2 = 25
6**2 = 36
7**2 = 49
8**2 = 64
9**2 = 81`
*/
public static void main(String[] args)
{
 Java for Mathematicians/Page 81

final int STOP = 16;

for (int i = 0; i < STOP; i++)
{
SquareNumber sq = new SquareNumber(i);
System.out.print(sq.getValue() + " ");
}
System.out.println();
}

That program produces these results: “0 1 4 9 16 25 36

49 64 81 100 121 144 169 196 225” which matches
sequence A000290 in OEIS.

Topic: Interfaces

Can a program list the value or gnomon of any

triangular number or square number that equals 73? For
that challenge the program loads all the instances of
triangular numbers and square numbers into an
ArrayList, but ArrayList is limited to one class type.
To solve that conundrum the program defines an
interface for polygonal numbers because both
triangular numbers and square numbers are examples of
figurate numbers, described at
http://en.wikipedia.org/wiki/Figurate_number, which are
polygonal numbers that can be described as a
collection of points arranged in a regular convex
polygon. For further information on polygonal numbers
go to http://en.wikipedia.org/wiki/Polygonal_Numbers. That
interface includes the methods common to all polygonal
number classes, such as getID, getSize, and getGnomon.
 Java for Mathematicians/Page 82

An interface sets expectations for any class

implementing that interface. The interface can contain
fixed variables and signatures of methods. Any class
implementing that interface can use those fixed
variables and must have a full method implementing the
expected methods.

// Polygonal Numbers

public interface PolygonalNumber

{
public int getID();
public int getValue();
public int getGnomon();
}

Both TriangleNumber and SquareNumber already have

those expected methods. When the class header includes
implements PolygonalNumber then instances of that
class can be included in the array list defined by

ArrayList<PolygonalNumber> polygonalNumbers = new

ArrayList<>();

Class TriangleNumber now begins with public class

TriangleNumber implements PolygonalNumber.

/**
* Triangular numbers can be represented by a group of
dots
* organized in an equilateral triangle.
* The first few triangular numbers are of size 0, 1,
3,
 Java for Mathematicians/Page 83

* 6, and 10. Those are the sizes of

* triangles with edge length of 0, 1, 2, 3, and 4.
* The edge lengths are the IDs for those
* triangular numbers. Only one instance should be
made for each ID.
* Class TriangleNumber represents a single triangular
number.
*/
public class TriangleNumber implements PolygonalNumber
{

private final int tn; // which triangle number

/**
* This default constructor makes instance zero of
class TriangleNumber.
*/
TriangleNumber()
{
tn = 0;
}

/**
* This fully parametrized constructor makes
instance n of class TriangleNumber.
*
* @param n the ID for an instance of
TriangleNumber of size n
*/
public TriangleNumber(int n)
{
tn = n;
}
 Java for Mathematicians/Page 84

/**
* Returns the ID for this instance of
TriangleNumber.
*
* @return returns the ID of this instance of
TriangleNumber
*/
public int getID()
{
return tn;
}

/**
* This pseudo accessor calculates the size of this
instance of
* class TriangleNumber according to the equation
found in
* http://en.wikipedia.org/wiki/Figurate_number.
*
* @return the value or size of the triangle
*/
public int getValue()
{
return (tn * (tn + 1)) / 2;
}

/**
* The gnomon is how much must be added to produce
the
* next triangular number. This is a pseudo-
accessor.
* @return gnomon
 Java for Mathematicians/Page 85

*/
public int getGnomon()
{
return tn + 1;
}

/**
* Overrides toString method to display tn: value
*
* @return returns String tn: size
*/
public String toString()
{
return "t" + getID() + ": " + getValue();
}

In a similar manner class SquareNumber begins with

public class SquareNumber implements PolygonalNumber.

/**
* Each instance of this class represents a square
number.
* They are 0, 4, 9, 16, etc.
*/
public class SquareNumber implements PolygonalNumber
{

private final int sqN; // square's ID

/**
* This default constructor creates an instance of
 Java for Mathematicians/Page 86

SquareNumber with ID of 0.
*/
public SquareNumber()
{
sqN = 0;
}

/**
* This fully parametrized constructor creates an
instance of
* SquareNumber with ID of n.
*/
public SquareNumber(int n)
{
sqN = n;
}

/**
* This accessor returns the instance's ID.
*
* @return the square number's ID
*/
public int getID()
{
return sqN;
}

/**
* This virtual accessor returns the size of the
square number.
*
* @return the size of the square number
*/
 Java for Mathematicians/Page 87

public int getValue()

{
return sqN * sqN;
}

/**
* This virtual accessor returns the additional
amount needed
* to increase to the next square number. This is
the edge across
* the top and on the right side of the current
square. Thus
* the gnomon is 2*n + 1.
*
* @return gnomon for this square
*/
public int getGnomon()
{
return (2 * sqN) + 1;
}

The program must define the target number and the

array list. Next the triangular numbers and square
numbers are added to that array list. Finally the
program searches its array list for the target number.
The target number is 73. Because 73 is a fixed value
the definition is final int TARGET = 73. Final
variables have names in
all capital letters.

Each square number has both a size and a gnomon. The

 Java for Mathematicians/Page 88

size is the square of the edge, so largest square

needed would be 10 because 10 times 10 equals 100
which exceeds the target of 73. Also the gnomon must
be considered. The gnomon for a square of edge n is 2n
+ 1, so a square of size 50 would have a gnomon of
101, which exceeds the target of 73. Adding additional
squares is not a problem. The program must include all
the square whose size or gnomon is less than 73. For
each ID less than 50 the program must create an
instance of square number and add that instance to the
array list. Repeatedly performing that task requires a
loop. Since we know the loop will be repeated 50 times
the program uses the standard for loop.

for (int i = 0; i < SQ_STOP; i++)

{
SquareNumber sq = new SquareNumber(i);
polygonalNumbers.add(sq);
}

Calculating the size and gnomon of a triangular number

can be done, but it's slightly more complex, so the
program uses a while loop. The gnomon is always less
than the size, so the program sets up a loop. While
the gnomon is less than the target size, the
triangular number is added to the array list. Notice
variable n has the triangular number's ID. Variable n
begins with a value of zero. Each iteration of the
loop increments n by one.

int n = 0;
TriangleNumber tri = new TriangleNumber(n);
int gnomon = tri.getGnomon ();
 Java for Mathematicians/Page 89

while ( gnomon < TARGET)

{
polygonalNumbers.add(tri);
n = n + 1;
tri = new TriangleNumber(n);
gnomon = tri.getGnomon();
}

This final step uses a for each loop to search the

entire array list for any polygonal number that has
either a value of gnomon of the right size.

for (PolygonalNumber p: polygonalNumbers)

{
if (p.getValue() == TARGET)
{
System.out.print(p.getClass().getName() +
" " + p.getID() + " ");
}
if (p.getGnomon() == TARGET)
{
System.out.print(p.getClass().getName() +
" " + p.getID() + " gnomon ");
}
}
System.out.println();

The entire program is shown below.

import java.util.ArrayList;

class FindNumberTypes
{
 Java for Mathematicians/Page 90

/*
OEIS: On-Line Encyclopedia of Integer Sequences
The squares (sequence A000290 in OEIS) smaller than
602 are:

0**2 = 0
1**2 = 1
2**2 = 4
3**2 = 9
4**2 = 16
5**2 = 25
6**2 = 36
7**2 = 49
8**2 = 64
9**2 = 81`
*/

/*
OEIS: On-Line Encyclopedia of Integer Sequences
The sequence of triangular numbers (sequence
A000217 in OEIS),
starting at the 0th triangular number, is:

0, 1, 3, 6, 10, 15, 21, 28, 36, 45, 55, 66, 78, 91,
105, 120
` */

/* For a given number list all its number types:

square, square's gnomon, square, and square gnomon.
*/

public static void main(String[] args)

 Java for Mathematicians/Page 91

{
final int TARGET = 73;

ArrayList<PolygonalNumber> polygonalNumbers =
new ArrayList<>();

// Generate SquareNumbers with values below 100.

// 10**2 = 100 so use a for loop interating from 0
to 9
// A square's of edge length n has a gnomon of size
2n + 1,
// so generate squares up to edge size 50.

final int SQ_STOP = 50;

for (int i = 0; i < SQ_STOP; i++)
{
SquareNumber sq = new SquareNumber(i);
polygonalNumbers.add(sq);
}

// Generate triangular numbers with values below 73

// The gnomon is always smaller than the size,
// so use a while loop
// stopping when the gnomon is greater than or
equal to 73.
int n = 0;
TriangleNumber tri = new TriangleNumber(n);
int gnomon = tri.getGnomon ();
while ( gnomon < TARGET)
{
polygonalNumbers.add(tri);
n = n + 1;
tri = new TriangleNumber(n);
 Java for Mathematicians/Page 92

gnomon = tri.getGnomon();
}

System.out.print(TARGET + ": ");

// Use a for each loop to search the array list
for (PolygonalNumber p: polygonalNumbers)
{
if (p.getValue() == TARGET)
{
System.out.print(p.getClass().getName() +
" " + p.getID() + " ");
}
if (p.getGnomon() == TARGET)
{
System.out.print(p.getClass().getName() +
" " + p.getID() + " gnomon ");
}
}
System.out.println();
}

The program's result is 73: SquareNumber 36 gnomon.

This program used all three loop methods: for each,

while, and for. Obviously each style of loop meets the
needs of different situations. A strong knowledge of
loops in java is essential for a java computer
language programmer.

Topic: Refactoring
 Java for Mathematicians/Page 93

The computer program works well with square and

triangular numbers, but concerns arise when the
program must begin processing pentagonal (5-sided),
hexagon(6-sided), heptagonal(7-sided), octagonal(8-
sided), nonagon(9-sided), decagonal(10-sided), and
more polygonal numbers as described at
http://en.wikipedia.org/wiki/Polygonal_Numbers. Each polygonal
number class currently requires a separate loop to
generate the polygonal numbers. Including more number
classes would make the program overly complex, so the
program should be re-factored. A factory method
generates upon request an instance of any of the
classes that implement an interface. For this computer
program a factory method could produce either Square
or Triangular numbers as requested by the main
program. Factory methods reduce the complexity of the
main method.
 Java for Mathematicians/Page 94

Now the main method of the UseFactory class creates an

instance of PolygonalNumber factory. The command is
PolygonalNumberFactory pnFactory = new
PolygonalNumberFactory(); Then the computer program
must loop through a list of the types of polygonal
numbers, so the names of those types of polygonal
numbers are stored in an array. The command is
String[] polygonalNames = { "TRIANGLE", "SQUARE" };
to produce all the instances of the square and
triangular polygonal numbers. For each such type of
polygonal number the program creates another instance
of that class with a larger ID until its value or
gnomon exceed the requires. The command to create the
polygonal number is PolygonalNumber polyNum =
pnFactory.createPolygonalNumber(pn, id); A while loop
implements that loop which continues creating new
polygonal numbers until the requirements are
satisfied. Here is the source code for that while loop.

int id = 0;
int gnomon = 0;
int value = 0;
while (( gnomon < TARGET ) ||
( value < TARGET))
{
PolygonalNumber polyNum =
pnFactory.createPolygonalNumber(pn, id);
polygonalNumbers.add(polyNum);
id++;
value = polyNum.getValue();
gnomon = polyNum.getGnomon();
// System.out.println("Add to array pn: " +
pn + " id: " + id + " value: " + value + "
 Java for Mathematicians/Page 95

gnomon: " + gnomon);

}

During each iteration of the while loop each

polygonal number is added to an array list named
polygonalNumbers. The id number must be incremented by
the command id++; so the next iteration will advance
to the next polygonal number. The value and gnomon are
retrieved from the instance of the polygonal number so
the while loop can test them for completion of the
loop. The test command is (( gnomon < TARGET ) ||
( value < TARGET)) so the loop continues to be
iterated until both the gnomon and the value exceed
the sentinel value in the constant named TARGET.

The main method contains this source code.

final int TARGET = 73;

ArrayList<PolygonalNumber> polygonalNumbers =
new ArrayList<>();

// Generate TriangleNumbers and SquareNumbers with

// values and gnomons below 100.
String[] polygonalNames = { "TRIANGLE", "SQUARE" };
PolygonalNumberFactory pnFactory = new
PolygonalNumberFactory();
for (String pn : polygonalNames)
{
int id = 0;
int gnomon = 0;
int value = 0;
while (( gnomon < TARGET ) ||
 Java for Mathematicians/Page 96

( value < TARGET))

{
PolygonalNumber polyNum =
pnFactory.createPolygonalNumber(pn, id);
polygonalNumbers.add(polyNum);
id++;
value = polyNum.getValue();
gnomon = polyNum.getGnomon();
// System.out.println("Add to array pn: "
+ pn + " id: " + id + " value: " + value + " gnomon: "
+ gnomon);
}
}

The for-each loop that searches through all the

polygonal numbers for matches with the TARGET value
remains the same. Observe the main method simply works
with polygonal numbers. No longer is the main method
concerned with creating square and triangular numbers.
All that work moved to the instance of the factory
class. Factory classes do make the main program easier
to write and understand.

/**
* Creates instances of the concrete classes that
* implement the PolygonalNumber interface.
*/
public class PolygonalNumberFactory
{

/**
* This method returns an instance of a concrete
* class that implements the PolygonalNumber interface.
 Java for Mathematicians/Page 97

*
* @return instance of concrete class implement
PolygonalNumber interface
*/
public PolygonalNumber createPolygonalNumber(String
concreteClass, int id)
{
PolygonalNumber pn = null;
switch(concreteClass)
{
case "TRIANGLE":
pn = new TriangleNumber(id);
break;
case "SQUARE":
pn = new SquareNumber(id);
break;
}
return pn;
}
}

The factory class uses a switch statement based on the

name of the concrete class. Adding pentagonal (5-
sided), hexagon(6-sided), heptagonal(7-sided),
octagonal(8-sided), nonagon(9-sided), decagonal(10-
sided), and more polygonal numbers requires simply to
add more case statements and write the source code for
those class descriptions of those types of polygonal
numbers.

switch(concreteClass)
{
case "TRIANGLE":
 Java for Mathematicians/Page 98

pn = new TriangleNumber(id);
break;
case "SQUARE":
pn = new SquareNumber(id);
break;
case "PENTAGONAL":
pn = new PentagonalNumber(id);
break;
case "HEXAGON":
pn = new HexagonNumber(id);
break;
case "HEPTAGONAL":
pn = new HeptagonalNumber(id);
break;
}
return pn;

Although the program is longer, the source code is

simple to expand and remains easy to comprehend.

Topic: Compiler warning

A computer programmer might be tempted to write the

factory class in a briefer fashion by immediately
using a return statement. For example, this erroneous
program demonstrates that mistake.

/**
* Creates instances of the concrete classes that
* implement the PolygonalNumber interface.
*/
public class PolygonalNumberFactoryv2
{
 Java for Mathematicians/Page 99

/**
* This method returns an instance of a concrete
* class that implements the PolygonalNumber
interface.
*
* @return instance of concrete class implement
PolygonalNumber interface
*/

public PolygonalNumber createPolygonalNumber(String

concreteClass, int id)
{
switch(concreteClass)
{
case "TRIANGLE":
return new TriangleNumber(id);
case "SQUARE":
return new SquareNumber(id);
}
}

The break command is no longer needed because the

return command immediately ends the switch. The
compiler studies this source code and provides this
error message.

PolygonalNumberFactoryv2.java:24: error: missing

return statement
ÏÏ§Ï }
 Java for Mathematicians/Page 100

ÏÏ§Ï ^
ÏϧÏ1 error
ÏϧÏ

The compiler is uncertain that the method contains the

return statement required by that method's declaration
of a return type of PolygonalNumber.

The solution to this compiler error message is to move

the return statement to the end of your method. When
your code produces the return value earlier a variable
temporarily holds the value. Then the break statement
drops the execution down to the last statement which
was the required return statement.

Always ending your method with the return statement

means a few more commands are needed but the code is
easy to understand.

Topic: Activity

Search Wikipedia's Polygonal Numbers topic for a

relationship between square and triangular numbers.
Write a java program to verify the first few values.
 Java for Mathematicians/Page 101

Chapter Five: java interface

Topic: Cartesian Coordinate System
Source: Wikipedia http://en.wikipedia.org/wiki/File:Cartesian-coordinate-
system.svg

The Cartesian coordinate system describing two

dimensional Euclidean space uses the real number
system to label x and y coordinates crossing at a
right angle on a flat plane. Each point is identified
as (x,y) where x and y are the coordinates. The origin
is at (0,0). For more information concerning this
coordinate system go to
http://en.wikipedia.org/wiki/Cartesian_coordinates.

