FUNNY IMAGES: Designing Active Server Pages

Table of Contents
Preface ................................................................................................................... vii
1. Introduction ................................................................................................ 1
What Is Application Design? ............................................................................ 1
What’s Wrong with ASP Design? ..................................................................... 3
Why Hasn’t ASP Design Advanced? ................................................................ 4
What Can Be Done to Improve ASP Design? ................................................. 5
Further Reading .............................................................................................. 18
2. Choosing a Server-Side Scripting Language ................................. 19
The Popularity of VBScript ............................................................................ 19
Specifying the Scripting Language ................................................................ 20
Creating ASP Pages with JScript .................................................................... 25
Creating ASP Pages with PerlScript ............................................................... 31
Creating ASP Pages with Python ................................................................... 37
Further Reading .............................................................................................. 37
3. Exception Handling ............................................................................... 39
A Bit of Terminology ..................................................................................... 40
Detecting When Exceptions Occur ............................................................... 42
Responding to Exceptions ............................................................................. 65
Creating Custom HTTP Error Pages .............................................................. 73
Further Reading .............................................................................................. 77
vi Table of Contents
Oracle 8i Internal Services for Waits, Latches, Locks, and Memory, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
4. Regular Expressions, Classes, and Dynamic Evaluation and
Execution ................................................................................................... 78
Using the RegExp Object ............................................................................... 79
Using Object-Oriented Programming with VBScript .................................... 85
Using Dynamic Evaluation and Execution ................................................... 99
Further Reading ............................................................................................ 103
5. Form Reuse ............................................................................................. 104
The Importance of Code Reuse .................................................................. 104
A Primer on Form Use ................................................................................. 106
Form Validation ............................................................................................ 111
Creating Reusable Server-Side Form Validation Routines .......................... 116
Developing Reusable Form Creation Routines ........................................... 124
The Practicality of Reuse ............................................................................. 136
Further Reading ............................................................................................ 137
6. Database Reuse ..................................................................................... 138
Examining Database Usage ......................................................................... 138
The Building Blocks for Creating Reusable Administration Pages ............ 140
Creating Reusable Administration Pages .................................................... 149
Further Reading ............................................................................................ 246
7. Using Components ............................................................................... 247
COM—A Quick Overview ........................................................................... 248
Lesser-Known Microsoft COM Components ............................................... 249
Enhancing Microsoft’s COM Components .................................................. 275
Building Components .................................................................................. 276
Further Reading ............................................................................................ 290
8. Enhancing Your Web Site with Third-Party Components ..... 291
Executing DOS and Windows Applications on the Web Server
with ASPExec ............................................................................................... 292
Obtaining Detailed Information About Your Users’s Browsers ................ 296
Grabbing Information from Other Web Servers ......................................... 303
Encrypting Information ................................................................................ 309
Uploading Files from the Browser to the Web Server ............................... 324
Why Reinvent the Wheel? ............................................................................ 335
Further Reading ............................................................................................ 337
Index .................................................................................................................... 339

vii
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Preface
The other day a friend called me with some problems he was having in creating
an ASP script. This friend, who is relatively new to ASP but has been writing
Visual Basic applications for several years, was in the midst of building a large,
data-driven web site. During our conversation, he frustratedly commented that
there seemed to be a lot of monotony involved in creating ASP pages.
After I asked him to elaborate, he explained there were several pages that did relatively
similar things for his site: one set of ASP pages served as an administration
tool for the database driving his site; another set of ASP pages allowed users to
enter information into the database. Externally, these pages looked and acted differently,
but their core functionality—accessing a database table and adding, editing,
and removing entries—was identical. Despite these similarities, my friend was
finding that he created separate ASP pages for each task, even if the tasks were
related.
This friend is not alone. While the popularity and use of Active Server Pages has
grown radically over the past couple of years, the quality of the code has not. As a
consultant, author, and editor and founder of 4GuysFromRolla.com—one of the
largest online Active Server Pages resource sites—I’ve created thousands of ASP
pages over the past three years. I’ve also worked in several teams designing large
web sites using Active Server Pages and have reviewed other developers’ scripts.
When developing ASP pages, I find myself (and other developers) continually
reinventing the wheel. Take a moment to think about how many database administration
pages you have created. About how many ASP scripts have you written to
perform server-side form validation? Why is it that we put so much time into
design when developing a Visual Basic or Visual C++ application but so little time
when developing ASP pages?
viii Preface
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Simply put, this book looks at why and how Active Server Page design is lacking
and examines the steps that can be taken to improve this design process. It is an
important topic that has received very little attention in the past.
Who This Book Is For
This book is intended for intermediate to advanced Active Server Pages developers
who have solid ASP skills and are interested in learning techniques for creating
reusable and robust ASP applications. Since this book focuses on writing
maintainable, reusable code, if you work on large-scale ASP applications, especially
with teams of ASP developers, you will find this book especially helpful.
This book is also for those who, like my friend, have found creating ASP pages to
be monotonous. By spending time carefully developing Active Server Pages before
writing the actual code, you will soon find yourself producing reusable, cleaner,
and less error-prone code.
Finally, this book is also intended for all the ASP pros out there—those who are
passionate about developing Active Server Pages, those who enjoy learning new
ways to create ASP pages, and those who pride themselves on their ASP skills.
How This Book Is Organized
This book is divided into eight chapters. The first chapter serves as an introduction
to the book, describing what technologies ASP offers to aid building robust,
reusable applications. This chapter introduces some of the new features in ASP 3.0
and the VBScript scripting engine that are used throughout the book to assist in
application design.
The next three chapters expound on the technologies briefly discussed in the
introduction. Chapter 2, Choosing a Server-Side Scripting Language, discusses what
scripting languages are available for designing ASP pages. The vast majority of
books and ASP resource web sites present ASP examples in VBScript only. This
chapter discusses some of the other languages that can be used and their advantages
and disadvantages. When building an ASP application with a team of developers,
there are times when a language other than VBScript needs to be used due
to the various developers’ past experiences. For example, if only one of five developers
is fluent with VBScript, perhaps using JScript or PerlScript as the server-side
scripting language would be a better choice.
Chapter 3, Exception Handling, examines exception-handling techniques as well as
the new ASPError object introduced with ASP 3.0. Too often, when an error occurs
in an ASP page, an unreadable, ugly message is displayed. With proper exceptionhandling
techniques, these unreadable error messages can be avoided. In certain
cases, errors that usually generate error messages can be “fixed” on the fly!
Preface ix
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
In Chapter 4, Regular Expressions, Classes, and Dynamic Evaluation and Execution,
we look at using several of the new features available in the VBScript Scripting
Engine Version 5.0. This chapter introduces VBScript classes, which are similar
to classes in Visual Basic.
Chapters 5 and 6 use VBScript classes intensively to build a reusable set of classes.
Chapter 5, Form Reuse, looks at building a server-side validation class, while
Chapter 6, Database Reuse, focuses on building a generic database administrationgeneration
class. At the conclusion of these two chapters you’ll have created a
very powerful and very reusable set of classes you can use in your own ASP applications!
Furthermore, classes hide the complexity of completing a task. Therefore,
if you are on a team of ASP developers that contains some beginning developers,
your senior developers can work on building reusable classes while your juniorlevel
developers can spend their time using instances of these classes within ASP
pages.
Chapter 7, Using Components, discusses the benefits of using Microsoft-provided
COM components to enhance your web site. A discussion on building custom COM
components using both high-level languages and script is presented as well.
Chapter 8, Enhancing Your Web Site with Third-Party Components, examines how
to use preexisting COM components to add a variety of features to your site. There
are literally hundreds of third-party COM components that can help you add various
common web site features. Chapter 8 looks at some of the most widely used
third-party components.
At the end of each chapter you’ll find a section titled “Further Reading.” This section
contains a list of URLs you can visit to obtain further information on the topics
discussed in the chapter.
ASP Information on the Web
One thing that fascinates me to no end is the sheer amount of free and useful
Active Server Pages information on the Internet. Furthermore, there are literally
thousands of ASP developers from around the world who actively participate in
providing free information, answering questions, and giving feedback and encouragement.
This group of people, the ASP community, is one of the nicest and most
helpful groups of computer experts I’ve ever had the opportunity to interact with.
Some of the greatest ASP information web sites available include:
4GuysFromRolla.com
This is the ASP web site I run. The site focuses on high-quality articles for
intermediate to advanced ASP developers.
LearnASP.com
If you are one who learns best by examining source code, seeing it run, and
tinkering with it (as opposed to reading articles), then LearnASP.com is for
x Preface
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
you. With thousands of code examples ready to run, LearnASP.com is the
place to go for ASP source code.
15Seconds.com
This was one of the first ASP sites on the Internet. 15Seconds.com serves as a
gateway to ASP information, containing hundreds of links to relevant articles
on a number of advanced topics.
As mentioned earlier, if you have a specific ASP question, I guarantee there are
several folks who will take the time to help you find the answer. There are several
web sites that focus on bringing members of the ASP community together to
help answer questions and solve problems. If you have an ASP-related question
that needs answering, be sure to check out these web sites:
ASPMessageboard.com
The Internet’s most popular ASP-related message board, receiving hundreds of
questions and answers each and every day! If you’ve got a question,
ASPMessageboard.com is a great place to ask.
ASPLists.com
If you prefer ListServs over forum web sites, be sure to check out ASPLists.
com. There are several different ListServs available from ASPLists.com, from
the very high-volume, general lists (ASP Free For All) to the very specific, lowvolume
lists (Fast Code, Advanced ADSI, etc.).
Obtaining the Sample Code
All of the example Visual Basic source code from Designing Active Server Pages is
downloadable free from the O’Reilly & Associates web site at http://vb.oreilly.com.
Conventions Used in This Book
Throughout this book, I have used the following typographic conventions:
Constant width
Indicates a language construct such as a language statement, a constant, or an
expression. Interface names appear in constant width. Lines of code also
appear in constant width, as do classes and function and method prototypes.
Italic
Represents intrinsic and application-defined functions, the names of system
elements such as directories and files, and Internet resources such as web documents.
New terms are also italicized when they are first introduced.
Preface xi
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Constant width italic
Indicates replaceable parameter names in prototypes or command syntax, and
indicates variable and parameter names in body text.
Constant width boldface
Indicates user input, as well as emphasized code.
How to Contact Us
The information in this book has been tested and verified, but you may find that
features have changed (or even find mistakes!). You can send any errors you find,
as well as suggestions for future editions, to:
O’Reilly & Associates, Inc.
101 Morris Street
Sebastopol, CA 95472
(800) 998-9938 (in the U.S. or Canada)
(707) 829-0515 (international/local)
(707) 829-0104 (fax)
You can send us messages electronically. To be put on the mailing list or request a
catalog, send email to:
info@oreilly.com
To ask technical questions or comment on the book, send email to:
bookquestions@oreilly.com
There is a web site for the book, where examples, errata, and any plans for future
editions are listed. You can access this page at:
http://www.oreilly.com/catalog/designasp
For more information about this book and others, see the O’Reilly web site:
http://www.oreilly.com
For technical information on Visual Basic programming, to participate in VB discussion
forums, or to acquaint yourself with O’Reilly’s line of Visual Basic books,
you can access the O’Reilly Visual Basic web site at:
http://vb.oreilly.com
Acknowledgments
This book has been a dream of mine for some time. I’ve always considered web
page design to be an extremely important topic, especially with ASP and have
xii Preface
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
wanted to write a book on the subject for some time. That being said, I’d like to
thank my editor, Ron Petrusha, who gave me the creative freedom and encouragement
needed to write such a book. Thanks, Ron!
In January 1998, I started working with ASP with Empower Consulting, Inc., in
Kansas City, Missouri. I instantly fell in love with ASP and decided to create a small
web site, 4GuysFromRolla.com, to showcase what cool things I had done with
ASP. Who knew my little hobby site would grow into one of the largest ASP information
sites on the Net, attracting over 15,000 ASP developers each day? A hearty
thank you to all those who visit 4Guys regularly; this book is dedicated to you!
Finally, I’d like to say thanks to my great family. Our life, our decisions, our attitudes,
our beliefs, and our dreams are based largely upon our upbringing. I was
very fortunate to have such a loving and encouraging family; where I am today
and what I am today is due, in large part, to the continual support, respect, and
advice provided by my parents, brother, and relatives.
I hope you enjoy reading this book as much as I enjoyed writing it! As always,
Happy Programming!
1
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Chapter 1 1
Introduction
As the Web has become a more integral part of everyday life and business, web
sites have matured from small, static sites into rich, data-driven, complex applications.
Several technologies, including dynamic scripting technologies, like Active
Server Pages and Perl, have aided along the way, making the Web a more viable
application medium. This book focuses on using Active Server Pages to quickly
and easily build powerful and dynamic web sites.
Although the server-side tools for building web sites have experienced a nice maturation,
the processes used for building these sites have not. What, specifically, is
Active Server Page design? What is currently wrong with ASP design? Why hasn’t
the art of designing ASP pages advanced? What can be done to improve ASP
design? The first three questions will be answered in this chapter. The last one,
however, is a meaty one; it is addressed in this chapter, and answered over the
next seven chapters.
This chapter not only addresses these questions, but also introduces functions and
programming styles that will be used extensively throughout this book.
What Is Application Design?
Designing a single ASP page is trivial. The challenges involved in designing ASP
pages arise when large web sites with hundreds of web pages are being crafted.
These large web sites are, in their own rights, full-scale applications. Therefore,
before we examine Active Server Page design, we will first take a step back and
discuss application design. As we’ll see shortly, application design is a long, arduous
process, involving much more than simply punching out code.
Imagine that you have just been assigned the task of creating a new program that
will be deployed to all of the employees in your company. This program will
2 Chapter 1: Introduction
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
allow the users to query a centralized database and place the results into a number
of formats that correspond to various interoffice forms your company uses.
What do you do first? How do you get started?
Designing, coding, testing, and deploying an application is known as application
development. Temporally, it can be viewed as the time span from when the program
was first conceptualized to when its first stable version was available for use.
(Of course, application development does not end with the first, stable release of a
product. Rather, it continues as long as updates and enhancements of the given
product are being made.) An entire branch of the computer science discipline is
dedicated to studying various application-development methodologies. While there
are a number of different methodologies, they all usually share a certain number
of phases, which include:
• Design
• Coding
• Testing
• Deployment
In classical software development, developers usually follow this flow, working on
the application’s design before moving on to writing the actual code. For example,
large software companies draft very detailed plans for their software projects
before a developer writes a single line of code.
Designing for the Web
With the Web, the design stage usually focuses on the aesthetics of the web site,
such as the look and feel of each web page. This makes sense, since in the Web’s
earlier days, the vast majority of web sites were static, their usefulness and allure
directly dependent upon their HTML design. As the Web matures, though, and
web sites become more dynamic, it is important that adequate time be spent on
the design of the scripts that run a site.
Simply put, the design phase in web site development focuses too much on the
layout and HTML issues and too little on the programmatic side. That’s not to say
that HTML design is unimportant: it is very important. The end users benefit
directly from the HTML design. Imagine a site that had terrible HTML design, a site
cumbersome to navigate and difficult to use. Chances are, after one visit to this
site, you’d not come back.
Script design, on the other hand, benefits the developers directly, and the end
users indirectly. For example, if your dynamic web site contains robust, easy-toupdate
scripts, the developers will directly benefit from reduced time spent in the
What’s Wrong with ASP Design? 3
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
coding phase when a new feature needs to be added. The end users will indirectly
benefit from such well-designed scripts, since the less code a developer has
to write, the less buggy her code will be. Additionally, well-designed scripts
require less maintenance time, making it easier to add new functionality to existing
scripts. (The benefits of code reuse and robust scripts are discussed in more
detail in the next section, “What’s Wrong with ASP Design?”) Since both HTML
design and script design are important, it is essential that adequate time be spent
on both.
Since ASP pages are scripts, ASP design is the art of crafting robust, reusable code.
Good design requires planning, documentation, and above all, patience. Before
you begin writing code for a particular ASP page, think of what the script needs to
accomplish. Is it similar to the functionality of other ASP pages on the site? Have
you created a page with similar functionality in a past project? How can you build
this page so that it can be easily reused in future projects?
What’s Wrong with ASP Design?
In my consulting experience, I have found that many developers don’t spend any
time working on the design of their scripts. Rather, they just start coding once they
have an understanding of what the web site needs to look like and what tasks it
needs to accomplish.
Such an approach is inefficient and error-prone. If no thought is dedicated to
determining the design of the ASP scripts before they are actually written, then for
each similar script, the developer essentially reinvents the wheel. For example,
imagine the developer has four ASP scripts that need to make modifications to a
database. While these scripts may not be identical, assume they all have many
functional similarities. Why should the developer spend the time to create each
page separately, when one generic page would do?
There is a direct correlation between the amount of code you write and the number
of bugs in your program. Typos and silly mistakes catch us all, and they occur
proportionally to the amount of raw source code actually written. If we can reduce
the number of total ASP scripts that need to be written by generalizing certain
scripts to handle the functionality present in the previous four, we will create less
buggy ASP applications.
Furthermore, intelligent ASP script design will not only save time and bugs in our
current project, but also in future projects. Once a robust module is written to handle
a specific task, if that task needs to be completed in a future project, we need
only reuse our existing code! With hasty design techniques, code is often written
without looking toward the future.
4 Chapter 1: Introduction
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Why Hasn’t ASP Design Advanced?
Despite the advantages of a lengthy ASP design stage, ASP pages are still typically
designed hastily. Rigorous ASP design hasn’t gained wide acceptance due to three
reasons:
• Businesses are operating on Internet time. Due to the fact that businesses must
operate at Internet time to stay competitive, many of the lengthier software
development phases that are enjoyed in classical software development are
rushed through or sidestepped completely for web applications.
• Active Server Pages are created using scripting languages. Scripting languages
are meant to solve small, discrete problems. Rarely does a developer consider
using a scripting language to tackle a formidable programming challenge.
Therefore, when coding with a scripting language, it may seem and feel unjustifiable
to spend any significant amount of time working on the design of the
script.
• Active Server Pages are easy to use, and can build powerful, dynamic web
sites quickly. This, of course, is an advantage of using ASP to develop a web
site, but it is a bit of a double-edged sword. Due to developers’ expectations
concerning the ease of development and quick time frame for creating an ASP
web site, a lengthy design process may seem out of place.
You may have noted a slight contradiction in the last couple of
pages. In “What’s Wrong with ASP Design?” I mention that ASP
scripts can be used to build large, data-driven web sites, yet in the
second bulleted item above, I state that rarely will a scripting language
be used for a large project.
For large web projects, compiled components should be created in
a high-level programming language like Visual C++, Java, or Visual
Basic, and used within ASP pages. In Chapter 7, Using Components,
we’ll look at combining ASP development with compiled
components.
While ASP design has been lacking, it is not because ASP makes such design difficult.
In fact, quite the opposite is true; several techniques can be used to create
robust, reusable ASP scripts. In the next section, “What Can Be Done to Improve
ASP Design?” and throughout the rest of this book, we’ll be looking at the tools
and methodologies to accomplish this.
What Can Be Done to Improve ASP Design? 5
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
What Can Be Done to Improve
ASP Design?
Even though ASP scripts are, obviously, scripts, there are several approaches that
can be taken to modularize your source code and encapsulate complex tasks.
These include using server-side includes, taking advantage of VBScript classes, and
making calls to the Server.Execute and Server.Transfer methods.
Server-Side Includes
One of the most common approaches to creating modularized code is to use
server-side includes (SSI). Server-side includes are used to import ASP code into an
ASP page. The benefit of being able to import code from one ASP page to another
is you can create individual ASP pages with common functions or classes, and
have these functions and class definitions imported into the pages that require
their use. Instead of having to cut and paste a particular function that is needed in
several ASP web pages, you can place that function in a single file, and then use a
server-side include to import the function definition into each ASP page that needs
to make use of that particular function.
For example, imagine that you run an e-commerce site. Conceivably, there are a
number of ASP pages in which you need to compute the sales tax. Rather than
hardcode a sales tax percentage in each of these pages, you could create a single
function—ComputeTotalWithSalesTax—that would accept the total less the sales
tax as a parameter, returning the new total including sales tax. Example 1-1 contains
an example of the ComputeTotalWithSalesTax function.
The above snippet of code places business logic—determining the
sales tax—within an ASP page. As we’ll discuss in Chapter 7, business
logic should be placed in custom components. The above code
snippet only serves to show an example of using server-side
includes.
If this function existed on each page that needed to calculate the sales tax, imagine
what would happen if the sales tax percentage changed. You would have to
Example 1-1. Determining the Sales Tax
Function ComputeTotalWithSalesTax(curTotalLessSalesTax)
Const curSalesTax = 0.0695
ComputeTotalWithSalesTax = curTotalLessSalesTax + _
curTotalLessSalesTax * curSalesTax
End Function
6 Chapter 1: Introduction
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
poke through every ASP page, checking to see if it referenced a hardcoded sales
tax value, and if it did, make the appropriate change. Of course, if you missed a
page, you’d be in for a headache when certain users were being charged less tax
than others.
Needless to say, such an approach is error-prone; furthermore, the developer who
chooses this approach is ulcer-prone. A wiser decision would be to place a single
instance of the ComputeTotalWithSalesTax function into a file, say /CODEREUSE/
ComputeSalesTax.asp, and then use server-side includes in each ASP page that
needs to reference this function.
When using a server-side include to import the contents of one file to another, the
text from the included file is just dumped straight into the file that issues the
server-side include. Using a server-side include is functionally identical to copying
the entire contents of the included file and pasting them into the file that initiated
the server-side include. Therefore, if you have Option Explicit declared in the
file that issues the server-side include (which you always should), you will need to
have every variable you use in the included file declared in the included file. Also,
the code in the file to be included should be placed within an ASP code block,
using either the <% and %> delimiters or the <SCRIPT RUNAT=SERVER
LANGUAGE="VBSCRIPT"> and </SCRIPT> delimiters.
ASP pages are not the only type of file that can execute server-side
includes. Later in this chapter, in the section “File types that can perform
server-side includes,” we’ll look at using server-side includes in
non-ASP pages.
To revisit our sales tax example, if we created a file to hold the sales tax computing
function and named that file /CODEREUSE/ComputeSalesTax.asp, we would
want to enter the following code into that file:
<%
Function ComputeTotalWithSalesTax(curTotalLessSalesTax)
Const curSalesTax = 0.0695
ComputeTotalWithSalesTax = curTotalLessSalesTax + _
curTotalLessSalesTax * curSalesTax
End Function
%>
Note that the above code is nearly identical to the code in Example 1-1. The only
difference is the code snippet above contains the <% and %> delimiters around the
ASP code. Remember, this is needed to be properly included.
Now, for each page that needs to access this function, we need to add a single
line of code. The following is an example ASP page that uses a server-side include
to make the ComputeTotalWithSalesTax function accessible:
What Can Be Done to Improve ASP Design? 7
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
<% @LANGUAGE="VBScript" %>
<% Option Explicit %>

