Expressions

Eric Roberts CS 106A April 11, 2012

Once upon a time . . .

Holism vs. Reductionism

In his Pulitzer-prizewinning book, computer scientist Douglas Hofstadter identifies two conceptsholism and reductionismthat turn out to be important as you begin to learn about programming. Hofstadter explains these concepts using a dialogue in the style of Lewis Carroll:
Achilles: I will be glad to indulge both of you, if you will first oblige me, by telling me the meaning of these strange expressions, holism and reductionism. Crab: Holism is the most natural thing in the world to grasp. Its simply the belief that the whole is greater than the sum of its parts. No one in his right mind could reject holism.

Anteater: Reductionism is the most natural thing in the world to grasp. Its simply the belief that a whole can be understood completely if you understand its parts, and the nature of their sum. No one in her left brain could reject reductionism.

Expressions

The Add2Integers Program

class Add2Integers extends ConsoleProgram { public void run() { println("This program adds two numbers."); int n1 = readInt("Enter n1: "); int n2 = readInt("Enter n2: "); int total = n1 + n2; println("The total is " + total + "."); } n1 n2 total } 17 25 42
Add2Integers

This program adds two numbers. Enter n1: 17 Enter n2: 25 The total is 42.

Expressions in Java
The heart of the Add2Integers program from Chapter 2 is the line
int total = n1 + n2;

that performs the actual addition.

The n1 + n2 that appears to the right of the equal sign is an example of an expression, which specifies the operations involved in the computation.
An expression in Java consists of terms joined together by operators. Each term must be one of the following:
A constant (such as 3.14159265 or "hello, world") A variable name (such as n1, n2, or total) A method call that returns a value (such as readInt) An expression enclosed in parentheses

Primitive Data Types

Although complex data values are represented using objects, Java defines a set of primitive types to represent simple data. Of the eight primitive types available in Java, the programs in this text use only the following four:
int

This type is used to represent integers, which are whole numbers such as 17 or 53.

double

This type is used to represent numbers that include a decimal fraction, such as 3.14159265.

boolean This type represents a logical value (true or false).

char

This type represents a single character.

Constants and Variables

The simplest terms that appear in expressions are constants and variables. The value of a constant does not change during the course of a program. A variable is a placeholder for a value that can be updated as the program runs. A variable in Java is most easily envisioned as a box capable The format of a constant depends on its type: of storing a value. consist of a string of digits, optionally preceded by a Integral constants

total minus sign, as in 0, 42, -1, or 1000000. Floating-point constants include a decimal point, as in 3.14159265 or (contains an int) 42 10.0. Floating-point constants can also be expressed in scientific notation by has the following attributes: Each variableadding the letter E and an exponent after the digits of the number, so that 5.646E-8 represents the number 5.646 x 10-8. A name, which enables you to differentiate one variable from another. The two constants of type boolean are true and false. A type, which specifies what type of value the variable can contain. Character and string constants are discussed in detail in Chapter 8. A value, which represents the current contentsaof the variable. consists For the moment, all you need to know is that string constant of a sequence of characters enclosed in double quotation marks, such The name and type of a variable are fixed. The value changes as "hello, world".

whenever you assign a new value to the variable.

Variable Declarations
In Java, you must declare a variable before you can use it. The declaration establishes the name and type of the variable and, in most cases, specifies the initial value as well. The most common form of a variable declaration is
type name = value;

where type is the name of a Java primitive type or class, name is an identifier that indicates the name of the variable, and value is an expression specifying the initial value. Most declarations appear as statements in the body of a method definition. Variables declared in this way are called local variables and are accessible only inside that method.

Variables may also be declared as part of a class. These are called instance variables and are covered in Chapter 6.

Operators and Operands

As in most languages, Java programs specify computation in the form of arithmetic expressions that closely resemble expressions in mathematics. The most common operators in Java are the ones that specify arithmetic computation: + Addition * Multiplication Subtraction / Division % Remainder Operators in Java usually appear between two subexpressions, which are called its operands. Operators that take two operands are called binary operators. The - operator can also appear as a unary operator, as in the expression -x, which denotes the negative of x.

Division and Type Casts

Whenever you apply a binary operator to numeric values in Java, the result will be of type int if both operands are of type int, but will be a double if either operand is a double. This rule has important consequences in the case of division. For example, the expression
14 / 5

seems as if it should have the value 2.8, but because both operands are of type int, Java computes an integer result by throwing away the fractional part. The result is therefore 2. If you want to obtain the mathematically correct result, you need to convert at least one operand to a double, as in
(double) 14 / 5

The conversion is accomplished by means of a type cast, which consists of a type name in parentheses.

The Pitfalls of Integer Division

Consider the following Java statements, which are intended to convert 100 Celsius temperature to its Fahrenheit equivalent:
double c = 100; double f = 9 / 5 * c + 32;

The computation consists of evaluating the following expression:

9 The problem arises from the fact that both 9 and 5 are of type int, which means that the result is also an int. / 5 * 100 1 9 / 5 * c + 32 c 132 32 +

The Pitfalls of Integer Division

You can fix this problem by converting the fraction to a double, either by inserting decimal points or by using a type cast:
double c = 100; double f = (double) 9 / 5 * c + 32;

The computation now looks like this:

212.0 180.0

1.8
9.0 (double) 9 / 5 * c + 32

The Remainder Operator

The only arithmetic operator that has no direct mathematical counterpart is %, which applies only to integer operands and computes the remainder when the first divided by the second:
14 % 5

returns 14 % 7 returns 7 % 14 returns

4 0 7

The result of the % operator make intuitive sense only if both operands are positive. The examples in the book do not depend on knowing how % works with negative numbers. The remainder operator turns out to be useful in a surprising number of programming applications and is well worth a bit of study.