Source: http://docs.oracle.com/javase/tutorial/2d/overview/coordinate.html

Topic: Java Graphics coordinates

The base class CartesianShape describes an arbitrary

shape. Only the quadrant with both positive x and
positive y coordinates are considered because Java
provides its own graphics system with that limitation.
After construction of a system of classes describing
points, line segments, triangles, rectangles, and
circles, another Java program can display this
computer program's results using Java's own graphic
system of geometric primitives, such as point, line,
triangle, rectangle, and oval. The base class
CartesianShape is so theoretical that it does not
describe any commonly named shape. It merely knows the
Cartesian (x,y) address top left corner point. Basic
methods for moving that point are included in the base
class.
 Java for Mathematicians/Page 102

Topic: Abstract Class

An abstract method for area is described because some

shapes, such as a point, or line, have no area. Also
the computation of area depends on the specific shape,
such as rectangle, square, circle, or triangle. Each
such sub class of shape has its own mathematical
equation for area. Nonetheless all those shapes have
an area, even if might be zero. Because area is an
abstract method, which means it has only a method
signature, not a body, the enclosing class,
CartesianShape, must also be labeled abstract. The
Java command, new, cannot be used to make an instance,
or object, of an abstract class.

/**
Use only the one quadrant of the Cartesian grid
with positive x and y coordinates. The origin (0,0) is
at the top left corner. The x-coordinate increases
from left to right. The y-coordinate increases from
top to bottom. This description does not match the
diagram above.
*/
public abstract class CartesianShape
{
private double topLeftX;
// X coordinate of top left corner of bounding
rectangle
private double topLeftY;
// Y coordinate of top left corner of bounding
rectangle
 Java for Mathematicians/Page 103

/**
default constructor. Shape is assumed to begin at
(0,0).
*/
public CartesianShape()
{
topLeftX = 0;
topLeftY = 0;
}

/**
full constructor.
*/
public CartesianShape(double x, double y)
{
topLeftX = x;
topLeftY = y;
}

/**
Return x coordinate of top left corner
of bounding rectangle.
*/
public double getTopLeftX()
{
return topLeftX;
}

/**
set x coordinate of top left corner of
bounding rectangle.
*/
public void setTopLeftX(int x)
 Java for Mathematicians/Page 104

{
topLeftX = x;
}

/**
Return y coordinate of top left corner
of bounding rectangle.
*/
public double getTopLeftY()
{
return topLeftY;
}

/**
set y coordinate of top left corner of
bounding rectangle.
*/
public void setTopLeftY(double y)
{
topLeftY = y;
}

/**
Increment x and y coordinates of top left
corner
of bounding rectangle.
*/
public void move(double x, double y)
{
topLeftX += x;
topLeftY += y;
}
 Java for Mathematicians/Page 105

/**
Change x and y coordinates of top left corner
of bounding rectangle.
*/
public void moveTo(double x, double y)
{
topLeftX = x;
topLeftY = y;
}

/**
Return area of shape. Zero if not closed
loop..
*/
public abstract double getArea();

Unlike normal practice no test class accompanies this

abstract class. The Java program cannot use the
command, new, to make an instance, or object, of an
abstract class. Therefore no test class was written to
test abstract class CartesianShape.

Another class can extend an abstract class. When all

the abstract methods are described, thus having a body
of statements for each method, then that new child
class is not abstract.

Class CartesianRectangle extends abstract class

CartesianShape and provides a full definition for the
area method, thus class CartesianRectangle is not an
abstract class. CartesianShape is the parent and
 Java for Mathematicians/Page 106

CartesianRectangle is the child class in that

relationship. A rectangle is defined by its length and
width, so CartesianRectangle has length and width as
attributes. Naturally constructors, accessors, and
mutators for each attribute are included.

/**
Extends CartesianShape to define a Cartesian grid
based rectangle.
The rectangle is always oriented along the x and y
axes.
*/
public class CartesianRectangle extends CartesianShape
{

private double length;

private double width;
// Length and width are never negative values.

/**
Default constructor
*/
public CartesianRectangle()
{
length = 0;
width = 0;
}

/**
Fully parametrized constructor
*/
public CartesianRectangle(double x, double y,
double len, double wide)
 Java for Mathematicians/Page 107

{
super(x,y);
if (len >= 0 && wide >= 0)
{
length = len;
width = wide;
}
}

/**
Accessor for attribute length
*/
public double getLength()
{
return length;
}

/**
Mutator for attribute length
*/
public void setLength(double len)
{
if (len >= 0)
{
length = len;
}
}

/**
Accessor for attribute width
*/
public double getWidth()
{
 Java for Mathematicians/Page 108

return width;
}

/**
Mutator for attribute width
*/
public void setWidth(double wide)
{
if (wide >= 0)
{
length = wide;
}
}

/**
Defines the abstract method of
CartesianShape.
GetArea acts likes an accessor, so
area is a pseudo-attribute.
*/
public double getArea()
{
return length * width;
}

The test class sets up a CartesianRectangle and

displays its data. A second such rectangle is moved to
another location and its information is printed.

public class TestCartesianRectangle

{
 Java for Mathematicians/Page 109

public static void main(String[] args)

{
CartesianRectangle rect = new
CartesianRectangle();
System.out.println("rectangle 1 x: " +
rect.getTopLeftX()
+ " y: " + rect.getTopLeftY()
+ " length: " + rect.getLength() + "
width: " + rect.getWidth() + " area:
" + rect.getArea());
CartesianRectangle rect2 =
new CartesianRectangle(5, 3, 20.5,
3.09);
System.out.println("rectangle 2 x: "
+ rect2.getTopLeftX() + " y: " +
rect2.getTopLeftY()
+ " length: " + rect2.getLength() + "
width: " + rect2.getWidth() + "
area: " + rect2.getArea());
rect2.moveTo(100,200);
System.out.println("rectangle 2 moved to
(100,200).");
System.out.println("rectangle 2 x: "
+ rect2.getTopLeftX() + " y: " +
rect2.getTopLeftY()
+ " length: " + rect2.getLength() + "
width: "
+ rect2.getWidth() + " area: " +
rect2.getArea())
}

}
 Java for Mathematicians/Page 110

The test class produces this result. The default

constructor produced the first rectangle so the
location of the top left corner is (0,0), the length
is zero, and the width is zero as expected from a
default constructor. The second rectangle was set at
(5,3) which was passed to the constructor of the
CartesianShape class. Also the length was set to 20.5
and width to 3.09. Area is a pseudo-attribute
resulting from the multiplication of length by width.

rectangle 1 x: 0.0 y: 0.0 length: 0.0 width: 0.0

area: 0.0
rectangle 2 x: 5.0 y: 3.0 length: 20.5 width: 3.09
area: 63.345
rectangle 2 moved to (100,200).
rectangle 2 x: 100.0 y: 200.0 length: 20.5 width:
3.09 area: 63.345

Not only does CartesianRectangle extend

CartesianShape, both of those classes extend class
Object. In fact all Java classes extend class Object.

Source:

http://docs.oracle.com/javase/7/docs/api/java/lang/Object.html
 Java for Mathematicians/Page 111

Class Object provides several methods including

toString.

Thus our test program can be simplified. Instead of a

long print command, the main method can simply use
System.out.println(rectangle);

public class Test2CartesianRectangle

{

public static void main(String[] args)

{
CartesianRectangle rect = new
CartesianRectangle();
System.out.println(rect);
CartesianRectangle rect2 =
new CartesianRectangle(5, 3, 20.5, 3.09);
System.out.println(rect2);
rect2.moveTo(100,200);
System.out.println("rectangle 2 moved to
(100,200).");
System.out.println(rect2);

}
}

The new simplified test class prints this cryptic

report.
 Java for Mathematicians/Page 112

CartesianRectangle@1cd8f55c
CartesianRectangle@67d479cf
rectangle 2 moved to (100,200).
CartesianRectangle@67d479cf

The toString method of class Object simply prints the

object's class name and address in memory. Naturally
such reports do not satisfy the user's needs to know
the location of each rectangle, nor their sizes.
Thankfully inherited methods can be overridden. The
child class simply includes a new definition for that
method. The method's return type and name must be the
same. For example, a new definition for toString is
shown below.

/**
Override default toString method inherited from
object named Object.
*/
public String toString()
{
return "rectangle at (" + getTopLeftX() + ","
+ getTopLeftY() + ") length: " + length + " width: "
+ width + " area: " + getArea();
}

Now the new test class displays this more informative

report.

rectangle at (0.0,0.0) length: 0.0 width: 0.0 area: 0.0

rectangle at (5.0,3.0) length: 20.5 width: 3.09 area: 63.345

rectangle 2 moved to (100,200).
rectangle at (100.0,200.0) length: 20.5 width: 3.09 area:
63.345
 Java for Mathematicians/Page 113

Many computer programmers routinely override the

toString method for every class, so testing work will
be simple.

Topic: Using Java graphics to study infinite series

Source: http://xkcd.com/994/

A most famous infinite series is from Zeno's paradox.

The series can be represented in modern mathematical
symbols as:
 Java for Mathematicians/Page 114

That series should equal 1. More information can be

found in Wikipedia at
http://en.wikipedia.org/wiki/Zeno%27s_paradoxes.

Topic: GUI

A graphical user interface (GUI) is a collection of

Java classes supporting the display of images, text,
and drawings on a two dimensional computer screen with
the possibility of users interacting with that screen
either through touch or mouse input.
http://en.wikipedia.org/wiki/GUI describes GUI
interfaces in further detail. The original Java
classes, contained in the Abstract Windows Toolkit
(AWT), are also known as heavyweight components
because of their dependency upon the native methods of
the local machine. The AWT is being replaced by the
Swing component classes, which mirror their AWT
classes, but are named lightweight components because
they do not depend on the machine's native GUI
resources. As a result Java has both a AWT Frame class
and a corresponding Swing JFrame class. Because AWT is
being phased out, all new programming should use Swing
classes. The Swing components can be classified into
three main groups: component classes, container
classes, and helper classes. Examples of the component
classes include JButton, JLabel, and JTextField which
define buttons, labels, and text fields. Examples of
container classes include JFrame, JPanel, and JApplet.
Applets are used with Web pages. Frames are used with
other Java programs. Panels are sub-containers that
can be used within a frame or an applet. Examples of
the helper classes are Graphics, Color, and Font. We
 Java for Mathematicians/Page 115

will use those classes in the following example

programs.

Topic: JFrame

The first step is to produce a plain frame within a

window. A frame acts like a small window, so the
variable is named window. The Java command JFrame
window = new Jframe(); creates a new frame named
window.

/**
Use JFrame for a GUI program
*/
import javax.swing.JFrame;

public class DrawCartesianObjects {

public static void main(String[] a) {

JFrame window = new JFrame();

window.setDefaultCloseOperation(JFrame.EXIT_ON_CLOSE);
int width = 300;
int length = 300;
window.setBounds(30, 30, width, length);
window.setTitle("Draw Cartesian Objects");
window.setVisible(true);
}

That program displays this window.

 Java for Mathematicians/Page 116

Because of the command

window.setDefaultCloseOperation(JFrame.EXIT_ON_CLOSE);
pressing the normal close box will close this window.
Forgetting that command means pressing the close box
will not close the window.

The commands
int width = 300;
int length = 300;
window.setBounds(30, 30, width, length);
 Java for Mathematicians/Page 117

set the top left corner of the frame at point (30,

30), the width to 300 pixels and the length to 300
pixels as shown below. Notice that
window.setTitle("Draw Cartesian Objects"); set the
title of the new frame.

Suppose you want to change the background color to

pink. Use the command
window.getContentPane().setBackground( Color.PINK ); A
JFrame has both a content pane and a menu bar,
jMenuBar. Buttons, labels, and text fields are
automatically added to the content pane. To change the
content pane, use the method getContentPane. To reduce
lines of code the method setBackground is applied to
the result of window.getContentPane(), thus one
command completes several actions. PINK is a static
value within the Color class, so the color is
described as Color.PINK.
 Java for Mathematicians/Page 118

Instead of the name window, many programmers would use

the name frame.

Next the program adds an object of class JComponent to

the frame. JComponent is an abstract class that is a
base class for all Swing components except top-level
containers. Thus our class named MyCanvas must extend
JComponent. Instances of MyCanvas must be added to a
top level container, such as JFrame, JDialog, or
Japplet. Those top-level Swing containers provide a
place for other Swing components to paint themselves.
Because class MyCanvas is of interest only to this
program that class is included within our java
program. Notice class MyCanvas lacks the attribute
public.

class MyCanvas extends JComponent {

protected void paintComponent(Graphics g) {

Graphics2D g2 = (Graphics2D) g;
 Java for Mathematicians/Page 119

int x = 10;
int y = 10;
int width = 200;
int height = 200;
// Draw outline of rectangle
Rectangle surround = new Rectangle(x, y,
width, height);
g2.draw(surround);
}
}

Our program's main method will create a new instance

of MyCanvas. When that instance is added to the frame
or when it is repainted the Swing toolkit
automatically creates a graphics objects of type
Graphics2D, which is an extension of the basic class
Graphics and passes that to the paintComponent method.
Graphics is a more primitive class than Graphics2D, so
the first command casts that object as an instance of
Graphics2D. Now the method can use the more advanced
methods, such as draw and fill, which automatically
draw any object implementing the Shape interface. Our
CartesianRectangle class does not yet implement the
Shape interface because that would involve
considerable effort, so our current method creates an
object of class Rectangle which does satisfy the Shape
interface. That new rectangle is named surround
because it will contain the planned series of
rectangles representing Zeno's series of one-half,
one-fourth, etc.

Topic: Anonymous Objects

 Java for Mathematicians/Page 120

The main method will must create an instance of

MyCanvas and attach it to the instance of JFrame named
window. Until now all programs have declared a new
variable, so the main method would have included this
command, MyCanvas canvas = new MyCanvas(); Next, the
program would get the content pane for the JFrame
named window, so the next command would have been
Container content = window.getContentPane(); Finally
the program would add the canvas to the content pane
with the command content.add(canvas); As a convenience
java allows programs to add items directly to the
frame without fetching the content pane, so the last
two commands can be condensed as window.add(canvas);
This abbreviation is permitted only with the add
method. Recall that getContentPane was necessary to
change the background color of the content pane of a
frame.

Java also allows the use of anonymous items. The

canvas object will only be used once in this program
when it is added to window, the JFrame object, so no
name is needed. In such a situation programmers use
anonymous variables which occur when you create an
object and immediately use it. Thus the three commands
are reduced to simply window.add(new MyCanvas()); The
new canvas has no name, thus it is anonymous. The use
of anonymous instances can greatly abbreviate your
program, reduce complexity, and make your program
easier to comprehend.

import java.awt.*;
import java.awt.Graphics;
import java.awt.Rectangle;
 Java for Mathematicians/Page 121

import javax.swing.JComponent;
import javax.swing.JFrame;

class MyCanvas extends JComponent {

protected void paintComponent(Graphics g) {

Graphics2D g2 = (Graphics2D) g;
int x = 10;
int y = 10;
int width = 200;
int height = 200;
// Draw outline of rectangle
Rectangle surround = new Rectangle(x, y,
width, height);
g2.draw(surround);
}
}

public class DrawRectSeries2 {

public static void main(String[] a) {

JFrame window = new JFrame();

window.setDefaultCloseOperation(JFrame.EXIT_ON_CLOSE);
int width = 300;
int length = 300;
window.setBounds(30, 30, width, length);
window.add(new MyCanvas());
window.setVisible(true);
}

}
 Java for Mathematicians/Page 122

Next expand the paintComponent method to draw a series

of nine stacked rectangles of heights ½, ¼, etc.
Include the following code.

// draw 9 rectangles of length 1/2, 1/4, etc

final int LIMIT = 9;
for (int i = 0; i < LIMIT; i++)
{
height = height / 2;
Rectangle box = new Rectangle(x,y,width,height);
g2.draw(box);
y = y + height;
}

Because the number of rectangles is known a for loop

was chosen, not a while loop, nor a for-each loop. The
limit of 9 is a constant thus it is defined with the
attribute of final and the variable's name is in all
capital letters. For each rectangle we reduce its
height by one-half. The program then creates a new
Rectangle of that size. Because class Rectangle
satisfies interface Shape, the Graphics2D method draw
can draw that rectangle. The final step is to adjust
the top-left point (x,y) downwards, so that point
(x,y) will become the top-left point of the next
rectangle. The newly expanded method produced this
result.
 Java for Mathematicians/Page 123

The addition of fill color for each rectangle would

improve our understanding of the situation. During the
for loop the counter varies from 0 to 9. By
alternating colors from orange to black based on
whether that loop counter is even or odd, the program
can produced alternating colors. The % operator
returns the remainder of an integer division. “i % 2”
returns either 0 or 1. The use of an if statement
would be as shown here.

Color color = Color.black;

…
if (i == 0)
{
color = Color.ORANGE;
}
else
{
color = Color.BLACK;
}
 Java for Mathematicians/Page 124

g2.setColor(color);

Programmers often prefer the switch statement because

it is so easy to read and comprehend. The case
condition must be an integer, a long integer, a byte
character, such as 'a', or a String. When a case
condition matches the control variable, which is an
anonymous variable holding the result of “i % 2” the
commands following the colon are executed. The break
command is essential, or all the commands in all
following cases will be executed also. The break
command exits control from the switch statement. Break
commands are also used to exit from a loop.

