BASIC ELEMENTS OF JAVA - Java Programming: From Problem Analysis to Program Design - page 49

To allocate memory, we use Java's declaration statements. The syntax to declare a named
constant is:

static final dataType IDENTIFIER = value;

2

In Java, static and final are reserved words. The reserved word final specifies that
the value stored in the identifier is fixed and cannot be changed.

In syntax, the shading indicates the part of the definition that is optional.

Because the reserved word static is shaded, it may or may not appear when a named
constant is declared. The section, Creating a Java Application Program, later in this
chapter, explains when this reserved word might be required. Also, notice that the
identifier for a named constant is in uppercase letters. This is because Java programmers
typically use uppercase letters for a named constant. (If the name of a named constant is a
combination of more than one word, called a run-together-word, then the words are
separated using an underscore; see the next example.)

EXAMPLE 2-11

Consider the following Java statements:

final double CENTIMETERS_PER_INCH = 2.54;
final int NO_OF_STUDENTS = 20;
final char BLANK = ' ';
final double PAY_RATE = 15.75;

The first statement tells the compiler to allocate enough memory to store a value of type
double , call this memory space CENTIMETERS_PER_INCH , and store the value 2.54 in it.
Throughout a program that uses this statement, whenever the conversion formula is
needed, the memory space CENTIMETERS_PER_INCH can be accessed. The other state-
ments have similar meanings.

As noted earlier, the default type of floating-point numbers is double . Therefore, if you
declare a named constant of type float , then you must specify that the value is of type
float as follows:

final float PAY_RATE = 15.75f;

otherwise, the compiler will generate an error message. Notice that in 15.75f , the
letter f at the end specifies that 15.75 is a float value. Recall that the memory
size for float values is 4 bytes; for double values, 8 bytes. We will mostly use
the type double to work with floating-point values.