Precedence
If an expression contains more than one operator, Java uses precedence rules to determine the order of evaluation. The arithmetic operators have the following relative precedence:
unary * /

(type cast)
%

highest

lowest

Thus, Java evaluates unary - operators and type casts first, then the operators *, /, and %, and then the operators + and -.
Precedence applies only when two operands compete for the same operator. If the operators are independent, Java evaluates expressions from left to right. Parentheses may be used to change the order of operations.

Exercise: Precedence Evaluation

What is the value of the expression at the bottom of the screen?
42

32

32

0
3 30

4
8

( 1 + 2 ) % 3 * 4 + 5 * 6 / 7 * ( 8 % 9 ) + 10

Assignment Statements
You can change the value of a variable in your program by using an assignment statement, which has the general form:
variable = expression;

The effect of an assignment statement is to compute the value of the expression on the right side of the equal sign and assign that value to the variable that appears on the left. Thus, the assignment statement
total = total + value;

adds together the current values of the variables total and value and then stores that sum back in the variable total. When you assign a new value to a variable, the old value of that variable is lost.

Shorthand Assignments
Statements such as
total = total + value;

are so common that Java allows the following shorthand form:

total += value;

The general form of a shorthand assignment is

variable op= expression;

where op is any of Javas binary operators. The effect of this statement is the same as
variable = variable op (expression);

For example, the following statement multiplies salary by 2.

salary *= 2;

Increment and Decrement Operators

Another important shorthand form that appears frequently in Java programs is the increment operator, which is most commonly written immediately after a variable, like this:
x++;

The effect of this statement is to add one to the value of x, which means that this statement is equivalent to
x += 1;

or in an even longer form

x = x + 1;

The -- operator (which is called the decrement operator) is similar but subtracts one instead of adding one.
The ++ and -- operators are more complicated than shown here, but it makes sense to defer the details until Chapter 11.

Extending Add2Integers
The next few slides extend the Add2Integers program from Chapter 2 to create programs that add longer lists of integers. These slides illustrate three different strategies:
Adding new code to process each input value Repeating the input cycle a predetermined number of times Repeating the input cycle until the user enters a sentinel value

The Add4Integers Program

You could easily change the Add2Integers into a program that added four integers just by adding additional variables, as shown in the following example:
public class Add4Integers extends ConsoleProgram { public void run() { println("This program adds four numbers."); int n1 = readInt("Enter n1: "); int n2 = readInt("Enter n2: "); int n3 = readInt("Enter n3: "); int n4 = readInt("Enter n4: "); int total = n1 + n2 + n3 + n4; println("The total is " + total + "."); } }

This strategy, however, is difficult to generalize and would clearly be cumbersome if you needed to add 100 values.

The Repeat-N-Times Pattern

One strategy for generalizing the addition program is to use the Repeat-N-Times pattern, which executes a set of statements a specified number of times. The general form of the pattern is
for (int i = 0; i < repetitions; i++) { statements to be repeated }

The information patternatic patterns repetitions is specified by As is true for allabout the number of in this book, the italicized the first line in the parts which is called the header change for words indicate the pattern,of the pattern you need to line. each statements to To use this pattern, for example, you the for The application. be repeated are called the body of need to replace repetitions with an expression giving the number of statement and are indented with respect to the header line. repetitions and include the statements to be repeated inside the A control statement that repeats a section of code is called a loop. curly braces. Each execution of the body of a loop is called a cycle.

The AddNIntegers Program

The body of in loop consistsworks correctly onlyThe first reads the variable This for loopthe thisthe Repeat-N-Times pattern toifcompute the program uses example of two statements. an integer from the to 0 number of integer loop. total is initialized user before executing the values, specified by sum of a predetermined into the variable value, and the second adds that value to the the named constant N.variable total.
public class AddNIntegers extends ConsoleProgram { public void run() { println("This program adds " + N + " numbers."); int total = 0; for (int i = 0; i < N; i++) { int value = readInt(" ? "); total += value; } println("The total is " + total + "."); } private static final int N = 100; }

The Repeat-Until-Sentinel Pattern

A better approach for the addition program that works for any number of values is to use the Repeat-Until-Sentinel pattern, which executes a set of statements until the user enters a specific value called a sentinel to signal the end of the list:
while (true) { prompt user and read in a value if (value == sentinel) break; rest of loop body }

You should choose a sentinel value that is not likely to occur in the input data. It also makes sense to define the sentinel as a named constant to make the sentinel value easy to change.

The AddIntegerList Program

This program uses the Repeat-Until-Sentinel pattern to add a list of integers, stopping when the user enters a value that matches the named constant SENTINEL.
public class AddIntegerList extends ConsoleProgram { public void run() { println("This program adds a list of integers."); println("Enter values, one per line, using " + SENTINEL); println("to signal the end of the list."); int total = 0; while (true) { int value = readInt(" ? "); if (value == SENTINEL) break; total += value; } println("The total is " + total + "."); } private static final int SENTINEL = 0; }

The AddIntegerList Program

public void run() { println("This program adds a list of integers."); println("Enter values, one per line, using " + SENTINEL); println("to signal the end of the list."); int total = 0; while (true) { int value = readInt(" ? "); if (value == SENTINEL) break; total += value; } value total println("The total is " + total + "."); 0 3 2 1 6 3 1 0 }
AddIntegerList

This program adds a list of integers. Enter values, one per line, using 0 to signal the end of the list. ? 1 ? 2 ? 3 ? 0 The total is 6.
skip simulation

The End