// draw 9 rectangles of length 1/2, 1/4, etc

// with alternating colors
final int LIMIT = 9;
Color color = Color.BLACK;
for (int i = 0; i < LIMIT; i++)
{
switch (i % 2)
{
case 0: color = Color.ORANGE;
break;
case 1: color = Color.BLACK;
break;
}
g2.setColor(color);
height = height / 2;
Rectangle box = new Rectangle(x,y,width,height);
g2.fill(box);
y = y + height;
 Java for Mathematicians/Page 125

The improved paint method now produces alternating

orange and black rectangles almost filling the entire
box. Thus the programmer can see that in this one
instance Zeno's series approaches one.

Topic: Activity

Write a java computer program to demonstrate the

correctness of other infinite series for the first few
iterations.

How can your knowledge of java computer programming

help you find more relationships in the field of
polygonal numbers?
 Java for Mathematicians/Page 126

Chapter Six: java interface

Topic: array

An array is a fixed-size collection of variables all

of the same type. The declaration int prime declares a
variable with the name prime and the type of int. The
symbol [ ] following a primitive or class type changes
the declaration from a single variable to an array of
variables of that same type. The declaration int[ ]
primes declares an array named primes containing a
collection of variables all of type int. The command
int[ ] primes = new int[ 10 ]; both declares variable
primes as the name of an array of variables all of
type int and creates a new array of ten variables.
Those variables are addressed from 0 to 9. The first
variable is identified as primes[0]. The last variable
is primes[9]. The Java Virtual Machine automatically
checks the range of your index, so if your java
program attempts to access the contents of array[10] ,
array[-1], or worse your program will receive an error
message.

All those variables in array primes are still empty.

You could painstakingly enter each prime number one at
a time by using commands such as primes[0] = 1; or
primes[4] = 7; Thankfully java provides us an easier
way to initialize all those variables using an array
initializer. When the array is declared equal to a
list the array's length is automatically set to the
length of that list and the values are all set up in
the array. For example
 Java for Mathematicians/Page 127

int[] primes = { 1, 2, 3, 5, 7, 9, 11, 13, 17, 19,

23, 25,
29, 31, 33, 35, 37, 41, 43, 47, 49, 53, 59, 61,
65,
67, 69, 71, 73, 77, 79, 83, 89, 91, 97, 101,
103, 107,
109, 113, 125, 127, 129, 131, 133, 137, 139,
141, 143,
145, 149, 151, 155, 157, 161, 163, 167, 169,
173, 179, 181 };

will declare a variable named primes as an array of

variables all of type int. Also the values will be set
up in that array. That single command has the same
result as the following list of commands.

int[] primes = new int[61];

primes[1] = 1;
…
primes[61] = 181;

The array declaration and array initializer must be

used in one command. You cannot use two commands. The
following sequence of commands will produce an error
message.

int[] primes;
primes = { 1, 2, 3, 5, 7, 9, 11, 13, 17, 19, 23,
25,
29, 31, 33, 35, 37, 41, 43, 47, 49, 53, 59, 61,
65,
67, 69, 71, 73, 77, 79, 83, 89, 91, 97, 101,
 Java for Mathematicians/Page 128

103, 107,
109, 113, 125, 127, 129, 131, 133, 137, 139,
141, 143,
145, 149, 151, 155, 157, 161, 163, 167, 169,
173, 179, 181 };

An array differs from an ArrayList and the object

named Array. Also an array is not a member of the java
collections framework. With an ArrayList object the add
method will always successfully add another item to
the array list. There is no add method for an array.
Once the size of an array has been set the array size
cannot be changed.

The following class sets up an array of int values

containing all the prime numbers from 1 to 181. It has
been known that prime numbers are infinite in size
since Euclid offered proof in his Elements that there
is no largest prime number.
(http://en.wikipedia.org/wiki/Euclid%27s_theorem)

OEIS provides access to an

encyclopedia of integer sequences.
Besides stimulating our
imagination, these integer
sequences allow us to check the
correctness of our java computer
program's results. Each sequence has its own internet
address. For example the prime numbers are listed at
http://oeis.org/A090332

/**
 Java for Mathematicians/Page 129

This class works with the prime numbers from 1

through 181.
*/
public class Primes
{
// An array of int primitive values, not object
instances of classes
// List of prime numbers. Source:
http://oeis.org/A090332
private int[] primes = { 1, 2, 3, 5, 7, 9, 11, 13,
17, 19, 23, 25,
29, 31, 33, 35, 37, 41, 43, 47, 49, 53, 59, 61,
65,
67, 69, 71, 73, 77, 79, 83, 89, 91, 97, 101,
103, 107,
109, 113, 125, 127, 129, 131, 133, 137, 139,
141, 143,
145, 149, 151, 155, 157, 161, 163, 167, 169,
173, 179, 181 };

/**
default constructor
*/
public Primes()
{

/*
This method works with the prime numbers.
*/
private void go()
{
 Java for Mathematicians/Page 130

System.out.println("I know these prime

numbers:");
for (int prime : primes) // for each grade
in grades
System.out.println(prime); // print that
grade
}

/**
main method
obj is a common abbreviation of "object".
*/
public static void main(String[] args)
{
Primes obj = new Primes();
obj.go();
}

The program produces a list of prime numbers.

Program Results

The program produces a list of prime numbers from 1 to

181.

I know these prime numbers:

1
2
3
...
181
 Java for Mathematicians/Page 131

The list has been abbreviated with an ellipsis (…)

replacing a large stretch of the listing.

Notice that the default constructor has no commands.

Removing that empty default constructor would shorten
your program and still produce the same results.
Commonly an IDE automatically adds such empty default
constructors. Programmers usually leave those empty
default constructors in their programs so that later
they can easily expand the empty constructor. Of all
the new programming languages java has the largest
amount of comments, requires everything be defined
before being used, and generally has the largest
source files.

Often for a short quick development a programmer will

combine the test class and the object class into one
class as shown above. But serious development requires
more diligence at following the rules of object-
oriented programming. For a more object-oriented
approach to this program, the prior class has been
divided into two classes: a class named Primes which
maintains a list of prime numbers and a separate class
named TestPrimes to exercise class Primes.

Topic: Class Primes

The class named Primes continues to use the first few

prime numbers from the integer sequence A090332 of
oeis.org to initialize an array named primes. In java
names are case sensitive so there is no confusion, at
least to the compiler, between the array primes and
 Java for Mathematicians/Page 132

the class Primes. The array's name begins with a

lowercase letter p and the class' name begins with a
capital letter P. Normally programmers avoid using
such similar names differing only by starting with a
capital letter. A better idea would be to name the
class PrimeNumbers, not Primes, then those names of
the array and class would not be so similar. Notice
below that the class name is PrimeNumbers, not Primes.
The class includes the method isPrime for deciding
whether an integer is a prime number, getPrime for
getting the prime number from a location in the array,
and getMaxPrime returns the largest prime number known
to this class.

/**
This class works with the prime numbers from 1
through 181.
*/
public class PrimeNumbers
{
// An array of int primitive values, not object
instances of classes
// List of prime numbers. Source:
http://oeis.org/A090332
private int[] primes = { 1, 2, 3, 5, 7, 9, 11, 13,
17, 19, 23, 25,
29, 31, 33, 35, 37, 41, 43, 47, 49, 53, 59, 61,
65,
67, 69, 71, 73, 77, 79, 83, 89, 91, 97, 101,
103, 107,
109, 113, 125, 127, 129, 131, 133, 137, 139,
141, 143,
145, 149, 151, 155, 157, 161, 163, 167, 169,
 Java for Mathematicians/Page 133

173, 179, 181 };

/**
default constructor
*/
public PrimeNumbers()
{

/**
Test if an integer is a prime.
*/
public boolean isPrime(int n)
{
boolean answer = false;
for (int prime : primes) // foreach prime
in primes
if (n == prime) // test prime
{
answer = true;
break; // interrupt loop
}
return answer;
}

/**
* Return prime from array.
*/
public int getPrime(int index)
{
return primes[index];
}
 Java for Mathematicians/Page 134

/**
* Return largest prime from array, which is the
last entry.
*/
public int getMaxPrime()
{
return primes[primes.length-1];
}

/**
* Return prime from array.
*/
public int getMaxCntPrimes()
{
return primes.length;
}

Topic: Scope

The array named primes is an attribute of the class,

so it is declared before any constructor or method.
All the methods have access to the array named primes
and can use that variable in their commands. The array
named primes is not limited to any method thus the
array named primes has a scope of the entire class.
Normally a variable's scope is as limited as possible.
In this class all the methods need access to the
variable named primes, so that variable has class
scope. Variables at the start of the class with class
scope and containing information that is the main goal
 Java for Mathematicians/Page 135

of the class are called attributes in object-oriented

terminology.

Method getMaxCntPrimes simply returns the length of

the array named primes. Each array has an associated
length. In our example the array named primes knows
its own length which the program can obtain by simply
using primes.length.

/**
* Return prime from array.
*/
public int getMaxCntPrimes()
{
return primes.length;
}

Method getPrime returns the prime number stored at a

certain location. The location of number in an array
is shown by its index, which for the first location is
zero. The next location is at index 1. For an array of
five entries, the index varies from 0 to 4.
Notice this method does not check that the index is
correct. Suppose the user asked for the prime number
at index 5. That presents an impossible situation.
Currently this method ignores that problem. Later
versions of this method will fix this situation.

/**
* Return prime from array.
*/
public int getPrime(int index)
{
 Java for Mathematicians/Page 136

return primes[index];
}

Method getMaxPrime returns the largest prime number in

the list. Since those prime number are conveniently
stored in ascending numeric order, this method simply
returns the last entry in the list. An array of
length five has five items with addresses from 0
through 4. The addresses of array entries begin with
zero, not one. So the last entry in primes is at
location primes.length - 1. Thus the method returns
primes[primes.length - 1]. For an array of length five
primes.length would be five, so primes.length – 1
would be four, which is the address of the last entry.
Then return primes[primes.length – 1]; is equivalent
to return primes[4]; When those numbers are stored in
ascending sequence return primes[4] will return the
largest number in that list.

/**
* Return largest prime from array, which is the
last entry.
*/
public int getMaxPrime()
{
return primes[primes.length - 1];
}

Method isPrime currently uses a for-each loop to

search the entire array for the target value. When the
target value is found the method returns a boolean
value of true. Otherwise the method returns the
default value of false. Some programmers use a return
 Java for Mathematicians/Page 137

statement to immediately return a value of true.

Other algorithms can find the answer quicker. This

topic of searching a list is an area of active
research in computer science.

/**
Test if an integer is a prime.
*/
public boolean isPrime(int n)
{
boolean answer = false;
for (int prime : primes) // for each prime
in primes
if (n == prime) // test prime
{
return true;
}
return answer;
}

Programming rule of thumb: Each method has only one

return statement which is at the end.

A common programming approach is for the last command

to return a default value. Thus the programmer is
certain the method always returns an acceptable value.

Many programmers follow a stricter demand. They

require that each method have only one return
statement, which then must be the last command in the
method. Thus our isPrime method sets the return value
and uses a break command to exit any loop.
 Java for Mathematicians/Page 138

/**
Test if an integer is a prime.
*/
public boolean isPrime(int n)
{
boolean answer = false;
for (int prime : primes) // foreach prime
in primes
if (n == prime) // test prime
{
answer = true;
break; // interrupt loop
}
return answer;
}

The Java compiler naturally requires a return

statement at the end of each method. Notice this
method does not return a value when the integer is
zero.

/**
* Return +1 for positive integers
* Return -1 for negative integers
* Print "zero" for integer zero.
*/
public class PositiveInteger
{

private int positiveInteger(int i)

{
if (i > 0) {
 Java for Mathematicians/Page 139

return 1;
}
else if (i < 0) {
return -1;
}
else {
System.out.println("zero");
}
}

The Java compiler produced this complaint.

----jGRASP exec: javac -g PositiveInteger.java

PositiveInteger.java:20: error: missing return

statement
}
^
1 error

The Java compiler does not produce a complaint for

this following method.

/**
* Return +1 for positive integers
* Return -1 for negative integers
* Print 0 for integer zero.
*/
public class PositiveInteger
{
 Java for Mathematicians/Page 140

private int positiveInteger(int i)

{
if (i > 0) {
return 1;
}
else if (i < 0) {
return -1;
}
else {
return 0;
}
}

Thus we can begin the understand the wisdom of the

programmer's rule of thumb that each method has only
one return statement and that the return statement is
at the end of the method. So we would rewrite the
method as follows. Because of the return statement at
the end of the method, outside of any if or loop
construct, the program certainly always returns a
value.

/**
* Return +1 for positive integers
* Return -1 for negative integers
* Print 0 for integer zero.
*/
public class PositiveInteger
{
 Java for Mathematicians/Page 141

private int positiveInteger(int i)

{
int answer = 0;
if (i > 0) {
answer = 1;
}
else if (i < 0) {
answer = -1;
}
else {
answer = 0;
}
return answer;
}

Topic: Big O

Method isPrime uses a for-each loop to search the

entire array. This method has a “Big O” classification
(http://en.wikipedia.org/wiki/Big_O_notation ) of O(n)
meaning as the array size of n grows the number of
computer steps to locate the target value also grows
like n. A computer algorithm of O(n) is amongst the
slowest of all algorithms and should be avoided, if
possible.

/**
Test if an integer is a prime.
*/
public boolean isPrime(int n)
{
 Java for Mathematicians/Page 142

boolean answer = false;

for (int prime : primes) // foreach prime
in primes
if (n == prime) // test prime
{
answer = true;
break; // interrupt loop
}
return answer;
}

When the array is in sequence the method can use a

binary search
(http://en.wikipedia.org/wiki/Binary_search_algorithm)
. A binary search begins by comparing the target value
with the middle item in the array. If the target value
is lower then the target is in the lower half of the
array, otherwise the target is in the higher half. The
search then focuses on the identified half. Thus a
binary search repeatedly divides the range of the
search in half until the target is located. The number
of comparisons required is log to the base 2 of n.
Thus the Big O notation for a binary search is O(log
n). Searching an array of 100 items would require on
average 50 comparisons using a sequential search. Two
raised to the tenth power is 1,024 so a binary search
requires at most only 10 comparisons. Computer
programming is a form of mathematical reasoning
applied to computing machinery. As you write your
computer programs you should ponder their computing
requirements.

This class exercises class Primes.

 Java for Mathematicians/Page 143

import java.util.Scanner;

/**
* Test class Primes
*/
public class TestPrimeNumbers
{

Primes p;

/**
* constructor
*/
public TestPrimeNumbers()
{
p = new PrimeNumbers();
}

/**
This method lists all the prime numbers in the
list.
*/
private void list()
{
System.out.println("prime numbers in list:");
int maxIndex = p.getMaxCntPrimes();
for (int i = 0; i < maxIndex; i++)
{
System.out.println(p.getPrime(i)); // print
prime
}
}
 Java for Mathematicians/Page 144

/**
This method prints the largest prime numbers in
the list.
*/
private void maxPrime()
{
System.out.println("The largest prime in this
list: " + p.getMaxPrime());
}

/**
This method prints the count of prime numbers in
the list.
*/
private void cntPrime()
{
System.out.println("The count of prime numbers
in this list: " + p.getMaxCntPrimes());
}

/**
This method checks whether or not an integer is
in the list of prime numbers.
*/
private void chkPrime()
{
// Check 4 as a prime number
System.out.print("prime candidate: ");
int candidate = 4;
System.out.print(candidate + " is ");
if (p.isPrime(candidate) == false)
{
 Java for Mathematicians/Page 145

System.out.print("not ");
}
System.out.println("a prime number.");
// Check 5 as a prime number
System.out.print("prime candidate: ");
candidate = 5;
System.out.print(candidate + " is ");
if (p.isPrime(candidate) == false)
{
System.out.print("not ");
}
System.out.println("a prime number.");
}

/**
This method works with the prime numbers.
*/
private void go()
{
Scanner input = new Scanner(System.in);
final int EXIT = 0;
final int LIST = 1;
final int MAX = 2;
final int CNT = 3;
final int CHK = 4;
int choice = -1;
System.out.print("Testing class Primes \nSelect
1-List 2-Max 3-Count 4-prime 0-stop: ");
choice = input.nextInt();
while (choice > 0)
{
switch(choice) {
case EXIT:
 Java for Mathematicians/Page 146

System.exit(0);
break;
case LIST:
list();
break;
case MAX:
maxPrime();
break;
case CNT:
cntPrime();
break;
case CHK:
chkPrime();
break;
default:
System.out.println("Enter 0, 1, 2, or
3.");
break;
}
System.out.print("Testing class Primes
\nSelect 1-List 2-Max 3-Count 4-Prime 0-stop: ");
choice = input.nextInt();
}
}

/**
main method
obj is a common abbreviation of "object".
*/
public static void main(String[] args)
{
TestPrimes obj = new TestPrimes();
obj.go();
 Java for Mathematicians/Page 147

Topic: Scanner

Our discussion of new material begins with the go

method. Scanner input = new Scanner(System.in);
produces a new object of the Scanner class that is
connected to System.in. Recall System.out is an output
stream connected to our console. Likewise System.in
is an input stream connected to our console. When you
press the enter key the choice you typed on the
console goes to the input stream named System.in. The
Scanner object named input will parse that console
string into its tokens which are separated by blank
spaces. A space is the default delimiter separating
tokens. The next command, choice = input.nextInt();,
tells that Scanner object to search the tokens from
the input string for the next integer. The switch
statement then branches depending on the integer of
your choice.