<%
Dim curTotal
curTotal = 46.72
'Output the total with sales tax
Response.Write "Cost Before Sales Tax: " & FormatCurrency(curTotal, 2)
Response.Write "<BR>Cost After Sales Tax: " & _
FormatCurrency(ComputeTotalWithSalesTax(curTotal), 2)
%>
The server-side include has two forms:

and:

If you use the file keyword, then filename is relative to the directory in which
the ASP page that issues the server-side include resides. If you use the virtual
keyword, filename is relative to the web server’s root directory. For example, if
from an ASP page in the /MyASPScripts directory you had:

then the file ComputeSalesTax.asp would need to exist in the /MyASPScripts directory;
if it did not, an error would result when attempting to use the server-side
include.
Remember that when using the virtual keyword, you need to specify filename
relative to the Web’s root directory, regardless of the directory that the ASP page
that issues the server-side include exists in. Therefore, if in an ASP page you used:

the file ComputeSalesTax.asp would need to exist in the /CODEREUSE directory,
regardless of what directory the ASP page that issued the server-side include
existed in.
Use the virtual Keyword Religiously
When using a server-side include, there is no reason why you should use the
file keyword. Commonly used modules should be placed in specific directories
(such as /CODEREUSE). Using the virtual keyword relieves any concern
about being able to correctly specify the path relative to the executing ASP
page.
8 Chapter 1: Introduction
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Your include files can contain HTML formatting and/or ASP code. Also, an include
file can use server-side includes itself to further modularize the code! However, if
you try to perform a cyclic include—that is, Page1.asp includes Page2.asp, and
Page2.asp includes Page1.asp—an error message will be generated. (More verbosely,
the only limitation for including files in included files is included files can’t
include the file that included them. Be sure to check out the sidebar, “An Interesting
Lingo.”) Forcing a cyclic include to generate an error makes sense, because if a
cyclic include did not generate an error, Page1.asp and Page2.asp would recursively
include one another until the web server ran out of available memory.
Example 1-2 contains the code for Page1.asp, while Example 1-3 contains the
code for Page2.asp. If you visit Page1.asp through a browser, the error message
shown in Figure 1-1 will be generated.
File types that can perform server-side includes
ASP pages are not the only files that can perform server-side includes. Any file
type that is mapped to asp.dll or ssinc.dll can perform server-side includes.
An Interesting Lingo
Using include files within include files allows for confusing, yet cool-sounding,
statements. Imagine what people unfamiliar with server-side includes might
think if they overheard you making a comment like: “Hey, Bob, in Report.asp,
did you want the file included by the include to include stats.asp, which
includes another included file, or do you want Report.asp not to include any
included files that include included files?”
Example 1-2. Page1.asp Uses a Server-Side Include to Import the Contents of Page2.asp
<%
Response.Write "Preparing to include Page2.asp<P>"
'Include the file Page2.asp
%>

Example 1-3. Page2.asp Uses a Server-Side Include to Import the Contents of Page1.asp
<%
Response.Write "Preparing to include Page1.asp<P>"
'Include the file Page1.asp
%>

What Can Be Done to Improve ASP Design? 9
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
When a client requests a file from an IIS web server, the web server first determines
the file extension being requested by the client. Next, the web server
checks to see if that extension is mapped to a particular ISAPI DLL. If there is a
mapping between that file extension and an ISAPI DLL, the ISAPI DLL is invoked.
This is, essentially, what happens when you request an ASP page through your
web browser.
ssinc.dll is an ISAPI DLL that allows server-side directives to be executed. A serverside
include is one facet of server-side directives. See the “Further Reading” section
for an article that discusses server-side directives in more detail. In IIS 5.0,
files with the extensions .shtml, .shtm, and .stm are, by default, mapped to ssinc.
dll, and therefore can perform server-side includes. Of course, files with these
extensions cannot process ASP code, unless you explicitly map these extensions to
be processed by asp.dll.
asp.dll is capable of performing only one type of server-side directive:
server-side includes. ssinc.dll, on the other hand, can perform
all server-side directives but cannot process ASP code. For more
information on all of the server-side directives, be sure to refer to the
article “Using Server-Side Directives” listed at the end of this chapter
in the “Further Reading” section.
You can explicitly map particular file extensions to particular ISAPI DLLs. To
change a file extension mapping in Windows 2000, go to Start . Programs .
Administrative Tools . Internet Services Manager (in Windows NT, open up the
IIS MMC). A listing of the web sites on your computer should appear. Right-click
on the web site whose file extension mapping you wish to alter, and click on
Properties. A tabbed Web Site Properties dialog box will appear; select the “Home
Directory” tab, and click the Configuration button. You should now be presented
with the Application Configuration dialog box shown in Figure 1-2.
Figure 1-1. Attempting to perform a cyclic include generates an error message
10 Chapter 1: Introduction
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Note that the .asp file extension is mapped to asp.dll, while the extensions .shtml,
.shtm, and .stm are mapped to ssinc.dll. You can add, edit, and remove these
mappings. For example, you could create a new extension mapping, having all
files with the extension of .scott map to ssinc.dll. With such a mapping, files on
your web site with the .scott extension would be functionally identical to files with
extensions of .shtml, .shtm, or .stm; in short, .scott files would be able to perform
server-side directives!
Be careful when adding new application mappings. There is a bug in
IIS 5.0 that can arise when the .htm extension is mapped to ssinc.dll
and the default document is a .htm file. For more information on this
bug, be sure to read: http://support.microsoft.com/support/kb/articles/
Q246/8/06.ASP.
Dynamic server-side includes
Server-side includes are executed before any ASP code is processed. Due to this
fact, you cannot specify a dynamic filename. For example, it would be nice to be
able to do something like:
Figure 1-2. The “App Mappings” tab of the Application Configuration dialog box lists each
file type’s corresponding ISAPI DLL
What Can Be Done to Improve ASP Design? 11
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
<%
Dim strPage
strPage = "/scripts/MyPage.asp"
%>

However, since the server-side includes are performed before the ASP code is processed,
the above code will cause IIS to complain that the file <%=strPage%> cannot
be found. There are some methods available to fake dynamic server-side
includes. One way is to use a Select Case statement, with a Case for each
potential server-side include. For example, if you knew you would need to process
the code in one of five potential include files, you could use the following
code:
<%
Select Case strPageToExecute
Case "IncludeFile1.asp" %>

<% Case "IncludeFile2.asp" %>

<% Case "IncludeFile3.asp" %>

<% Case "IncludeFile4.asp" %>

<% Case "IncludeFile5.asp" %>

<% End Select %>
Each server-side include is issued immediately before any of the ASP code in the
ASP page is run. That means all of the five server-side includes will be executed
and their contents imported into the ASP page.
The downsides of this approach are that you are limited to a finite number of
potential include files, and for each potential include file, you need to hardcode a
case statement. As well, whenever you use an #include directive, the entire contents
of filename are inserted into the ASP script before processing. If you have a
large number of potential include files and these files contain large amounts of
code, this could cause a performance bottleneck.
Another way to “fake” dynamic server-side includes is to use the FileSystemObject
object model to read in the contents of the include file you are interested in inserting
into your ASP page. The following code allows for a file to be inserted into an
ASP script:
<% Option Explicit %>
<%
'************************* DESCRIPTION **************************
'* Output the entire contents of a text file, whose name can be *
'* dynamically generated. *
'****************************************************************
Dim strFileName, objFSO, objFile
12 Chapter 1: Introduction
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
'Create an instance of the FSO object
Set objFSO = Server.CreateObject("Scripting.FileSystemObject")
'What file do we want to import?
strFileName = "C:\InetPub\wwwroot\Adages.htm"
'Open a text file, using the OpenTextFile method
Set objFile = objFSO.OpenTextFile(strFileName)
'Output the entire contents of the text file
Response.Write objFile.ReadAll
objFile.Close 'Close the file...
'Clean up
Set objFile = Nothing
Set objFSO = Nothing
%>
Note that the FileSystemObject expects full physical filenames (C:\
InetPub\wwroot\Adages.htm) and not virtual filenames (/Adages.htm).
The above code may seem like it accomplishes a dynamic server-side include, but
it is a little misleading. Since the text file cannot be opened and read until the ASP
script is running, if there is any ASP code within the file specified by
strFileName, it won’t be executed; rather, the ASP code will be output just like
any other HTML content in the file! This method is an acceptable one if you only
need a dynamic server-side include to read in an HTML header or footer. However,
any ASP code in the file specified by strFileName will not be executed!
With ASP 2.0, truly dynamic server-side includes were impossible. However, with
ASP 3.0 and the new Execute method of the Server object, true dynamic includes
are possible! Server.Execute is discussed in the section “Using Server.Execute.”
Naming your include files
When grouping common functions into a particular file to be included by other
ASP pages, be sure to give the file an .asp extension. If you fail to do this, and a
web surfer guesses the correct filename of an include file, the user can view the
contents of the include file! Files with an .asp extension are safe from prying eyes
because when an ASP page is requested, IIS steps in and executes the ASP page,
turning the ASP code into HTML. Files without an .asp extension are not processed
by IIS and are sent directly to the client. This can reveal the source code of
your include files.
What Can Be Done to Improve ASP Design? 13
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
If for some reason, you must use include files with a non-ASP extension, be sure
to place these files in a directory that has HTTP read permissions turned off. With
the HTTP read permissions turned off, ASP pages can still import the contents of
these include files, but if someone attempts to request these files directly through
an HTTP-request (i.e., entering the full URL of the include file in their browser),
they will receive an error message indicating that HTTP reads are not permitted.
Since developers often place constants, database connections, and common functions
in include files, it is imperative that the contents of include files remain away
from prying eyes.
It’s true there is not a great chance that anyone will be able to exactly guess the
name of an include file. However, that does not mean your include files are safe.
Jerry Walsh has detected a potential security bug that can occur when using serverside
includes. This security bug happens when you create an include file that has a
programmatic error in it. For example, imagine a file named /CODEREUSE/dbConn.
inc that contained the following code:
<%
Dim objConn
Set objConn = Server.CreateObject("ADODB.Conection")
objConn.ConnectionString = "DRIVER={Microsoft Access (*.mdb)};" & _
"DBQ=" & Server.MapPath("/MyDatabase.mdb")
objConn.Open
%>
Note that there is an error in the above code snippet: the class ID for the Connection
object is misspelled, with “Conection” missing an “n.” Many ASP sites use an
include file to have one database connection file. However, making such a page
with an error can yield disastrous results, as we’ll see shortly. Now, if some page
on your site uses this include file like so:
<% @LANGUAGE="VBScript" %>
<% Option Explicit %>

<%
'Do stuff with the database connection...
%>
an error will occur when a user visits the page, since there is an error in /CODEREUSE/
dbConn.inc. Specifically, the user will see the error message:
Microsoft VBScript runtime error '800a004'
Invalid Class String
/CODEREUSE/dbConn.inc, line 3
Notice that the full path to the include file is displayed in the error message! Since
the include file does not contain an .asp extension, anyone who stumbles across
14 Chapter 1: Introduction
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
this erroneous page can now visit your database connection script, which contains
the path to your Access database (MyDatabase.mdb). The user can now
download directly!
You might think you are safe if you do not provide some mechanism for your
users to reach the ASP page that uses the erroneous include file. That is, if you
provide no direct hyperlinks to this page, no user will see the error message and
no user will know the include file’s path. While it is true that no user will likely
stumble across the ASP page, search engines may still index them!
Again, you might think that this is a minor problem and that there is an incredibly
low probability that a search engine will index this odious page. However, if you
go to AltaVista and enter the following search terms:
+"Microsoft VBScript runtime error" +".inc, "
you’ll find that an alarming number of pages are returned that display the complete
path to an include file whose contents can be read by any visitor!
The moral of this story: don’t put files whose contents will be imported into ASP
pages via a server-side include on your production web site until they have been
thoroughly tested and do not contain any errors that will reveal their location. Or,
more simply, just make sure you give your include files an .asp extension or place
them in a directory where HTTP read permissions are turned off. Another potential
solution is discussed in Chapter 3, Exception Handling. In that chapter, we’ll
look at how to have include files (and other ASP pages) handle errors more gracefully
than simply spitting out an error message.
According to an article on Microsoft’s web site (http://msdn.
microsoft.com/library/tools/aspdoc/iiwainc.htm), “…it is good programming
practice to give included files an .inc extension to distinguish
them from other types of files.” Be sure not to follow this
suggestion unless you place these files in a directory that has read
permissions disabled.
Protecting the contents of your include files from prying eyes
There are some simple steps you can take to ensure that your include files’ contents
won’t be seen by prying eyes. The two simplest options are:
• Create a directory for all your include files, and remove the read permissions
from this directory. This will prevent HTTP read requests on the specified
directory. If a web visitor attempts to retrieve any file from that directory, they
will receive a Permission Denied error.
What Can Be Done to Improve ASP Design? 15
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
• Give all include files an .asp extension. If a user tries to view a file with an .asp
extension, IIS will attempt to process the file first with ASP.DLL and will send
the resulting HTML instead of the actual source.
More about server-side includes
Server-side includes are extremely useful for developing and using code modules.
Placing common code in modules allows for easier code reuse. Throughout this
book, server-side includes are used quite frequently. Often classes will be created
that are used among many ASP pages. Rather than copying the class definition into
each ASP page, it makes much more sense to define the class in a single file and
to use a simple server-side include in all of the pages that need to utilize the given
class.
Since server-side includes are used so frequently throughout this book, it is important
you have a solid understanding of how they work. If you are a little rusty on
server-side includes, I highly recommend that you take the time to read the following
two articles:
• “The low-down on #includes,” found at http://www.4guysfromrolla.com/
webtech/080199-1.shtml.
• “Security Alert—Using includes Improperly from non-Debugged ASP Pages
can allow Visitors to View your source code,” found at http://www.
4guysfromrolla.com/webtech/020400-2.shtml.
VBScript Classes
Since Version 5.0 of the VBScript scripting engine, developers have had the opportunity
to use VB-like classes in their VBScript code. Classes provide for an objectoriented-
like programming approach when developing ASP pages, which greatly
enhances the reusability of a particular ASP page.
Classes are great for creating black-box modules. For other developers to use a
black-box module, they do not need to know any of the specific implementation
details; rather, they just use the black box through its publicly accessible methods
and properties.
We will discuss classes in detail in Chapter 4, Regular Expressions, Classes, and
Dynamic Evaluation and Execution. Classes are used extensively in Chapter 5,
Form Reuse, and Chapter 6, Database Reuse, as we look at code reuse techniques.
Since classes encapsulate complexity, hide implementation details, and function as
a black box for the developer, classes assist greatly when creating reusable code.
16 Chapter 1: Introduction
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
In Chapters 5 and 6 we will be examining reusable scripts that make
heavy use of VBScript classes.
Using Server.Execute
Server.Execute can also be used to modularize your ASP code. Server.Execute
branches the control flow from one ASP page to another. When the page that was
called via the Server.Execute completes processing, control is returned to the page
that issued the Server.Execute command. Figure 1-3 illustrates the semantics of
Server.Execute.
Server.Execute’s main advantage is its ability to perform truly dynamic includes.
For example, using Server.Execute, you could do the following:
<%
Dim strPage
strPage = "/scripts/MyPage.asp"
Server.Execute(strPage)
%>
When an ASP page uses Server.Execute to branch control to another ASP page, all
of its built-in ASP objects are passed along. For example, if Page1.asp issues a
Server.Execute("Page2.asp"), Page2.asp will have access to Page1.asp’s
intrinsic ASP objects. Remember that the Request object is an intrinsic ASP object
and that it contains the Form and QueryString collections. Since all the intrinsic
objects are shared from Page1.asp to Page2.asp, when Page2.asp attempts to read
Figure 1-3. Server.Execute branches the control flow to a separate ASP page, runs the page,
and returns the control flow to the original page
OriginalPage.asp
<%
Server.Execute("Page2.asp")
'Continue processing ASP script
%>
Page2.asp
Page2.asp is run in
its entirety. Once it
completes, control is
returned to the line
immediately following
the Server.Execute in
OriginalPage.asp.
Control is temporarily transferred to
Page2.asp, which will execute in its entirety.
Control is returned to the ASP page
that originated the Server.Execute.
What Can Be Done to Improve ASP Design? 17
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
from either of these Request object collections, it is reading the values from Page1.
asp’s Request object collections. Now that’s a mouthful!
For example, if Page1.asp simply had the code:
Server.Execute("Page2.asp")
and Page2.asp simply had the code:
Response.Write Request("Age")
if we visited Page1.asp through a browser, entering the URL http://localhost/Page1.
asp?Age=21, the Response.Write in Page2.asp would output 21.
To summarize: Server.Execute can be used in place of server-side includes when
dynamic includes are needed. If you don’t need dynamic includes, and a vanilla
server-side include would suffice, I would recommend sticking with the serverside
include.
Using Server.Transfer
Server.Transfer can also be used to improve ASP script design, although it cannot
serve as a modularization technique like Server.Execute and server-side includes.
Server.Transfer is similar to Server.Execute. If Page1.asp performs a:
Server.Transfer("Page2.asp")
the control flow is transferred to Page2.asp. When Page2.asp finishes executing,
control is not returned to Page1.asp. Figure 1-4 illustrates the semantics of Server.
Transfer.
Figure 1-4. Server.Transfer branches the control flow to a separate ASP page, runs the page to
completion, and stops executing
OriginalPage.asp
<%
Server.Transfer("Page2.asp")
%>
Page2.asp
Once Page2.asp completes
its execution, control is not
returned to OriginalPage.asp.
Using a Server.Transfer in
OriginalPage.asp is similar to
using a Server.Execute
as the last statement in
OriginalPage.asp, since once
the Server.Transfer is executed,
no further code is executed
on OriginalPage.asp.
Control is branched to Page2.asp. The control
flow is not returned to OriginalPage.asp once Page2.asp
finishes executing.
18 Chapter 1: Introduction
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
As with Server.Execute, when Page1.asp performs a Server.Transfer(“Page2.asp”),
Page1.asp’s built-in ASP objects are passed along to Page2.asp. Being able to
access the Request.Form and Request.QueryString collections from Page1.asp in
Page2.asp is an incredibly useful feature, which we’ll capitalize on in Chapter 5
and in Chapter 6.
Server.Transfer improves ASP design by providing a mechanism to seamlessly
move from one ASP page to another. Part of good application design is creating
robust, reusable code, which usually results in the creation of several generic
helper ASP pages. With Server.Transfer, we can “plug into” these helper pages,
moving from one to the next. In Chapter 5 we’ll look at applying Server.Transfer
to gracefully hop from one ASP page to another!
Further Reading
At the end of each chapter you’ll find a “Further Reading” section. These sections
include links to articles that relate to the topics discussed within the chapter. These
articles aren’t prerequisites to understanding the topics covered in the chapter;
rather, they are auxiliary readings, either describing some of the chapter topics in
finer detail or providing information on related topics. In either case, I think you’ll
find these suggested readings worthwhile and beneficial.
• For an incredibly thorough discussion on server-side includes, be sure to check
out “The low-down on #includes,” available at http://www.4guysfromrolla.com/
webtech/080199-1.shtml.
• For a good summary of the new features added to IIS 5.0 and ASP 3.0, visit
http://www.microsoft.com/mind/0499/iis5/iis5.htm.
• Interested in learning more about HTML design? Check out this article: http://
webreview.com/pub/98/10/30/bookends/index.html. Also, every developer should
often visit Jakob Nielsen’s usability site, http://www.useit.com/.
• Server-side includes form just one facet of server-side directives. To learn more
about the other server-side directives (which, unfortunately, can’t be used
through an ASP page), be sure to check out, “Using Server-Side Directives” at
http://www.4guysfromrolla.com/webtech/082599-1.shtml.
• For more information on using Server.Execute and Server.Transfer to alter flow
control between ASP pages, be sure to read http://msdn.microsoft.com/library/
psdk/iisref/eadg4d9v.htm.
• Need to brush up on the FileSystemObject? If so, check out the FileSystemObject
FAQ at http://www.aspfaqs.com/webtech/faq/faqtoc.shtml#FileSystemObject.
19
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Chapter 2 2
Choosing a Server-Side
Scripting Language
When creating ASP pages, developers have a number of scripting languages to
choose from. The most frequently used server-side scripting language is VBScript,
but ASP is not limited to VBScript alone. In fact, any ActiveX scripting engine can
be used as the scripting language for ASP. Microsoft provides two ActiveX scripting
engines: VBScript and JScript. PerlScript and Python are two additional serverside
scripting languages that can be used in an ASP page.
This chapter introduces these rarely used scripting languages, demonstrating how
to accomplish some common ASP tasks. This chapter is not a language reference
or tutorial. Extensive online resources and reference books covering the details of
these scripting languages already exist.
The Popularity of VBScript
If you’ve read many other Active Server Pages books or are a regular at any of the
large ASP resource sites on the Internet, you’ve no doubt noticed that the vast
majority of code examples use VBScript as the server-side scripting language. In
fact, this book uses VBScript for its code examples. I believe the reason that
VBScript is used so frequently in code examples is that a large number of ASP
developers were previous Visual Basic developers. Also, a good number of beginning
ASP developers are web developers with extensive HTML skills but few programming
skills. VBScript is relatively easy to learn and understand, especially for
those who are new to programming.
VBScript did not achieve such popularity among ASP developers solely because it
is a simple language. When ASP started to grow into a widely used tool for creating
dynamic web sites, there were only two scripting engines: VBScript and
20 Chapter 2: Choosing a Server-Side Scripting Language
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
JScript. At the time, VBScript had several advantages over JScript. For one thing,
VBScript allowed for error handling, while earlier versions of JScript did not.
Current versions of VBScript and JScript, though, both contain error handling, as
well as a number of other enhancements. Several notable improvements were
added to the Version 5.0 scripting engines; in fact, Chapter 4, Regular Expressions,
Classes, and Dynamic Evaluation and Execution, discusses these improvements in
detail.
Just because VBScript is the most popular server-side scripting language does not
mean VBScript is the right scripting language for you. Each scripting language
offers its own strengths and weaknesses. Also, by supporting a plethora of languages,
ASP does not favor only those developers who have experience in a particular
language. If your strengths lie in Perl, use PerlScript as the server-side
scripting language. If you are a JavaScript guru, use JScript instead.
Specifying the Scripting Language
Since any compliant ActiveX scripting engine can be used to parse an ASP page,
when a web server receives a request for an ASP page, it must first determine
what ActiveX scripting engine needs to be used to parse the ASP page’s code. If
the ASP code consists of valid JScript syntax, but the web server attempts to have
the ASP page parsed with the VBScript engine, errors will abound.
An ASP page can explicitly indicate what scripting language was used through the
Language directive. For example, the following ASP snippet uses the Language
directive to indicate that JScript is the scripting language used:
<% @LANGUAGE = "JScript" %>
<%
var strQuote;
strQuote = "Hello, world!";
Response.Write(strQuote);
%>
The Language directive, like all other ASP directives, is preceded by @. Furthermore,
directives must appear before all other ASP code. Failure to do so will result
in the following error:
Active Server Pages error 'ASP 0140'
Page Command Out Of Order
The @ command must be the first command within the Active Server Page
Specifying the Scripting Language 21
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
If the Language directive is excluded (which it commonly is), the default scripting
language is used. When first installed, both IIS and PWS set the default scripting
language to VBScript. The default scripting language can be changed, though.
To change the default scripting language in IIS, visit the Internet Information Services
by going to Start . Programs . Administrative Tools . Internet Services
Manager. The web sites on your machine should be listed. Right-click and select
Properties for the web site whose default scripting language you wish to change.
A tabbed dialog box should appear; select the tab titled “Home Directory.”
Figure 2-1 shows this dialog box with the correct tab selected.
From the “Home Directory” tab, click the “Configuration…” button; a new tabbed
dialog box will appear. Select the second tab, “App Options.” One of the options
will read “Default ASP language.” Enter the scripting language of your choice here.
Figure 2-2 displays a picture of this tab with VBScript entered as the “Default ASP
language.”
Figure 2-1. The Home Directory tab contains home directory and web application options
22 Chapter 2: Choosing a Server-Side Scripting Language
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
If you have multiple web sites configured on your machine, you can
easily set the default scripting language for all of the sites in one fell
swoop. Rather than right-clicking and selecting the Properties option
for a particular web site, right-click and select the Properties option
for the machine name (the parent node of the various web site
nodes). Next, choose to Edit the Master WWW properties. You will
be presented with a dialog box identical to the one in Figure 2-1.
Simply follow the same sequence of steps as you would in changing
a single site’s default scripting language.
While the Language directive is not required, it is recommended that you always
explicitly use it. This will lead to less ambiguity in your ASP code. Advanced visual
editors, such as Microsoft’s Visual InterDev, automatically add the Language directive
to new ASP pages.
Specifying the Server-Side Scripting Language
with SCRIPT Blocks
The <% and %> delimiters are commonly used to indicate ASP code blocks. However,
SCRIPT blocks can also be used. For example, an ASP page could consist of:
<SCRIPT LANGUAGE="VBSCRIPT" RUNAT="SERVER">
Response.Write "Hello, World!"
</SCRIPT>
When using SCRIPT blocks, however, you must specify the LANGUAGE attribute;
leaving it off will generate an error. One incentive for using SCRIPT blocks is they
Figure 2-2. The default scripting language can be set through the “App Options” tab in the
Application Configuration dialog
Specifying the Scripting Language 23
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
allow for multiple server-side languages in one ASP page. Example 2-1 demonstrates
the use of a VBScript script block and a JScript script block, intertwined
with <% and %> delimited code:
A big disadvantage of using script blocks (especially script blocks with different
server-side languages) is that the order of script block execution is erratic. For
example, one would expect that the output of Example 2-1 would be:
Hello, from the VBScript block!
Hello from the HTML Block!
Hello from the <% ... %> block!
Hello, from the JScript block!
However, the output from Example 2-1 is quite different, and can be seen in
Figure 2-3. Note that the script block that does not contain the specified scripting
language operates first. Next, the HTML and <%- and %>-delimited code operate
serially; since the HTML output precedes the <%- and %>-delimited output, the
HTML output is displayed first. Finally, the scripting block that contains the specified
server-side scripting language is executed.
Due to this confusing fact, I find that writing ASP code using <% and %> delimiters
is not only easier to read, but it is easier to follow the control flow. For this
Example 2-1. Multiple Script Blocks Can Be Used in an ASP Page
<% @LANGUAGE="VBSCRIPT" %>
<% Option Explicit %>
<HTML>
<BODY>
<SCRIPT LANGUAGE="VBSCRIPT" RUNAT="SERVER">
'Display a message from the VBScript script block
Response.Write "Hello, from the VBScript block!<BR>"
</SCRIPT>

