Transcript Chapter4

Java Software Solutions
Chapter 4
Objects Revisited
1
Objects
 Objects are created from classes.
 A class is a template that defines the attributes and
behavior of the objects it describes.
 The values of an object’s variables describe the
object’s attributes.
 The “variables” mentioned here refer to the instance
variables defined in the object’s class.
 The collection of attributes ( values stored in the
instance variables) define an object’s state of being.
 The methods that can be invoked using the object
define the object’s behaviors.
 Here “methods” refers to the methods defined in the
object’s class.
2
Another Way to Say It
 Classes are what you create when you write
a program.
 Objects are created in the computer during
program execution.


When your program is executing, your code
instantiates objects whose state is stored in
the object’s instance variables and whose
behavior (changes in the object’s state) is
manipulated using the object’s methods.
The only state and behavior an object can
exhibit is that defined in the classes you wrote.
3
What Happens in Memory
During Program Execution
 Each object has its own state.
 Each object has memory allocated for its own instance
variables.
 All objects of a particular class can access the same
methods to control their behavior.

The code for a class’s methods exists for the use of
any object from the class and only needs to be stored
one place in memory.
 An object’s behavior often modifies its state.
 When an object’s method executes, it often results in
a change in the values stored in its instance
variables.
4
Objects can Represent
Items that are Intangible
 Software objects often represent tangible
items such as people, cars, etc. (things you
can touch).

But objects can represent intangibles such as
error messages, mathematical concepts and
operations, etc.
 A common mistake for programmers new to
object-oriented programming is to limit their
possibilities for creating objects to tangible
entities.
5
Classes
 An object is defined by a class
 A class represents a pattern from which an
object is created during program execution.
 In general, no space to store data values is
reserved in a class.



To allocate space to store data values during
program execution, we must instantiate one
or more objects from the class.
Each object is an instance of its class.
Each object has space in memory for its own
data, which is why each object can have its
own state.
6
Anatomy of a Class
 A class contains:
 declarations of the data (instance variables)
that will be stored in each instantiated object.
 declarations of the methods that can be
invoked during program execution by using an
instantiated object.

Methods are invoked using the commands of the
form:
objectName.methodName(arguments)
 Collectively, the data and methods are
referred to as the members of the class.
7
Demonstration Program
CountFlips.java
 CountFlips.java (page 213 of our text)


The CountFlips program simulates the
flipping of a coin 1000 times to see how often
it comes up heads or tails.
Note that the CountFlips class contains the
static method main().

Objects cannot be instantiated from this class.
Rather the main() method exists for the purpose
of controlling execution by instantiating a Coin
object and using that object to invoke methods
defined by the Coin class.
8
Demonstration Program
Coin.java
 Coin.java (page 214 of our text)
 The Coin class has two integer constants,
HEADS and TAILS and one integer variable
named face.
 The first method is a constructor that is used
in the main() method of CountFlips to
instantiate a Coin object.
 The remaining methods, flip() and isHeads()
are invoked by CountFlips’ main() method to
flip the Coin object and to check if heads or
tails comes up.
9
Instance Data
 In the Coin class, the constants HEADS and TAILS
and the variable face are declared inside the class,
but not inside any method.



By being declared at the class level (not within any
method), these variables and constants can be
referenced by any method of the class.
The location at which a variable is declared defines its
scope, which is the area within a program in which that
variable can be referenced.
Variables declared with class scope are called
instance data because memory space is created for
each instance (instantiated object) of the class.
10
String Concatenation and Objects
 When an object reference variable is used as
an operand of the string concatenation
operator (+), that object’s toString() method
is automatically called to get a string
representation of the object.

If no toString() method is defined for a
particular class, a default version is called that
returns a string that contains the name of the
class, together with other information.

It is usually a good idea to define a specific
toString() method for a class.
11
More on Class Level Variables
 Java automatically initializes any variables
declared at the class level.


Numeric variables such as int and double are
initialized to zero.
However, it is good practice to initialize
variables explicitly (usually by using a
constructor) so that anyone reading the code
will clearly understand the intent.
12
Demonstration Program
FlipRace.java
 FlipRace.java (on pages 217 & 218 of our
text)

FlipRace instantiates two Coin objects.



Note that the flip() method is called with coin1
and then again with coin2.
Even though the same method is called both
times, it is only the Coin object that invoked the
method that is flipped.
Also, note that the instance variables count1
and count2 are explicitly initialized to zero,
even though that would have happened
automatically.
13
Unified Modeling Language (UML)
Diagrams
 UML Class Diagrams
 Consist of one or more classes, each with three
sections for the class name, attributes, and
methods.
 Example using the FlipRace program:
Coin
FlipRace
1
2
face : int
main(args : String[]) : void
flip() : void
isHeads() : boolean
toString() : String
14
Explanation of UML Class Diagram
 This is a diagram representing two classes.