Topic: Scope

Notice how the Scanner class object named input is

defined in the go method. The Scanner class object
named input is used only in the go method, so its
defined in the go method. Thus only the java commands
within method go can use the Scanner class object
named input. The scope of input is the method named
go. When a variable or object can only be seen, thus
 Java for Mathematicians/Page 148

used, only in one method, then that variable or object

has a scope of that method.

Programmers define their variables and objects within

the smallest possible scope. Restricting the scope of
variables prevent accidental misuse of that variable.

Topic: Magic Values

Instead of “magic numbers” such as the integers 0, 1,

2, and 3 programmers use constants. More discussion on
this topic is available on Wikipedia at
http://en.wikipedia.org/wiki/Magic_number_(programming)
under the title of “Unnamed numerical constants”. Thus
in the program listed above the switch statement has
cases using constants, not integers. Those constants
are defined using final, which means the program can
only assign a value to that variable once. From then
on that value cannot be changed.

final int EXIT = 0;

final int LIST = 1;
final int MAX = 2;
final int CNT = 3;
final int CHK = 4;

Using named constants is preferable to using integer

values as shown below.

/**
This method works with the prime numbers.
*/
 Java for Mathematicians/Page 149

private void go()

{
Scanner input = new Scanner(System.in);
System.out.print("Testing class Primes \nSelect
1-List 2-Max 3-Count 4-Prime 0-stop: ");
choice = input.nextInt();
while (choice > 0)
{
switch(choice) {
case 0:
System.exit(0);
break;
case 1:
list();
break;
case 2:
maxPrime();
break;
case 3:
cntPrime();
break;
case 4:
chkPrime();
break;
default:
System.out.println("Enter 0, 1, 2, 3 or
4.");
break;
}
System.out.print("Testing class Primes
\nSelect 1-List 2-Max 3-Count 4-Prime 0-stop: ");
choice = input.nextInt();
}
 Java for Mathematicians/Page 150

Using constants instead of magic numbers makes your

program easier to understand. Most of the work of
computer programming is updating and maintaining
computer programs. Seldom does a computer programmer
develop an entirely new computer program. Using case
statements for 0, 1, 2, 3, and 4 is easier to type,
but using named constants saves effort in the long
term.

The default case can appear anywhere in the list of

cases, but is most commonly used as the last case.

The method maxPrime displays the sentence: “The

largest prime in this list is 161”. To produce that
line maxPrimed called the getMaxPrime method of the
PrimeNumbers object named p.
/**
This method prints the largest prime numbers in
the list.
*/
private void maxPrime()
{
System.out.println("The largest prime in this
list: " + p.getMaxPrime());
}

Because method maxPrime has only one command that

method could have been eliminated by simply moving the
one command to the switch statement.

switch(choice) {
 Java for Mathematicians/Page 151

case EXIT:
System.exit(0);
break;
case LIST:
list();
break;
case MAX:
// maxPrime();
System.out.println("The largest prime in
this list: " + p.getMaxPrime());
break;
case CNT:
cntPrime();
break;
case CHK:
chkPrime();
break;
default:
System.out.println("Enter 0, 1, 2, or
3.");
break;
}

Most programmers prefer to keep the case statements as

brief as possible. Also using a method call makes the
program easier to read and understand. Using a
separate method instead of writing the print command
in the case statement is a question of judgment.
Certainly programmers avoid large amounts of code in a
case statement.

Topic: When to use the for, while, and for-each

 Java for Mathematicians/Page 152

commands.

The list method displays all the prime numbers in the

array. This method gets the size of the array from the
getMaxCntPrimes method of the PrimeNumbers object.
Because the size of the array is known, this method
uses a for loop as shown below.

/**
This method lists all the prime numbers in the
list.
*/
private void list()
{
System.out.println("prime numbers in list:");
int maxIndex = p.getMaxCntPrimes();
for (int i = 0; i < maxIndex; i++)
{
System.out.println(p.getPrime(i)); // print
prime
}
}

The go method uses a while loop because the number of

iterations needed is unknown. The while loop continues
iterating the loop until the sentinel value is met. In
this case the while loop continues until the user
selects a choice of 0 to exit the program.

Another method named getArrayPrimes could return the

address of the array of prime numbers.

/**
 Java for Mathematicians/Page 153

This method the entire array of the prime

numbers.
*/
public int[] getArrayPrimes()
{
return primes;
}

The corresponding method in the test class calls the

getArrayPrimes method of the PrimeNumbers object. Then
the list method can use a for-each command to loop
through the contents of that array. The source code is
shown here.

/**
This method lists all the prime numbers in the
list.
*/
private void list()
{
System.out.println("prime numbers in list:");
int[] primes = p.getArrayPrimes();
for (int prime : primes)
{
System.out.println(prime); // print prime
}
}

Overall these two classes have shown examples of the

for, while, and for-each commands.

Topic: pass by reference versus pass by value

 Java for Mathematicians/Page 154

When a primitive value, such as an int, is returned a

copy of the value is returned. Thus the calling object
gets the value, but cannot change the value stored in
the called object. For example the original list
method received each prime number as an int value.

Example: pass by value

Then TestPrime could erroneously include a command to

attempt to reset each prime number to zero.

/**
This method lists all the prime numbers in the
list.
This method contains an error.
*/
private void list()
{
System.out.println("prime numbers in list:");
int maxIndex = p.getMaxCntPrimes();
for (int i = 0; i < maxIndex; i++)
{
int prime = p.getPrime(i);
System.out.println(prime); // print prime
/* The following command changes the value of
the local variable named prime. It does
not change the value in the PrimeNumbers
object.
*/
prime = 0;
}
}
 Java for Mathematicians/Page 155

The command, prime = 0;, attempts to reset the prime

number in object PrimeNumbers, but only changes the
local variable named prime. The prime number in the
PrimeNumbers object named p is not changed. When
returning a primitive value, such as an int, the
method getPrime of object PrimeNumbers returns a copy
of the value, so the calling object named
TestPrimeNumbers can only change its local variable,
not the original prime number in object PrimeNumbers.

Example: pass by reference

The getArrayPrimes method of the PrimeNumbers object

returns a reference, also known as a pointer, to the
actual array of prime numbers held within the
PrimeNumbers object. Then the following method can and
will change all the prime numbers to zero.

/**
This method lists all the prime numbers in the
list.
*/
private void list()
{
System.out.println("prime numbers in list:");
int[] primes = p.getArrayPrimes();
for (int i = 0; i < primes.length; i++)
{
int prime = primes[i];
System.out.println(prime); // print prime
// Change the prime to zero.
primes[i] = 0;
}
 Java for Mathematicians/Page 156

The primes[i] = 0; command within class

TestPrimeNumbers directly changes the value of the
array element within the other object named p of class
type PrimeNumbers. The command int[] primes =
p.getArrayPrimes(); meant primes got a reference to
the array of prime numbers in object p, which was of
class type PrimeNumbers, so the command primes[i] = 0;
could actually change values of the prime numbers
stored in the PrimeNumbers object named p.

This method produced these results during testing.

First the prime numbers are listed. The second request
for a list produced an equally long list of zeroes.

Testing class Primes

Select 1-List 2-Max 3-Count 4-prime 0-stop: 1
prime numbers in list:
1
2
3
...
181
Testing class Primes
Select 1-List 2-Max 3-Count 4-Prime 0-stop: 1
prime numbers in list:
0
0
0
…
0
 Java for Mathematicians/Page 157

Testing class Primes

Select 1-List 2-Max 3-Count 4-Prime 0-stop:

Clearly pass by value produces different results from

pass by reference. Thus programmers carefully limit
their use of pass by reference. For that reason the
getArrayPrimes method was not included in the original
version of class PrimeNumbers.

The power of object-oriented programming comes from

using objects. Methods do normally return objects,
thus methods do normally use pass by reference.

Topic: Test the limits

Currently the test class only checks if the numbers 4

and 5 are prime numbers. The method's commands have
been rewritten as follows.

/**
This method checks whether or not an integer is
in the list of prime numbers.
*/
private void chkPrime()
{
System.out.print("prime candidate: ");
int candidate = input.nextInt();
System.out.print(candidate + " is ");
if (p.isPrime(candidate) == false)
{
System.out.print("not ");
}
 Java for Mathematicians/Page 158

System.out.println("a prime number.");

}

The command int candidate = input.nextInt(); gets the

next int value from the scanner object named input,
which is linked to the input stream System.in. The JVM
displays the following compiler error message.

TestPrimeNumbers.java:73: error: cannot find symbol

int candidate = input.nextInt();

symbol: variable input

location: class TestPrimeNumbers
1 error

The scanner object named input is defined within the go

method, so its scope is limited to that method. Moving
the declaration of scanner input to the top of the
class resolves the compilation error.

import java.util.Scanner;

/**
* Test class Primes
*/
public class TestPrimeNumbers
{

PrimeNumbers p;
Scanner input;

/**
 Java for Mathematicians/Page 159

* constructor
*/
public TestPrimeNumbers()
{
p = new PrimeNumbers();
input = new Scanner(System.in);
}

Also the constructor now initializes scanner input to

a new scanner object linked to the input stream named
System.in. Because the scanner object named input now
has class-wide scope, the class compiles correctly.

A large amount of effort is required to ensure your

computer program's answers are correct. The field of
testing is a major research area in its own right.

Beginning programmers normally exercise their computer

programs for a few sample values. When those samples
agree with your predicted values then they declare
their computer programs to be correct. While checking
your computer program you should test the limits. For
this problem the limits are the smallest prime number
1 and the largest prime number in the list 181. Notice
that the console's output for this computer program
includes both the limits and a few sample numbers when
checking for prime numbers.

Testing class Primes

Select 1-List 2-Max 3-Count 4-Prime 0-stop: 4
prime candidate: 1
1 is a prime number.
 Java for Mathematicians/Page 160

Testing class Primes

Select 1-List 2-Max 3-Count 4-Prime 0-stop: 4
prime candidate: 181
181 is a prime number.
Testing class Primes
Select 1-List 2-Max 3-Count 4-Prime 0-stop: 4
prime candidate: 4
4 is not a prime number.
Testing class Primes
Select 1-List 2-Max 3-Count 4-Prime 0-stop: 4
prime candidate: 23
23 is a prime number.
Testing class Primes
Select 1-List 2-Max 3-Count 4-Prime 0-stop:

Hint: Text-based test programs can lead to rapid

debugging

Since text-based programs can be rapidly developed

they are an excellent method for quickly checking and
debugging your computer programs.

Topic: Recursion

Recursion means a method repeatedly calls itself. The

binary search algorithm lends itself to recursion. The
algorithm begins searching the entire range from the
first to the last by checking the middle location. If
the candidate equals that middle prime number then the
search is finished. Next the algorithm decides whether
to search the lower or higher section of the list. So
the algorithm repeated performs the same task simply
on smaller portions of the list of prime numbers.
 Java for Mathematicians/Page 161

A recursive algorithm normally begins by testing for

the end state. This algorithm has two end states. When
the candidate equals the middle item in the list then
the algorithm returns a boolean completion code of
true. Thus the top line calculates the location of the
middle item. Next the comparison and return statements.

int middle = low + ((high - low) / 2);

if (candidate == primes[middle])
{
return true;
}

Next the algorithm checks for the state when the

candidate is not a prime. The algorithm continues
checking smaller and smaller intervals designated by
the two variable low and high. Normally during each
recursive call low approaches high. If low is greater
than or equal to high, then the candidate never
matched the middle item, so it's not in the list.

if (low >= high)

{
return false;
}

This program has just completed two different tests

for completion. Notice that the previous two tests
must be completed in the sequence described.

Next the algorithm must decide whether to search the

lower half or the higher half of the range. Comparing
 Java for Mathematicians/Page 162

the candidate with the middle item gives the answer.

If the candidate is less than the middle item, then
the candidate must be in the lower range.

if (candidate < primes[middle])

{
return binarySearchPrimes(low, middle-1,
candidate);
}

The next range is from the low to middle – 1, not low

to middle because the algorithm has already checked
the middle item in the list.

Likewise checking the upper portion of the range would

be from the middle + 1 item to the top item.

return binarySearchPrimes(middle+1, high,

candidate);

Because of the conditional statements the algorithm is

written as a series of nested if statements.

/**
* Use a binary search to find whether or not an
integer is a prime number.
*/
private boolean binarySearchPrimes(int low, int
high, int candidate)
{
int middle = low + ((high - low) / 2);
if (candidate == primes[middle])
{
 Java for Mathematicians/Page 163

return true;
}
else
{
if (low >= high)
{
return false;
}
else
{
if (candidate < primes[middle])
{
return binarySearchPrimes(low, middle-
1, candidate);
}
else
{
return binarySearchPrimes(middle+1,
high, candidate);
}
}
}
}

Because the nested if statements march off the page to

the right, thus making the program difficult to read
and comprehend, many programmers combine the else and
if commands on one line. Then the nested if statements
march down the page, not across. This style of coding
only works when the program uses only one of the two
result branches for each if statement.

/**
 Java for Mathematicians/Page 164

* Use a binary search to find whether or not an

integer is a prime number.
*/
private boolean binarySearchPrimes(int low, int
high, int candidate)
{
boolean rtnValue = true;
int middle = low + ((high - low) / 2);
// Check for completion when match is found
if (candidate == primes[middle])
{
rtnValue = true;
}
// Check for completion when candidate is not in
list
// This test must be after, not before, the
previous test
else if (low >= high)
{
rtnValue = false;
}
else if (candidate < primes[middle])
{
// Recursive call using lower half of list
rtnValue = binarySearchPrimes(low, middle-1,
candidate);
}
else
{
// Recursive call using higher half of list
rtnValue = binarySearchPrimes(middle+1, high,
candidate);
}
 Java for Mathematicians/Page 165

return rtnValue;
}

The block symbols { and } are only needed when more

than one statement is in the then or else group. Thus
the code can be written much more briefly as follows.

/**
* Use a binary search to find whether or not an
integer is a prime number.
*/
private boolean binarySearchPrimes(int low, int
high, int candidate)
{
boolean rtnValue = true;
int middle = low + ((high - low) / 2);
if (candidate == primes[middle])
rtnValue = true;
else if (low >= high)
rtnValue = false;
else if (candidate < primes[middle])
rtnValue = binarySearchPrimes(low, middle-1,
candidate);
else
rtnValue = binarySearchPrimes(middle+1, high,
candidate);
return rtnValue;
}

Surrounding the then and else statements in block

symbols { } makes your code safer. Adding a statement
to either a then or else group of statements would
require adding the block symbols { and }. Forgetting
 Java for Mathematicians/Page 166

those { and } symbols would result in a compilation

error. Thus the norm for an if statement is shown
below.

if (comparison)
statement;
else
statement;

Most programmers and IDEs routinely add the block

symbols { and }.

if (condition)
{
statement;
}
else
{
statement;
}

Many programmers find recursive programs easier to

write and understand. The model is two steps. First
test for the state of completion. If the computation
is not complete then update the parameters for the
next call of the recursive program.

To better understand the stack of execution calls to

the recursive method programmers will include a
leading series of dots before each print message. As
the recursions increase in depth of calls the string
of dots grows larger.
 Java for Mathematicians/Page 167

Topic: Include print statements to display

program execution.

Notice these print statements.

System.out.print("candidate: " + candidate + "

low: " + low + " high: " + high);
System.out.println(" middle: " + middle);

They are not required by the algorithm. They help the

programmer find any errors in the method's
implementation of the algorithm. Each time the method
is called the print statements display the values in
important variable such as the list of parameters.
Programmers either depend on the IDE's debugging
tools, or they include print statements. Either method
helps the programmer trace the execution of the
method's statements. Any discrepancy between the
printed low, high, and middle values and the expected
values indicates an error.

To help us better comprehend recursion the program

prints the low, high, and middle values. Also an
offset, which is two dots, is printed at the start of
each recursive call. As the recursion continues the
offsets grows by an additional two dots. The recursive
method has been modified to also display the return
values.

/**
* Use a binary search to find whether or not an
 Java for Mathematicians/Page 168

integer is a prime number.