Hello from the HTML Block!<BR>
<%
'Display a message from the &lt% ... %&gt block
Response.Write "Hello from the <% ... %> block!<BR>"
%>
<SCRIPT LANGUAGE="JSCRIPT" RUNAT="SERVER">
// Display a message from the JScript script block
Response.Write("Hello, from the JScript block!<BR>");
</SCRIPT>
</BODY>
</HTML>
24 Chapter 2: Choosing a Server-Side Scripting Language
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
reason, the code examples presented throughout the remainder of this book use
these delimiters instead of SCRIPT blocks.
Choosing the Right Scripting Language
Now that we’ve looked at how to use different scripting languages, let’s look at
why one might choose to use a scripting language other than VBScript. In the next
three sections, “Creating ASP Pages with JScript,” “Creating ASP Pages with
PerlScript,” and “Creating ASP Pages with Python,” a discussion of the languages,
benefits and drawbacks will be presented, along with code samples of each scripting
language.
When deciding what scripting language to use, remember there is no correct
scripting language. Of course, developers may feel they are being encouraged to
use VBScript. After all, VBScript is the initial default scripting language for both IIS
and PWS. Also, the vast majority of books on Active Server Pages use VBScript
extensively in examples and sample code. However, VBScript isn’t necessarily
always the best choice. If it were, there wouldn’t be so many other scripting languages
available!
When deciding what scripting language to use, choose a language the developers
working on the project are familiar with. If the developers know more than one of
the available scripting languages, choose the scripting language whose features
closely match the needs of the project at hand.
Finally, understand that VBScript and PerlScript are not identical to the Visual
Basic and Perl languages. Rather, these scripting languages are subsets of the complete
VB language and Perl. Therefore, a developer experienced in Perl should
have no problem writing Active Server Pages using PerlScript.
Figure 2-3. The output of script blocks is not serial
Creating ASP Pages with JScript 25
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Creating ASP Pages with JScript
Web developers who are familiar with writing client-side JavaScript code already
have a good understanding of JScript’s syntax and structure. JScript’s syntax and
control structures are also very similar to C’s. For example, JScript’s control
structures—if … else, switch, while, do … while, and for statements—are
syntactically identical to C’s.
Statement Termination
JScript handles its statement termination a bit differently than C. In C, a semicolon
is needed to end a statement; in JScript, either a semicolon or a newline character
will suffice. Therefore, you can have a JScript statement end without a semicolon
as long as the next statement begins on a new line. You can have multiple statements
on one line, but then a semicolon must delimit each of these statements.
For example, the following code fragment illustrates the legal and illegal use of
semicolons and newline characters:
<% @LANGUAGE = "JScript" %>
<%
Response.Write("Each new-line character represents")
Response.Write("a new statement in JScript. So does a semicolon.<P>")
// This is legal code:
Response.Write("Hello, "); Response.Write("World!");
// This is not:
Response.Write("Hello, ") Response.Write("World!");
%>
I highly recommend always ending each statement with a semicolon, regardless of
whether the next statement starts on a new line. JScript code examples in this
book will adhere to this strict use of semicolons.
JScript’s Variables and Datatypes
JScript does not require that you explicitly declare your variables. There is no
Option Explicit-type command in JScript. To declare a variable, use the var
keyword. Variable names must begin with a letter, an underscore ( _ ), or a dollar
sign ($); subsequent characters can be letters, numbers, underscores, or dollar
signs.
JScript, like VBScript, is a loosely typed language. That is, every variable is of type
Variant, meaning a single variable can be a string, an object, a number, or any
other type. There are six datatypes available with JScript: string, number, object,
Boolean, null, and undefined. The first five datatypes are fairly straightforward.
The undefined datatype, however, warrants further discussion.
26 Chapter 2: Choosing a Server-Side Scripting Language
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
When a variable is first created (either implicitly or explicitly with the var statement),
the variable is undefined. When using an undefined variable in a mathematical
expression, the resulting answer will be NaN, which stands for Not a
Number. If you attempt to concatenate a string variable and an undefined variable,
the concatenated portion of the undefined variable will read “undefined.”
The following code example illustrates this point:
<% @LANGUAGE = "JScript" %>
<%
var x, y; // x and y are undefined
y = "Hello " + x; // y now equals "Hello undefined"
Response.Write(y + "<P>");
y = 4 * x; // y now equals NaN
Response.Write(y);
%>
In this example, the results would be identical if you replaced x in the two assignment
statements with an implicitly created variable.
Case Sensitivity
JScript, like JavaScript, is case-sensitive. For this reason, it is important to know the
case for the various ASP intrinsic objects and their methods and properties. Each
ASP intrinsic object has the following case: Response, Request, Server, Session,
Application, ObjectContext, and ASPError. Using improper case will result in an
error. Try running the following code, in which the R in Response is not
capitalized:
<% @LANGUAGE = "JScript" %>
<% response.Write("Case is important!!"); %>
Since the JScript engine does not recognize response as a valid object name, an
error message is displayed when viewing the ASP page through a browser, as
Figure 2-4 illustrates.
If you are a VBScript developer learning JScript, you may find this
case-sensitivity issue to be an annoyance. Not only do you need to
know the case of the intrinsic ASP objects, but you also need to know
the case of all of their methods and properties! The general rule is all
of the methods and properties of the built-in objects have the first letter
of each word capitalized. If the property or method contains just
one word (like the Write method of the Response object), then just
the first letter is capitalized. However, if the property or method contains
two words (like the ExpiresAbsolute method of the Response
object), then the first letter of each word is capitalized.
Creating ASP Pages with JScript 27
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Regular Expressions
JScript contains a number of intrinsic objects, such as the Math, String, and Number
objects. These objects contain useful properties and methods that make
accomplishing various tasks much easier. For example, the Math object contains
methods like cos and sin, which can be used to calculate the cosine and sine of a
number.
The String object is especially useful, since it contains methods that allow for
regular expression searching and replacing. Regular expressions are a powerful
string-searching technique. If you are unfamiliar with the specifics of regular
expressions, be sure to pick up a copy of O’Reilly’s Mastering Regular Expressions
by Jeffrey E. F. Friedl. This section will not attempt to teach regular expressions;
rather, it will just show how to use regular expression matching and replacing in
JScript.
A short regular-expression tutorial is presented in Chapter 4.
The String object contains three methods for regular expression searching and
matching: match, replace, and search. All three of these methods expect a Regular
Expression object instance as one of their parameters. A Regular Expression object
contains a valid pattern expression, and optionally, a switch. An instance of the
Regular Expression object should be declared like this:
var RegularExpressionInstance = /pattern/[switch];
Figure 2-4. Error message caused by incorrect case of Response object
28 Chapter 2: Choosing a Server-Side Scripting Language
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
where pattern is a valid regular expression pattern and switch is one of the following
values:
i Ignore case.
g Perform a global search for all occurrences of pattern.
gi Perform a global search for all occurrences of pattern, ignoring case.
Once you have a valid Regular Expression object instance, you can use one of the
regular expression methods available in the String object.
The match method, which expects a Regular Expression object instance as its single
parameter, will return an array with the zeroth element of the array containing
the last matched characters. Array elements one through N will contain matches to
any parenthesized substrings in pattern.
The search method of the String object, like the match method, expects a single
parameter containing a valid Regular Expression object instance. search returns a
number, indicating the position in the string where the first match of the regular
expression occurs. If no such match exists, search returns –1.
The replace method expects two parameters, a Regular Expression object instance
and a string containing text to replace any matches with. The replace method simply
searches for any matches, replacing all occurrences with the string passed in as
its second parameter.
Regular expression matching can be very useful for ASP pages in server-side form
validation. If we want to collect and store user information in a database, form validation
can help ensure database integrity. For example, if a user submits a form
with her name and telephone number, it would be nice to ensure that the telephone
number is in a valid format before inserting the information into a database.
Example 2-2 demonstrates several uses of JScript’s regular expression
capabilities. Portions of Example 2-2’s code could easily be extended for use as
server-side form validation. In Chapter 5, Form Reuse, we discuss server-side and
client-side form validation in detail.
Example 2-2. Regular Expressions in JScript
<% @LANGUAGE = "JScript" %>
<%
// First, examine the following statement:
var strColdFusion;
strColdFusion = "Using Cold Fusion is a fun, exciting way to create " +
"dynamic Web pages! I really like Cold Fusion.";
// Now, let's tailor that sentence to this book!
// We will replace all occurrences of "Cold Fusion" with "ASP"
// Note that we have to use the global switch (g) to replace all
Creating ASP Pages with JScript 29
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
You can also perform regular expression searches using VBScript.
With Version 5.0 of the scripting engines, Microsoft released a Regular
Expression COM object. This object is covered in greater detail in
Chapter 4.
// instances of Cold Fusion with ASP
var objRegExpr = /Cold Fusion/g;
var strASP = strColdFusion.replace(objRegExpr, "ASP");
// We have now just used the replace function to replace all instances of
// Cold Fusion with ASP. Output the corrected sentence!
Response.Write(strASP + "<P>");
// Now, we will use the search method to make sure a particular string
// has a particular format. If we want to make sure that the string
// strPhoneNumber has the format ###-###-####, we can do the following:
var strPhoneNumberValid = "555-555-1234";
var strPhoneNumberInvalid = "67-325533-324";
// Our regular expression object instance will search for three numbers,
// a hyphen, three numbers, a hyphen, and then four numbers.
var objRegExprPhone = /\d{3}-\d{3}-\d{4}/;
// If the phone number is valid, a match should be found starting at the
// zeroth character of the string, and the length of the string should be
// exactly 12 characters
if (strPhoneNumberValid.length == 12 &&
strPhoneNumberValid.search(objRegExprPhone) == 0)
// Valid phone number!
Response.Write(strPhoneNumberValid + " is a valid phone number!");
else
// Invalid phone number!
Response.Write(strPhoneNumberValid + " is <B>not</b> a valid phone number!");
Response.Write("<BR>");
if (strPhoneNumberInvalid.length == 12 &&
strPhoneNumberInvalid.search(objRegExprPhone) == 0)
// Valid phone number!
Response.Write(strPhoneNumberInvalid + " is a valid phone number!");
else
// Invalid phone number!
Response.Write(strPhoneNumberInvalid + " is <B>not</B> a valid phone number!");
%>
Example 2-2. Regular Expressions in JScript (continued)
30 Chapter 2: Choosing a Server-Side Scripting Language
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Error Handling
The ASPError object, new to ASP 3.0, provides a mechanism to catch any errors
that occur either directly from the code in an ASP page, or in an external object
that is instantiated within an ASP page. This object works independently of the
server-side scripting language being used.
Before ASP 3.0 and the ASPError object, though, all error handling needed to be
performed by the scripting language used. VBScript has long had error-handling
capabilities in the form of On Error Resume Next, but JScript has not had such
capabilities until Version 5.0 of the scripting engine. JScript employs an errorhandling
syntax identical to C++’s try … catch method.
With the ASPError object, however, scripting language-specific error handling has
become obsolete for use in Active Server Pages development. However, these
error-handling routines still matter for the developer using ASP 2.0.
Chapter 3, Exception Handling, is dedicated to error handling, both through the
use of the ASPError object and through the VBScript and JScript scripting engines.
To become acquainted with JScript’s error-handling mechanism, be sure to read
Chapter 3.
An ASP Example Using JScript
Now that we’ve examined a number of useful JScript features, it’s time to present
some code. This code is intended to give you a flavor of the JScript syntax and to
provide a working, useful example of JScript code.
In Example 2-3, an ASP page is presented that will display detailed information for
each file in a particular web directory. This ASP page shows this information only
for the Web’s root directory; however, only a slight modification to the code
would be needed to allow the user to pass in the particular directory he is interested
in viewing through the query string.
Example 2-3. A JScript Example
<% @LANGUAGE = "JScript" %>
<%
// What Web directory are we interested in?
var strDirectory = "/";
// Translate the virtual directory into a physical directory
var strPhysicalDirectory = Server.MapPath(strDirectory);
// Create an instance of the FileSystemObject object
var objFSO = Server.CreateObject("Scripting.FileSystemObject");
var objFolder, objFile;
Creating ASP Pages with PerlScript 31
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
For an extensive technical reference on the JScript syntax, along with the latest
version of the JScript scripting engine, visit: http://msdn.microsoft.com/scripting.
Creating ASP Pages with PerlScript
If you are an experienced Perl developer, you may find PerlScript more comfortable
to use than VBScript or JScript. PerlScript is syntactically similar to Perl. The
PerlScript scripting engine is maintained by ActiveState, a company that describes
its role as “assist[ing] with the transition of [Perl] scripts between Windows and
Unix based systems” (http://www.activestate.com/corporate/ ). The PerlScript engine
can be downloaded for free from ActiveState’s web site, http://www.activestate.com.
// Use FSO's GetFolder method to obtain a copy of the
// folder strPhysicalDirectory
objFolder = objFSO.GetFolder(strPhysicalDirectory);
// Now, enumerate through each file in the folder, outputting its
// details...
var objFiles = new Enumerator(objFolder.Files);
Response.Write("<HTML>\n<BODY>\n");
Response.Write("<TABLE BORDER=1 ALIGN=CENTER CELLSPACING=1>\n");
Response.Write("<TR><TH>Name</TH><TH>Date Created</TH><TH>");
Response.Write("Date Last Modified</TH><TH>Size (in Bytes)</TH>");
Response.Write("<TH>Type</TH></TR>\n");
var iFileCount = 0;
while (!objFiles.atEnd())
{
objFile = objFiles.item();
// Output the various detailed file properties
Response.Write ("<TR>\n");
Response.Write("\t<TD><B>" + objFile.Name + "</B></TD>\n");
Response.Write("\t<TD>" + objFile.DateCreated + "</TD>\n");
Response.Write("\t<TD>" + objFile.DateLastModified + "</TD>\n");
Response.Write("\t<TD>" + objFile.Size + "</TD>\n");
Response.Write("\t<TD>" + objFile.Type + "</TD>\n");
Response.Write ("</TR>\n");
objFiles.moveNext();
++iFileCount;
}
Response.Write("</TABLE>\n\n");
Response.Write("There were " + iFileCount + " files in " + strDirectory);
Response.Write("\n\n</BODY>\n</HTML>\n\n");
%>
Example 2-3. A JScript Example (continued)
32 Chapter 2: Choosing a Server-Side Scripting Language
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
With PerlScript, like with Perl and C/C++, each statement must end with a semicolon.
PerlScript’s control structures, which are identical to Perl’s, include if …
elsif … else, for, while, foreach, and foreach … until. If you are new to
Perl, I highly recommend that you obtain a copy of O’Reilly’s Learning Perl, Second
Edition by Randal Schwartz and Tom Christiansen.
PerlScript’s Variables and Datatypes
Each PerlScript variable needs to contain a certain prefix. The prefix determines
the variable’s datatype. With PerlScript, as with Perl, there are three major
datatypes, as illustrated by Table 2-1.
There are a number of ways to declare variables. The method you use determines
the scope of the variable. To limit a variable’s scope to the block of code in which
it is declared, use the my keyword. For example, the following ASP code snippet
creates and initializes a scalar variable, an array, and an associative array:
<% @LANGUAGE = "PerlScript" %>
<%
my $name = "Scott";
my @array = ("Microsoft", 74.5, -5, "4GuysFromRolla.com");
my %grades = ("English" => 94.5, "Math" => 100.0, "Science" => 91.3);
%>
The second way to declare a variable is to use the local keyword. When declaring
a variable as local, it is accessible within the block of code in which it was
declared, as well as in subroutines called from the block of code where it was initialized.
If you declare a variable with the global keyword, or do not declare the
variable before using it, the variable has global scope.
Creating global-scoped variables in an ASP script can lead to some
daunting problems. It is highly recommended that you declare all of
your variables using the my keyword.
Take a moment to look over the previous code example. Note that the => operator
is used to assign a key in the associative array to a value. Also, when you want
Table 2-1. PerlScript’s Datatypes
Datatype Prefix Description
Scalar $ Contains simple datatypes, such as strings and numbers
Array @ Contains a variable number of scalar elements indexed by
number; arrays are indexed starting at zero
Associative Array % Contains a variable number of scalar elements indexed by a
string value; also referred to as a hash, and functionally
identical to a Dictionary object
Creating ASP Pages with PerlScript 33
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
to write or read a particular value from an array or associative array, you need to
reference the array or associative array with a dollar sign ($), since you want to
deal with a particular scalar value in the array instead of the entire array. For
example, imagine we wanted to assign the zeroth item from @array in the previous
code snippet to a variable named $BigCompany. The following line of code
would accomplish that:
my $BigCompany;
$BigCompany = $array[0];
Case Sensitivity and Object Reference
PerlScript, like VBScript, is not case-sensitive. When accessing the intrinsic ASP
objects, the object name must be preceded by a dollar sign ($). Furthermore,
rather than using the period to reference a particular object’s method, you must
use the arrow (a dash and the greater-than sign) (->). And just to make matters
more confusing, to reference an object’s property, you must use the arrow and
surround the property name with braces. The following code snippet demonstrates
setting the Response object’s Buffer property using PerlScript:
<% @LANGUAGE = "PerlScript" %>
<%
# Turn off Buffering
$Response->{Buffer} = 0;
$Response->Write("Buffering has been disabled.");
%>
Additionally, to access an element from an object’s collection, you should use the
following syntax:
$ObjectName->CollectionName("KeyName")->Item;
For example, if you expected a variable named Age to be to be passed through
the querystring, you could access the variable on your ASP page using:
$Request->QueryString("Age")->Item;
There are a lot of interesting intricacies of Perl. If you are someone
who likes to have ten different ways to accomplish the same task,
then you’ll absolutely love Perl. An acronym used often in the Perl
community is TMTOWTDI (There’s More Than One Way To Do It).
Entire books are dedicated to Perl’s many intricacies and the plethora
of ways to accomplish a given task. If you are interested in
learning PerlScript, I can’t stress enough the importance of owning a
copy of one of these books! Again, this chapter is not going to even
attempt to dive into the specifics of PerlScript. Rather, it will touch
upon a few of the unique features PerlScript brings to the table and
present a useful ASP application using PerlScript code.
34 Chapter 2: Choosing a Server-Side Scripting Language
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Reading and Writing Files with PerlScript
PerlScript, unlike VBScript and JScript, has inherent support for file access. To
open a file, use the open function. This function expects two parameters: a
filehandle and a filename. The filename is simply a string, and the mode
with which the file is opened is determined by an optional prefix on the filename.
The following prefixes can be used to indicate what mode to open the file in:
< Read
> Write (erases file contents upon opening)
+> Read and write
+< Read and write (erases file contents upon opening)
>> Append
If no prefix is added, the file is opened for reading. If you wanted to create an
ASP page that would open the file C:\Scott\MyFile.txt for reading, you could use
the following code:
open(FILEHANDLE, "<C:\\Scott\\MyFile.txt");
The less-than sign preceding the filename indicates to open the file for reading.
Also, note the double backslashes. Perl, like JScript, uses a backslash as the escape
character; therefore, to insert a literal backslash, you must use two backslashes. Be
sure to close a file with the close function once you are finished reading or writing
from the file:
close(FILEHANDLE);
PerlScript reads the contents of a file one line at a time. To loop through and display
the entire contents of a file, you can use the following code:
<% @LANGUAGE = "PerlScript" %>
<%
open(FILEHANDLE, "C:\\global\\WinTop.inf");
while (<FILEHANDLE>)
{
$Response->Write("$_<BR>");
}
close(FILEHANDLE);
%>
The $_ symbol is a special variable in PerlScript that is beyond the scope of this
book. Again, I strongly suggest you pick up a copy of a book dedicated to Perl if
you decide to use PerlScript regularly in your ASP pages.
Creating ASP Pages with PerlScript 35
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Regular Expressions
PerlScript, like JScript, offers inherent regular-expression capabilities; however,
PerlScript’s implementation of regular expressions differs syntactically from
JScript’s. To use regular expressions with PerlScript, use the =~ operator, which
performs a regular expression match or replace to the scalar variable to the left of
the =~ operator. To the right of the =~ operator, the following format needs to be
presented:
operator/pattern/replacement/switch
The operator defines the action to be taken. For example, if you want to replace
all matches of the pattern to the string variable to the left of the =~ operator, use
the s operator, denoting substitution. pattern contains the string you are trying
to match in the variable on the left side of the =~ operator. If you plan on replacing
all matches found with some string, enter the string to replace the matches
with in the replacement section. Finally, the switch section is identical to
JScript’s switches in its regular-expression implementation.
If you haven’t used regular expressions in Perl before, the last paragraph is probably
as clear as mud. Hopefully an example will help. In Example 2-2, JScript’s
regular-expression capabilities were used to perform a number of tasks. Let’s perform
the first task—replacing a substring (Cold Fusion) with a different substring
(ASP)—using PerlScript’s regular expression:
<% @LANGUAGE = "PerlScript" %>
<%
# First, examine the following statement:
my $strColdFusion;
$strColdFusion = "Using Cold Fusion is a fun, exciting way to create ";
$strColdFusion .= "dynamic Web pages! I really like Cold Fusion.";
# Now, let's tailor that sentence to this book!
# We will replace all occurrences of "Cold Fusion" with "ASP"
# Note that we have to use the global switch (g) to replace all
# instances of Cold Fusion with ASP
$strColdFusion =~ s/Cold Fusion/ASP/g;
# Output the new value of $strColdFusion
$Response->Write("$strColdFusion");
%>
A couple of quick points: the .= operator performs string concatenation, and to
output a string variable, you must include it in a string delimited by double quotes
(for example, "My name is $Name").
36 Chapter 2: Choosing a Server-Side Scripting Language
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Other PerlScript Features
One of PerlScript’s greatest benefits is Perl modules. Modules are prewritten code
that you can easily incorporate into your PerlScript scripts. ActiveState maintains a
robust library of Perl modules that you can freely use in your PerlScripts. For more
information visit: http://www.activestate.com/packages/.
PerlScript also allows for object-oriented programming, similar to VBScript’s ability
to use classes (for more information, see Chapter 4). For more information, be sure
to obtain a book on Perl; it’s well worth the investment!
An ASP Example Using PerlScript
Now that we’ve worked through a number of useful PerlScript features, it’s time to
look at some code. As with the JScript code in Example 2-2, this code is intended
to give you a flavor of the PerlScript syntax and provide a working, useful example
of PerlScript code.
Many web sites today have some way to allow users to interact with one another.
For example, you might have a chat application, message board, or guest book on
your site. If you run a family-oriented site, it would be nice to ensure that the messages
posted by your users contain no profane words. Using regular expressions,
you can easily remove particular offensive words from your users’ posts.
Example 2-4 provides some code that will censor a particular string. Perhaps it
could be used on Microsoft’s site, for it censors words like Linux, Unix, Sun, and
Java! The PerlScript expects a variable named txtMessage to be passed through
the Request.Form collection. The script then censors the string and displays it. You
could easily modify the code in Example 2-3 to insert the record into a database,
or perform other operations on the submitted message.
Example 2-4. Censoring with PerlScript
<% @LANGUAGE = "PerlScript" %>
<%
# Read in the string from the Request.Form named txtMessage
my $strMessage = $Request->$Form("txtMessage")->Item();
# Now, use regular expression to remove offensive words
$strMessage =~ s/linux/Windows/gi;
$strMessage =~ s/UNIX/Windows/gi;
$strMessage =~ s/Java/Visual Basic/gi;
$strMessage =~ s/Sun/Microsoft/gi;
# Display the new, censored Message
$Response->Write("<HTML>\n<BODY>\n\n");
$Response->Write("\tYour <i>new</i> message reads:<BR>$strMessage\n");
Further Reading 37
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
To learn more about using PerlScript within ASP pages, visit http://www.
perlscripters.com. To obtain the latest version of PerlScript and to access the many
Perl modules freely available, visit http://www.activestate.com.
Creating ASP Pages with Python
Python is an object-oriented scripting language that incorporates the best features
from an array of languages, such as Java, C/C++, Perl, and awk. If you’re not
familiar with object-oriented programming, you may find the Python syntax a bit
daunting. Python is the only scripting language discussed in this chapter that
requires you to write object-oriented code! PerlScript and VBScript have objectoriented
capabilities, but do not require that you use such techniques.
To use Python in your ASP pages, you need to download the latest Python
ActiveX scripting engine. You can download this for free from http://www.python.
org/download/. Be sure to download and install both the Python interpreter and
the Win32 COM extensions. Also, you have to register the Python ActiveX scripting
engine on the web server before you can use it. Consult the online Python
documentation for more information.
To learn more about Python, be sure to visit http://www.python.org. Extensive documentation
exists at http://www.python.org/doc/. Finding good references on using
Python with ASP is far from easy. There are a couple of short articles available at
Python.org, but those articles are targeted for experienced Python developers
interested in learning ASP as opposed to experienced ASP developers interested in
learning Python!
Further Reading
Unfortunately, there is little in the way of documentation on these alternative
server-side scripting languages. While there are a vast number of ASP-related web
sites, nearly all of them focus exclusively on VBScript. The vast majority of ASPrelated
books focus on VBScript as well. Despite this, there are some useful articles,
tutorials, and sites that deal with these alternative scripting languages.
• This tutorial is geared towards those who are more familiar with Perl than with
ASP. Nevertheless, it is a very thorough tutorial on PerlScript and worth a read:
http://www.fastnetltd.ndirect.co.uk/Perl/Articles/PSIntro.html.
• As I mentioned earlier in the chapter, the source for PerlScript and ASP information
is definitely PerlScripters.com, available at http://www.perlscripters.com.
$Response->Write("</BODY>\n</HTML>\n\n");
%>
Example 2-4. Censoring with PerlScript (continued)
38 Chapter 2: Choosing a Server-Side Scripting Language
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
• If you want to ask your Jscript questions with fellow ASP developers, be sure
to sign up for the JScript ListServ at http://www.asplists.com/learn/javascript.asp.
• Read a good tutorial discussing JScript in terms of objects at http://www.
asptoday.com/articles/19990316.htm.
• Learn how to incorporate both VBScript and JScript code in your ASP pages.
This article, available at http://www.asptoday.com/articles/19990420.htm, even
demonstrates how to call VBScript functions from JScript code and vice versa.
• One of VBScript’s advantages is its many inherent formatting functions, such
as FormatCurrency, FormatDateTime, and others. Mosey on over to http://
www.4guysfromrolla.com/webtech/vb2java.shtml to see some JScript translations
of some of the more common and useful VBScript formatting functions.
• For more information on the order in which SCRIPT blocks execute within an
ASP page, refer to http://msdn.microsoft.com/workshop/server/feature/morqa.
asp#order.
39
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Chapter 3 3
Exception Handling
Errors happen. Regardless of how skilled a given programmer might be, there is
no way he can create truly bug-free code. Programming errors can be broken
down into two broad categories: logical errors and implementation errors. Before
starting to actually write code, a developer must have a solid understanding of
what application she is creating, and the logic behind this application.
For example, when creating an e-commerce web application, a developer might
decide that once a user decides to purchase an item, he is taken directly to a page
to enter his billing and shipping information. Many users today are used to a shopping
cart system, in which they can purchase multiple items online and then proceed
to a “check out” page, where they can review a summary of the goods being
purchased and enter the needed billing and shipping information. By not providing
a shopping cart system, your users may find your site confusing and burdensome
to use. Logical errors are often high-level errors, costing immense time and
money to fix, since they are usually not caught until late in the product development
cycle.
Implementation errors are those errors residing in the actual code. In an e-commerce
site, if not all of the items selected by the user appear in the summary listing on the
“check out” page, an implementation error has occurred. To summarize: flaws existing
within the high-level view of the application are logical errors; errors or bugs
resulting from the implementation of an idea or feature into code are implementation
errors. Implementation errors are easy to detect: if the program crashes or produces
incorrect output, an implementation error is at fault. Logical errors, however,
are much more sinister and difficult to detect.
This chapter does not focus on how to reduce the number of errors you, as a
developer, commit when writing code. Rather, it looks at how to gracefully handle
these errors when they inevitably occur. We’ve all seen the patented Microsoft
40 Chapter 3: Exception Handling
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Blue Screen of Death one too many times, and all know the frustration involved
when a program fails to gracefully handle an error. Since bugs and errors in any
decent-sized application are certain, it is vitally important to be able to respond to
these errors smoothly when they occur.
A Bit of Terminology
We’ve already looked at the two broadest categories of errors: logical and implementation
errors. Implementation errors are often described in terms of halting
errors and non-halting errors. Halting errors cause the premature termination of a
running program, whereas nonhalting errors produce unreliable or unpredicted
output but allow the continuation of the program. Halting errors can be further
divided into runtime errors and compile-time errors. Runtime errors are those
errors that occur during the execution of a program, while compile errors occur
during the compilation process of the program.
When an ASP script is requested, it goes through a two-phase process. First, the
script is compiled into an executable version. Any syntactical errors are caught
here; these are referred to as compile-time errors. For example, the following
VBScript syntax, which is illegal, will generate a compile-time error:
<% @LANGUAGE="VBSCRIPT"%>
<%
Dim strScott
strScott = "Mitchell 'This will cause a compile-time error since the
'assignment to the string strScott is missing a
'closing quotation mark
%>
A syntactically correct script may be free of compile-time errors, but it can still
contain runtime errors. Runtime errors occur when some unexpected event happens
during the execution of the script. For example, dividing a number by zero is
legal syntax in VBScript (therefore not generating a compile-time error), but will
nevertheless generate a runtime error since the result of division by zero is undefined.
The following code snippet will generate a runtime error:
<% @LANGUAGE="VBSCRIPT"%>
<%
Dim iUndefined
iUndefined = 1 / 0 'This will cause a runtime error since the division by
'zero is undefined
%>
Runtime errors are not always the fault of the programmer who developed the
script that generates an error. For example, imagine that you create an ASP page
to display the contents of a particular SQL Server database table. Even though your
script may be flawless, containing no logic or implementation errors, if the database
table being queried is renamed or the entire database is deleted, your onceA
Bit of Terminology 41
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
perfect script will now abruptly terminate when attempting to query a non-existent
database table.
A more general term for a halting error is exception. An exception, in the most
general terms, is something unexpected. For this chapter, the definition of exception
is narrowed to an unexpected event that halts script execution. Therefore,
both an implementation error and an unexpected event (like a database table
being renamed) qualify as exceptions. It is every programmer’s goal to gracefully
handle exceptions.
You may have noticed that I’ve been using the phrase “gracefully handle an
exception” quite a bit, and as a curious reader, you may be wondering what I
mean, exactly, by “gracefully.” When an exception occurs, a number of things can
happen. Since the developer does have a bit of control over what happens following
an exception, the worst course to follow once an error occurs is to display
some illegible, confusing error message to a user who may not be computerliterate.
Rather than presenting the end user with a perplexing error message, one option
is to provide a more readable, friendlier message. This message would not need to
alert the end user to the technical aspects of the error, but rather would let the
user know a problem has occurred and the developers have been made aware of
the problem. It could also suggest other web pages to visit, or could contain a link
for more information (if you add the above) and the email address of the technical
support team.
Error messages are usually unreadable to the end user because they
usually contain information to help the developer detect the cause of
the error. While an error message like:
General Protection Fault in module 256_1024.DRV
is not readable for the end user, it lets the developer know the
driver in which the error occurred. In this chapter we will examine
how to create readable, friendly error messages for the end user and
how to provide detailed error information to the developer!
Displaying an understandable error message to the end user is usually the best
thing that can happen when an exception occurs. However, there are rare times
when an exception can be circumvented. That is, if a particular exception is
detected, a sequence of steps can be undertaken to dance around the problem,
continuing the execution flow, so that to the end user, it appears as though no
exception had occurred. For example, imagine that every week your database’s
data was backed up to a secondary database. If, in your ASP page, you have trouble
connecting to the primary database, rather than displaying an error message,
42 Chapter 3: Exception Handling
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
you could show the user the backup data. Granted, this data might be out of date,
and you would want to let the user know they are viewing dated information, but
showing the user something useful is better than showing them an error message.
Detecting When Exceptions Occur
To be able to handle errors gracefully, it is imperative that some mechanism exists
to inform us developers when an error arises. Errors occurring in an ASP page can
be detected in one of two ways: through the scripting language’s error-handling
interface, or through ASP 3.0’s new ASPError object.
The VBScript and JScript 5.0 scripting engines both contain robust runtime errordetection
mechanisms. ASP pages (or other scripts) created using either of these
scripting languages can use the scripting language’s inherent error-handling capabilities