A FlipRace class with no attributes and just a main()
method.
A Coin class with an integer attribute named face and three
methods, flip(), isHeads(), and toString(), each of which
takes no arguments. flip() returns void, isHeads() returns a
boolean value, and toString() returns a String.
Coin
FlipRace
1
2
face : int
main(args : String[]) : void
The line between the class diagrams
represents an association that shows
FlipRace associated with two Coins.
flip() : void
isHeads() : boolean
toString() : String
15
Comments about UML Diagrams
 UML diagrams always show the type of an
attribute (instance variable) and the
parameters and return value (after a colon)
of the methods.


UML is intended to be language independent,
thus its syntax is not specific to Java or any
other language.
UML is the world’s most popular notation for
the design of object-oriented software.
16
UML Object Diagrams
 A UML object diagram consists of one or
more instantiated objects.


A snapshot of the objects at a given point in
the executing programs.
Example using the FlipRace Program:
coin1 : Coin
face = 0
coin2 : Coin
face = 1
17
Additional Comments about UML
Diagrams
 UML notation is primarily a language-independent
mechanism for visualizing and capturing the design
of a program before it is written.


It not intended to describe a program after it is written.
A common mistake of beginning (and some
experienced) programmers is to start writing their
programs before doing the design work.
 We will see additional UML notations that will help us
visualize the classes and objects we use in our
programs.
18
Two Ways of
Thinking about Objects
 When designing and implementing an object:
 We need to think about how the object works.

Define the variables that will be held in the object
and write the methods that make the object
useful.
 When designing a solution to a larger
problem that involves multiple objects:

We need to think about how objects in the
program will interact.

At this level, we only need to think about the
services an object provides, not about the details
of how those services are provided.
19
Focusing on the Larger Picture
 Objects provide a level of abstraction that
allows us to focus on the larger picture when
we need to.

For the abstract to work, we must maintain the
boundaries between objects.


Only the methods within an object should have
direct access to the variables in that object.
We should make it difficult for the code outside
of a class to change the value of a variable
that is declared inside the class.
20
Encapsulation and Visibility
Modifiers
 An object should be encapsulated from the
rest of the system.


An object should interact with other parts of
the program (methods in other objects) only
through the specific set of methods which
define the services the object provides.
An object’s methods define the interface
between that object and the program that uses
it (i.e., the methods in other objects within the
program).
21
Accomplishing Encapsulation
 We accomplish object encapsulation using
modifiers.


Modifiers are reserved words in Java.
Various modifiers will be discussed at
appropriate points in this course.

All modifiers are summarized in Appendix E.
 Some Java modifiers are called visibility
modifiers because they control access to the
members of a class.
22
Visibility Modifiers
 The reserved words public and private are
visibility modifiers that can be applied to the
variables and methods of a class.

If a member (variable or method) of a class
has:

public visibility
 It can be directly referenced from outside of the
object.

private visibility
 It can be accessed only by the methods of the class.
 This makes the objects created from that class selfgoverning.
23
Public Variables
Violate Encapsulation
 Instance data should be defined with private
visibility.


So while it can still be used anywhere inside
the class definition, it cannot be referenced
externally.
If instance variables were declared public,
they could be directly modified by other
object’s (and possibly other program’s)
methods during program execution.

Which would make it much easier for hackers to
obtain information
24
Methods are Usually Public
 An object that needs to use the services of
another object is called a client of the other
object.
 Methods that provide services to the clients of
a class must be declared with public visibility
so they can be invoked by the client.

The only purpose of a private method is to
help the other methods of the class do their
job.

Such methods are referred to as support
methods.
25
Effects of
Public and Private Visibility
public
private
Variables
Violate
Encapsulation
Enforce
Encapsulation
Methods
Provide Services
to Clients
Support other
methods in the
class
26
Public Visibility for Constants
 Giving constants public visibility is generally
considered acceptable.


Because constants cannot be changed, they
do not need to be protected from malicious
attempts to change them.
However, sometimes constants need to be
protected from viewing. If so, they can be
declared as private.
27
Additional Comments Regarding
Visibility
 A client can still access or modify private data by
invoking public service methods.


A class must provide public service methods for valid
client operations.
The code for those methods must be carefully
designed to permit only appropriate access and valid
changes.
 UML diagrams reflect the visibility of a class member
with special notations.


A member with public visibility is preceded by a plus
sign ( + ).
A member with private visibility is preceded by a
minus sign ( - ).
28
Higbee’s Program 3B
UML Class Diagram
Program3B
QuizAverage
1
+main(args : String[]) : void
*
- fullName : String
- quizAvg : double
+ getData( ) : void
- getMessage( ) : String
+ toString( ) : String
Indicates a one to many
association in which
multiple quiz averages
can be calculated and
displayed.
29
Methods
 A class contains:


declarations of the data (instance variables)
that will be stored in each instantiated object.
declarations of the methods that can be
invoked during program execution by using an
instantiated object.