*/
private boolean binarySearchPrimes(String offset,
int low, int high, int candidate)
{
int middle = low + ((high - low) / 2);
System.out.print(offset + "candidate: " +
candidate + " low: " + low + " high: " + high);
System.out.println(" middle: " + middle);
// Check for completion when match is found
if (candidate == primes[middle])
{
System.out.println(offset + "return true");
return true;
}
else
{
// Check for completion when candidate is not in
list
// This test must be after, not before, the
previous test
if (low >= high)
{
System.out.println(offset + "return
false");
return false;
}
else
{
offset += "..";
if (candidate < primes[middle])
{
// Recursive call using lower half of list
 Java for Mathematicians/Page 169

boolean rtnValue =
binarySearchPrimes(offset, low, middle-1, candidate);
System.out.println(offset + rtnValue);
return rtnValue;
}
else
{
// Recursive call using higher half of list
boolean rtnValue =
binarySearchPrimes(offset, middle+1, high, candidate);
System.out.println(offset + rtnValue);
return rtnValue;
}
}
}
}

At the top of the method the print statements display

the list of parameters: offset, low, and high. Also
the location of the middle item is shown.

System.out.print(offset + "candidate: " +

candidate + " low: " + low + " high: " + high);
System.out.println(" middle: " + middle);

When the candidate matches the middle item a print

statement displays “return true”.

if (candidate == primes[middle])
{
System.out.println(offset + "return true");
return true;
}
 Java for Mathematicians/Page 170

Before recursively calling itself the method increase

the size of offset by an additional two dots.

offset += "..";
if (candidate < primes[middle])
{
boolean rtnValue = binarySearchPrimes(offset,
low, middle-1, candidate);
System.out.println(offset + rtnValue);
return rtnValue;
}

The revised method produces this report.

Testing class Primes

Select 1-List 2-Max 3-Count 4-prime 0-stop: 4
prime candidate: 5
5 is
..candidate: 5 low: 0 high: 60 middle: 30
....candidate: 5 low: 0 high: 29 middle: 14
......candidate: 5 low: 0 high: 13 middle: 6
........candidate: 5 low: 0 high: 5 middle: 2
..........candidate: 5 low: 3 high: 5 middle: 4
............candidate: 5 low: 3 high: 3 middle: 3
............return true
............true
..........true
........true
......true
....true
a prime number.
 Java for Mathematicians/Page 171

Chapter Seven: java interface

Topic: UML

Unified Modeling Language (UML), described in

Wikipedia at
http://en.wikipedia.org/wiki/Unified_Modeling_Language
, is a standard for describing class diagrams. This
chapter's project will work with multiple classes, so
a UML diagram would help us understand the
interactions between those classes.

UML class diagrams display the class name at the top.

A horizontal line separates that class name from a
list of attributes. An additional horizontal line then
separates the methods from the attribute list. So
class diagrams have three main sections: name,
attributes, and methods.

Before each attribute and method is a indicator of the

access limitations. A dash (-) indicates private
access. Only methods within that same class can access
private items, such as attributes and other methods. A
 Java for Mathematicians/Page 172

plus symbol (+) stands for public access. Other

objects and their methods can use items with public
access. And, a # is used for protected access, which
limits access to objects and their methods in the same
package. A java package corresponds to a computer's
folder.

The Triangle class uses three points to describe each

triangle. Since those three points are attributes, the
access is set to private. Both a default constructor
and fully parametrized constructors are available. The
default constructor creates three points all at
location (0,0). The fully parametrized constructor
simply uses the three points available in the list of
arguments.

The Point class uses the java standard Point2D.Double

class using double values for x and y. A different
Point class uses int values for x and y. Point2D is an
abstract class. The java Application Program
Interfaces (APIs) describe this hierarchy of classes
at http://docs.oracle.com/javase/7/docs/api/.

Each Triangle object contains three objects of class

Point so the connecting line from class diagram for
Triangle to class diagram for Point begins with a
special symbol. Other authors may use different
symbols to represent the connecting relationship in
their UML diagrams. The UML diagram shown above was
drawn using the community edition of Visual Paradigm
described at http://www.visual-paradigm.com/. Much
more detail can be included in UML class diagrams. The
UML diagram shown above is for introducing use of UML.
 Java for Mathematicians/Page 173

The standard strategy for developing a computer

application begins with drawing UML class diagrams for
the core classes, but not the test classes. Utility
classes such as Math and Scanner are also not
included. Commercial versions of UML can automatically
produce java source code for those classes, which
greatly reduces the programmer's effort.

The next stage of development writes and tests java

programs to the core classes. Here is the source code
for the Triangle class.

import java.awt.geom.*;

/**
* Class Triange uses three Points to describe its
triangle.
*/

public class Triangle

{

// Attributes
Point2D.Double pointA, pointB, pointC;

// Constructors

/**
* default constructor
*/
public Triangle()
{
 Java for Mathematicians/Page 174

pointA = new Point2D.Double(); // by default set

to (0,0)
pointB = new Point2D.Double();
pointC = new Point2D.Double();
}

/**
* fully parametrized constructor
*/
public Triangle(Point2D.Double a, Point2D.Double b,
Point2D.Double c)
{
pointA = a;
pointB = b;
pointC = c;
}

// Accessors and mutators

/**
* getPointA returns the Point object for attribute
pointA
*/
public Point2D.Double getPointA()
{
return pointA;
}

/**
* setPointA sets pointA to the Point provided in
the list of arguments */
public void setPointA(Point2D.Double point)
{
 Java for Mathematicians/Page 175

pointA = point;
}

/**
* getPointB returns the Point object for attribute
pointB
*/
public Point2D.Double getPointB()
{
return pointB;
}

/**
* getPointB sets pointB to the Point provided in
the list of arguments */
public void setPointB(Point2D.Double point)
{
pointB = point;
}

/**
* getPointC returns the Point object for attribute
pointC
*/
public Point2D.Double getPointC()
{
return pointC;
}

/**
* getPointA sets pointA to the Point provided in
the list of arguments
*/
 Java for Mathematicians/Page 176

public void setPointC(Point2D.Double point)

{
pointC = point;
}

/**
* get length of side A - B
*/
public double getSideAB()
{
// Point2D.Double extends Point2D which provides
the distance method
return pointA.distance(pointB);
}

/**
* get length of side A - B
*/
public double getSideBC()
{
return pointB.distance(pointC);
}

/**
* get length of side A - B
*/
public double getSideAC()
{
return pointA.distance(pointC);
}

/**
* get perimeter of triangle
 Java for Mathematicians/Page 177

*/
public double getPerimeter()
{
return getSideAB() + getSideBC() + getSideAC();
}

/**
* calculate area of triangle
* In geometry, Heron's (or Hero's) formula, named
after Heron of Alexandria,
* states that the area T of a triangle whose sides
have lengths a, b, and c is
* area = sqrt( s(s-a)(s-b)(s-c) )
* where s is the semiperimeter of the triangle.
* Source: http://en.wikipedia.org/wiki/Hero
%27s_formula
*/
public double getArea()
{
double s = getPerimeter() / 2;
double a = getSideAB();
double b = getSideBC();
double c = getSideAC();
double area = Math.sqrt(s * (s-a) * (s-b) * (s-
c));
return area;
}

This command import java.awt.geom.*; allows the

compiler to freely import classes from the geom
directory of the Abstract Windows Toolkit (AWT)
 Java for Mathematicians/Page 178

directory. The asterisk * means all the classes within

the java.awt.geom directory are available for
inclusion as needed. Classes are only included when
necessary. Thus programmers often use the asterisk,
since only the needed classes are included. This
differs from other computer languages.

The Triangle class has three attributes all of class

Point2D.Double, thus the following declaration is
allowed.

// Attributes
Point2D.Double pointA, pointB, pointC;

That one declaration is equivalent to the following

three declarations.

// Attributes
Point2D.Double pointA;
Point2D.Double pointB;
Point2D.Double pointC;

The class name Point2D.Double includes a dot (.)

because class Double is defined within class Point2D
as can be seen in the java docs for that class.
 Java for Mathematicians/Page 179

Notice the API's statement that the enclosing class is

Point2D, thus class Double is defined within class
Point2D. The java code would appear as shown below.

public class Point2D

{
…
public class Double
{
…
}
 Java for Mathematicians/Page 180

…
}

Next in the java class Triangle are the constructors.

The default constructor creates three separate points
and assigns them to the variables point A, point B,
and point C.

/**
* default constructor
*/
public Triangle()
{
pointA = new Point2D.Double(); // by default set
to (0,0)
pointB = new Point2D.Double();
pointC = new Point2D.Double();
}

The default constructor for the Double class, which is

defined within class Point2D, automatically creates a
point at location (0,0). Although all three points are
at (0,0), the constructor must create three different
point objects. The constructor cannot simply set all
three variables pointA, pointB, and pointC to a single
point.

/**
* default constructor
* This constructor contains a serious ERROR
*/
public Triangle()
{
 Java for Mathematicians/Page 181

pointA = new Point2D.Double();

pointB = pointA;
pointC = pointA;
}

The fully parametrized constructor must individually

define each argument in its list of arguments, even
though all are of the same type: Point2D.Double.

/**
* fully parametrized constructor
*/
public Triangle(Point2D.Double a, Point2D.Double b,
Point2D.Double c)
{
pointA = a;
pointB = b;
pointC = c;
}

The rules of Java do not permit any abbreviation of

the list of arguments such as shown below.

/**
* fully parametrized constructor has an ERROR in
* its list of arguments
* Each argument requires its own type declaration.
*/
public Triangle(Point2D.Double a, b, c)
{
pointA = a;
pointB = b;
pointC = c;
 Java for Mathematicians/Page 182

The access methods for the lengths of the sides of the

triangle all depend on the distance method provided by
the Point2D class. Each point is of subclass
Point2D.Double so they inherit the distance method
from its parent class Point2D.

/**
* get length of side A - B
*/
public double getSideAB()
{
// Point2D.Double extends Point2D which provides
the distance method
return pointA.distance(pointB);
}

The getArea method implements Hero's formula. Many

other equations are available for calculating the area
of a triangle.

/**
* calculate area of triangle
* In geometry, Heron's (or Hero's) formula, named
after Heron of Alexandria,
* states that the area T of a triangle whose sides
have lengths a, b, and c is
* T = sqrt( s(s-a)(s-b)(s-c) )
* where s is the semiperimeter of the triangle.
* Source: http://en.wikipedia.org/wiki/Hero
%27s_formula
*/
 Java for Mathematicians/Page 183

public double getArea()

{
double s = getPerimeter() / 2;
double a = getSideAB();
double b = getSideBC();
double c = getSideAC();
double area = Math.sqrt(s * (s-a) * (s-b) * (s-
c));
return area;
}

The test program uses a text-based console program

because such programs are quick and easy to generate.

import java.util.Scanner;
import java.awt.geom.*;

/**
* Test calculator of triangle facts
*/

public class TestTriangle

{

/**
* Default constructor
*/
public TestTriangle()
{

}
 Java for Mathematicians/Page 184

/**
* Go method
*/
private void go()
{
System.out.println("Start test of Triangle
object.");
Point2D.Double pointA = new Point2D.Double(0,0);
Point2D.Double pointB = new Point2D.Double(0,1);
Point2D.Double pointC = new Point2D.Double(1,1);
Triangle tri = new Triangle(pointA, pointB,
pointC);
Point2D.Double a = tri.getPointA();
Point2D.Double b = tri.getPointB();
Point2D.Double c = tri.getPointC();
System.out.println("Point A: (" + a.getX() + ","
+ a.getY() + ")" );
System.out.println("Point B: (" + b.getX() + ","
+ b.getY() + ")" );
System.out.println("Point C: (" + c.getX() + ","
+ c.getY() + ")" );
System.out.println("side AB: " + tri.getSideAB()
);
System.out.println("side BC: " + tri.getSideBC()
);
System.out.println("side AC: " + tri.getSideAC()
);
System.out.println("perimeter: " +
tri.getPerimeter());
System.out.println("area: " + tri.getArea() );
}
 Java for Mathematicians/Page 185

/**
* Main method
*/
public static void main(String[] args)
{
System.out.println("Start test run of
calculator.");
TestTriangle obj = new TestTriangle();
obj.go();
}

The test program produces the following output.

Start test run of calculator.

Start test of Triangle object.
Point A: (0.0,0.0)
Point B: (0.0,1.0)
Point C: (1.0,1.0)
side AB: 1.0
side BC: 1.0
side AC: 1.4142135623730951
perimeter: 3.414213562373095
area: 0.49999999999999983
Ï
The test triangle is a right triangle with sides of 1,
1, and square root of 2, so the results for side AB
and side BC are correct. Side AC should be the value
for the square root of two, which Wikipedia quotes
OEIS as
1.4142135623730950488016887242096980785696718753769480
7317667973799... (sequence A002193 inOEIS), so the
 Java for Mathematicians/Page 186

program's answer is correct to the number of decimal

places displayed. The area should be one-half, since
the area is one-half of a one by one square, thus the
answer for the area is wrong beginning with the
seventeenth and eighteenth decimal positions, which
contain 83. All the float and decimal operations
contain small errors, which accumulate as the
calculations are performed. Let's focus on the trail
of square root operations. Triangle class called the
Math class to calculate the square root of s(s-a)(s-b)
(s-c). Variables a, b, and c had been calculated by
calling the distance method of the Point2D.Double
class, which might have called the square root method,
so getArea's calculation might have involved three
more square root operations to calculate a, b, and c
for a total of four square root operations. Instead of
having each slight error balance out another, using
the same square root operation could have led to a
steady increase of the total error.

Topic: printf

Many programmer use the printf command to display

numeric values to a reasonable amount of accuracy. The
printf command appears in many other computer
programming languages, with java being one of the
languages most recently adding support for printf.
Wikipedia provides an overview at
http://en.wikipedia.org/wiki/Printf.

The test triangle class now includes the following

command.
 Java for Mathematicians/Page 187

System.out.printf("printf area: %7.5f",

tri.getArea() );

The above command begins with a string enclosed in

quotation marks: “printf area: %7.5f”, which displays
each character on the print line until an % symbol is
encountered. The % symbol provides special processing
commands. “7.5” means the output will be seven
characters wide including a decimal point and five
decimal places. The “f” means a non-integer number,
such as java's double or float numbers. Of course
numbers are rounded to the specified decimal place.

The test program now produces these results.

Start test run of calculator.

Start test of Triangle object.
Point A: (0.0,0.0)
Point B: (0.0,1.0)
Point C: (1.0,1.0)
side AB: 1.0
side BC: 1.0
side AC: 1.4142135623730951
perimeter: 3.414213562373095
area: 0.49999999999999983
printf area: 0.50000

The Open Group at

http://www.opengroup.org/susv3xsh/fprintf.html
provides a detailed description of the options
available for displaying data using the printf method.
For the next iteration of the text-based test class
includes a loop, so more than one triangle can be
 Java for Mathematicians/Page 188

tested. Also a Scanner class is used to get the user's

x and y coordinates for each point of the triangles.

import java.util.Scanner;
import java.awt.geom.*;

/**
* Test calculator of triangle facts
*/

public class TestTriangleV2

{
// attributes
Triangle tri;

Scanner scan;

/**
* Default constructor
*/
public TestTriangleV2()
{
// initialize scanner
scan = new Scanner(System.in);
// Create default triangle
Point2D.Double pointA = new Point2D.Double(0,0);
Point2D.Double pointB = new Point2D.Double(0,1);
Point2D.Double pointC = new Point2D.Double(1,1);
Triangle tri = new Triangle(pointA, pointB,
pointC);
}

/**
 Java for Mathematicians/Page 189

* makePoint sets up a new Point2D.Double

*/
private Point2D.Double makePoint()
{
double x = 0;
double y = 0;
Point2D.Double point = new Point2D.Double(0,0);
System.out.println("Creating new (x,y) point");
System.out.print("Enter x:");
x = scan.nextDouble();
System.out.print("Enter y:");
y = scan.nextDouble();
point = new Point2D.Double(x,y);
return point;
}

/**
* makeTriangle sets up a new triangle.
*/
private void makeTriangle()
{
Point2D.Double a = makePoint();
Point2D.Double b = makePoint();
Point2D.Double c = makePoint();
tri = new Triangle(a,b,c);
}

/**
* listTriPoints displays the (x,y) coordinates of
all three vertices
*/
private void listTriPoints()
{
 Java for Mathematicians/Page 190

Point2D.Double a = tri.getPointA();
Point2D.Double b = tri.getPointB();
Point2D.Double c = tri.getPointC();
System.out.println("Triangle");
System.out.printf("Vertex\tx\ty\n");
System.out.printf("%c\t%5.3f\t%5.3f\n", 'A',
a.getX(), a.getY());
System.out.printf("%c\t%5.3f\t%5.3f\n", 'B',
b.getX(), b.getY());
System.out.printf("%c\t%5.3f\t%5.3f\n", 'C',
c.getX(), c.getY());
}

/**
* Go method
*/
private void go()
{
boolean done = false;
System.out.println("Start test of Triangle
object.");
System.out.print("Enter 0-exit 1-setup 2-list 3-
perimeter 4-area: ");
int choice = scan.nextInt();
while (choice != 0)
{
switch(choice)
{
case 0:
System.out.println("bye");
System.exit(0);
case 1:
makeTriangle();
 Java for Mathematicians/Page 191

break;
case 2:
listTriPoints();
break;
case 3:
System.out.printf("perimeter: %.3f\n",
tri.getPerimeter());
break;
case 4:
System.out.printf("area: %.3f\n",
tri.getArea());
break;
default:
System.out.println("Enter 0, 1, 2, 3,
or 4");
}
System.out.print("Enter 0-exit 1-setup 2-list
3-perimeter 4-area: ");
choice = scan.nextInt();
}

/**
* Main method
*/
public static void main(String[] args)
{
System.out.println("Start test run of
calculator.");
TestTriangleV2 obj = new TestTriangleV2();
obj.go();
}
 Java for Mathematicians/Page 192

Now the program produced this report.