to detect when errors occur. As we’ll see in “Detecting Exceptions with
Scripting Language Error-Handling Mechanisms,” using either VBScript’s or JScript’s
error-handling mechanisms often results in verbose, inelegant, and unreadable
code when compared to the elegance of the ASPError object.
The ASPError object, which we’ll detail in the section “Detecting Exceptions with
the ASPError Object,” provides detailed information about an error that just
occurred in an ASP page, including a description and line number. However, the
ASPError object can only be used with IIS 5.0/ASP 3.0, which, of course, requires
Windows 2000 as the web server. Therefore, if you are using Windows NT as your
web server’s operating system, you must rely on the scripting languages’ errorhandling
techniques to detect and gracefully respond to exceptions.
If you are using Windows 2000, though, you are in luck. To use the ASPError
object, a separate error-handling web page is needed. This separate page is
responsible for receiving the detailed error information and providing an understandable
error message to the end user. Using this approach, an error that occurs
on any line in any ASP page will result in a call to the error-handling web page.
Furthermore, the ASPError object provides more detailed information on the error
that occurred than either of the VBScript or JScript error-handling objects.
Using the ASPError object approach alone makes it next to impossible
to detect an error, circumvent it, and continue processing. Once
an error-handling web page is set up, when an error occurs, the user
is sent to the error-handling web page via a Server.Transfer.
Therefore, you can’t fix the error on the error-handling web page
and resume processing on the ASP page where the error originated.
We’ll look at the ASPError object in finer detail later in this chapter
in “Detecting Exceptions with the ASPError Object.”
Detecting When Exceptions Occur 43
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Detecting Exceptions with Scripting Language
Error-Handling Mechanisms
Both the VBScript and JScript 5.0 scripting engines provide runtime error-handling
routines to assist in detecting and gracefully handling implementation errors.
VBScript’s error handling is very similar to Visual Basic’s, using the On Error
Resume Next notation and the Err object. JScript’s error handling is similar to
C++’s and Java’s, using try ... catch blocks. Regardless of the scripting language
used, providing adequate error handling usually results in bloating one’s code
severely.
If your web site is not running on IIS 5.0, then unfortunately, you do not have
access to the ASPError object, and therefore must rely on the techniques discussed
in this section to detect errors. If, however, you are using IIS 5.0, and all
you plan on doing when an error occurs is displaying a friendly error message to
the end user, then it is highly recommended that you rely on the error handling
provided by the new ASP 3.0 ASPError object.
The main advantage the scripting language error-handling techniques have over
the ASPError object is the ability to recover from an error. Using these techniques,
a developer can detect when an error occurs and decide on some alternative
course of action, keeping the script running smoothly from the perspective of the
end user.
VBScript and JScript error-handling mechanisms can only detect
runtime errors. You cannot use these techniques to detect, recover
from, or display readable error messages about compile-time errors.
Detecting errors with VBScript
If you are familiar with error handling in Visual Basic, you’ll find VBScript error
handling easy to pick up. By default, error handling is disabled. This means if an
error occurs during the execution of a VBScript script, the script is halted and the
user is presented with an error message. Usually these error messages are quite
cryptic and daunting for the end user. Imagine a visitor to your web site who is
presented with an error similar to:
Server object error 'ASP 0177 : 800401f3'
Server.CreateObject Failed
/SomePage.asp, line 14
Invalid class string
44 Chapter 3: Exception Handling
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Would you expect this customer to return to your site in the future? This daunting
error message could be modified to a more friendly error message that is less
likely to scare the user away. To be able to determine when an error occurs in a
script, though, error handling must be enabled.
In VBScript, error handling is enabled and disabled using the On Error statement.
To enable error handling, enter the following command at the top of your ASP
page:
On Error Resume Next
This informs the scripting engine that when an error occurs, it should proceed to
the next line of code rather than abruptly stopping and displaying an error message.
To disable error handling, use the following statement:
On Error Goto 0
Simply enabling error handling is not enough to detect when an error occurs,
however. After each statement that might cause an error, an If statement needs to
be used to determine whether or not an error did occur. This technique is known
as
inline error handling. I find this constant inquiring whether or not an error has
occurred to be rather verbose and annoying, and liken it to an inefficient manager.
Imagine that you are a manager for one employee. This employee performs a
number of tasks, and as a manager, it is your job to ensure that these tasks do not
cause any errors. After each task this employee performs that might cause an
error, you take a moment out of your schedule to ask the employee whether or
not an error occurred. Such a model is far from efficient and is very verbose, containing
many unneeded communications between manager and employee. Ideally,
the employee would know when an error occurs, and would be able to
approach you, the manager, when something has gone awry. However, with
VBScript and JScript error handling, we do not have this option, and must continually
be bothered with having to inquire whether or not an error has occurred.
How often should you, the manager, ask your employee if an error
has occurred? Ideally, after every statement that might generate an
error, the employee should be queried if an error occurred. Of
course, what statements might generate an error? Deciding when to
check for an error is not always easy. This topic is discussed in
greater detail in the next section, “Deciding when to check if an
exception occurred.”
Example 3-1 contains an example of error handling. Note that at the start of the
script, On Error Resume Next is issued, enabling error handling. Then, after each
Detecting When Exceptions Occur 45
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
statement that might cause an error, a conditional statement is used to determine
whether or not an error has occurred. If an error has occurred, a short error message
is presented to the visitor.
VBScript provides an Err object to handle runtime errors. When error handling is
enabled and an error occurs, the Err object’s five properties are automatically populated
with the error information from the latest error. These properties are presented
in Table 3-1.
Example 3-1. Using VBScript to Detect an Error
<%@ LANGUAGE = "VBSCRIPT" %>
<% Option Explicit %>
<%
'Enable error handling
On Error Resume Next
'Create two objects
Dim objConn, objRS
Set objConn = Server.CreateObject("ADODB.Connection")
'Check to see if an error occurred
If Err.Number <> 0 then
Response.Write "Error in creating Connection object.<BR>"
Response.Write "Description: " & Err.Description
Response.End
End If
' The ProgID is invalid, so this line should generate an error
Set objRS = Server.CreateObject("Database Recordset")
'Check to see if an error occurred
If Err.Number <> 0 then
Response.Write "Error in creating Recordset object.<BR>"
Response.Write "Description: " & Err.Description
Response.End
End If
'If we've reached here, no errors!
Set objRS = Nothing
Set objConn = Nothing
%>
Table 3-1. The Err Object’s Properties
Property Description
Number A numeric value or code uniquely identifying the error
Source The name of the application or object that generated the error
Description A lengthy description of the error
HelpFile A fully qualified path to a help file
HelpContext A context ID for a topic in a help file
46 Chapter 3: Exception Handling
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
To determine whether an error has just occurred, check the Number
property of the Err object. If no error has occurred, this property will
equal zero. If an error has occurred, the property’s value will be
unequal to (i.e., either greater than or less than) zero.
The Err object contains two methods: Clear and Raise. The Clear method simply
resets all of the property values. If you plan on trying to recover from an error,
you will need to use the Clear method once you detect an error and find some
alternative path around the error. If the Clear method is not used, the next time we
check Err.Number we will find it is not equal to zero (since it still equals the
value from our first error). Therefore, if you plan on continuing execution after
handling an error, be sure to issue an Err.Clear after recovering from each error.
The following code snippet presents a case when you’d use the Clear method:
If Err.Number = SomeSpecificErrorNumber then
'Some specific error occurred, circumvent the problem somehow
Problem Circumvented Somehow...
'Now that we took some alternative approach, we want to clear the
'Err object
Err.Clear 'Cleared the error... continue on with the script
End If
The Raise method generates a runtime error, and has the following definition:
Err.Raise Number, Source, Description, HelpFile, HelpContext
When developing reusable scripts, often the developer who writes a given reusable
script is not the only one who uses it in practice. For example, in Chapter 5,
Form Reuse, we will look at how to create a reusable server-side form-validation
script. While I wrote the code and use it, I don’t expect to be the only developer
who will use it. The server-side form-validation script presented in Chapter 5
expects certain inputs from the developer using the script. If these inputs are out
of acceptable ranges, an error is raised so the developer is notified immediately
when trying to execute the script.
When raising errors in your own objects, Microsoft suggests that you add the
VBScript constant vbObjectError to the custom error code. Examples in this
book using the Raise method add vbObjectError to the error number, like so:
Err.Raise vbObjectError + ErrorNumber, ...
There is nothing magical about vbObjectError. It is simply a constant
with a value of 80040000 hexadecimal.
Detecting When Exceptions Occur 47
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Since this book focuses on building reusable, robust scripts, throughout this book
we’ll use the Raise method of the Err object extensively.
Deciding when to check if an exception occurred
Virtually any line of code can cause an exception. A given exception is either consistent
or inconsistent. A consistent exception is one that occurs every time the
script is executed. For example, the following line of code is an example of a consistent
exception:
Dim 4GuysFromRolla 'variable names cannot begin with a number
This exception is consistent, since every single time an ASP page is visited with the
above line of code, an error will be generated. The following lines of code demonstrate
an inconsistent exception:
'Connect to a database
Dim objConn
Set objConn = Server.CreateObject("ADODB.Connection")
objConn.ConnectionString = "DRIVER={Microsoft Access Driver (*.mdb)};" & _
"DBQ=C:\InetPub\wwwroot\datastores\NWind.mdb;" & _
"UID=admin;PWD=;"
objConn.Open 'Open a connection to the database
An error would occur if the database file NWind.mdb did not exist in the path
specified in the ConnectionString property (C:\InetPub\wwwroot\datastores\).
Such an exception is inconsistent because the above script would work flawlessly
so long as the database was in the right directory. Imagine, though, that NWind.
mdb was moved or renamed. Once that occurred, the script would start producing
error messages when the Open method was reached.
Consistent exceptions are simple to detect. When testing a script that contains:
Dim 4GuysFromRolla
a developer can quickly diagnose and remedy the problem. Since inconsistent
exceptions happen more unexpectedly, it is these exceptions that error detection
should be employed for. After each line that may generate an inconsistent exception,
be sure to place an If statement to check if Err.Number is not equal to
zero. Those statements that might cause inconsistent exceptions are usually limited
to methods of COM objects, such as the Open method of the ADO Connection
object. Therefore, when using COM objects in your ASP scripts, be sure to
check for errors immediately after calling one of the COM object methods.
In the later section “Responding to Exceptions,” we look at how to gracefully handle
exceptions that occur using VBScript exception handling.
48 Chapter 3: Exception Handling
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Error handling in functions and subroutines
If your ASP page contains any functions or subroutines, error handling becomes
somewhat trickier. If you enable error handling at the start of your ASP page and
then call a function or subroutine that contains a runtime error, the function or
procedure will halt and pass control back through the call stack until a procedure
with enabled error handling is found. Figure 3-1 illustrates this concept
graphically.
When a function or subroutine is called, error handling is deactivated. Therefore, if
you want a function or subroutine to be able to perform error handling on its own
(that is, if you don’t want it to halt and pass control back to the calling procedure),
you must explicitly turn on error handling in the function or subroutine
with On Error Resume Next. Example 3-2 examines VBScript’s error-handling
conventions with functions and subroutines.
Figure 3-1. When an error occurs in a non-error handling-enabled procedure, control is
passed back through the call stack until a procedure with error handling enabled is found
Procedure1 is called. If an error occurs in Procedure1 (or any procedures called by Procedure1)
and error handling is not enabled, control will be returned to the ASP page. If error handling is
on in the ASP page, the error can be handled (test for an error with "If Err.Number<>0"). If error
handling is disabled, an error will be generated, presenting the user with an error message.
ASP Page
Procedure1
Procedure1 receives the error. If it can't
handle it, it returns control to the section
of the ASP page where the procedure was
invoked.
Procedure1 calls some
procedure...
ProcedureN-1
ProcedureN-1 receives the error from
ProcedureN. If ProcedureN-1 has error
handling enabled, it handles the error;
otherwise, it passes it up the call stack.
ProcedureN-1 calls
ProcedureN
ProcedureN
An error occurs in ProcedureN, which
does not have error handling enabled.
ProcedureN halts and passes the error up
the call stack (that is, the error is passed
to the procedure that called ProcedureN).
ProcedureN halts and
passes up the runtime
error information to
ProcedureN-1
ProcedureN-1 halts
and passes back the
error information to
its calling Procedure...
Control is returned to
the ASP page
The ASP page calls Procedure1
Detecting When Exceptions Occur 49
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Note that the error-handling code after the call to Procedure1 is tripped for the
error that occurred in Procedure2. The output of the code in Example 3-2 is as
expected:
An error occurred!
500 - Variable is undefined
Streamlining VBScript error-handling code
With VBScript error handling, after each statement that might cause an exception,
the developer needs to determine whether an error occurred. The following code
is usually found after every potential error-causing statement:
If Err.Number <> 0 then
Display an error message
End If
Example 3-2. When an Error Occurs in a Function or Subroutine, Control Is Passed Up the Call
Stack Until an Error-Handling Enabled Function or Subroutine Is Found
<% @LANGUAGE="VBSCRIPT" %>
<% Option Explicit %>
<%
'Enable error handling
On Error Resume Next
'Call Procedure1
Procedure1
'We can check to see if an error occurred
If Err.Number <> 0 then
Response.Write "An error occurred!<BR>"
Response.Write Err.Number & " - " & Err.Description
End If
Sub Procedure1()
'No error handling here
'Call Procedure2
Procedure2
End Sub
Sub Procedure2()
'No error handling here either
'Whoops, an error occurred! (Using an undeclared variable)
strScott = strMitchell
'The above error will halt Procedure2, returning control to Procedure1. Since
there is no error handling implemented there, ontrol is returned to the'line
'immediately following the call 'to Procedure1. Since error handling is enabled
'there, it will handle the error. (Otherwise, the user would be shown
'an error message.)
End Sub
%>
50 Chapter 3: Exception Handling
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
This approach can be streamlined a bit by using a subroutine to determine if an
error has occurred or not. This subroutine, CheckForError, should have the following
definition:
Sub CheckForError(strErrorMessage)
strErrorMessage is a string containing the error message that should be displayed
to the user if an error occurs. After each line of code that may cause an
error, the clunky:
If Err.Number <> 0 then
Display an error message
End If
can be replaced with:
CheckForError strErrorMessage
The CheckForError subroutine contains fairly straightforward code, simply checking
to see if Err.Number is nonzero. If it is, strErrorMessage is displayed and
execution is halted with Response.End. Example 3-3 contains the source code for
CheckForError.
Unfortunately, this subroutine needs to exist in all of your ASP pages that use
scripting-language error handling. Of course, server-side includes should be used
so that only one instance of this function is needed. In the later section “Responding
to Exceptions,” we look at how to improve this subroutine to include the ability
to notify the support team of the exception’s details.
A list of VBScript runtime error numbers can be found on Microsoft’s
scripting site at: http://msdn.microsoft.com/scripting/default.htm?/
scripting/vbscript/doc/vsmscRunTimeErrors.htm.
VBScript syntax error numbers are listed at: http://msdn.microsoft.com/
scripting/default.htm?/scripting/vbscript/doc/vsmscSyntaxErrors.htm.
Example 3-3. The CheckForError Subroutine Allows for More Streamlined Error-Handling
Code in VBScript
Sub CheckForError(strErrorMessage)
If Err.Number <> 0 then
'An error occurred! Display the error message
Response.Write strErrorMessage
Response.End
End If
End Sub
Detecting When Exceptions Occur 51
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Detecting errors with JScript
JScript’s error handling is similar to C++’s and Java’s, using try ... catch blocks.
Error handling is fairly new to JScript; it’s only available in the JScript scripting
engine Version 5.0 or greater. To download the free, latest version of the JScript
and VBScript scripting engines, direct your browser to http://msdn.microsoft.com/
scripting.
If you have IIS 5.0 installed as your web server, the Version 5.0
scripting engines are already installed on your computer. If you are
using IIS 4.0, you may have an older scripting engine version, and
should upgrade your scripting engine at Microsoft’s Scripting Site
(http://msdn.microsoft.com/scripting).
In VBScript, error handling is enabled with the On Error Resume Next command.
If an exception is raised, the Err object is populated with the error information. To
detect an error, after each line of code that may be guilty of causing an error, a test
is performed, checking to see if Err.Number is nonzero. In the earlier section
“Detecting errors with VBScript,” these actions were likened to an inefficient manager,
continually checking on an employee to see if he had performed some erroneous
task.
JScript’s error handling is a bit different. In JScript, there is no single command
(like VBScript’s On Error Resume Next) that indicates that error handling should
be enabled. Furthermore, JScript’s error handling takes a more direct approach
than VBScript’s. To relate it to the manager/employee analogy used earlier, rather
than asking our employee if an error occurred after he has attempted a task, we
cautiously tell our employee to try a certain set of tasks. The employee then
attempts to perform these tasks, and gets back to us if any of these tasks caused
an error.
To let our dutiful employee know what tasks to perform, a try block is used,
which has the following syntax:
try {
possibly erroneous statements
}
A try block must be immediately followed by a catch block. A catch block contains
code that is executed only if an error occurred in the commands within the
preceding try block.
catch (errorObjectInstance) {
error handling code
}
If an error occurs in one of the commands issued within a try block, the details
of the error are encapsulated within an Error object. An instance of this object is
52 Chapter 3: Exception Handling
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
assigned to the variable specified by errorObjectInstance immediately following
the catch statement in the catch block.
The JScript Error object is similar to VBScript’s Err object, but only contains two
properties and no methods. The two properties of the Error object can be seen in
Table 3-2.
The way Microsoft chose to represent their error numbers is, in my opinion, a little
confusing. The easiest way to pick apart an error number is to look at it in
hexadecimal format. All error codes you’ll receive through an ASP page will have
the following hexadecimal format:
800FNNNN
The F denotes the facility code, which indicates who raised the error. VBScript and
JScript use a facility code of 10 (A, in hex). The NNNN represents the actual error
number. Therefore, using JScript’s number property, to get the NNNN portion of the
32-bit error number, use the following:
try {
some code...
} catch (err) {
// display the error number, less the 800F
Response.Write("The error generated the following error code: ");
Response.Write(err.number & 0xFFFF);
}
Likewise, to obtain just the facility code, use the following:
try {
some code...
} catch (err) {
// display the error number, less the 800F
Response.Write("The error generated the following facility code: ");
Response.Write(err.number >> 16 & 0xF);
}
Table 3-2. The JScript Error Object Contains Two Properties
Property Description
description A string containing detailed information on the exception.
Number A 32-bit number containing both a facility code and the actual error
number. The upper 16 bits are the facility code, while the lower 16 bits
are the error number.
Detecting When Exceptions Occur 53
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Extracting Bits
JScript provides a number of bitwise operators that perform bit-level manipulation.
The bitwise operators supported by JScript include &, |, ~, ^, >>, and
<<. Respectively, these operators perform bitwise AND, bitwise OR, bitwise NOT,
bitwise XOR, bitwise shift-right, and bitwise shift-left.
Bitwise operators perform operations on bits. The bitwise AND, OR, and XOR
operators are binary operators, taking two bits as arguments and returning a
single bit. The bitwise AND operator, which is used to extract the error number
and facility code, returns a 1 if both input bits are 1; otherwise, the AND operator
returns a 0. Due to this property, the bitwise AND operator is often used to
extract particular bits from a numeric variable. For example, if we had an integer
with the bitwise value 01100011, we could extract bits 0 through 3 by bitwise
ANDing that value with 00001111. The resulting integer would be
00000011, containing the value of bits 0 through 3 of the integer.
Therefore, to extract bits 0 through 15 of the 32-bit error number, we bitwise
AND the error number with 1111111111111111, or FFFF in hexadecimal. We
examined how to do this in JScript:
err.number & 0xFFFF
The shift-right bitwise operator shuffles around the bits in an integer, shifting
the bits to the right. For example, if we have an integer with the value
01100011 and shift-right the bits by four bits (denoted 01100011 >> 4), the
result would be 00000110. Note that the four rightmost bits were truncated,
and the four leftmost bits were moved over four spots. When extracting the
facility code, we first need to shift-right the error number by 16 bits, since the
facility code resides in bits 16 through 19. Once this has been completed, a
bitwise AND is used to extract the 4 bits that make up the facility code. For
example:
// Assume the error number is 800XYYYY, where X is the facility code and
// YYYY is the actual error number. The error number, which is a 32-bit
// number can be represented in binary with:
// 100000000000XXXXYYYYYYYYYYYYYYYY.
// The XXXX are the four bits that represent the facility code, while the
// string of Ys represents the 16-bit error number. To extract the Ys,
// a bitwise AND with the 32-bit error number and a 16-bit integer
// of 1s (FFFF)
var errorNumber = err.number & 0xFFFF;
// To grab the facility code, the XXXX must first be shifted over to
// become the rightmost bits.
var facilityNumber = err.number >> 16;
—Continued—
54 Chapter 3: Exception Handling
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
A list of JScript runtime error numbers can be found on Microsoft’s
scripting site at http://msdn.microsoft.com/scripting/default.htm?/
scripting/jscript/doc/jsmscRunTimeErrors.htm.
JScript syntax error numbers are listed at http://msdn.microsoft.com/
scripting/default.htm?/scripting/jscript/doc/jsmscSyntaxErrors.htm.
The try block can contain several lines of code; as soon as an error occurs, the
corresponding catch block is visited. Therefore, a separate set of try ... catch
blocks should be used for each statement that might generate an exception. As
discussed in “Deciding when to check if an exception occurred,” method calls to
COM objects are the usual suspects for inconsistent exceptions; therefore, a try ...
catch block should exist for each COM object method call.
Recall that VBScript’s Err object contains a Raise method, which, as its name
implies, raises an error. JScript’s Error object does not contain any such method; to
raise an error in JScript, use the throw statement. throw has the following syntax:
throw errorObjectInstance;
errorObjectInstance needs to be an Error object instance with its number and
description properties already set. Example 3-4 demonstrates how to use the
throw statement.
// Now, facilityNumber equals:
// 0000000000000000100000000000XXXX
// A bitwise AND with a 4-bit integer containing
// 4 1s is needed (F, in hexadecimal)
facilityNumber = facilityNumber & 0xF;
Example 3-4. The throw Statement Raises an Error
<% @LANGUAGE="JSCRIPT" %>
<%
function ThrowError()
{
// throw an error...
var err;
err = new Error(); // Create an Error object instance
err.description = "Egad, an error!";
err.number = 0x800A1234;
throw err; // throw the error (similar to Err.Raise)
}
// try calling the function ThrowError
try {
ThrowError();
Detecting When Exceptions Occur 55
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Nested try ... catch blocks
try ... catch blocks can be nested within one another. For example, the following
is perfectly legal syntax:
try { // outer try block
try { // inner try block
// something
} catch (err) {
// this catch block is only visited if an error occurs in the
// code within the inner try block
}
} catch (err) {
// this catch block is only visited if an error occurs in the
// code within the outer try block
}
While explicit try ... catch block nesting isn’t often used, implicit try ... catch
blocks are quite common. Example 3-5 illustrates a pair of try ... catch blocks
nested implicitly. The inner try ... catch block is within the function foobar; the
outer try ... catch block contains the code that calls foobar.
In Example 3-5, the catch block in the outer try ... catch block will never be
visited, since if any error occurs in foobar, the inner catch block will respond.
There may be times, however, when the inner try ... catch block in foobar might
} catch (err) {
// If we've reached here, an error occurred. Display the error info...
Response.Write("An error occurred! Here are the following details:");
Response.Write("<P><UL>Description: " + err.description);
Response.Write("<BR>Error Number:</B> " + (err.number & 0xFFFF));
Response.Write("<BR>Facility Code:</B> " + (err.number >> 16 & 0xF));
}
%>
Example 3-5. try ... catch Blocks Can Be Implicitly Nested
function foobar()
{
// this is the inner try ... catch block
try {
// execute some code that might throw an exception...
} catch (err) {
// handle an error
}
}
// this is the outer try ... catch block
try {
foobar();
} catch (err) {
// handle any errors thrown by foobar()
}
Example 3-4. The throw Statement Raises an Error (continued)
56 Chapter 3: Exception Handling
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
not want to respond to an error. For example, when creating reusable scripts using
JScript, there may be instances when you don’t want your reusable script to be
responsible for handling all possible errors. In Chapter 6, Database Reuse, we’ll
examine a reusable script that creates generic database administration pages.
These scripts work tightly with a database that is specified by the end developer
(that is, the developer who is utilizing these reusable scripts). If the end developer
specifies a database connection string that is invalid, it might be best to let
the end developer handle the error rather than attempting to have the reusable
script do so. The end developer might want to try a different connection string,
perhaps, and if passed the error, might be able to recover from it.
To hand off the responsibility of handling an error, simply re-throw the error in
the catch block. In the following code snippet, the catch block is only interested
in working with a specific error, 0x800A1234; all other errors are passed on:
try {
some code that might throw an exception
} catch (err) {
// only handle error 0x800A1234
if (err.number == 0x800A1234) {
handle the error
} else {
// pass off the responsibility to the outer try ... catch block
throw err;
}
}
Example 3-6 demonstrates how the responsibility of handling an error can be
passed from an inner try ... catch block to an outer one.
Example 3-6. To Pass Off the Responsibility of Handling an Exception from an Inner to an
Outer try ... catch Block, Re-throw the Exception
<% @LANGUAGE="JSCRIPT" %>
<%
function foobar()
{
// foobar tries to connect to a database. If it cannot because of an
// an invalid data source name, the catch block re-throws the error, passing
// off the responsibility to the outer catch block...
try {
// attempt to connect to the database...
var objConn;
objConn = Server.CreateObject("ADODB.Connection");
objConn.ConnectionString = "DSN=NonExistentDSN";
objConn.Open();
} catch(err) {
if ((err.number & 0xFFFF) == 16389)
throw err; // pass off the responsibility
else {
// not a connection problem... handle the error
}
Detecting When Exceptions Occur 57
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
VBScript’s error handling versus JScript’s error handling
Now that we’ve examined both VBScript and JScript’s error-handling mechanisms,
which one is best? I find JScript’s implementation of error handling to be much
more elegant and readable than VBScript’s. Since VBScript’s error-handling
approach requires the developer to inquire whether an exception occurred or not
after the fact, the approach is more error-prone that JScript’s, where the developer
must set up the proper error-handling calls prior to executing the questionable
statements.
I also find the JScript syntax far more elegant than VBScript’s error-handling syntax.
Of course, to enjoy the benefits of JScript error handling, you do have to use
JScript as the server-side scripting language, something only a handful of ASP
developers do.
Detecting Exceptions with the ASPError Object
As veteran web surfers, we’ve all stumbled across a broken link, which took us to
a nonexistent web page. When clicking on a broken link, an HTTP 404 error is
received. This error is generated by the web server when it cannot find a
requested document.
Whenever something goes awry deep within a web server, an HTTP error is generated.
These HTTP errors are three digits in length, ranging from 100–505. The
first number in the three-number HTTP error code denotes the type of HTTP error.
400-level errors are client errors, such as 404 (file not found), 401 (unauthorized),
}
}
// try calling the function ThrowError
try {
// Execute foobar; if there is a problem, the catch block will be visited
foobar();
} catch (err) {
// at this point, we could check to see if the DSN name was bad, and, if
// so, retry foobar or call some other function
// Instead, for this example, we'll just output the error information
Response.Write("An error occurred! Here are the following details:");
Response.Write("<P><UL><B>Description:</B> " + err.description);
Response.Write("<BR>Error Number:</B> " + (err.number & 0xFFFF));
Response.Write("<BR>Facility Code:</B> " + (err.number >> 16 & 0xF));
}
%>
Example 3-6. To Pass Off the Responsibility of Handling an Exception from an Inner to an
Outer try ... catch Block, Re-throw the Exception (continued)
58 Chapter 3: Exception Handling
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
and 414 (request URL too long). 500-level errors indicate intrinsic web server
errors. As we’ll see shortly, IIS 5.0 uses a 500-level HTTP error to signal that an
error has occurred on an ASP page.
A complete listing of HTTP error codes and their associated meanings
can be seen at http://www.webpartner.com/html/AlertsandErrors.htm.
Prior to ASP 3.0, exception handling on an ASP page was limited to the exceptionhandling
capabilities of the scripting language used. Due to the total lack of error
handling in JScript (until Version 5.0) and the verbose nature of error handling in
VBScript, ASP was long overdue for a simpler and more effective exceptionhandling
approach. With the release of ASP 3.0 (available exclusively with IIS 5.0
and Windows 2000), a new intrinsic object, ASPError, was added to assist in
exception handling. The ASPError object contains nine properties, which are
detailed in Table 3-3.
When an error occurs in IIS 5.0, whether it’s from a COM object, the scripting language,
or an internal ASP error, an HTTP 500;100 error is generated. (The 100 indicates
that it is a special kind of HTTP 500 error.) Whenever an HTTP error occurs,
IIS checks an error table to see what it should do in response to the error. Its
choices are limited to:
• Sending a default message to the client
• Sending the client the contents of a particular file (usually an HTML file)
• Redirecting the client to an error-handling web page
Table 3-3. The Intrinsic ASPError Object Assists in Detecting and Handling Exceptions
Property Description
ASPCode Returns a rather detailed error code that is generated by IIS
Number Returns the error number
Source Returns the line of code that caused the error, if available
Category Indicates what type of error occurred; this property will indicate if
the exception was caused by the scripting language, an external
COM component, or an internal ASP error
File Returns the name of the ASP page that generated the error
Line Returns the line number the error occurred on
Column Returns the column the error occurred on
Description Returns a detailed description of the error
ASPDescription Returns a detailed description for those errors that fall under the
internal Active Server Pages error Category
Detecting When Exceptions Occur 59
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Through the Internet Information Services, a developer can specify how IIS should
handle particular HTTP errors. Start by opening the Internet Information Services
(in Windows 2000, go to Start . Programs . Administrative Tools . Internet Services
Manager). Right-click on the web site whose HTTP errors you wish to modify
and choose Properties. The Web Site Properties dialog box will open; the HTTP
error-handling information is under the Custom Errors tab. Figure 3-2 shows a listing
of these HTTP errors for IIS on my Windows 2000 machine.
Notice that in Figure 3-2 the 500;100 HTTP error redirects the user to the URL
/CODEREUSE/Error.asp. To adjust your settings for this HTTP error, select the particular
error and click the Edit Properties button. You should now be presented
with the Error Mapping Properties dialog box that contains a list box and a text
box. The list box contains the three various options IIS can take when an HTTP
error occurs: send the default message to the client (Default), send the client a file
(File), or redirect the client to a URL (URL). Figure 3-3 contains a screenshot of the
Error Mapping Properties dialog box for the HTTP 500;100 error.
With the settings shown in Figure 3-3, whenever an internal web server error
occurs in an ASP page, the HTTP 500;100 error will be tripped. This will send the
Figure 3-2. The Custom Errors tab in the Web Site Properties dialog box specifies how IIS
responds to various HTTP errors
60 Chapter 3: Exception Handling
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
user to /CODEREUSE/Error.asp via an implicit Server.Transfer. Figure 3-4
graphically depicts this branch in execution.
When an error occurs on an ASP page, not only is a HTTP 500;100 error generated,
but the details of the error are packaged into an ASPError object instance.
This particular ASPError object instance can be retrieved in the error-handling web
page using the GetLastError method of the Server object. Example 3-7 contains the
source code for our error-handling web page, /CODEREUSE/Error.asp.
Figure 3-3. The Error Mapping Properties dialog defines how IIS handles particular server
errors
Figure 3-4. When an error occurs on an ASP page, an HTTP 500;100 error is generated,
branching execution to /CODEREUSE/Error.asp
SomeAspPage.asp
If any given ASP web page on
the site causes an exception
(either through invalid scripting
language syntax, through an
internal ASP error, or through an
external COM component), an
HTTP Error 500;100 is generated,
and control is branched to the
error-handling page indicated in
the Custom Errors tab in the
Internet Services Manager.
/CODEREUSE/Error.asp
When an error occurs, this page is
implicitly called via a Server.Transfer.
This page's responsibility is to output
the error message in a readable, friendly
format for the end user.
When an error occurs, control is
immediately branched to the errorhandling
web page (in our example,
/CODEREUSE/Error.asp).
Detecting When Exceptions Occur 61
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Example 3-7. The Error-Handling Web Page Obtains Error Information Using
Server.GetLastError
<% @LANGUAGE = "VBSCRIPT" %>
<% Option Explicit %>
<%
'Grab the instance of the ASPError object that contains info on
'the error that brought us to this error-handling web page
Dim objError
Set objError = Server.GetLastError()
%>
<FONT SIZE=+1><B>D'oh!</B></FONT><HR NOSHADE>
An unexpected error occurred. We apologize for this inconvenience.
<P>

