Object-Oriented Programming

Object-Oriented Programming
with Visual Basic .NET
J.P. Hamilton
Beijing • Cambridge • Farnham • Köln • Paris • Sebastopol • Taipei • Tokyo

Introduction
To understand the world of object-oriented programming, look at the world around
you for a moment. You might see vacuum cleaners, coffee makers, ceiling fans, and a
host of other objects. Everywhere you look, objects surround you.
Some of these objects, such as cameras, operate independently. Some, such as telephones
and answering machines, interact with one another. Some objects contain
data that persists between uses, like the address book in a cell phone. Some objects
contain other objects, like an icemaker inside of the freezer.
Many objects are similar in function but different in purpose. Bathtubs and kitchen
sinks, for example, both provide water and are used for cleaning. But it is a rare occasion
when you will take a bath in the kitchen sink or wash your dishes in the tub.
However, the bathtub and the kitchen sink in your house probably share the same
plumbing. Certainly, they share a common interface: hot and cold water knobs, a
faucet, and a drain.
When you think about it, what is the difference between a sink and a bathtub? The
location? The size of the basin? Their heights off the ground? How many more similarities
are there than differences?
Sometimes the same action causes an object to do different things depending on the
context of the situation. When you press Play on the remote, the DVD might play a
movie on the television. But if a CD is in the player, it plays music out of the speakers.
Same button, same action—different results. When you flip the switch on the
back porch, the light comes on. But the switch in the kitchen turns on the garbage
disposal. You use the same kind of switch, but obtain different results.
You can think about many objects around you in terms of black boxes. You comprehend
the fundamentals of these objects and possess a basic understanding of what
makes them work, but the specifics of their operation are unknown to you. And you
like it that way. Do you really want to have to know the inner mechanisms of every
object in your house in order to use it?

Consider that light bulb on the back porch. The filament in the bulb is nothing more
than a simple resistor. When the 100-watt bulb is “on,” the filament’s temperature is
about 2550 degrees Celsius. The resulting thermal radiation, which is proportional
to the length of the filament (but not the diameter), produces about 1750 lumens
worth of visible light at a wavelength of about 555 nanometers. And by the way, the
filament is made out of tungsten.
Do you really want to know these minute details, or do you just want the light to
come on when you flick the switch?
Any object has two inherent properties: state and behavior. The light bulb on the
back porch has state. It can be on or off. It has a brand name and a life expectancy. It
has been in use for a certain number of hours. It has a specified number of hours left
before the irregular evaporation of its tungsten filament causes it to burn out. Behaviorally,
it provides light; it shines.
But an object is rarely an island unto itself.
Many objects participate collectively in a system. The television and surrounding
sound speakers are a part of a system called a home theater. The refrigerator and
oven belong to a system called a kitchen. These systems, in turn, are a part of a larger
system that is called an apartment. Collections of apartments make up a system
known as a complex. Apartments and houses belong to neighborhoods and so on, ad
infinitum.
In essence, this book discusses systems. Building and designing objects is one aspect
of this process of building a system. Determining how these objects interact with one
another is another. Understanding both phases of development is crucial when
building any system that has more than a modicum of complexity.
Generally, you can think of this process of developing a system as object-oriented
programming and object-oriented design. Specifically, though, you are really working
toward an understanding of the objects you build and the system in which they
participate. Component-based programming forms the basis of this system.
Programming objects in software doesn’t require an object-oriented language, and
just because you use an object-oriented programming language doesn’t mean that
your code is object-oriented. Languages can only assist the process; they can’t make
any guarantees. The ability to write object-oriented software was always available
with VB. Writing it just hasn’t always been easy because the language wasn’t always
oriented in that direction. Developing binary reusable components in VB has been
possible for some time now, but using these components across languages used to be
considered somewhat of a black art—until now.
Today, Visual Basic .NET is a cutting-edge, object-oriented language that runs inside
of a state-of-the-art environment. It is feature-rich and designed to take advantage of
the latest developments in object-oriented programming. Writing software and
building components has never been easier.
Visual Basic .NET and Object-Oriented
Programming
Visual Basic .NET is a fully object-oriented programming language, which means it
supports the four basic tenets of object-oriented programming: abstraction, encapsulation,
inheritance, and polymorphism.
We have already conceptualized many of these object-oriented concepts by just looking
at the objects that surround us in our everyday lives. Let’s look more closely at
these terms and see what they actually mean and what they do for developers of
object-oriented software.
Abstraction
A radio has a tuner, an antenna, a volume control, and an on/off switch. To use it,
you don’t need to know that the antenna captures radio frequency signals, converts
them to electrical signals, and then boosts their strength via a high-frequency amplification
circuit. Nor do you need to know how the resulting current is filtered,
boosted, and finally converted into sound. You merely turn on the radio, tune in the
desired station, and listen. The intrinsic details are invisible. This feature is great
because now everyone can use a radio, not just people with technical know-how.