Start test run of calculator.

Start test of Triangle object.
Enter 0-exit 1-setup 2-list 3-perimeter 4-area: 1
Creating new (x,y) point
Enter x:0
Enter y:0.0
Creating new (x,y) point
Enter x:2.5
Enter y:3.1
Creating new (x,y) point
Enter x:7.25
Enter y:8.3725768
Enter 0-exit 1-setup 2-list 3-perimeter 4-area: 2
Triangle
Vertex x y
A 0.000 0.000
B 2.500 3.100
C 7.250 8.373
Enter 0-exit 1-setup 2-list 3-perimeter 4-area: 3
perimeter: 22.154
Enter 0-exit 1-setup 2-list 3-perimeter 4-area: 4
area: 0.772
Enter 0-exit 1-setup 2-list 3-perimeter 4-area: 0

The list points method contains printf commands to

display a table. Of special note are the tab “\t” and
new line “\n” commands. The backslash character, known
as escape characters, allows special formatting
 Java for Mathematicians/Page 193

controls.

/**
* listTriPoints displays the (x,y) coordinates of
all three vertices
*/
private void listTriPoints()
{
Point2D.Double a = tri.getPointA();
Point2D.Double b = tri.getPointB();
Point2D.Double c = tri.getPointC();
System.out.println("Triangle");
System.out.printf("Vertex\tx\ty\n");
System.out.printf("%c\t%5.3f\t%5.3f\n", 'A',
a.getX(), a.getY());
System.out.printf("%c\t%5.3f\t%5.3f\n", 'B',
b.getX(), b.getY());
System.out.printf("%c\t%5.3f\t%5.3f\n", 'C',
c.getX(), c.getY());
}

Notice the list points method produced this table.

Triangle
Vertex x y
A 0.000 0.000
B 2.500 3.100
C 7.250 8.373

The command printf("%c\t%5.3f\t%5.3f\n", 'A',

a.getX(), a.getY()); produced the line “A 0.000
0.000”. The printf's list of arguments begins with a
command string: "%c\t%5.3f\t%5.3f\n". For each of the
 Java for Mathematicians/Page 194

following arguments, the command string includes a

corresponding placeholder designated by a % symbol.
The %c placeholder matches with the 'C' character. The
\t escape character is a tab command. The two %f
placeholders match the following two numbers provided
by methods a.getX() and a.getY(). The %5.3f
placeholder matches a java double or float number. The
5.3 specifies a total width of five columns with three
digits to the right of the decimal point. The whole
number portion is thus restricted to a single digit.
Obviously that is not sufficient. The use of tabs
contributes to the appearance of a table with a title
line, column header line, and three rows of data. Each
row describes a single vertex of the triangle.

Topic: Exception

Continued testing produced the error message displayed

next.

Start test of Triangle object.

Enter 0-exit 1-setup 2-list 3-perimeter 4-area: 2.3
Exception in thread "main"
java.util.InputMismatchException
at java.util.Scanner.throwFor(Scanner.java:909)
at java.util.Scanner.next(Scanner.java:1530)
at java.util.Scanner.nextInt(Scanner.java:2160)
at java.util.Scanner.nextInt(Scanner.java:2119)
at TestTriangleV2.go(TestTriangleV2.java:80)
at TestTriangleV2.main(TestTriangleV2.java:116)

The main loop had expected a singled digit integer

 Java for Mathematicians/Page 195

from zero to four, but the user typed 2.3, which the
nextInt() was unable to process. The JVM produced an
input-mismatch-exception. The Java 7 API for that
exception is shown next.

When the input does not match the the scanner's

expectations, the JVM makes an input-mismatch-
exception. Your programs are expected to handle such
error situations. That exception acts like a
communication telling your program to handle the
 Java for Mathematicians/Page 196

error. Notice the InputMismatchException is an

extension of Exception, which has an expectation that
a program would normally handle such an error.

Topic: Try Catch

The try block acts like a method call. The try block
shown below attempts to execute the following two
commands.

System.out.print("Enter 0-exit 1-setup 2-list

3-perimeter 4-area: ");
choice = scan.nextInt();

Those commands were the source of the recent error

message, when a user mistakenly entered 2.3, instead
of a integer from 0 to 4. A try block for those two
command appears as shown below.

try
{
System.out.print("Enter 0-exit 1-setup 2-list
3-perimeter 4-area: ");
choice = scan.nextInt();
}

Following that try block is an associated catch block,

which catches only exceptions of class
InputMismatchException.

Since exceptions form a family tree as shown in the

API for InputMismatchException, programs commonly
include a series of catch blocks beginning with the
 Java for Mathematicians/Page 197

lowest levels of the tree of exceptions and ending

with a general catch for all exceptions as shown below.

catch (InputMismatchException ime)

{
System.out.println("Enter only a single digit
number from 0 to 4.");
}
catch (Exception e)
{
System.out.println("Check your choice for
error.");
}

The API for the nextInt method of the Scanner class

lists three possible exceptions that it could throw.

Illegal state exception means the scanner is closed,

which is unlikely except for a serious program error.
Such an error would be detected during testing of the
application because the program should not close the
 Java for Mathematicians/Page 198

scanner until the program has completed all

processing.

No such element exception is more appropriate when

scanning a file, not when waiting for a user to type a
choice. Thus only input mismatch exception is
appropriate for the test program.

The while loop requires two nextInt commands. Before

the while loop begins the first nextInt gets the
choice, because the top line of the while loop tests
the value of the choice. At the end of the while loop
the user also makes the same choice. To avoid
repeating the try block and cat block the nextInt
command was moved to a separate method.

The getChoice method begins by setting the value of

choice to EXIT, so when the user's choice is invalid
the program displays an error messages and ends
execution. This continues the model of always starting
each method by assigning a valid answer to the return
variable. Thus the return statement always returns a
valid answer. The value of EXIT was chosen to avoid
any confusing responses to the user's choice. Also an
error message is displayed informing the user of the
error. Any other choice than EXIT would mean the main
loop would perform another action, which would most
likely not be the user's choice, thus any other choice
than EXIT would confuse the user. EXIT was not chosen
to punish the user.

Exiting a program at the first indication of an error

frustrates the user, so an immediate exit is not the
 Java for Mathematicians/Page 199

best response. Later programs will include a loop, so

the user will have multiple chances to enter a valid
choice.

Also the getChoice method checks that the input choice

is within the acceptable range of zero to four and
displays an appropriate error message.

/**
* getNextInt get the next int value from the scanner
*/
private int getChoice()
{
int choice = EXIT;
try
{
System.out.print("Enter 0-exit 1-setup 2-list
3-perimeter 4-area: ");
choice = scan.nextInt();
if (choice < EXIT || choice > AREA)
{
System.out.println("Your choice of " +
choice + " is not in the acceptable range of 0 ... 4");
choice = EXIT;
}
}
catch (InputMismatchException ime)
{
System.out.println("Enter only a single digit
number from 0 to 4.");
}
catch (Exception e)
{
 Java for Mathematicians/Page 200

System.out.println("Check your choice for

error.");
}
// scan.next();
return choice;
}

/**
* Go method
*/
private void go()
{
int choice = EXIT;
System.out.println("Start test of Triangle
object.");
choice = getChoice();
while (choice != EXIT)
{
switch(choice)
{
case EXIT:
System.out.println("bye");
System.exit(0);
case MAKE:
makeTriangle();
break;
case LIST:
listTriPoints();
break;
case PERIMETER:
System.out.printf("perimeter: %.3f\n",
tri.getPerimeter());
break;
 Java for Mathematicians/Page 201

case AREA:
System.out.printf("area: %.3f\n",
tri.getArea());
break;
default:
System.out.println("Enter 0, 1, 2, 3,
or 4");
}
choice = getChoice();
}

A catch block can handle more than one type of

exception. The programmer simply types the list of
exceptions separated by vertical bars |. For example
to catch all three exceptions of the Scanner class the
catch block becomes as shown below.

/* This catch block contains an ERROR! */

catch (InputMismatchException | NoSuchElementException
| IllegalStateException se)
{
System.out.println("Enter only a single digit
number from 0 to 4.");
}
catch (Exception e)
{
System.out.println("Check your choice for
error.");
}

Unfortunately that produces an error message because

 Java for Mathematicians/Page 202

InputMismatchException extends NoSuchElementException.

TestTriangleV3.java:100: error: Alternatives in a

multi-catch statement cannot be related by subclassing
catch (InputMismatchException | NoSuchElementException
| IllegalStateException se)

Alternative InputMismatchException is a subclass of

alternative NoSuchElementException
1 error

The API for the Input-mismatch-exception is shown

below.

The API shows InputMismatchException extends

NoSuchElementException, so the catch block should list
only NoSuchElementException, not both of those
exceptions. The reason is that InputMismatchException
 Java for Mathematicians/Page 203

is also a NoSuchElementException because it extends

NoSuchElementException.

By observing that family tree of exceptions the

programmer understands that catching an exception also
catches all exceptions listed below that exception.
For example a catch Throwable would also catch
Throwable, Exception, RuntimeException,
NoSuchElementException, and InputMismatchException.

Notice in the program the separate catch Exception

below the catch for NoSuchElementException and
InputMismatchException. The reason for two different
catch block is the different actions performed. The
error messages differ, so different catch blocks are
needed. Because the catch for Exception would catch
them all, that catch must be below the specialized
catch for NoSuchElementException and
InputMismatchException.

The catch blocks for this program are shown below.

catch (NoSuchElementException | IllegalStateException

se)
{
System.out.println("Enter only a single digit number
from 0 to 4.");
}
catch (Exception e)
{
System.out.println("Check your choice for error.");
}
 Java for Mathematicians/Page 204

Now the program prints an error message and ends the

processing when the input is of the wrong type, such
as 2.3 not being an int.

Enter 0-exit 1-setup 2-list 3-perimeter 4-area: 2.3

Enter only a single digit number from 0 to 4.

The test program is shown below.

import java.util.*;
import java.awt.geom.*;

/**
* Test calculator of triangle facts
*/

public class TestTriangleV2

{
// attributes
Triangle tri;

Scanner scan;

final int EXIT = 0;

final int MAKE = 1;
final int LIST = 2;
final int PERIMETER = 3;
final int AREA = 4;

/**
* Default constructor
*/
public TestTriangleV2()
 Java for Mathematicians/Page 205

{
// initialize scanner
scan = new Scanner(System.in);
// Create default triangle
Point2D.Double pointA = new Point2D.Double(0,0);
Point2D.Double pointB = new Point2D.Double(0,1);
Point2D.Double pointC = new Point2D.Double(1,1);
tri = new Triangle(pointA, pointB, pointC);
}

/**
* makePoint sets up a new Point2D.Double
*/
private Point2D.Double makePoint()
{
double x = 0;
double y = 0;
Point2D.Double point = new Point2D.Double(0,0);
System.out.println("Creating new (x,y) point");
System.out.print("Enter x:");
x = scan.nextDouble();
System.out.print("Enter y:");
y = scan.nextDouble();
point = new Point2D.Double(x,y);
return point;
}

/**
* makeTriangle sets up a new triangle.
*/
private void makeTriangle()
{
Point2D.Double a = makePoint();
 Java for Mathematicians/Page 206

Point2D.Double b = makePoint();
Point2D.Double c = makePoint();
tri = new Triangle(a,b,c);
}

/**
* listTriPoints displays the (x,y) coordinates of
all three vertices
*/
private void listTriPoints()
{
Point2D.Double a = tri.getPointA();
Point2D.Double b = tri.getPointB();
Point2D.Double c = tri.getPointC();
System.out.println("Triangle");
System.out.printf("Vertex\tx\ty\n");
System.out.printf("%c\t%5.3f\t%5.3f\n", 'A',
a.getX(), a.getY());
System.out.printf("%c\t%5.3f\t%5.3f\n", 'B',
b.getX(), b.getY());
System.out.printf("%c\t%5.3f\t%5.3f\n", 'C',
c.getX(), c.getY());
}

/**
* getNextInt get the next int value from the scanner
*/
private int getChoice()
{
int choice = EXIT;
try
{
System.out.print("Enter 0-exit 1-setup 2-list
 Java for Mathematicians/Page 207

3-perimeter 4-area: ");

choice = scan.nextInt();
if (choice < EXIT || choice > AREA)
{
System.out.println("Your choice of " +
choice + " is not in the acceptable range of 0 ... 4");
choice = EXIT;
}
}
catch (NoSuchElementException |
IllegalStateException se)
{
System.out.println("Enter only a single digit
number from 0 to 4.");
}
catch (Exception e)
{
System.out.println("Check your choice for
error.");
}
return choice;
}

/**
* Go method
*/
private void go()
{
int choice = EXIT;
System.out.println("Start test of Triangle
object.");
choice = getChoice();
while (choice != EXIT)
 Java for Mathematicians/Page 208

{
switch(choice)
{
case EXIT:
System.out.println("bye");
System.exit(0);
case MAKE:
makeTriangle();
break;
case LIST:
listTriPoints();
break;
case PERIMETER:
System.out.printf("perimeter: %.3f\n",
tri.getPerimeter());
break;
case AREA:
System.out.printf("area: %.3f\n",
tri.getArea());
break;
default:
System.out.println("Enter 0, 1, 2, 3,
or 4");
}
choice = getChoice();
}

/**
* Main method
*/
public static void main(String[] args)
 Java for Mathematicians/Page 209

{
System.out.println("Start test run of
calculator.");
TestTriangleV2 obj = new TestTriangleV2();
obj.go();
}

}
 Java for Mathematicians/Page 210

Chapter Eight: event-driven programming

Topic: GUI and Event-driven programming

In this chapter your Java computer program will allow

users to interact with the graphical display. Java
provides the Jbutton class to produce buttons, the
JList class for lists, the JTextField for text fields,
JCheckBox for a check box and JMenuItem for a menu
item. Those items can be added to a GUI object such as
a JPanel object. For example this simple Java GUI
program has a single object, a button.

import java.awt.*;
import javax.swing.*;
import java.awt.event.*;

public class ChameleonBtn extends JFrame

{
public ChameleonBtn()
{
JPanel panel = new JPanel();
final JButton btn = new JButton("change color");
panel.add(btn);
// A button listener is needed.
add(panel);
}

public static void main(String[] args)

{
JFrame frame = new ChameleonBtn();
frame.setTitle("chameleon");
 Java for Mathematicians/Page 211

frame.setLocationRelativeTo(null);

frame.setDefaultCloseOperation(JFrame.EXIT_ON_CLOSE);
frame.setSize(820, 800);
frame.pack();
frame.setVisible(true);
}

Class ChameleonBtn extends JFrame so instances of

ChameleonBtn produce a GUI interface with a frame. You
can set the title of the frame, its locations, and
size. Also setDefaultCloseOperation with parameter
value of JFrame.EXIT_ON_CLOSE allows the frame close
button labeled X to work properly. The final step is
frame.setVisible(true); Additional changes made after
that step might not be included in the GUI display.

The constructor for class ChameleonBtn creates an

instance of JButton which is attached to an instance
of JPanel. Finally that panel is attached to the frame
by a simple add command. Pressing the change color
button produces no result. Next our Java program must
include a listener to react to the ActionEvent
resulting from pressing that button.

Every time a user uses the mouse to press a button,

select from a list of choices, or move the scroll bar
that user action is an event. Likewise pressing a key
is an event. When the user interacts with a source
object or source component that object or component
 Java for Mathematicians/Page 212

produces an event. When you select from a JList

object a ListSelectionEvent is produced.

All event types are extensions of the core

EventObject. The AWTEvent, ListSelectionEvent, and
ChangeEvent are direct extensions of the EventObject.
ActionEvent, AdjustEvent, ComponentEvent, ItemEvent,
and TextEvent are extensions of the AWTEvent object.
The ComponentEvent continues to support extensions for
the ContainerEvent, FocusEvent, InputEvent,
PaintEvent, and WindowEvent. Finally the InputEvent is
extended by the MouseEvent and KeyEvent objects. Thus
an entire family of event objects has been developed
for each possible interaction with a Java GUI program.
The core EventObject provides getSource and toString
methods. When one listener is attached to several
instances of JButton objects then the listeners
commonly use getSource to identify which button was
pushed.

When you press a Jbutton an ActionEvent is produced.

Your java computer program must create an action
listener and register that action listener with the
button by using the addActionListener method. The
ActionListener interface requires implementation of
the
actionPerformed method.