<TABLE ALIGN=CENTER BORDER=1 CELLSPACING=1>
<TR><TH COLSPAN=2>Detailed Error Information</TH></TR>
<TR>
<TH>Property</TH>
<TH>Value</TH>
</TR>
<TR>
<TD>ASPCode</TD>
<TD><%=objError.ASPCode%></TD>
</TR>
<TR>
<TD>Number</TD>
<TD><%=objError.Number%></TD>
</TR>
<TR>
<TD>Source</TD>
<TD><%=objError.Source%></TD>
</TR>
<TR>
<TD>Category</TD>
<TD><%=objError.Category%></TD>
</TR>
<TR>
<TD>File</TD>
<TD><%=objError.File%></TD>
</TR>
<TR>
<TD>Line</TD>
<TD><%=objError.Line%></TD>
</TR>
<TR>
<TD>Column</TD>
<TD><%=objError.Column%></TD>
</TR>
<TR>
<TD>Description</TD>
<TD><%=objError.Description%></TD>
</TR>
62 Chapter 3: Exception Handling
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
The error-handling code in Example 3-7 uses the Server.GetLastError method to
obtain information on the error that caused the HTTP 500;100 error. Next, the
properties of the ASPError object are displayed in an HTML table. As we discussed
earlier, the end user shouldn’t be bothered with the technical details of the
error, and usually we wouldn’t want to display this detailed information.
As illustrated in Table 3-3, one of the ASPError object’s properties is Category,
which indicates which one of three categories the error falls under. The three possible
error categories are:
• Scripting language error
• Internal ASP error
• External COM error
For the remainder of this section, we will examine code snippets that generate an
error in a unique category. After each code snippet, we’ll examine the output generated
by the error-handling script in Example 3-7.
Scripting language errors
A scripting language error occurs when code that violates the scripting language’s
syntax is encountered. For example, in VBScript an If statement has the following
syntax:
If condition then
Statements
[ElseIf condition then
ElseIf statements]
[Else
Else statements]
End If
Or
If condition then Statements [Else Else Statements]
Writing an If statement that violates this expected syntax will result in a scripting
language error. Example 3-8 contains the source code to ScriptingLanguageError.
asp, which is a script with an illegal If statement. When ScriptingLanguageError.
asp is visited through a browser, the user is automatically redirected to
<TR>
<TD>ASPDescription</TD>
<TD><%=objError.ASPDescription%></TD>
</TR>
</TABLE>
<%
Set objError = Nothing 'Clean up...
%>
Example 3-7. The Error-Handling Web Page Obtains Error Information Using
Server.GetLastError (continued)
Detecting When Exceptions Occur 63
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
/CODEREUSE/Error.asp. A screenshot of a browser visiting ScriptingLanguageError.
asp can be seen in Figure 3-5.
Detecting compile-time errors with the ASPError object
Unlike the scripting language error-handling methods, the ASPError object can be
used to detect compile-time errors. This is evident from the code in Example 3-8,
which generates a compile-time error. Note that, according to the Category property
value listed in Figure 3-5, an improperly formatted If statement results in a
Example 3-8. A C-Style If Statement Generates a Scripting Language Error When Using
VBScript
<% @LANGUAGE="VBSCRIPT" %>
<% Option Explicit %>
<%
if (1 = 2)
Response.Write "What!? 1 = 2?"
%>
Figure 3-5. A scripting language error produces these values in the ASPError object’s
properties
64 Chapter 3: Exception Handling
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
“Microsoft VBScript compilation” error. This makes sense, since illegal syntax generates
compile-time errors.
Internal ASP errors
When creating ASP pages, there are certain guidelines that must be followed
regardless of the scripting language used. For example, server-side includes must
have a certain syntax and must reference an existing file; when using the <% and
%> script delimiters, these delimiters must match (that is, for every <% there must
be a corresponding %>).
Internal ASP errors are a lower level of errors than scripting language errors.
Scripting language errors are dependent upon the syntax of the particular scripting
language used. On the other hand, internal ASP errors are scripting languageindependent.
When an internal ASP error occurs, the ASPDescription property of
the ASPError object contains extra information further describing the error.
Example 3-9 contains the source code to InternalASPError.asp, a script that tries to
include a file that does not exist. Figure 3-6 contains the output of the script when
viewed through a browser. Note that the ASPDescription property contains information
that was not present in the scripting language error case (Figure 3-5).
External COM object errors
One of the reasons ASP is such a powerful tool for creating dynamic web pages is
its ability to utilize COM components. COM components, like ADO objects, may
raise errors if used improperly, or if an external exception is encountered. If such
an error occurs in a COM component, the ASPError object classifies it as an external
COM object error. The Category property is set to the programmatic identifier
of the offending COM component.
Example 3-10 contains the source code for COMError.asp, a script that raises an
external COM object error due to the fact that it attempts to close an ADO Connection
object before it has been opened. The output of COMError.asp, when visited
through a browser, can be seen in Figure 3-7.
Example 3-9. Using a Server-Side Include to Import the Contents of a Nonexistent File
Generates an Internal ASP Error
<% @LANGUAGE="VBSCRIPT" %>
<% Option Explicit %>
<%
'An internal ASP error will be generated since we are attempting to include
'a nonexistent ASP page...
%>