Hiring a consultant to come to your home every time you wanted to listen to the
radio would become awfully expensive. In other words, you can say that the radio is
an object that was designed to hide its complexity.
If you write a piece of software to track payroll information, you would probably
want to create an Employee object. People come in all shapes, sizes, and colors. They
have different backgrounds, enjoy different hobbies, and have a multitude of beliefs.
But perhaps, in terms of the payroll application, an employee is just a name, a rank,
and a serial number, while the other qualities are not relevant to the application.
Determining what something is, in terms of software, is abstraction.
In object-oriented software, complexity is managed by using abstraction. Abstraction
is a process that involves identifying the crucial behavior of an object and eliminating
irrelevant and tedious details. A well thought-out abstraction is usually
simple, slanted toward the perspective of the user (the developer using your objects),
and has probably gone through several iterations. Rarely is the initial attempt at an
abstraction the best choice.
Remember that the abstraction process is context sensitive. In an application that
will play music, the radio abstraction will be completely different from the radio
abstraction in a program designed to teach basic electronics. The internal details of
the latter would be much more important than the former.
Encapsulation
Programming languages like C and Pascal can both produce object-like constructs.
In C, this feature is called a struct; in Pascal, it is referred to as a record. Both are
user-defined data types. In both languages, a function can operate on more than one
data type. The inverse is also true: more than one function can operate on a single
data type. The data is fully exposed and vulnerable to the whims of anyone who has
an instance of the type because these languages do not explicitly tie together data
and the functions that operate on that data.
In contrast, object-oriented programming is based on encapsulation. When an
object’s state and behavior are kept together, they are encapsulated. That is, the data
that represents the state of the object and the methods (Functions and Subs) that
manipulate that data are stored together as a cohesive unit.
Encapsulation is often referred to as information hiding. But although the two terms
are often used interchangeably, information hiding is really the result of encapsulation,
not a synonym for it. They are distinct concepts. Encapsulation makes it possible
to separate an object’s implementation from its behavior—to restrict access to its
internal data. This restriction allows certain details of an object’s behavior to be hidden.
It allows us to create a “black box” and protects an object’s internal state from
corruption by its clients.
Encapsulation is also frequently confused with abstraction. Though the two concepts
are closely related, they represent different ideas. Abstraction is a process. It is
the act of identifying the relevant qualities and behaviors an object should possess.
Encapsulation is the mechanism by which the abstraction is implemented. It is the
result. The radio, for instance, is an object that encapsulates many technologies that
might not be understood clearly by most people who benefit from it.
In Visual Basic .NET, the construct used to define an abstraction is called a class.
The terms class and object are often used interchangeably, but an object is actually an
instance of a class. A component is a collection of one or more object definitions, like
a class library in a DLL.
Inheritance
Inheritance is the ability to define a new class that inherits the behaviors (and code)
of an existing class. The new class is called a child or derived class, while the original
class is often referred to as the parent or base class.
Inheritance is used to express “is-a” or “kind-of” relationships. A car is a vehicle. A
boat is a vehicle. A submarine is a vehicle. In OOP, the Vehicle base class would provide
the common behaviors of all types of vehicles and perhaps delineate behaviors
all vehicles must support. The particular subclasses (i.e., derived classes) of vehicles
would implement behaviors specific to that type of vehicle. The main concepts
behind inheritance are extensibility and code reuse.
relationship
is created by using composition. Composition, which is sometimes referred
to as aggregation, means that one object contains another object, rather than inheriting
an object’s attributes and behaviors. Naturally, a car has an engine, but it is not a
kind of engine.
C++ supports a type of reuse called multiple inheritance. In this scenario, one class
inherits from more than one base class. But many C++ programmers will tell you
that using multiple inheritance can be tricky. Base classes with identical function
names or common base classes can create nightmares for even the most experienced
programmers.
VB.NET, like Java, avoids this problem altogether by providing support only for single
inheritance. But don’t worry, you aren’t missing out on anything. Situations that
seem ideal for multiple inheritance can usually be solved with composition or by
rethinking the design.
When it comes to proper object-oriented design, a deep understanding of inheritance
and its effects is crucial. Deriving new classes from existing classes is not always
as straightforward as it might initially appear. Is a circle a kind of ellipse? Is a square
a kind of rectangle? Mistakes in an inheritance hierarchy can cripple an object model.
Polymorphism
Polymorphism refers to the ability to assume different forms. In OOP, it indicates a
language’s ability to handle objects differently based on their runtime type.
When objects communicate with one another, we say that they send and receive messages.
The advantage of polymorphism is that the sender of a message doesn’t need
to know which class the receiver is a member of. It can be any arbitrary class. The
sending object only needs to be aware that the receiving object can perform a particular
behavior.