Methods are invoked using the commands of the
form:
objectName.methodName(arguments)
30
Anatomy of a Method
 The header of a method includes:
 the type of the return value
 the method name, possibly preceded by
some modifiers (public, static, etc.)
 within parentheses, a list of parameters that
the method accepts when it is invoked
 The statements that make up the body of
the method are defined in a block delimited
by braces.
 See the syntax diagram on page 224.
31
Invoking a Method
 The method declaration consists of the method
header and the method body.

The declaration is written within the method’s class.
 A method invocation is the code that causes the
method to execute.

The invocation may be in the same object as the
method or in a different object belonging to a different
class.
 If an invoked method is part of the same object, only
the method name and parentheses enclosing any
actual parameters need be used to invoke it.
 If an invoked method is part of an object from a
different class, that object’s name must be used
followed by a period, the method name, and
parentheses enclosing any actual parameters.
32
Flow of Control Following
Method Invocations
 Figure 4.8 shows how a method is invoked from an
object in the same class doThis() invoking helpMe()
and from a different class main() invoking doThis().
a class
main( )
obj.doThis()
object from another class
doThis( )
helpMe( )
helpMe()
33
Driver Programs and main( )
 The class that contains the main( ) method is
usually a driver class (often referred to as a
driver program).



No object is ever instantiated from the class.
All the class does is instantiate and use
objects from other classes created as part of
the overall application.
Sometimes driver classes are created for the
sole purpose of testing the other classes.
34
The Banking Demo Program
 Banking.java on page 226.


An example of a driver program.
Calls the constructor and other methods in
the Account class.
 Account.java on pages 227-229.


Methods of the Account class perform
various services on a bank account.
Contains examples of “get” methods and a
toString() method.
35
UML Object Diagrams
for Objects of the Banking Program
acct1 : Account
acct2 : Account
- name = “Ted Murphy”
- acctNumber = 72354
- balance = 102.56
- name = “Jane Smith”
- acctNumber = 69713
- balance = 40.00
acct3 : Account
- name = “Edward Dempsey”
- acctNumber = 93757
- balance = 759.32
36
The Return Statement
 The return type specified in the method header can
be a primitive type, a class name, or the reserved
word void.



The return statement consists of the reserved word
return followed by an expression whose value is
returned.
The data type of the expression must be consistent
with the return type in the method header.
 For example, you can’t return an integer if the return
type shown in the method header is String.
A method that returns a value must have a return
statement.
37
When can you Not have a
Return Statement
 A method that does not return a value does
not usually have a return statement.



Such a method will have void as its return
type in the method header.
However, a method with a void return type
may contain a return statement with no
expression.
A method with no return statement will
automatically return when the end of the
method is reached.
38
Other Considerations Regarding
return
 Constructors do not have a return type at all.

Not even void.
 A return value can be ignored.

Often the value returned to a calling statement
is used in some way.


Such as being assigned to a variable or used in a
control statement (e.g., when a boolean value is
returned).
In other cases, the value returned is simply
ignored.
39
Parameters
 The parameter list in a method header
specifies the types of the values that are
passed and the names by which the called
method will refer to those values.
 The names of the parameters in the header
of a method declaration are called formal
parameters.
 The values passed into a method when it is
invoked are called actual parameters.
40
Invoking versus Declaring
 Beginning students are often unclear about the
difference between a method declaration and a
method invocation.


A method declaration is located in the class to which
the method belongs and consists of the method
header and method body. This is “the method”.
A method invocation is often in a different class and
consists of the method name (often preceded by an
object name and a period) and parentheses
enclosing the actual parameters.
 The method invocation does not include a method
body.
41
Formal and Actual Parameters
 Formal parameters are identifiers that serve as
variables inside the method



The values in the formal parameters come from the
actual parameters in the method invocation.
If an expression is used as an actual parameter, it is
fully evaluated before the method is called and the
result is passed to the formal parameter.
The formal parameters and actual parameters must
correspond in number and type including the order
they occur in their respective parentheses.
42
Constructors
 When we define a class, we usually define
one or more constructors to help us set up
an object when one is instantiated during
program execution.
 A constructor differs from a regular method in
two ways:


A constructor has the same name as the
class.
A constructor does not have a return type
specified in the method header (not even
void).
43
Using Constructors
 A constructor is generally used to initialize the
instance variables of a newly instantiated
object.
 We don’t have to define a constructor for
every class.

Each class has a default constructor that takes
no parameters and is used if we don’t provide
an explicit constructor.

This default constructor generally has no effect on
the newly created object.
44
Local Data versus Instance Data
 A variable that is declared inside a method is
called a local variable.