Responding to Exceptions 65
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Responding to Exceptions
In the earlier section “Detecting When Exceptions Occur,” we examined two methods
to determine when exceptions occurred: through the server-side scripting
Figure 3-6. An internal ASP error produces these values in the ASPError object’s properties
Example 3-10. Attempting to Close a Closed Connection Object Results in an External COM
Object Error
<% @LANGUAGE="VBSCRIPT" %>
<% Option Explicit %>
<%
'The following code snippet will generate a COM object error, since we
'can't close a database connection if it's already closed!
Dim objConn
Set objConn = Server.CreateObject("ADODB.Connection")
objConn.Close
%>
66 Chapter 3: Exception Handling
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
language, and for ASP 3.0, through the ASPError object. Exception handling
involves more than just detecting when an exception occurs; it also involves an
appropriate response. In this section, we’ll look at various ways to respond to an
exception.
We’ll start with a look at displaying understandable error messages. Next, a means
for notifying the technical support team of the error’s details is covered. Finally,
we’ll look at ways to recover from an error. Of course, these three approaches to
responding to exceptions are not mutually exclusive. That is, when an exception
occurs, you can display a readable error message, notify the support team, and
attempt to recover from the error.
Displaying an Understandable Error Message
Errors are inevitable. They will happen. Your code contains errors. Therefore, be
ready to respond to errors gracefully. Do not trivialize exception handling, expecting
your code to be error-free. The worst thing that can happen when an error
occurs is premature termination of the script followed by an illegible error
Figure 3-7. An external COM object error produces these values in the ASPError object’s
properties
Responding to Exceptions 67
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
message. As we discussed earlier, it is vitally important to provide readable, understandable
error messages for your users.
Good error messages, in my opinion, contain the following four elements:
• Acknowledgement that an error occurred.
• A high-level explanation of the error.
• An apology.
• Links to other sections on a site. Links are how the Web is navigated. An error
message without links to other site resources serves as a dead end, encouraging
the user to leave your site altogether.
For example, a good error message for a database connection exception would be
something like:
“Ah! We could not display the information you requested due to a temporary database
problem. Our development team has been notified of this problem and is
working diligently to resolve it as quickly as possible. We apologize for this inconvenience.
In the meantime, why not check out these sections on our web site?”
Here a list of applicable links would follow.
Personally, I find that good error messages usually contain the four elements listed
above. However, I am no expert on readable error messages. For a more in-depth
discussion on creating error messages, be sure to check out O’Reilly’s Developing
Windows Error Messages by Ben Ezzell.
Notifying the Support Team When an Error Occurs
Good error messages let the user know the support team has been made aware of
the problems. By informing the support team when an error occurs, bugs can be
quickly resolved. When an error occurs, a plethora of information is available,
either through the ASPError object or the VBScript Err object. This technical information
can be easily sent to the support team via email using one of the common
free ASP email components.
Notifying the support team with scripting language
error-handling techniques
In the earlier section “Streamlining VBScript error-handling code” we looked at a
subroutine, CheckForError, which contained code that checked to see if Err.
Number was nonzero. If Err.Number was nonzero, indicating that an error had
occurred, the appropriate error message was displayed. Let’s look at how this
function can be expanded to include an error report email to be sent automatically
to the support team.
For this example, we’ll use the CDONTS email component. CDONTS is a free,
lightweight email component from Microsoft that is shipped with IIS. There are
68 Chapter 3: Exception Handling
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
plenty of good articles around the Net with information on how to use CDONTS.
One such article, “Sending Emails in ASP Using CDONTS,” can be found at: http://
www.4guysfromrolla.com/webtech/112298-1.shtml.
Example 3-11 contains the new code for CheckForError. The error report contains
the Description and Number properties from the Err object, as well as the time the
error occurred, and the complete set of ServerVariables. This combined information
should be enough to get the support team started on fixing this error.
Since error handling is turned on, if an error occurs in sending the
email message (for example, if CDONTS is not installed on the web
server), the support team will never know about errors encountered
by the user. For this reason, make sure CDONTS works properly on
your web server before using it in this context.
Example 3-11. CheckForError Automatically Emails an Error Report to the Support Team in the
Event of an Error
Sub CheckForError(strErrorMessage)
If Err.Number <> 0 then
'An error occurred! Display the error message
Response.Write strErrorMessage
'Send error-report email to the support team
Dim objCDO
Set objCDO = Server.CreateObject("CDONTS.NewMail")
objCDO.To = "support@yourdomain.com"
objCDO.From = "support@yourdomain.com"
objCDO.Subject = "Error Report: Error Generated at " & Now()
objCDO.Importance = 2
Dim strBody, strName
strBody = "An error occured on " & Request.ServerVariables("SCRIPT_NAME") & _
" at " & Now() & vbCrLf & vbCrLf & "Description: " & _
Err.Description & vbCrLf & "Number: " & Err.Number & _
vbCrLf & vbCrLf
'Display the ServerVariables
For Each strName in Request.ServerVariables
strBody = strBody & strName & " - " & Request.ServerVariables(strName) & vbCrLf
Next
objCDO.Body = strBody
objCDO.Send 'Send the email!
Set objCDO = Nothing 'Clean up...
Response.End
End If
End Sub
Responding to Exceptions 69
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Here is a shortened example of the error report:
An error occured on /CheckForError.asp at 4/5/2000 2:38:18 PM
Description: Division by zero
Number: 11
ALL_HTTP - HTTP_ACCEPT:image/gif, image/x-xbitmap, image/jpeg,
image/pjpeg, application/vnd.ms-excel, application/msword, */*
HTTP_ACCEPT_LANGUAGE:en-us
HTTP_CONNECTION:Keep-Alive
HTTP_HOST:localhost
HTTP_USER_AGENT:Mozilla/4.0 (compatible; MSIE 5.01; Windows NT 5.0)
HTTP_ACCEPT_ENCODING:gzip, deflate
ALL_RAW - Accept: image/gif, image/x-xbitmap, image/jpeg, image/pjpeg,
application/vnd.ms-excel, application/msword, */*
Accept-Language: en-us
Connection: Keep-Alive
Host: localhost
User-Agent: Mozilla/4.0 (compatible; MSIE 5.01; Windows NT 5.0)
Accept-Encoding: gzip, deflate
' ... error report shortened for brevity ...
A similar approach for sending error-report email can be used with JScript. Either
use the CDONTS component in the catch block, or call a function similar to
CheckForError in the catch block.
Notifying the support team with the ASPError object
Since the ASPError object contains more properties than the Err object, using the
ASPError object approach will generate a more detailed error report for the support
team. To send email to the support team, code similar to that in Example 3-11
will be used. Of course, the email code will exist in the error-handling web page,
/CODEREUSE/Error.asp. Example 3-12 contains a new version of /CODEREUSE/
Error.asp, one that will automatically send an error report to the support team.
(The code that displayed the error information to the user has been removed for
brevity. To see this code, consult Example 3-7.)
Example 3-12. /CODEREUSE/Error.asp Shoots an Error Report to the Support Team
<% @LANGUAGE = "VBSCRIPT" %>
<% Option Explicit %>
<%
'Grab the instance of the ASPError object that contains info on
'the error that brought us to this error-handling web page
Dim objError
Set objError = Server.GetLastError()
'Send error-report email to the support team
Dim objCDO
Set objCDO = Server.CreateObject("CDONTS.NewMail")
70 Chapter 3: Exception Handling
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Recovering from an Error
Ideally, an application would be smart enough to analyze an error when it happens,
take some set of actions to sidestep the error, and continue processing. Of
course compile-time errors can’t be recovered from (it’s hard to recover from illegal
syntax), but there are several runtime errors that might have viable
workarounds that allow for the script to continue processing.
Recovering from an error is not possible using the ASPError method, since when
an error occurs, script execution is abruptly shifted from the erroneous ASP page
to the error-handling ASP page. When setting up a web site to redirect control to
the error-handling page when an HTTP 500;100 error occurs, the change
objCDO.To = "support@yourdomain.com"
objCDO.From = "support@yourdomain.com"
objCDO.Subject = "Error Report: Error Generated at " & Now()
objCDO.Importance = 2
Dim strBody, strName
strBody = "An error occured on " & Request.ServerVariables("SCRIPT_NAME") & _
" at " & Now() & vbCrLf & vbCrLf
'Display the ASP Error properties
strBody = strBody & "ASPCode: " & objError.ASPCode & vbCrLf
strBody = strBody & "Number: " & objError.Number & vbCrLf
strBody = strBody & "Source: " & objError.Source & vbCrLf
strBody = strBody & "Category: " & objError.Category & vbCrLf
strBody = strBody & "File: " & objError.File & vbCrLf
strBody = strBody & "Line: " & objError.Line & vbCrLf
strBody = strBody & "Column: " & objError.Column & vbCrLf
strBody = strBody & "Description: " & objError.Description & vbCrLf
strBody = strBody & "ASPDescription: " & objError.ASPDescription & vbCrLf
strBody = strBody & vbCrLf & vbCrLf
'Display the ServerVariables
For Each strName in Request.ServerVariables
strBody = strBody & strName & " - " & Request.ServerVariables(strName) & vbCrLf
Next
objCDO.Body = strBody
objCDO.Send 'Send the email!
Set objCDO = Nothing 'Clean up...
%>