A classic example of polymorphism can be demonstrated with geometric shapes.
Suppose we have a Triangle, a Square, and a Circle. Each class is a Shape and each
has a method named Draw that is responsible for rendering the Shape to the screen.
With polymorphism, you can write a method that takes a Shape object or an array of
Shape objects as a parameter (as opposed to a specific kind of Shape). We can pass
Triangles, Circles, and Squares to these methods without any problems, because
referring to a class through its parent is perfectly legal. In this instance, the receiver is
only aware that it is getting a Shape that has a method named Draw, but it is ignorant
of the specific kind of Shape. If the Shape were a Triangle, then Triangle’s version of
Draw would be called. If it were a Square, then Square’s version would be called, and
so on.
small graphics package and we need to draw several shapes on the screen at one
time. To implement this functionality, we create a class called Scene. Scene has a
method named Render that takes an array of Shape objects as a parameter. We can
now create an array of different kinds of shapes and pass it to the Render method.
Render can iterate through the array and call Draw for each element of the array, and
the appropriate version of Draw will be called. Render has no idea what specific kind
of Shape it is dealing with.
The big advantage to this implementation of the Scene class and its Render method is
that two months from now, when you want to add an Ellipse class to your graphics
package, you don’t have to touch one line of code in the Scene class. The Render
method can draw an Ellipse just like any other Shape because it deals with them
generically. In this way, the Shape and Scene classes are loosely coupled, which is
something you should strive for in a good object-oriented design.
This type of polymorphism is called parametricpolymorphism , or generics. Another
type of polymorphism is called overloading. Overloading occurs when an object has
two or more behaviors that have the same name. The methods are distinguished only
by the messages they receive (that is, by the parameters of the method).
Polymorphism is a very powerful concept that allows the design of amazingly flexible
applications. Chapter 4 discusses polymorphism in more depth.
The .NET Framework
The objects you construct with VB.NET will live out their lives within the .NET
Framework, which is a platform used to develop applications. The platform was
designed from the ground up by using open standards and protocols like XML,
HTTP, and SOAP. It contains a rich standard library that provides services available
to any language running under its protection.
The impetus behind its creation was the desire to develop a platform for building,
deploying, and running web-based services. In spite of this goal, the framework is
ideal for developing all types of applications, regardless of the design. The .NET
Framework makes child’s play of some of programming’s most sophisticated concepts,
giving you the ability to take advantage of today’s cutting-edge architectures:
• Distributed computing using open Internet standards and protocols such as
HTTP, XML, and SOAP
• Enterprise services such as object pooling, messaging, security, and transactions
• An infrastructure that simplifies the development of reusable cross-language
compatible components that can be deployed over the Internet
• Simplified web development using open standards
• Full language integration that make it possible to inherit from classes, catch
exceptions, and debug across different languages.
Deployment is made simpler because settings are stored in XML-based configuration
files that reside in the application directory; there is no need to go to the registry.
Shared DLLs must have a unique hash value, locale, and version, so physical
filenames are no longer important once these considerations are met. Not having
physical filenames makes it possible to have several different versions of the same
DLL in use at the same time, which is known as side-by-side execution. All dependencies
and references are stored within the executable in a section called the
manifest. In a sense, we’re back to the days of DOS because to deploy an application,
you only need to xcopy it from one directory to another.
This book explores many aspects of .NET in order to gain a complete understanding
of the components you write and the world in which they live. Doing it any other
way is impossible. The .NET Framework provides so many services your components
will use that discussing one without referring to the other is literally impossible—
they are that closely tied together.
Two major elements of the .NET Framework will be addressed repeatedly throughout
this book. The first is the Common Language Runtime (CLR), which provides
runtime services for components running under .NET.
The second element is the .NET class library, a vast toolbox containing classes for
everything from data access, GUI design, and security to multithreading, networking,
and messaging. The library also contains definitions for all primary data types,
such as bytes, integers, and strings. All of these types are inherently derived from a
base class called System.Object, which you can think of as a “universal” data type;
there is no distinction between the types defined by the system and the types you create
by writing classes or structures. Everything is an object!
The term .NET means many things to many different people. When
the term is used in this book, it always refers to the .NET Framework—
the Common Language Runtime and the .NET class library.
In the past, passing a string from a component written in VB to one written in C++
(or vice versa) could be frustrating. Strings in VB weren’t the same as the strings in
C++. In fact, under some circumstances, using a component written in C++ from VB
was downright impossible because of issues involving data types. VB just doesn’t
know what to do with an LPSTR! Every language under .NET uses the same data
types defined in the base class library, so interoperability problems of the past are no
longer an issue.
This book touches on several major areas of the library and focuses on the development
of components using VB.NET. However, if you follow the examples, you might
be surprised at just how much you know.