Stage 1: use a listener to display a message on

the standard output stream

The implementation included here simply displays a

 Java for Mathematicians/Page 213

message that the button was pushed onto system output.

public void actionPerformed(ActionEvent e)

{
System.out.println("Button was pushed");
}

The ButtonListener is in a separate class.

import java.awt.event.*;

public class ButtonListener implements ActionListener

{
public void actionPerformed(ActionEvent e)
{
System.out.println("Button was pushed");
}

Now the PushMeBtn class can create an instance of

ButtonListener and use the addActionListener method to
register that button listener with the button.

ButtonListener btnListener = new

ButtonListener();
btn.addActionListener( btnListener );

The main class demonstrates the role of the button

producing an event which will be handled by the button
listener.

import java.awt.*;
import javax.swing.*;
 Java for Mathematicians/Page 214

import java.awt.event.*;

public class PushMeBtn extends JFrame

{
public PushMeBtn()
{
JPanel panel = new JPanel();
final JButton btn = new JButton("push");
panel.add(btn);
ButtonListener btnListener = new
ButtonListener();
btn.addActionListener( btnListener );
add(panel);
}

public static void main(String[] args)

{
JFrame frame = new PushMeBtn();
frame.setTitle("push me");
frame.setLocationRelativeTo(null);

frame.setDefaultCloseOperation(JFrame.EXIT_ON_CLOSE);
frame.setSize(820, 800);
frame.pack();
frame.setVisible(true);
}

The java computer program produces this GUI.

 Java for Mathematicians/Page 215

Pressing the button produced this statement on system

output.

Stage 2: share a parameter to change the

button's color

A GUI program does not normally print on the standard

output stream. Instead results are displayed on the
screen. For our next iteration of this program the
button should simply change color. Now the
actionPerformed method needs access to that button.
The button is defined in a different class, so the
button must be passed as an argument to the
ButtonListener class. The actionPerformed method is
defined by the ActionListener interface, so our
computer program cannot change that method's signature
to include the button. Instead a new constructor for
the ButtonListener will accept the button and save it
as the value of a new attribute named btn.

import java.awt.*;
import javax.swing.*;
import java.awt.event.*;
 Java for Mathematicians/Page 216

/**
* implements interface ActionListener
*/

public class ButtonListener implements ActionListener

{

// attribute
JButton btn;

/**
* fully parametrized constructor
*
* @param JButton button
*/
public ButtonListener(JButton _btn)
{
btn = _btn;
}

/**
* interface ActionListener requires implementation
of this method
*
* @see java.awt.event.ActionListener
*/
public void actionPerformed(ActionEvent e)
{
btn.setBackground(Color.ORANGE);
}

}
 Java for Mathematicians/Page 217

The test program is also changed to include the button

in the call to construct a new instance of
ButtonListener. Instead of new ButtonListener() the
command is new ButtonListener(btn).

import java.awt.*;
import javax.swing.*;
import java.awt.event.*;

public class ChameleonBtn extends JFrame

{
public ChameleonBtn()
{
JPanel panel = new JPanel();
final JButton btn = new JButton("push");
panel.add(btn);
ButtonListener btnListener = new
ButtonListener(btn);
btn.addActionListener( btnListener );
add(panel);
}

public static void main(String[] args)

{
JFrame frame = new ChameleonBtn();
frame.setTitle("chameleon");
frame.setLocationRelativeTo(null);

frame.setDefaultCloseOperation(JFrame.EXIT_ON_CLOSE);
frame.setSize(820, 800);
frame.pack();
frame.setVisible(true);
 Java for Mathematicians/Page 218

When the button is pressed it changes color to orange.

Stage 3: use an inner class for the listener

A further refactoring of this program is to move the

listener class into the test class. A class within a
class is known as an inner class. The main reason for
this refactoring is that an inner class automatically
is aware of the attributes of the surrounding class.
For our program the listener as an inner class would
automatically be aware of the button as an attribute.
Then the button would not need to be passed as an
argument to the listener. Be aware that the button
must become an attribute.

import java.awt.*;
import java.awt.*;
import javax.swing.*;
import java.awt.event.*;

public class InnerButtonListener extends JFrame

{
private JButton btn;
 Java for Mathematicians/Page 219

public InnerButtonListener()
{
JPanel panel = new JPanel();
btn = new JButton("push");
panel.add(btn);
ButtonListener btnListener = new
ButtonListener();
btn.addActionListener( btnListener );
add(panel);
}

public static void main(String[] args)

{
JFrame frame = new InnerButtonListener();
frame.setTitle("chameleon");
frame.setLocationRelativeTo(null);

frame.setDefaultCloseOperation(JFrame.EXIT_ON_CLOSE);
frame.setSize(820, 800);
frame.pack();
frame.setVisible(true);
}

/**
* inner class button listener implements interface
ActionListener
*/

public class ButtonListener implements

ActionListener
{
 Java for Mathematicians/Page 220

/**
* interface ActionListener requires
implementation of this method
*
* @see java.awt.event.ActionListener
*/
public void actionPerformed(ActionEvent e)
{
btn.setBackground(Color.ORANGE);
}

Step 4: Use an anonymous inner class for the

button listener

Since the button listener is only used with the button

programmers often use an anonymous inner class for the
listener. Then the btn.addActionListener() method
begins with the new ActionListener() method followed
by a class definition with the curly braces {}. That
class definition does not include any declaration such
as public class Anonymous. Instead the class
definition simply includes the methods. The
ActionListener interface requires the actionPerformed
method to be declared so the anonymous class includes
only that method which sets the background color of
the button to orange.
 Java for Mathematicians/Page 221

import java.awt.*;
import javax.swing.*;
import java.awt.event.*;

public class AnonymousInnerClass extends JFrame

{
public AnonymousInnerClass()
{
JPanel panel = new JPanel();
final JButton btn = new JButton("push");
panel.add(btn);
btn.addActionListener(
new ActionListener() {
public void actionPerformed(ActionEvent
e)
{
btn.setBackground(Color.ORANGE);
}
} );
add(panel);
}

public static void main(String[] args)

{
JFrame frame = new AnonymousInnerClass();
frame.setTitle("chameleon");
frame.setLocationRelativeTo(null);

frame.setDefaultCloseOperation(JFrame.EXIT_ON_CLOSE);
frame.setSize(820, 800);
frame.pack();
frame.setVisible(true);
}
 Java for Mathematicians/Page 222

When you examine the directory containing your java

computer program you will notice your java file, the
associated class file, and another class file ending
with $1. Each inner class has $number attached with
the number increasing from 1 for each inner class.

Anonymous inner classes are popular because the action

performed method is closely associated with the button
or other GUI object. Also the coding is briefer.

Topic: Layout managers

A layout manager controls the placement of objects on

the screen. Java provides several layout managers such
as FlowLayout, GridLayout, and BorderLayout, all of
which will be used in developing this application for
studying triangles.

Topic: FlowLayout
 Java for Mathematicians/Page 223

The setLayout command sets the layout manager for the

application. Until now our graphics programs used a
Jpanel which has the default FlowLayout manager, the
simplest layout manager. Being the default our java
programs have not included the setLayout(new
FlowLayout(FlowLayout.LEFT, 10, 20)) command which
creates an anonymous instance of FlowLayout having the
command to align objects to the left with a horizontal
gap between objects of 10 pixels and a vertical gap
between objects of 20 pixels. FlowLayout sets the
objects from left to right across the screen. When a
row is full FlowLayout advances down to the next row.
LEFT is a static constant of the FlowLayout class.
This simple java program will demonstrate the use of
flow layout.

import javax.swing.*;
import java.awt.*;

public class DemoFlowLayout extends JFrame

{

/**
* constructor
*/
public DemoFlowLayout()
{
setTitle("Demo FlowLayout");
setSize(200, 200);
setLocationRelativeTo(null); // Center the frame
setDefaultCloseOperation(JFrame.EXIT_ON_CLOSE);

setLayout(new FlowLayout(FlowLayout.LEFT, 5,
 Java for Mathematicians/Page 224

10));
JPanel panel = new JPanel();
panel.setLayout(new
FlowLayout(FlowLayout.LEFT,5,5));
panel.add(new JButton("1"));
panel.add(new JButton("2"));
panel.add(new JButton("3"));
panel.add(new JButton("4"));
panel.add(new JButton("5"));
panel.add(new JButton("6"));
add(panel);
setVisible(true);
}

/**
* main method
*/
public static void main(String[] args)
{
DemoFlowLayout demo = new DemoFlowLayout();
}

}
 Java for Mathematicians/Page 225

Topic: BorderLayout
The BorderLayout layout manager divides the screen
into five areas: East, South, West, North, and Center.
To add a new button to the east side of the screen the
command would be add(new JButton(“East”),
BorderLayout.EAST); The component requiring the most
space is usually placed in the center. This java
program demonstrates the use of border layout.

import javax.swing.*;
import java.awt.*;

public class DemoBorderLayout extends JFrame

{

/**
* constructor
*/
public DemoBorderLayout()
{
setTitle("Demo Border Layout");
setSize(200, 200);
setLocationRelativeTo(null); // Center the frame
setDefaultCloseOperation(JFrame.EXIT_ON_CLOSE);

//Set BorderLayout with horizontal gap 5 and

vertical gap 10
JPanel panel = new JPanel();
panel.setLayout(new BorderLayout(5, 10));
panel.add(new JButton("center"),
BorderLayout.CENTER);
panel.add(new JButton("east"),
BorderLayout.EAST);
 Java for Mathematicians/Page 226

panel.add(new JButton("west"),
BorderLayout.WEST);
panel.add(new JButton("north"),
BorderLayout.NORTH);
panel.add(new JButton("south"),
BorderLayout.SOUTH);
add(panel);
setVisible(true);
}

/**
* main method
*/
public static void main(String[] args)
{
DemoBorderLayout demo = new DemoBorderLayout();
}

Topic: GridLayout
 Java for Mathematicians/Page 227

The GridLayoutManager divides the container into a

rectangular grid. The command setLayout(new
GridLayout(3,2,5,5)) will divide the container into a
grid of three rows each containing two columns. Both
the horizontal and vertical separation between objects
will be ten pixels.

This example java program will demonstrate the use of

grid layout.

import javax.swing.*;
import java.awt.*;

public class DemoGridLayout extends JFrame

{

/**
* constructor
*/
public DemoGridLayout()
{
setTitle("Demo GridLayout");
setSize(200, 200);
setLocationRelativeTo(null); // Center the frame
setDefaultCloseOperation(JFrame.EXIT_ON_CLOSE);

setLayout(new FlowLayout(FlowLayout.LEFT, 5,
10));
JPanel panel = new JPanel();
panel.setLayout(new GridLayout(3,2,5,5));
panel.add(new JButton("1"));
panel.add(new JButton("2"));
 Java for Mathematicians/Page 228

panel.add(new JButton("3"));
panel.add(new JButton("4"));
panel.add(new JButton("5"));
panel.add(new JButton("6"));
add(panel);
setVisible(true);
}

/**
* main method
*/
public static void main(String[] args)
{
DemoGridLayout demo = new DemoGridLayout();
}

Topic: Event-driven GUI programming

Next we will use our new knowledge of event-driven

 Java for Mathematicians/Page 229

programming to develop a GUI program to draw and study

triangles. Development of the program begins with an
empty JFrame.

The corresponding java computer program uses a class

that extends JFrame so the class definition begins
with public class TriangleToolV1 extends JFrame
followed by the body of the class within curly braces
{} also known as curly brackets. JFrame requires an
import statement for java.swing.JFrame or more
generally import java.swing.*;. A main method is
required to begin execution of your GUI program. The
constructor has a setTitle(“Triangle Tool”) that sets
the frame's title. Next the setSize(800,600) command
sets the frame's size and the command
setLocationRelativeTo(null) sets the frame's location.

The layout is set to BorderLayout. Because the screen

area for displaying a triangle would demand the most
space that screen area can be set in the center and
command the largest share of the screen's space.

import javax.swing.*;
import java.awt.*;
 Java for Mathematicians/Page 230

/**
* Tool for working with triangles
*/
public class TriangleToolV1 extends JFrame
{

/**
* default constructor for triangle tool
*/
public TriangleToolV1()
{
// Set up GUI
setTitle("Triangle Tool");
setSize(800, 600);
setLocationRelativeTo(null); // Center the frame
setDefaultCloseOperation(JFrame.EXIT_ON_CLOSE);

//Set BorderLayout with horizontal gap 5 and

vertical gap 10
setLayout(new BorderLayout(5, 10));
setVisible(true);
}

/**
* starts the program
*/
private void go()
{

/**
 Java for Mathematicians/Page 231

* main method
*/
public static void main(String[] args)
{
TriangleToolV1 tool = new TriangleToolV1();
tool.go();
}

} // end class TriangleTool

The program produces an empty frame.

For the next step of this application's development

our program adds a panel to the center region of the
screen. An inner class named SketchPanel extends JPanel
so the paintComponent method can be overwritten with
our own code to draw triangles.

/**
* inner class SketchPanel for drawing triangles,
 Java for Mathematicians/Page 232

etc
*/
class SketchPanel extends JPanel
{

/**
* Override JPanel's paintComponent method
*
* @param Graphics object g
*/
protected void paintComponent(Graphics g)
{
super.paintComponent(g);
g.drawString("triangles will be drawn here",
0, 40);
}
}

The GUI update method within the java virtual machine

automatically calls this paintComponent method and
supplies the graphics object named g. The
super.paintComponent(g) command calls the
paintComponent method of the superclass, or parent
class, so the screen area will be erased and made
ready for our new graphics command. Instead of drawing
a triangle this program simply draws a string stating
“triangles will be drawn here”.

Next the Triangle tool's constructor will create an

instance of the sketch pad class and add that new
panel to the center of the frame.

// Add sketch pad to draw triangle

 Java for Mathematicians/Page 233

JPanel sketchPad = new SketchPanel();

sketchPad.add(new JLabel("sketch pad"));
add(sketchPad, BorderLayout.CENTER);

The new label named “sketch pad” temporarily shows

that the panel has been added. That label will be
removed later.

The entire program is here.

import javax.swing.*;
import java.awt.*;

/**
* Tool for working with triangles
*/
public class TriangleToolV2 extends JFrame
{

/**
* default constructor for triangle tool
*/
public TriangleToolV2()
{
// Set up GUI
setTitle("Triangle Tool");
setSize(800, 600);
setLocationRelativeTo(null); // Center the frame
setDefaultCloseOperation(JFrame.EXIT_ON_CLOSE);

//Set BorderLayout with horizontal gap 5 and

vertical gap 10
 Java for Mathematicians/Page 234

setLayout(new BorderLayout(5, 10));

// Add sketch pad to draw triangle

JPanel sketchPad = new SketchPanel();
sketchPad.add(new JLabel("sketch pad"));
add(sketchPad, BorderLayout.CENTER);

setVisible(true);
}

/**
* starts the program
*/
private void go()
{

/**
* main method
*/
public static void main(String[] args)
{
TriangleToolV2 tool = new TriangleToolV2();
tool.go();
}

/**
* inner class SketchPanel for drawing triangles,
etc
*/
class SketchPanel extends JPanel
{
 Java for Mathematicians/Page 235

/**
* Override JPanel's paintComponent method
*
* @param Graphics object g
*/
protected void paintComponent(Graphics g)
{
super.paintComponent(g);
g.drawString("triangles will be drawn here",
0, 40);
}
}

} // end class TriangleTool

The triangle application now displays the label

“sketch pad” and the message “triangles will be drawn
here”.

Of course the next step is to draw a triangle. That

 Java for Mathematicians/Page 236

requires establishing a model of the triangle. The

easiest model is to keep the x and y coordinates of
the three vertices of the triangle. For that purpose a
separate class named TriangleCartesian will store the
x and y coordinates of each vertex as Point2D.Double
objects.