<%
Set objError = Nothing 'Clean up...
%>
Example 3-12. /CODEREUSE/Error.asp Shoots an Error Report to the Support Team (continued)
Responding to Exceptions 71
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
permeates to all web pages within the site. However, you can disable this on a
page-by-page basis by turning on the scripting language’s error handling.
For example, if you have successfully set up IIS to redirect to an error page upon
HTTP 500;100 errors, the following ASP script, when viewed through a browser,
will display any output from the error-handling page:
<% @LANGUAGE="VBSCRIPT" %>
<% Option Explicit %>
<%
Dim iSomeVariable
iSomeVariable = iSomeOtherVariable 'This will generate a runtime error since
'iSomeOtherVariable was not explicitly
'declared (and Option Explicit was stated)
%>
However, if the scripting language error handling is turned on (as it is in the code
snippet below), the error-handling page won’t be visited:
<% @LANGUAGE="VBSCRIPT" %>
<% Option Explicit %>
<%
On Error Resume Next 'The error-handling page will not be called
'since the script language error handling is
'enabled.
Dim iSomeVariable
iSomeVariable = iSomeOtherVariable 'This will generate a runtime error
since
'iSomeOtherVariable was not explicitly
'declared (and Option Explicit was stated)
%>
To designate certain portions of your code to be handled by the scripting language
error handling and other portions to be handled by the ASPError object
approach, simply enable and disable the scripting language error handling at will.
To enable error handling in VBScript, use On Error Resume Next;
to disable error handling, use On Error Goto 0. In JScript, there is
no “enabling” error handling. Code within a try block is handled by
the scripting language, while code outside of a try block is handled
by the ASPError object approach.
As previously discussed, ideally a program could recover from any runtime error.
However, creating such an intelligent, error-fixing program is next to impossible,
and far beyond the scope of this book. It is not so far-fetched, though, to assume
we can recover from a certain set of defined errors. A classic example is a backup
or mirror database system. If the primary database is unreachable for some reason,
rather than displaying a database connection error, an attempt can be made
to connect to the backup database.
72 Chapter 3: Exception Handling
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Example 3-13 contains source code to recover from such an error. The code
assumes that the System DSN PrimaryDatabase specifies the primary database
and that a secondary database is available through the System DSN
SecondaryDatabase. The code in Example 3-13 is straightforward; it enables
error handling in VBScript and attempts to connect to the PrimaryDatabase. If
there is an error in the connection, it tries to connect to the SecondaryDatabase.
If either database connection attempt is successful, error handling is disabled so
any other errors in the script will trip the HTTP 500;100 error and redirect the user
to the error-handling page.
When attempting to recover from an error, detailed information must be known
about the recovery path. Recovering from errors is rarely an easy task. However,
recovering from an error is the best course of action an application can take when
an error occurs.
Example 3-13. If You Have a Backup or Secondary Database, You Can Recover Nicely from
Connection Errors to the Primary Database
<% @LANGUAGE="VBSCRIPT" %>
<% Option Explicit %>
<%
'Enable error handling
On Error Resume Next
Dim objConn
Set objConn = Server.CreateObject("ADODB.Connection")
objConn.ConnectionString = "DSN=PrimaryDatabase"
objConn.Open
'Now, if there was an error, we need to try the SecondaryDatabase
If Err.Number <> 0 then
Err.Clear 'clear out the error information
objConn.ConnectionString = "DSN=SecondaryDatabase"
objConn.Open
End If
'If we have an error connecting to this database as well, then that's really bad.
If Err.Number <> 0 then
Response.Write "Both databases are off-line."
Response.End
End If
'Disable error handling, so that future errors are handled by the error-handling
'web page
On Error Goto 0
'...
%>
Creating Custom HTTP Error Pages 73
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Creating Custom HTTP Error Pages
As discussed in the earlier section “Detecting Exceptions with the ASPError
Object,” when an ASP error occurs on a web page, an HTTP 500;100 error is signaled.
If you correctly set up an error-handling web page through the Internet Services
Manager, IIS will automatically send the user to the specified error-handling
web page via a Server.Transfer.
Of course there are several other types of HTTP errors besides 500;100. One of the
most common errors is the HTTP 404 error, which results when a client requests a
page not found on the web server. There is nothing as annoying or frustrating as
the default HTTP 404 error page, which is shown in Figure 3-8.
The default 404 error page is ugly and serves as a dead end for your users. A
much better approach is to provide your visitors with a custom 404 error page that
apologizes for the page not being found and provides links to relevant sections on
the site. Relevant sections might include a search page or a listing of popular
resources on the site. The vast majority of large web sites realize the importance of
custom 404 error pages. You can see some examples of real-world custom 404
error pages at the following URLs:
• http://www.yahoo.com/NotAValidURL
• http://espn.go.com/NotAValidURL
• http://www.4guysfromrolla.com/NotAValidURL
Figure 3-8. The default 404 error page serves as a dead end, frustrating users
74 Chapter 3: Exception Handling
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Custom 404 pages, unlike custom HTTP 500;100 error pages, can be implemented
in both IIS 4.0 and IIS 5.0. As with creating a custom HTTP 500;100 error page, to
create a custom HTTP 404 error page, start by going to the Internet Services Manager.
You should see a listing of all of the web sites on your web server. Rightclick
on the web site for which you’d like to create a custom 404 error page and
choose Properties. Choose the Custom Errors tab and locate the 404 HTTP error,
as seen in the screenshot in Figure 3-9.
Next, click the Edit Properties button. You should be presented with a dialog
box similar to the one shown in Figure 3-3. Select URL from the Message Type
list box, and enter /CODEREUSE/404.asp as the URL to redirect to. Now whenever
a 404 error occurs on your site, your users will be automatically redirected
to /CODEREUSE/404.asp. When the custom 404 error page is called, it is passed
the following information in the querystring:
404;URL_That_Generated_The_404
Through this querystring, you can determine the exact URL the user tried to enter
that resulted in a 404 error. An industrious developer could create a custom 404
error page that checks the URL the user entered against valid URLs on the site.
Figure 3-9. Select the 404 HTTP Error from the Custom Errors tab in the Default Web Site
Properties dialog box
Creating Custom HTTP Error Pages 75
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
This script could rate how close the invalid URL was to various valid URLs and
automatically redirect the user to the valid URL.
Such a dynamic custom 404 error page is used at LearnASP.com, an ASP information
resource site run by Charles Carroll. The goal of Charles’s custom 404 page is
to allow ASP articles and tutorials to move directories, or to change names without
having users’ bookmarks become invalidated.
A database table keeps track of web page filenames when they are moved or
renamed. This database table’s definition can be seen in Table 3-4.
If a user enters an invalid URL, thereby causing an HTTP 404 error, the custom 404
error page is called. In the 404 error page, the invalid URL is checked against the
OldURL column in the database table. If a match is found, the Frequency column
is incremented and the user is automatically redirected to the URL specified by
NewURL.
Charles is a skilled, industrious programmer, and has added many bells and whistles
to this custom 404 error page. For example, the OldURL field can contain wildcard
fields, such as /Tutorials/*, with a corresponding NewURL of /Learn/*. That
way, if a user enters a URL that contains the /Tutorials directory (such as /Tutorials/
Lesson1.asp), he or she is automatically redirected to /Learn/Lesson1.asp.
Charles has also added numerous other noteworthy features. To
learn more about the custom 404 error page on LearnASP.com, be
sure to visit http://www.learnasp.com/how/notfoundintro.asp, which
includes a discussion on the motivation of the code, as well as the
full source!
I, however, am not as industrious as Charles! The custom 404 page presented in
Example 3-14 displays a message to the user indicating that the page could not be
found. It also sends email to the webmaster, letting her know a 404 was encountered.
The email contains information on the URL entered and the referring URL.
That way, if the 404 came via a broken link on the site, the webmaster will be
notified and able to quickly fix the broken link.
Table 3-4. A Database Table Maintains the Old and New URLs for Moved or Renamed
Web Pages
Column Name Description
OldURL The URL of the web page before it was moved or renamed
NewURL The URL of the web page after it was moved or renamed
Frequency A counter indicating how often a user requested the old URL and had to
be redirected to the new URL
76 Chapter 3: Exception Handling
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Here is an example email the webmaster might receive:
At 4/5/2000 8:09:30 PM a 404 error was encountered when the user attempted
to visit: http://localhost/NotAValidURL. The referring URL was:
http://localhost/SomeURL.asp
This automatically generated email serves as a quick locator of broken links on
one’s web site. Rather than requiring notification from users or expensive thirdparty
software to detect broken links, you can let your custom 404 error pages
indicate when a broken link is found.
Example 3-14. A Custom 404 Error Page Should Display an Informative Message
with Relevant Links
<% @LANGUAGE = "VBSCRIPT" %>
<% Option Explicit %>
<%
'Grab the URL of the 404. The Querystring is in the form:
' 404;URL
Dim strURL
strURL = Right(Request.QueryString, Len(Request.QueryString) - 4)
'Send email to the webmaster
Dim objCDO
Set objCDO = Server.CreateObject("CDONTS.NewMail")
objCDO.To = "webmaster@yourdomain.com"
objCDO.From = "webmaster@yourdomain.com"
objCDO.Subject = "404 Error Encountered on " & Now()
objCDO.Body = "At " & Now() & " a 404 error was encountered when the user " & _
"attempted to visit: " & strURL & ". The referring URL was: " & _
Request.ServerVariables("HTTP_REFERER")
objCDO.Send
Set objCDO = Nothing
%>
<HTML>
<HEAD>
<TITLE>Egad! We could not find the page you requested!</TITLE>
</HEAD>
<BODY>
<CENTER><FONT SIZE=+2><B>
We Could Not Find the Page You Requested
</B></FONT></CENTER>
<P><HR><P>
Ah! We could not find the page that you have requested! Our webmaster has been
automatically sent an email detailing this missing document. Here are some links
we think you will find helpful:
<P><UL>
<LI><A HREF="/search/">Search the Site</A><BR>
... 
</UL>
</BODY>
</HTML>
Further Reading 77
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Further Reading
While there are many good online articles on VBScript error handling, there are far
too few JScript error-handling articles. Part of the reason, of course, is error handling
wasn’t available in JScript until Version 5.0. Furthermore, there are not many
online articles discussing how to use the ASPError object for exception handling,
also due, no doubt, to its relative newness. Keeping that in mind, here are some
selected readings on the various exception-handling approaches discussed
throughout this chapter.
• For a good read on using VBScript error-handling techniques to present
understandable error messages to your visitors, check out “Error Handling in
ASP,” at http://www.4guysfromrolla.com/webtech/060399-1.shtml.
• A rather detailed discussion of JScript’s error handling can be seen on
Microsoft’s site at http://msdn.microsoft.com/workshop/languages/jscript/
handling.asp.
• On Microsoft’s site, there is a rather lengthy article on new ASP 3.0 features.
Unfortunately, only a short part of the article discusses using the ASPError
object for exception handling. The article, “Internet Information Services 5.0,”
can be found at http://www.microsoft.com/mind/0499/iis5/iis5.htm.
• The most thorough article on the ASPError object I could find was at http://
www.asptoday.com/articles/19990308.htm, in a piece entitled “Error Handling
in IIS 5.0.”
• To learn more on creating custom 404 error pages, be sure to read “Creating a
Custom 404 Error Page,” at http://www.4guysfromrolla.com/webtech/061499-1.
shtml.
78
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Chapter 4 4
Regular Expressions,
Classes, and Dynamic
Evaluation and Execution
With the introduction of the VBScript 5.0 scripting engine, VBScript developers
now have three very powerful techniques at their disposal which were previously
unavailable with VBScript:
• The availability of a Regular Expression object, RegExp. Regular expressions
provide for advanced string matching and parsing. If you are new to regular
expressions, you’ll soon realize their tremendous advantages and wonder how
you lived without them!
• The ability to create object-oriented code! VBScript now supports classes! I am
very excited about this new feature, for it allows the creation of robust and
reusable ASP pages. It’s also great for those working in large development
teams where certain developers aren’t as experienced as others with ASP. The
more experienced developers can create classes that encapsulate some of the
more difficult functionality, and the more novice developers can simply use
these classes to accomplish the common ASP tasks!
• Dynamic evaluation and execution. Dynamic evaluation allows a code snippet
contained in a string to be evaluated as though it had been entered
directly by the programmer creating the script. Dynamic execution allows a
code snippet contained in a string to be executed as through it had been
entered directly by the programmer. By employing the use of dynamic evaluation
and execution, a script can achieve flexibility not previously available.
We’ll discuss how to perform dynamic evaluations and executions in the section
“Using Dynamic Evaluation and Execution.”
If you are running ASP 3.0, you already have the VBScript 5.0 scripting engine. If
you are running an older version of ASP/IIS, be sure to download the latest
Microsoft scripting engines from: http://msdn.microsoft.com/scripting/.
Using the RegExp Object 79
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Using the RegExp Object
The RegExp object provides the VBScript engine with a means to perform regular
expression pattern matching. We’ve discussed regular expressions previously; in
Chapter 2, Choosing a Server-Side Scripting Language, a discussion on using regular
expressions in JScript and PerlScript was presented.
RegExp’s Properties
Take a moment to look back at Chapter 2 to see JScript’s implementation of regular
expressions. Note that the regular expression syntax contains a pattern and an
optional switch. The switch can have one of three values:
i Ignore case
g Perform a global search for all occurrences of pattern
gi Perform a global search for all occurrences of pattern, ignoring case
The RegExp object contains three properties that allow you to set the pattern and
these switches for regular expression usage in VBScript. These three properties are
Pattern, IgnoreCase, and Global. Pattern expects a string, and is the regular expression
pattern to search for. Global is a Boolean value indicating whether the regular
expression search should match all occurrences in a string or just the first one.
If not specified, Global defaults to False. IgnoreCase is also a Boolean value,
indicating whether or not a regular expression search is case-sensitive. By default,
IgnoreCase is set to False.
Legal Regular Expression Syntax
The Pattern property contains the regular expression. A regular expression pattern
is not restricted to just simple strings. The pattern can also contain special characters,
which allow for much more sophisticated searches. Table 4-1 contains a listing
of these special characters and their meanings.
Table 4-1. Special Characters in Regular Expression Patterns
Symbol Description
Any
alphanumeric
character
Matches the alphanumeric character(s) literally.
\ Indicates that the following character is a special character or a literal.
For example, a pattern containing “b” matches the character “b,”
while “\b” matches a word boundary.
^ Matches the beginning of a string. For example, “^A” would match
only the first “A” in “ASP is Awesome.”
$ Matches only the end of a string. For example “$d” would match the
last “d” in “Todd is mad.”
80 Chapter 4: Regular Expressions, Classes, and Dynamic Evaluation and Execution
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
\b Matches a word boundary. A word boundary exists between two characters,
where one of the characters is a word character and the other
is not. Furthermore, the beginning and end of a string are considered
word boundaries. For example, if you searched for “\bscience\b” in
“science has no conscience,” only the first word of the string would be
returned. The “science” in “conscience” would not be matched since
“science” is not preceded by a word boundary.
\B The opposite of \b. Matches any word boundary.
[abc…] Matches any single character that exists between the braces. For
example, “[aeiou]” would match the first vowel found in a string. You
can also use the hyphen for a range of characters. “[a-m]” would
match the first occurrence of a character belonging in the first half of
the alphabet.
[^abc…] Matches any single character not between the braces. For example,
“[^aeiou]” matches the first consonant found in a string. (You can also
use the hyphen to represent a range of characters.)
\w Matches any word character. A word character is one that contains an
alphanumeric character or an underscore.
\W Matches any nonword character.
\d Matches any digit. Functionally identical to [0–9].
\D Matches any nondigit.
\s Matches any space character (including a space, a newline character,
a carriage return, or a tab).
\S Matches any nonspace character.
. Matches any character other than \n: functionally identical to [^\n].
\n Matches a newline character.
\r Matches a carriage return.
\t Matches a tab.
{n} Matches exactly n occurrences of a regular expression. For example,
“\w{10}” matches 10 consecutive word characters.
{n,} Matches n or more occurrences of a regular expression. For example,
“\d{2,}” matches two or more consecutive digits.
{n,m} Matches between n and m occurrences of a regular expression. For
example, “\w{2,4}” matches either two, three, or four consecutive
word characters.
? Matches zero or one occurrences; functionally identical to {0,1}. For
example, “\w\d?” matches a word character followed by zero or one
digits.
* Matches zero or more occurrences; functionally identical to {0,}.
+ Matches one or more occurrences; functionally identical to {1,}.
( ) Used to group a series of symbols. For example, “xyz?” matches “xy”
and “xyz,” while “x(yz)?” matches “x” and “xyz.”
| Matches either one of two groups. For example, “(Scott)|( James)”
matches “Scott” or “James.”
Table 4-1. Special Characters in Regular Expression Patterns (continued)
Symbol Description
Using the RegExp Object 81
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
To search for a literal that is also used as a special symbol, precede
the literal with a backslash. For example, to match a question mark
character in a string, use the pattern “\?”, since the question mark is
a special symbol for regular expressions.
A regular expression can contain any number of the special symbols and literals
listed in Table 4-1. Regular expressions provide a powerful tool for validating
input. For example, imagine that a user will enter his social security number. The
expected format is ###-##-####. This can be validated with the regular expression:
\d{3}-\d{2}-\d{4}
which checks for three digits, followed by a dash, followed by two digits, followed
by a dash, followed by four digits. A regular expression that could validate
a phone number in either ###-###-#### or (###) ###-#### format could be:
(\d{3}-\d{3}-\d{4})|($\d{3}$ \d{3}-\d{4})
Note that when matching a literal left or right parenthesis, the parenthesis needs to
be preceded by a backslash.
RegExp’s Methods
The RegExp object contains three methods: Test, Replace, and Execute. The first
method, Test, accepts one parameter, which is the string to apply the regular
expression to. If a match is found, Test returns True; otherwise, it returns False.
For example, if you asked a user to input her name, you want to make sure only
letters, apostrophes, and hyphens exist within the person’s name. As Example 4-1
illustrates, you can use the Test method to quickly determine whether the name
entered by the user consists of any characters other than the set of accepted
characters.
Example 4-1. Using the Test Method to Validate a String
<% @LANGUAGE = "VBScript" %>
<% Option Explicit %>
<%
Dim objRegExp
Set objRegExp = New RegExp 'Create a RegExp object instance
'Set the pattern (allow all letters, apostrophes, and hyphens)
objRegExp.Pattern = "[^a-z' \-]"
'Ignore case
objRegExp.IgnoreCase = True
82 Chapter 4: Regular Expressions, Classes, and Dynamic Evaluation and Execution
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
We set the RegExp’s Pattern property to search for the occurrence of a character
that is not a letter, not an apostrophe, not a space, and not a hyphen. We want to
ignore case, so we set the IgnoreCase property to True. Then we use the Test
method to see if any of the illegal characters exist within the string strName. If Test
returns True, then the name is invalid; if Test returns False, then the name is
valid. (In the above example, the first two names are valid while the last is invalid.)
The Replace method expects two string parameters. The first parameter is the
string to which to apply the regular expression; the second parameter contains the
text to be used to replace matching occurrences; a new string reflecting the appropriate
substitutions is then returned by the function. For example, if we wanted to
replace all instances of the acronym “asp” with “ASP,” we could use the Replace
method as shown in Example 4-2.
'See if the regular expression is found in strName
Dim strName
strName = "Scott Mitchell"
If objRegExp.Test(strName) then
Response.Write strName & " is not a valid name!<BR>"
Else
Response.Write strName & " is a valid name!<BR>"
End If
strName = "Roger O'Grady"
If objRegExp.Test(strName) then
Response.Write strName & " is not a valid name!<BR>"
Else
Response.Write strName & " is a valid name!<BR>"
End If
strName = "Tim 7sten"
If objRegExp.Test(strName) then
Response.Write strName & " is not a valid name!<BR>"
Else
Response.Write strName & " is a valid name!<BR>"
End If
%>
Example 4-2. Using the Replace Method
<% @LANGUAGE = "VBScript" %>
<% Option Explicit %>
<%
Dim objRegExp
Set objRegExp = New RegExp 'Create a RegExp object instance
'Set the pattern (allow all letters, apostrophes, and hyphens)
objRegExp.Pattern = "\basp\b"
Example 4-1. Using the Test Method to Validate a String (continued)
Using the RegExp Object 83
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Note that the regular expression used did not simply search for “asp,” but rather
for “\basp\b.” Recall that the “\b” special symbol searches for a word boundary. If
the regular expression contained just “asp,” the “asp” in “aspire” would also have
been capitalized.
The third and final method of the RegExp object is Execute. The Execute method
takes one parameter like the Test method, a string to which to apply the regular
expression. The Execute method returns a Matches collection, which contains a
Match object for each successful regular expression match.
The Match object contains three read-only properties: FirstIndex, Length, and
Value. FirstIndex contains the position in the string where the match occurred.
Unfortunately, FirstIndex is zero-based; VBScript, as you probably know, indexes
its strings starting at one. In other words, you have to add one to the value of
FirstIndex to actually identify the starting position of the substring found by the
Match object. As its name suggests, Length contains the total length of the matched
string. The final property, Value, contains the matched text.
In Example 4-2, we used the Replace method to find all the instances of “asp” and
replace them with “ASP.” We can use the Execute method to grab all of the
matches. The code in Example 4-3 uses the Execute method to return a Matches
collection; next, the script iterates through the Matches collection, outputting the
properties of each of the individual Match objects.
objRegExp.IgnoreCase = True 'Ignore case
objRegExp.Global = True 'Make all possible changes
Dim strSentence
strSentence = "Asp is a fun language. I aspire to learn asp!"
Response.Write "<B>Before Replace</B><BR>" & strSentence
Response.Write "<P><B>After Replace</B><BR> "
'Use the replace method!
strSentence = objRegExp.Replace(strSentence, "ASP")
Response.Write strSentence
%>
Example 4-3. Using the Execute Method and the Matches Collection
<% @LANGUAGE = "VBScript" %>
<% Option Explicit %>
<%
Dim objRegExp
Set objRegExp = New RegExp 'Create a RegExp object instance
Example 4-2. Using the Replace Method (continued)
84 Chapter 4: Regular Expressions, Classes, and Dynamic Evaluation and Execution
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Since the Execute method returns an object, be sure to use the Set keyword when
assigning the Matches collection returned by Execute to a variable. Also, since the
Matches collection is, after all, a collection, you have access to the basic methods
and properties of a collection, such as the Count property. Furthermore, you can
completely iterate through the Matches collection using a For Each ... Next loop.
The code in Example 4-3 will generate the output shown in Figure 4-1. Note that
the first instance of “Asp” is at the beginning of the string strSentence, so the
Match object reports its FirstIndex property as zero instead of the more VBScriptfriendly
one.
Personally, I find regular expressions to be neat and fun to use, partially because
no other method allows for such powerful string parsing with such convoluted
code. For example, what does the following regular expression match?
^\s*((\$\s?)|(£\s?))?((\d+(\.(\d\d)?)?)|(\.\d\d))\s*(UK|GBP|GB|USA|US|USD)?)\s*$
(This regular expression is an example from Microsoft’s web site: http://msdn.
microsoft.com/workshop/languages/clinic/scripting051099.asp.)
'Set the pattern (allow all letters, apostrophes, and hyphens)
objRegExp.Pattern = "\basp\b"
objRegExp.IgnoreCase = True 'Ignore case
objRegExp.Global = True 'Make all possible changes
Dim strSentence
strSentence = "Asp is a fun language. I aspire to learn asp!"
'Use the execute method to obtain a matches collection
Dim objMatches, objMatch
Set objMatches = objRegExp.Execute(strSentence)
Response.Write "There were " & objMatches.Count & " matches.<BR>"
Response.Write "<P><HR><P>"
'Loop through each Match object in the Matches collection
Dim iCount
iCount = 1
For Each objMatch in objMatches
Response.Write "<B>Match " & iCount & "</B><BR>"
Response.Write "FirstIndex = " & objMatch.FirstIndex & "<BR>"
Response.Write "Length = " & objMatch.Length & "<BR>"
Response.Write "Value = " & objMatch.Value & "<BR>"
Response.Write "<P><HR><P>"
iCount = iCount + 1
Next
Set objRegExp = Nothing
%>
Example 4-3. Using the Execute Method and the Matches Collection (continued)
Using Object-Oriented Programming with VBScript 85
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
I think you’ll find regular expressions are much easier to build than to analyze
after the fact. If you know what patterns you wish to search for, you will most
likely be able to create a valid regular expression. However, if presented with an
unwieldy regular expression, I find it a bit more difficult to work backwards and
predict what type of data the regular expression was intended to validate.
That about wraps up using regular expressions in VBScript. While we looked at
the code needed to perform a regular expression search in VBScript, we only
touched upon how to effectively use regular expressions to achieve powerful
string parsing. To truly master regular expressions, you’ll almost certainly need an
entire book dedicated to the subject, such as O’Reilly’s Mastering Regular Expressions.
You can also visit Microsoft’s scripting site (http://msdn.microsoft.com/
scripting/ ) for more information on the RegExp object.
Using Object-Oriented Programming
with VBScript
With the introduction of the VBScript 5.0 scripting engine, developers now have
the ability to create classes in VBScript, much as they can in VB. The next section
Figure 4-1. The web page produced by Example 4-3
86 Chapter 4: Regular Expressions, Classes, and Dynamic Evaluation and Execution
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
is intended for those new to object-oriented programming, and provides a quick
introduction on the topic. The section following that discusses the VBScript syntax
for creating and using classes.
Object-Oriented Programming 101
Object-oriented programming (OOP) is a programming methodology in which
entities known as bjects are used. Objects contain properties and methods; as the
names suggest, a property describes the features of the object, while a method performs
some action involving the object.
True object-oriented programming is further defined as the ability for
new objects to inherit properties and methods from existing objects,
and for dynamic binding of derived object methods. VBScript’s
implementation of OOP doesn’t support these two additional
requirements, and is therefore not a true object-oriented programming
language.
As an ASP developer, you’ve already used object-oriented code written by others.
For example, ADO is nothing more than a collection of objects that can be used to
access a database. Having the ability to treat a complex task as a black box is
indeed beneficial. When using ADO, you don’t have to worry about what protocol
is required to establish a connection to a database; rather, you simply use the
Open method of the ADO Connection object.
An object and an instance
To fully understand object-oriented programming, it is essential to understand the
difference between an object and an instance of an object.
An object, such as ADO’s Connection object, is an abstraction. It does not physically
exist. Rather, it simply serves as a template for creating instances.
An instance is a physical entity of an object. For example, in the following code,
ADODB.Connection is an object, while objConn is an instance of the ADODB.
Connection object:
'Create an instance of the ADODB.Connection object
Dim objConn
Set objConn = Server.CreateObject("ADODB.Connection")
You must have an instance of an object to make any method calls or to set or get
any properties. A real-world example of an object/instance relationship can be
seen with automobiles. A “car” is an object, a template describing the features and
functionality of an automobile. A 1986 Mercury Sable GS is an instance of a car,
able to have its properties modified and its methods implemented.
Using Object-Oriented Programming with VBScript 87
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Encapsulating complexity
Object-oriented programming can be used to encapsulate the complexity associated
with particular tasks. For example, imagine that on your web site you needed
to be able to open a text file, count the number of lines in the file, and then log
that information in another file. Assume you didn’t have an object-oriented
approach. Each time you wanted to accomplish this task, you’d need to perform
the following steps:
1. Open a particular text file.
2. Count the number of lines in that text file.
3. Close the text file.
4. Open up the logging file.
5. Log the number of lines from the file in Step 2.
6. Close the log file.
Granted, these steps are not complex or too lengthy, but imagine that a novice
developer is expected to perform these steps in various ASP pages. We can simplify
the process by encapsulating the above steps into a single object. Let’s take a
look at creating such an object.
Let’s call this object LogTextFileLines; it will have one property, TextFilePath,
which contains the physical path to the log file. LogTextFileLines will also contain
one method, UpdateLog, which is responsible for counting the number of lines in
a specified text file and recording the line count in the appropriate log file.
UpdateLog will have the following definition:
Function UpdateLog(strFileName)
where strFileName is the path to the text file whose lines need to be counted
and logged. UpdateLog could return a Boolean value to indicate whether or not
the operation was successful.
Once this object is created, the six-step task outlined earlier becomes much
simpler:
'Create an instance of the LogTextFileLines object
Dim objLog
Set objLog = New LogTextFileLines