As opposed to an instance variable that is
also declared inside a class, but not in one of
the methods.
 The scope of a local variable is limited to
the method in which it is declared.

As opposed to an instance variable whose
scope is the entire class (any method in the
class can refer to it).
45
Local Variables
 It is possible to declare a local variable
inside a method using the same name as an
instance variable declared at the class level.

When you use the name inside that method, it
refers to the local variable.


Anywhere in the class outside that method, the
name would refer to the instance variable.
Our text suggests that you avoid doing this.
46
Formal Parameter Names
 The formal parameter names in a method
header serve as local variables.

They don’t exist until the method is called,
and they cease to exist when the method is
exited.
47
Method Overloading
 In object-oriented languages, you can use the
same method name with different
parameter lists for multiple methods within
the same class.


This technique is called method overloading.
Method overloading is useful when you need
to perform similar methods on different types
of data.

For example, you may want to compute the
weekly pay for both salaried and hourly
employees.
48
Method Signatures
 A method’s name along with the number,
type, and order of its parameters is called
the method’s signature.


Two methods can have the same name (can
be overloaded) as long as something else is
different in the signature.
Note that the return type of a method is not
part of the method signature.

Two overloaded methods cannot differ only by
their return type.
49
Constructors and Overloading
 Constructors are a primary candidate for
overloading.

Recall that constructors always have the same
name as the class.


Thus the only way a class can have multiple
constructors is to overload them.
By providing multiple versions of a constructor,
we can provide several ways to initialize an
object’s instance variables.
50
Example of Method Overloading
The Die Class
 SnakeEyes.java on page 236.

Provides the main() method. Calls two
different overloaded Die() constructors.
 Die.java on pages 236-237.

Contains the overloaded constructors:


public Die( )
public Die (int faces)
51
Method Decomposition
 Whenever a method becomes large or
complex, we should consider decomposing
it into multiple methods to create a more
understandable class design.
 First, however, we must consider how other
classes and objects can be defined to create
better overall system design.

In object-oriented design, method
decomposition must be subordinate to
object decomposition.
52
UML Class Diagram for the
PigLatin Program
PigLatin
1 1
PigLatinTranslator
+ main (args: String[]) : void
 Note: translating an entire
sentence into Pig Latin in one
big method would be long and
difficult. Instead the translate()
method is decomposed into
several support methods.
+ translate (sentence : String) : String
- translateWord (word : String) : String
- beginsWithVowel (word : String) : boolean
- beginsWtihBlend (word : String) : boolean
53
The PigLatin Program
 PigLatin.java on page 238.
 Contains main(). Assumes the input will have
no punctuation.
 PigLatinTranslator.java on pages 240-241.
 The translateWord(), beginsWithVowel(),
and beginsWithBlend() methods are
declared with private visibility and exist only to
help the public translate() method, which is
the only true service method in the PigLatin
class. The task of translating a sentence has
been decomposed into these four methods.
54
Object Relationships
 Classes, and their associated objects, can
have particular types of relationships to each
other.



Now, we consider the idea of a general
association.
We then extend that concept to include
associations between objects of the same
class.
Then we will explore aggregation, in which
one object is composed of other objects.
55
Association
 Two classes have a general association when those
classes are “aware” of each other.


Objects of those classes may use each other for the
specific services that each provides.
This is referred to as a use relationship.
 An association relationship is intended to be very
general and therefore very versatile.

Associations will be introduced throughout the course.
 One association could be an Author object writes a
Book object.
56
UML Diagram for an Association
 The association connections between two
classes in a UML diagram can be annotated
with comments referred to as adornments.
 Associations can have a multiplicity
associated with them.


On a UML diagram, the asterisk can be used
to indicate a general zero-or-more value.
An appropriate range of values could be given,
such as 1...5.
57
Association between Objects of the
Same Class
 Some associations occur between two
objects in the same class.

A method of one object takes as a parameter
another object of the same class.

The operation performed often involves the
internal data of both objects. An example is:
 str3 = str1.concat(str2);

The Rational.java and
RationalNumber.java programs on
pages 243-249 provide other examples.
58
Aggregation
 Some objects are made up of other objects.
 A car is an aggregation – it is composed of
other objects (a car has a chassis).
 Aggregation can be described as a has-a
relationship.
 An aggregate object is any object that
contains references to other objects as
instance data.
 The StudentBody.java , Student.java and
Address. java programs on pages 250 - 253
demonstrate an aggregation.
59
The UML Representation of an
Aggregation
2
StudentBody
1
Student
- firstName : String
- lastName : String
- homeAddress : Address
- schoolAddress : Address
+ main (args : String[ ]) : void
Address
- streeAddress : String
- city : String
- state : String
- zipCode : Long
+ toString( ) : String
60