As an aside a common paradigm for developing GUI

programs is named model-view-controller and described
at
http://en.wikipedia.org/wiki/Model_view_controller .
Although this program uses a variable named model to
represent the triangle, this application is not an
example of the model-view-controller paradigm.

import java.awt.geom.*;
import java.awt.*;

/**
* Class CartesianTriangle uses three Points to
describe its triangle.
*/

public class TriangleCartesian

{

// Attributes
Point2D.Double pointA, pointB, pointC;

// Constructors

/**
* default constructor
 Java for Mathematicians/Page 237

*/
public TriangleCartesian()
{
pointA = new Point2D.Double(); // by default set
to (0,0)
pointB = new Point2D.Double();
pointC = new Point2D.Double();
}

/**
* fully parametrized constructor
*
* @param Point2D.Double three vertices of triangle
*/
public TriangleCartesian(Point2D.Double a,
Point2D.Double b, Point2D.Double c)
{
pointA = a;
pointB = b;
pointC = c;
}

// Accessors and mutators

/**
* getPointA returns the Point object for attribute
pointA
*
* @return Point2D.Double for vertex A
*/
public Point2D.Double getPointA()
{
return pointA;
 Java for Mathematicians/Page 238

/**
* setPointA sets pointA to the Point provided in
the list of arguments
*
* @param the point for this vertex
*/
public void setPointA(Point2D.Double point)
{
pointA = point;
}

/**
* getPointB returns the Point object for attribute
pointB
*
* @return vertex B
*/
public Point2D.Double getPointB()
{
return pointB;
}

/**
* getPointB sets pointB to the Point provided in
the list of arguments
*
* @param the point for this vertex
*/
public void setPointB(Point2D.Double point)
{
pointB = point;
 Java for Mathematicians/Page 239

/**
* getPointC returns the Point object for vertex
pointC
*
* @return vertex C
*/
public Point2D.Double getPointC()
{
return pointC;
}

/**
* getPointA sets pointA to the Point provided in
the list of arguments
*
* @param the point for this vertex
*/
public void setPointC(Point2D.Double point)
{
pointC = point;
}

/**
* get length of side A - B
*/
public double getSideAB()
{
// Point2D.Double extends Point2D which provides
the distance method
return pointA.distance(pointB);
}
 Java for Mathematicians/Page 240

/**
* get length of side B-C
*
* @return length of side BC
*/
public double getSideBC()
{
return pointB.distance(pointC);
}

/**
* get length of side A - C
*
* @return length of side AC
*/
public double getSideAC()
{
return pointA.distance(pointC);
}

/**
* get perimeter of triangle
*
* @return perimeter
*/
public double getPerimeter()
{
return getSideAB() + getSideBC() + getSideAC();
}

/**
* calculate area of triangle
 Java for Mathematicians/Page 241

* In geometry, Heron's (or Hero's) formula, named

after Heron of Alexandria,
* states that the area T of a triangle whose sides
have lengths a, b, and c is
* T = sqrt( s(s-a)(s-b)(s-c) )
* where s is the semiperimeter of the triangle:
*
* @return area
*/
public double getArea()
{
double s = getPerimeter() / 2;
double a = getSideAB();
double b = getSideBC();
double c = getSideAC();
double area = Math.sqrt(s * (s-a) * (s-b) * (s-
c));
return area;
}

/**
* drawTriangle uses a graphics object g to draw
the triangle
*
* @param graphics object
*
* @see java.awt.Polygon
*/
public void drawTriangle(Graphics g)
{
final int SCALE = 10; // Scale by 10 then
truncate to int value for locations
Polygon poly = new Polygon();
 Java for Mathematicians/Page 242

poly.addPoint((int)(SCALE * pointA.getX()),
(int)(SCALE * pointA.getY()));
poly.addPoint((int)(SCALE * pointB.getX()),
(int)(SCALE * pointB.getY()));
poly.addPoint((int)(SCALE * pointC.getX()),
(int)(SCALE * pointC.getY()));
g.drawPolygon(poly);
}

Notice that the model stores the x and y coordinates

as double values, but the draw method truncates those
double values to int values before drawing the
triangle. Also class Polygon provides a draw method
which this class wisely employs. The strength of java
is the immense collection of classes provided for our
use.

Since java's coordinate system uses only the quadrant

of the Cartesian space where both x and y coordinates
have positive values our triangle will be required to
appear in that quadrant. Also (0,0) is located in the
top-left corner. The X coordinate increases from left
to right, but the Y coordinate increases from top to
bottom, so our triangle will display upside down.

Developing a more realistic coordinate system is a

different problem left to the reader.

The triangle tool must define the triangle model,

named triangle, as an attribute so the paint method of
the inner class will have access to that model. The
 Java for Mathematicians/Page 243

command to create that model uses class Triangle

Cartesian which require three double valued points.

// Make model of triangle using vertices.

triangle = new TriangleCartesian(
new Point2D.Double(0.0, 0.0),
new Point2D.Double(22.3, 34.9),
new Point2D.Double(35.7, 50.2)
);

The draw component method of the sketch panel must now

draw that triangle. The additional command is simply
triangle.drawTriangle(g).
The main class now includes the following commands.
import javax.swing.*;
import java.awt.*;
import java.awt.event.*;
import java.awt.geom.*;

/**
* Tool for working with triangles
*/
public class TriangleToolV3 extends JFrame
{
TriangleCartesian triangle;

/**
* default constructor for triangle tool
*/
public TriangleToolV3()
{
// Set up GUI
setTitle("Triangle Tool");
 Java for Mathematicians/Page 244

setSize(800, 600);
setLocationRelativeTo(null); // Center the frame
setDefaultCloseOperation(JFrame.EXIT_ON_CLOSE);

//Set BorderLayout with horizontal gap 5 and

vertical gap 10
setLayout(new BorderLayout(5, 10));

// Add sketch pad to draw triangle

JPanel sketchPad = new SketchPanel();
// sketchPad.add(new JLabel("sketch pad"));
add(sketchPad, BorderLayout.CENTER);

// Make model of triangle using vertices.

triangle = new TriangleCartesian(
new Point2D.Double(0.0, 0.0),
new Point2D.Double(22.3, 34.9),
new Point2D.Double(35.7, 50.2)
);

setVisible(true);
}

/**
* starts the program
*/
private void go()
{

/**
 Java for Mathematicians/Page 245

* main method
*/
public static void main(String[] args)
{
TriangleToolV3 tool = new TriangleToolV3();
tool.go();
}

/**
* inner class SketchPanel for drawing triangles,
etc
*/
class SketchPanel extends JPanel
{

/**
* Override JPanel's paintComponent method
*
* @param Graphics object g
*/
protected void paintComponent(Graphics g)
{
super.paintComponent(g);
triangle.drawTriangle(g);
}
}

} // end class TriangleTool

The program now draws a triangle as shown here.

 Java for Mathematicians/Page 246

Also the triangle program must create a panel for the

button to draw the triangle. The action performed
method of that anonymous listener simply call repaint.
That method calls the paint component method which is
redefined to draw the triangle. The additional command
is simply triangle.drawTriangle(g).

The next enhancement will create a data panel for the

user to enter coordinates for all three vertices and
 Java for Mathematicians/Page 247

to view calculated results such as area and length of

the perimeter.

// Create data panel for points and results at

right (EAST) side
dataPanel = new JPanel();
dataPanel.setLayout(new BorderLayout(5, 10));

// Add x,y addresses of three defining points of

triangle
TriangleTfGrid = new JPanel();
TriangleTfGrid.setLayout(new GridLayout(4, 2, 5,
5));
TriangleTfGrid.add(new JLabel("x"));
TriangleTfGrid.add(new JLabel("y"));
// Add 2 x 3 array of JTextField
trianglePointsTf = new ArrayList<PointXYTf>();

for (int i = 0; i < VERTEX_CNT; i++)

{
PointXYTf ptTf = new PointXYTf();
trianglePointsTf.add(ptTf);
TriangleTfGrid.add(ptTf.getTfX());
TriangleTfGrid.add(ptTf.getTfY());
}
dataPanel.add(TriangleTfGrid,
BorderLayout.NORTH);

The class named PointTf holds the text fields for the
x and y coordinates of a single vertex of the
triangle. Since the pair of x and y coordinates for a
vertex represent the coordinates of an object they
should be themselves stored in an object. All those
 Java for Mathematicians/Page 248

rows of text fields are stored in an array list named

trianglePointsTf.

The PointTf class is listed here.

import javax.swing.*;
import java.awt.geom.*;

/**
* Maintain textFields for the x and y coordinates of a
point
* Always maintain the point and the textfields with
equivalent values
*/
public class PointXYTf
{
JTextField tfX, tfY;
final int JTF_WIDTH = 8;
int jtfWidth;
final String JTF_INIT_VALUE = "0.0";

/**
* Default constructor
*/
public PointXYTf()
{
jtfWidth = JTF_WIDTH;

tfX = new JTextField();

tfX.setText(JTF_INIT_VALUE);

tfY = new JTextField();

tfY.setText(JTF_INIT_VALUE);
 Java for Mathematicians/Page 249

/**
* fully parametrized constructor using JTextFields
*/
public PointXYTf(JTextField tfx, JTextField tfy)
{
tfX = tfx;
tfY = tfy;
}

/**
* fully parametrized constructor using Point2D.Double
*
* @param Point2D.Double the point matching these
text fields
*/
public PointXYTf(Point2D.Double pt)
{
// point = pt;
double x = pt.getX();
String xStr = new Double(x).toString();
tfX = new JTextField(xStr);
double y = pt.getY();
String yStr = new Double(y).toString();
tfY = new JTextField(yStr);
}

/**
* Set contents of tfX and tfY for point
*
* @param Point2D.Double point matching these text
fields
 Java for Mathematicians/Page 250

*/
public void setTfXY(Point2D.Double p)
{
double x = p.getX();
Double dX = new Double(x);
String xString = dX.toString();
tfX.setText(xString);
double y = p.getY();
Double dY = new Double(y);
String yString = dY.toString();
tfY.setText(yString);
}

/**
* Set x value
*
* @param double x coordinate
*/
public void setTfX(double x)
{
Double dX = new Double(x);
String xString = dX.toString();
tfX.setText(xString);
}

/**
* Set y value
*
* @param double y coordinate
*/
public void setTfY(double y)
{
Double dY = new Double(y);
 Java for Mathematicians/Page 251

String yString = dY.toString();

tfY.setText(yString);
}

/**
* Get textfield X
*
* @return JTextField for X coordinate
*/
public JTextField getTfX()
{
return tfX;
}

/**
* Get textfield Y
*
* @return JTextField for Y coordinate
*/
public JTextField getTfY()
{
return tfY;
}

/**
* return Point2D.Double
*
* @return Point2D with X,Y coordinates
*/
public Point2D.Double getPoint2D()
{
String strX = tfX.getText();
double x = new Double(strX).doubleValue();
 Java for Mathematicians/Page 252

String strY = tfY.getText();

double y = new Double(strY).doubleValue();
Point2D.Double point = new Point2D.Double(x,y);
return point;
}

/**
* toString method
*
* @return description of x,y coordinates
*/
public String toString()
{
return "[" + tfX.getText() + "," + tfY.getText()
+"]";
}

}
Notice the PointXYTf class can return the vertex as a
Point 2D.Double object which satisfies the
requirements of the Polygon class. So these classes
cooperate well.

The results include area and perimeter. A new grid

layout holds the text fields for those results.

// Add 2 x 2 array of calculated results

results = new JPanel();
results.setLayout(new GridLayout(2, 2, 5, 5));
results.add(new JLabel("area"));
jtfArea = new JTextField(8);
jtfArea.setText("0.0");
area = 0;
 Java for Mathematicians/Page 253

results.add(jtfArea);
results.add(new JLabel("perimeter"));
jtfPerimeter = new JTextField(8);
jtfPerimeter.setText("0.0");
results.add(jtfPerimeter);
perimeter = 0;
dataPanel.add(results, BorderLayout.SOUTH);

That data panel is now added to the surrounding frame

using this command add(dataPanel, BorderLayout.EAST);

A control panel is created for the button named

calculate, which will both perform the calculations
and draw the new triangle.

// Add control panel to the bottom (SOUTH) of

the window
controls = new JPanel();
controls.setLayout(new
FlowLayout(FlowLayout.LEFT, 10, 20));
jbtnCalc = new JButton("calculate");
jbtnCalc.addActionListener(
new ActionListener()
{
public void actionPerformed(ActionEvent e)
{
calcTriangle();
repaint();
}
}
);
controls.add(jbtnCalc);
add(controls, BorderLayout.SOUTH);
 Java for Mathematicians/Page 254

Finally the method to perform the calculations, named

calcTriangle, must be written. It gets the vertex from
each row of the table of vertices. Those

The final program is shown here.

1 import javax.swing.*;
2 import java.awt.*;
3 import java.awt.event.*;
4 import java.awt.geom.*;
5 import java.util.*;
6
7 /**
8 * Tool for working with triangles
9 */
10 public class TriangleToolV4 extends JFrame
11 {
12 TriangleCartesian triangle;
13 ArrayList<PointXYTf> trianglePointsTf;
14 JTextField jtfArea, jtfPerimeter;
15 double area, perimeter;
16
17 /**
18 * default constructor for triangle tool
19 */
20 public TriangleToolV4()
21 {
22 // Set up GUI
23 setTitle("Triangle Tool");
24 setSize(800, 600);
25 setLocationRelativeTo(null); // Center the
frame
 Java for Mathematicians/Page 255

26
setDefaultCloseOperation(JFrame.EXIT_ON_CLOSE);
27 //Set BorderLayout with horizontal gap 5 and
vertical gap 10
28 setLayout(new BorderLayout(5, 10));
29
30 // Add sketch pad to draw triangle
31 JPanel sketchPad = new SketchPanel();
32 // sketchPad.add(new JLabel("sketch pad"));
33 add(sketchPad, BorderLayout.CENTER);
34
35 // Create data panel for points and results
at right (EAST) side
36 JPanel dataPanel = new JPanel();
37 dataPanel.setLayout(new BorderLayout(5, 10));
38
39 // Add x,y addresses of three defining
points of triangle
40 JPanel triangleTfGrid = new JPanel();
41 triangleTfGrid.setLayout(new GridLayout(4,
2, 5, 5));
42 triangleTfGrid.add(new JLabel("x"));
43 triangleTfGrid.add(new JLabel("y"));
44 // Add 2 x 3 array of JTextField
45 trianglePointsTf = new
ArrayList<PointXYTf>();
46 final int VERTEX_CNT = 3;
47
48 for (int i = 0; i < VERTEX_CNT; i++)
49 {
50 PointXYTf ptTf = new PointXYTf(new
Point2D.Double(0,0));
51 trianglePointsTf.add(ptTf);
 Java for Mathematicians/Page 256

52 triangleTfGrid.add(ptTf.getTfX());
53 triangleTfGrid.add(ptTf.getTfY());
54 }
55 dataPanel.add(triangleTfGrid,
BorderLayout.NORTH);
56
57 // Add 2 x 2 array of calculated results
58 JPanel results = new JPanel();
59 results.setLayout(new GridLayout(2, 2, 5,
5));
60 results.add(new JLabel("area"));
61 jtfArea = new JTextField(8);
62 jtfArea.setText("0.0");
63 area = 0;
64 results.add(jtfArea);
65 results.add(new JLabel("perimeter"));
66 jtfPerimeter = new JTextField(8);
67 jtfPerimeter.setText("0.0");
68 results.add(jtfPerimeter);
69 perimeter = 0;
70 dataPanel.add(results, BorderLayout.SOUTH);
71
72 add(dataPanel, BorderLayout.EAST);
73
74 // Add control panel to the bottom (SOUTH)
of the window
75 JPanel controls = new JPanel();
76 controls.setLayout(new
FlowLayout(FlowLayout.LEFT, 10, 20));
77 JButton jbtnCalc = new JButton("calculate");
78 jbtnCalc.addActionListener(
79 new ActionListener()
80 {
 Java for Mathematicians/Page 257

81 public void
actionPerformed(ActionEvent e)
82 {
83 calcTriangle();
84 repaint();
85 }
86 }
87 );
88 controls.add(jbtnCalc);
89 add(controls, BorderLayout.SOUTH);
90 triangle = new TriangleCartesian();
91 setVisible(true);
92 }
93
94 /**
95 * Calculate triangle's area and perimeter
96 */
97 private void calcTriangle()
98 {
99 Point2D.Double a, b, c;
100 System.out.println("Starting calcTriangle");
101 ArrayList<Point2D.Double> vertices = new
ArrayList<>();
102 for (PointXYTf row : trianglePointsTf)
103 {
104 System.out.println(row);
105 vertices.add(row.getPoint2D());
106 }
107 System.out.println("number of points: " +
vertices.size());
108 a = vertices.get(0);
109 b = vertices.get(1);
110 c = vertices.get(2);
 Java for Mathematicians/Page 258

111 triangle =
112 new TriangleCartesian(a, b, c);
113 area = triangle.getArea();
114 String areaStr = String.format("%8.3f",
area);
115 jtfArea.setText(areaStr);
116 perimeter = triangle.getPerimeter();
117 String perimeterStr = String.format("%8.3f",
perimeter);
118 jtfPerimeter.setText(perimeterStr);
119 }
120
121
122 /**
123 * starts the program
124 */
125 private void go()
126 {
127
128 }
129
130 /**
131 * main method
132 */
133 public static void main(String[] args)
134 {
135 TriangleToolV4 tool = new TriangleToolV4();
136 tool.go();
137 }
138
139 /**
140 * inner class SketchPanel for drawing
triangles, etc
 Java for Mathematicians/Page 259

141 */
142 class SketchPanel extends JPanel
143 {
144
145 /**
146 * Override JPanel's paintComponent method
147 *
148 * @param Graphics object g
149 */
150 protected void paintComponent(Graphics g)
151 {
152 super.paintComponent(g);
153 triangle.drawTriangle(g);
154 }
155 }
156
157 } // end class TriangleTool

The final program produces this GUI.

 Java for Mathematicians/Page 260

After entering values for the three vertices the

results are displayed.

The goal of this book is to demonstrate how java

computer programs can help you discover new properties
of mathematical objects. In this situation you would
expand this simple program to display more properties
of a triangle such as the inner circle touching all
three sides and the outer circle touching all three
vertices. Next you would add the ability to click and
drag on a vertex so you can play with the triangle.
Only then will constants be discovered.