'Set the LogFile path
objLog.LogTextFileLines = "C:\Log\linecount.log"
'Count and log the number of lines in C:\TXT\Scott.txt
If objLog.UpdateLog("C:\TXT\Scott.txt") then
Response.Write "C:\TXT\Scott.txt was logged successfully!"
Else
88 Chapter 4: Regular Expressions, Classes, and Dynamic Evaluation and Execution
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Response.Write "There was an error when attempting to log C:\TXT\Scott.txt"
End if
Being able to treat the entire process like a black box leads to less development
time, and most importantly, fewer bugs.
Using Classes in VBScript
If you are not familiar with VBScript classes, but perhaps are familiar with C++’s
implementation of classes, you’ll no doubt find VBScript’s use of classes somewhat
disappointing. VBScript classes can only contain a single constructor (a constructor
is a procedure that’s executed automatically when an instance of the class
is created); unfortunately, VBScript’s constructor can accept no parameters. Furthermore,
there is no support for inheritance in VBScript.
Inheritance is an OOP technique used to group objects into a logical
hierarchy corresponding to the relationships among objects. For
example, if you had a Mammal object, two of its children in this hierarchy
might be the Canine and Feline objects. These latter two
objects were inherited from Mammal. As an inherited class, Canine
and Feline would have the basic methods and properties of a
Mammal, as well as their own unique, specialized methods and
properties.
When creating classes, keep in mind you are creating a tool to be used by other
developers. In the discussion of classes throughout this book, there will be certain
times when a distinction between the developer who created the class and the
developer who is using the class is needed. The developer who created a class for
use by other developers will be referred to as the class developer, while the developer
using the created class will be referred to as the end developer.
Since VBScript classes are not compiled into binary objects like COM components,
for an end developer to use a class in an ASP script, she must have the class
defined within that ASP script. Rather than force the end developers to cut and
paste a class definition into each ASP page they use that needs that particular class
definition, it is best to create a text file that contains the class definition. Then any
end developer who needs to use a class can simply use a server-side include to
import the class definition into their ASP page.
As the class developer, once you have created a class, it is important that any
changes to the class do not “break” any of the class’s existing methods or properties.
For example, if you wanted to add a new feature to an existing class, make
sure that when adding it existing code utilizing that class will not break! If you
need to add a new feature that changes the existing class’s methods or properties
Using Object-Oriented Programming with VBScript 89
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
to the point that errors will occur in existing code, you should create an entirely
new class that utilizes the new feature. That way, old code still functions.
Creating classes
To create a class, use the Class statement, which has the following syntax:
Class ClassName
'Define the class's properties and methods here
End Class
where ClassName is the name you choose to assign to your class; your name
must follow standard VBScript variable naming conventions.
The Initialize and Terminate events
When creating classes, there are two important event handlers to be aware of: Initialize
and Terminate. The Initialize event fires when a class instance is created
using the New keyword. For example, we could create an instance of the
SomeObject class using:
Dim objSomeObjectInstance
Set objSomeObjectInstance = New SomeObject
The Terminate event occurs when a class instance is freed. A class instance can be
freed either explicitly with Set objSomeObjectInstance =Nothing , or implicitly,
when it goes out of scope.
We can define these event handlers in our classes. The Initialize event handler can
be used to initialize the class’s properties or perform other needed start-up tasks.
The Initialize event is often referred to as the class’s constructor. The Terminate
event can be used to perform any shutdown tasks, and is commonly referred to as
the class’s destructor. Example 4-4 contains the definition of a class with its single
constructor and destructor.
Example 4-4. The Initialize and Terminate Event Handlers
Class SomeObject
Private Sub Class_Initialize()
'This event is called when an instance of the class is instantiated
'Initialize properties here and perform other start-up tasks
End Sub
Private Sub Class_Terminate()
'This event is called when a class instance is destroyed
'either explicitly (Set objClassInstance = Nothing) or
'implicitly (it goes out of scope)
End Sub
End Class
90 Chapter 4: Regular Expressions, Classes, and Dynamic Evaluation and Execution
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Properties, methods, member variables, and member functions
From the end developer’s standpoint, a class contains properties and methods.
Properties are variables that the end developer can set that determine the state of
the class. Methods are functions of the class that the end developer can call to
have the class perform a given task. For example, the ADODB.Connection object
contains properties that describe the state of the object instance, such as
ConnectionString, ConnectionTimeout, and Provider, and methods that perform
actions, such as Open and Close.
A class, though, can contain variables and functions that the end developer cannot
directly call. These “hidden” variables and functions are referred to as member
variables and member functions, respectively.
The terminology may seem a bit confusing, or at least overly verbose. To put it
another way, a class contains functions and variables. If the end developer can call
a class function, the function is referred to as a method; otherwise, it is referred to
as a member function. Similarly, if the end developer can call a class variable, it is
referred to as a property; otherwise, it is referred to as a member variable.
The public and private statements
As discussed earlier in the section “Encapsulating complexity,” one of the goals of
object-oriented programming is to provide a black box, hiding the implementation
details from the end developer. To assist with this encapsulation of complexity,
VBScript allows you to hide methods and properties of an object. Remember,
an object should serve as a black box for the end developer; as the creator of an
object, you may wish to prevent the end user from directly calling specific methods
or setting certain properties.
To create member functions and member variables (methods and properties that
are hidden from the end developer), precede the member variable or member
function definition with the Private keyword. A member function can only be
executed from within another one of the class’s methods or member functions. A
member variable can only be modified or read by code in a class method, member
function, or through a Property Get, Property Let, or Property Set statement
(which we’ll discuss shortly).
Note that the Initialize and Terminate event handlers in Example 4-1
are declared as Private. This prevents the end developer from
explicitly triggering these events.
Using Object-Oriented Programming with VBScript 91
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
To create a property or method, precede the variable or function definition with
the Public keyword. If you don’t explicitly specify whether a function or variable
should be public or private, the function or variable will be made public.
The following code creates a class with a public method and property and a private
method and property:
Class MyClass
'Create a public method
Public Sub A_Public_Method()
'Call the private method
A_Private_Method
End Sub
'Create a public property
Public A_Public_Property ' Not a good thing to do
'Create a private method
Private Sub A_Private_Method()
'...
End Sub
'Create a private property
Private A_Private_Property
End Class
Note that when using either the Public or Private keywords with variables, you
leave off the Dim statement. If you decide not to explicitly specify whether or not
a variable is Public or Private, you will need to precede the property name
with the Dim statement, which will make the variable Public. However, it is
highly recommended that you always explicitly indicate whether or not a variable
or function is either Public or Private.
Trying to access private methods or properties through an instance of MyClass will
result in an error (see Example 4-5 and Figure 4-2). However, the method A_
Public_Method can execute private methods of MyClass; for example, in the previous
code snippet, A_Public_Method calls A_Private_Method.
Example 4-5. Accessing Public and Private Methods and Properties
'MyClass definition defined in above code snippet
'Create an instance of MyClass
Dim objMyClass
Set objMyClass = New MyClass
'Since A_Public_Method is Public, this is valid code:
objMyClass.A_Public_Method
'Since A_Public_Property is Public, this is valid code:
92 Chapter 4: Regular Expressions, Classes, and Dynamic Evaluation and Execution
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Using Property Get
Since all variables in VBScript are Variants, if we create a public property, we have
no control over what type of information the user enters. For example, imagine
that we had a class with a property named PhoneNumber. Obviously this variable
is intended to store a phone number. However, the developer working with our
object could just as easily assign an array to this variable, or an object instance, or
a currency value.
In languages like C++, the solution would be to make PhoneNumber a private
property and to provide a public method, setPhoneNumber. This public method
would expect one parameter: the value of the phone number the end developer
wants to assign to PhoneNumber. The setPhoneNumber method would contain
code to ensure that the developer couldn’t insert an erroneous or improperly formatted
phone number. Since PhoneNumber was made private, another public
method, getPhoneNumber, would be needed to allow the developer to read the
value of the PhoneNumber property.
An identical method can be implemented using VBScript; however, there is a
cleaner way to do this in VB/VBScript. As with the C++ method, start by creating
all of your properties as Private. Then if you wish to allow the end developer to
objMyClass.A_Public_Property = 7
'Since A_Private_Method is Private, this is code is invalid,
'and will cause an error:
objMyClass.A_Private_Method ' Invalid
'Since A_Private_Proverty is Private, this is code is invalid,
'and will cause an error:
objMyClass.A_Private_Property = 100 ' Invalid
'Since A_Private_Proverty is Private, this is code is invalid too
Dim iValue
iValue = objMyClass.A_Private_Property ' Invalid
Figure 4-2. An error message generated by Example 4-5
Example 4-5. Accessing Public and Private Methods and Properties (continued)
Using Object-Oriented Programming with VBScript 93
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
read the value of the property, use a Property Get statement in your class. A
Property Get statement has the following syntax:
[Public | Private] Property Get PropertyName [(arglist)]
'... assign the value of the private property to PropertyName ...
End Property
Let’s examine the phone number property we discussed earlier. First things first:
we need to create our private property, strPhoneNumber:
Class Information
'Create a private property to hold the phone number
Private strPhoneNumber
End Class
Now, if we wanted to allow the developer using our object to be able to read the
value of strPhoneNumber, we can add a Property Get like so:
Class Information
'Create a private property to hold the phone number
Private strPhoneNumber
Public Property Get PhoneNumber()
PhoneNumber = strPhoneNumber
End Property
End Class
The developer can now read the value of strPhoneNumber using the following
code:
'Assume the class Information is defined
Dim objInfo
Set objInfo = New Information
'This is a legal way to read the property
Response.Write "Phone Number = " & objInfo.PhoneNumber
'This is illegal, since strPhoneNumber is private
Response.Write "Phone Number = " & objInfo.strPhoneNumber
Often, developers will give their private properties a Hungarian-notation-like prefix
(e.g., strPhoneNumber), and will give the corresponding Property Get statements
more English-like names. For example, ADO’s Connection object properties
have names like ConnectionString and Timeout, not strConnectionString or
iTimeout.
Be sure to give your private properties and Property Get statements
different names. If the property and Property Get statement
share the same names, you will get a “Name redefined” error.
94 Chapter 4: Regular Expressions, Classes, and Dynamic Evaluation and Execution
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
If the property returned by Property Get is an object, you must use the Set
statement when assigning the Property Get name to the private property. For
example, if strPhoneNumber was to contain a Dictionary object instead of a
string, in the Property Get statement we’d need to use:
Public Property Get PhoneNumber()
Set PhoneNumber = strPhoneNumber
End Property
The arglist component of the Property Get statement should be present when
the property requires an argument or index to be properly used. For example, if
you had an array as a private property, and wanted to allow the end developer to
read a single element from the array as opposed to the entire array, you could use
the following code:
Class Information
'Create an array to hold the US States
Private aStates()
Private Sub Class_Initialize()
'ReDim the array to hold 50 states, then populate the array
ReDim aStates(50)
aStates(0) = "Alabama"
' ... and so on ...
End Sub
'Create a Property Get statement to grab a certain index from
'the aStates array
Public Property Get States(iIndex)
States = aStates(iIndex)
End Property
End Class
Once you have an instance of the Information class, use the following code to
read a particular value from the aStates array:
'Class Information defined above...
'Create an instance
Dim objInfo
Set objInfo = New Information
'Display the zeroth state
Response.Write objInfo.States(0)
Set objInfo = Nothing
Using Property Let
Property Get allows an end developer to retrieve a private property; Property
Let enables the end developer to assign a value to a private property. Property
Let statements provide assurances against property corruption. Recall our example
Using Object-Oriented Programming with VBScript 95
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
at the start of “Using Property Get.” If you created a class with a public property
named strPhoneNumber, the developer utilizing your class could easily assign
anything to strPhoneNumber. If the developer wanted strPhoneNumber to contain
“Hey there, Bob,” he could assign this value to the property. If the developer
wanted strPhoneNumber to be set to an ADO Recordset object, he could easily
assign this value to the property as well.
Of course, you don’t want to allow end developers to enter any kind of information
into your class’s properties. Rather, you want to make sure they enter an
accepted datatype with an accepted format. strPhoneNumber might require a
string datatype with the format ###-###-####. The Property Let statement
allows for datatype and format checking when an end developer attempts to
assign a value to a property.
Do not use a Property Let statement for properties of type Object.
Rather, use the Property Set statement, which is discussed in detail
in the next section, “Using Property Set.”
Nearly syntactically identical to the Property Get statement, the Property Let
statement has the following format:
[Public | Private] Property Let PropertyName([arglist,] value)
'... statements: check to see if value is of the correct datatype
'... and format. If it is, assign value to the private property
End Property
To assign a value to the private property strPhoneNumber, we could use the following
Property Let statement:
Class Information
'Create a private property to hold the phone number
Private strPhoneNumber
Public Property Let PhoneNumber(strPhone)
'Performs no format or type checking, simply assigns the
'value passed in by the developer to strPhoneNumber
strPhoneNumber = strPhone
End Property
End Class
The above Property Let statement does absolutely no format or type checking
on the value the end developer wants to assign to strPhoneNumber. The value
parameter (strPhone) of the Property Let statement is the value entered by
the end developer to the right of the equals sign. In the following code snippet,
96 Chapter 4: Regular Expressions, Classes, and Dynamic Evaluation and Execution
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
“555-4343” would be the value parameter passed to the PhoneNumber Property
Let statement:
'Create an instance of the object
Dim objInfo
Set objInfo = New Information
'Assign a value to the phone number
objInfo.PhoneNumber = "555-4343"
Of course, performing no format or type checking is silly, since if we don’t care
what types of data are stored in a particular property, then that property doesn’t
need to be defined as private. Since we do care about the format of
strPhoneNumber, let’s add some format-checking code in our Property Let
statement.
When creating format-checking code, you have to decide which formats are
acceptable and what to do if the end developer tries to assign an invalid format.
For this example, we will accept a telephone format of ###-###-####, and if the
end developer enters an invalid format, we’ll raise an error. Example 4-6 contains
the new class definition for Information.
Example 4-6. A More Robust Property Let Statement for PhoneNumber
Class Information
'Create a private property to hold the phone number
Private strPhoneNumber
Public Property Let PhoneNumber(strPhone)
'Ensures that strPhone is in the format ###-###-####
'If it is not, raise an error
If IsObject(strPhone) then
Err.Raise vbObjectError + 1000, "Information Class", _
"Invalid format for PhoneNumber. Must be in ###-###-#### format."
Exit property
End If
Dim objRegExp
Set objRegExp = New regexp
objRegExp.Pattern = "^\d{3}-\d{3}-\d{4}$"
'Make sure the pattern fits
If objRegExp.Test(strPhone) then
strPhoneNumber = strPhone
Else
Err.Raise vbObjectError + 1000, "Information Class", _
"Invalid format for PhoneNumber. Must be in ###-###-#### format."
End If
Set objRegExp = Nothing
Using Object-Oriented Programming with VBScript 97
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
If the end developer attempts to execute code such as:
Dim objInfo
Set objInfo = New Information
objInfo.PhoneNumber = "This is an invalid phone number!"
the error message shown in Figure 4-3 will be displayed when viewed through a
browser.
In the Property Get, Property Let, and Property Set statements,
the command Exit Property can be used to immediately
exit these three statements.
The Property Let statement also has an optional arglist. This arglist must
be identical to the arglist in the property’s corresponding Property Get statement,
if it exists.
Using Property Set
Classes can have properties that are objects, but special care needs to be taken
when returning an object through a Property Get, or when using Property Set
End Property
Public Property Get PhoneNumber()
PhoneNumber = strPhoneNumber
End Property
End Class
Figure 4-3. Error message produced by an invalid phone number
Example 4-6. A More Robust Property Let Statement for PhoneNumber (continued)
98 Chapter 4: Regular Expressions, Classes, and Dynamic Evaluation and Execution
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
to assign an object a reference. Recall from “Using Property Get” that when returning
an object reference, it is essential to use the Set keyword. For example:
Public Property Get PropertyName()
Set PropertyName = objPrivatePropertyObject
End Property
VBScript provides a Property Set statement to allow the end developer to assign
an object instance to a property. The Property Set statement has the following
syntax:
[Public | Private] Property Set PropertyName([arglist,] reference)
'... Perform any needed checks here, then use
'Set objPrivateProperty = PropertyName
End Property
The format of Property Set is nearly identical to that of Property Let. The only
functional difference between the two is Property Let assigns non-object values
to private properties, while Property Set assigns object instances to private properties.
For example, imagine we had a class that contained a private property
named objConn that was expected to be an ADO Connection object. The class
definition, with the Property Set and Property Get statements, might look
something like:
Class MyConnectionClass
'Create a private property to hold our connection object
Private objConn
Public Property Get Connection()
Set Connection = objConn
End Property
Public Property Set Connection(objConnection)
'Assign the private property objConn to objConnection
Set objConn = objConnection
End Property
End Class
The end developer would use the Property Set statement in the following
manner:
'Create an instance of MyConnectionClass
Dim objMyClass, objMyRecordset
Set objMyClass = New MyConnectionClass
Set objConnection = Server.CreateObject("ADODB.Connection")
'Assign objConnection to the Connection property
Set objMyClass.Connection = objConnection
As with the Property Let statement, the Property Set statement has an
optional arglist. This arglist must be identical to the corresponding
Property Get’s arglist.
Using Dynamic Evaluation and Execution 99
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Creating read-only or write-only properties
Using the Property Get and Property Let/Property Set statements enables
you to make your private properties editable and readable. However, there is
nothing requiring you to use both a Property Get and a Property Let/Property
Set for a private property. In fact, you can create read-only and write-only
properties by using either just Property Get or just Property Let/Property,
respectively.
Note that while read-only properties are fairly common, write-only
properties are not. There are some objects, though, that employ
write-only properties, such as the NewMail class of the Collaborative
Data Objects for NT Server (CDONTS) object. Write-only properties
can lead to headaches for developers using your classes,
especially if they are unaware the property is write-only. If the property
is write-only, an error will be generated if the end developer
attempts to do something like the following:
objClass.WriteOnlyProperty = 5
Response.Write objClass.WriteOnlyProperty
The second line will generate an error, since the value of
WriteOnlyProperty cannot be read.
Creating Useful, Reusable Code
In this chapter we discussed the fundamentals of object-oriented programming and
looked at how to create classes using VBScript. In the next several chapters, we’ll
look at how to tie together the lessons learned in this and the previous chapter to
create useful, reusable code. Having a solid understanding of object-oriented programming
and the associated VBScript syntax is essential, since the next chapters
will use classes extensively.
Using Dynamic Evaluation
and Execution
As a programmer, you have performed static evaluation and static execution
countless times. For example, the following code snippet performs both static
evaluation and static execution:
<% @LANGUAGE="VBSCRIPT" %>
<%
Dim iAnswer
iAnswer = 22 / 7 'A rough approximation of pi
%>
100 Chapter 4: Regular Expressions, Classes, and Dynamic Evaluation and Execution
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
The first line, Dim iAnswer, is an example of static execution. When a browser
visits the ASP page that the above code snippet resides in, the first line of code is
executed, causing a memory location to be set aside to store the value of
iAnswer. The second line, iAnswer =22 / 7, is an example of both static evaluation
and static execution. Initially, the expression 22 / 7 is evaluated, returning the
result 3.14285714285714. The execution occurs when this value is stored into
the variable iAnswer.
This type of code evaluation and execution is referred to as static because the
statement being evaluated and the commands being executed are hardcoded into
the script. The only way the expression 22 / 7 will change is if a developer edits
the actual ASP file by entering a new expression.
The VBScript 5.0 scripting engine offers two functions that facilitate dynamic evaluation
and execution. Let us examine each of these issues separately.
Dynamic evaluation
Imagine you wanted to present your users with a form with a single text box, into
which they could enter a mathematical expression. Once the form was submitted,
the result would be displayed. For example, the user might enter something like
(8 * (5 / 3.5)) – 34. Example 4-7 contains the code to create this form, and
Figure 4-4 displays the form when viewed through a browser.
When this form is submitted, SolveMathProblem.asp is called and is passed the
user’s mathematical expression in the form element Expression.
SolveMathProblem.asp dynamically evaluates the user’s input using the Eval function.
The Eval function has the following definition:
[EvaluationResult = ] Eval(expression)
where expression is a string variable that contains a valid VBScript expression.
Eval returns the result of the evaluated expression as though it had been hardcoded
in the script. The code for SolveMathProblem.asp, without any error checking
or validation code, is shown in Example 4-8.
Example 4-7. A Form to Solve a Mathematical Expression
<HTML>
<BODY>
<FORM METHOD=POST ACTION="SolveMathProblem.asp">
Enter a mathematical expression (like
<CODE>5 + 4 * (9 / 4 - 10.5) + 45/2</CODE>):<BR>
<INPUT TYPE=TEXT NAME=Expression SIZE=40>
<P>
<INPUT TYPE=SUBMIT VALUE="Solve this Expression!">
</FORM>
</BODY>
</HTML>
Using Dynamic Evaluation and Execution 101
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Eval will generate an error if the expression parameter is not a valid VBScript
expression. If the user entered a non-valid VBScript expression into the text box
(such as Scott Mitchell) in the form in Example 4-7, SolveMathProblem.asp would
output a syntax error message, as Figure 4-5 illustrates.
Allowing your users to directly enter the commands that will be used
in dynamic evaluation or dynamic execution is a security risk, to say
the least. If you permit your users to enter input that is dynamically
executed, they can easily enter malicious code, such as a series of
commands that will delete all of the web pages on your site using
the FileSystemObject object model. Even allowing your users to
directly enter commands that are dynamically evaluated poses a risk.
For example, a user could output the contents of your application
and session variables, which might contain connection strings or
other sensitive information.
Figure 4-4. The form in Example 4-7 when viewed through a browser
Example 4-8. SolveMathProblem.asp Dynamically Evaluates the User’s Input
<% @LANGUAGE="VBSCRIPT" %>
<% Option Explicit %>
<%
'Read in the user's expression
Dim strExpression
strExpression = Request("Expression")
'Output the result
Response.Write "The mathematical result of:<BR><CODE>"
Response.Write strExpression & "</CODE><P>is:<BR><CODE>"
Response.Write Eval(strExpression) & "</CODE>"
%>
102 Chapter 4: Regular Expressions, Classes, and Dynamic Evaluation and Execution
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Dynamic execution
The Eval function only allows for expression evaluation. If you need to execute a
statement—such as an assignment statement, a variable declaration, or a loop—
you will need to use the Execute statement. The Execute statement expects a
string parameter that contains one or more statements for execution. If there are
multiple statements, they must be delimited by a legal VBScript statement delimiter,
namely the carriage return or the colon.
The following code snippet dynamically executes an assignment operation, assigning
the value of “Hello, World!” to strWelcomeMessage:
Dim strStatement
strStatement = "strWelcomeMessage = ""Hello, World!"""
Execute strStatement
This next code snippet demonstrates how to execute multiple statements with one
call to Execute. Note that each statement is delimited by a colon in the first call to
Execute, while each statement is delimited by a carriage return in the second call
to Execute:
Dim strStatement
'Delimit the statements using a colon
strStatement = "Dim iAge : iAge = 4 : Response.Write iAge"
Execute strStatement
'Delimit the statements using carriage returns
strStatement = "Dim dtBirthdate" & vbCrLf & _
Figure 4-5. The Eval function expects a valid VBScript expression; an invalid expression
results in a syntax error
Further Reading 103
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
"dtBirthDate = DateSerial(1978, 8, 1)" & vbCrLf & _
"Response.Write ""I was born on "" & FormatDateTime(dtBirthDate)"
Execute strStatement
This final code snippet demonstrates how to create a variable-sized array without
using Redim:
'Create an array named aPerfectSizedArray
'How many elements should the array contain?
Dim strElementsInArray
strElements = "10"
'Create the array
Dim strStatement
strStatement = "Dim aPerfectSizedArray(" & strElements & ")"
Execute strStatement
In VBScript, the equals operator has two functions—logical equivalence
and assignment. When using the equals sign with the Eval
statement, the equals sign serves as the logical equivalence operator.
When using an equals sign within an Execute statement, the
equals sign serves as the assignment operator.
Further Reading
To learn more about the topics covered in this chapter, be sure to read these additional
resources:
• For a full list of the VBScript scripting engine version in which particular
VBScript language features were implemented, check out http://msdn.
microsoft.com/scripting/default.htm?/scripting/vbscript/doc/vbsversion.htm.
• To learn more about object-oriented programming, be sure to check out http://
java.sun.com/docs/books/tutorial/java/concepts/. While the site does focus on
Java technology, it contains worthwhile articles on OOP design and concepts.
• For a good beginner-level discussion on regular expressions, visit http://
hotwired.lycos.com/webmonkey/geektalk/97/33/index3a.html.
• The best article I could find on Microsoft’s site discussing regular expressions
and VBScript is at http://msdn.microsoft.com/workshop/languages/clinic/
scripting051099.asp. This article contains a good explanation on how to use
the RegExp object with VBScript, and supplies some basic regular expression
theory.
• To obtain the latest version of the VBScript and JScript scripting engines, visit
http://msdn.microsoft.com/scripting/default.htm?/scripting/vbscript/download/
vbsdown.htm.
104
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Chapter 5 5
Form Reuse
In general, the Web would be a pretty useless place if it weren’t for HTML forms.
Forms provide an interface for the client to submit information to the web server.
The most common form interfaces include search engine interfaces, e-commerce
interfaces, and data administration interfaces.
This chapter and the next focus primarily on data administration interface forms.
Data administration interfaces are common in many data-driven web sites. For
example, if you ran a web site that sold widgets, you’d have a pretty extensive
database structure, comprised of many tables containing an assortment of relevant
information. This site would also likely contain an extensive set of administration
pages, where each administration page permitted administrators to update, insert,
and delete records into the various database tables.
This chapter does not focus on creating HTML forms, or reading form variables
through an ASP page—it’s assumed you are familiar with those topics. (If you are
a little rusty, don’t worry; there’s a brief discussion on each of these topics early in
the chapter.) What we will be focusing on heavily throughout this chapter and the
next is the art of form reuse.
The Importance of Code Reuse
Code reuse has many advantages, the paramount one being that developers who
practice code reuse become more efficient developers, writing less buggy code.
Code reuse obviously saves time in the long run, since you don’t have to continuously
reinvent the wheel. Code reuse also produces less buggy code. There’s a
positive correlation between the number of lines of code one writes and the number
of bugs present. Ergo, writing fewer lines of code (through reuse) leads to
fewer bugs. Furthermore, the more often a particular piece of code is reused, the
The Importance of Code Reuse 105
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
more certain you can be that the reused code is bug-free. This benefits developers
in the debugging phase; if there is a bug in the system, there’s a much greater
probability it came from a new piece of code than from a code module that has
been used in several previous projects without fail.
Code reuse also benefits the end user. For example, all the Microsoft Office applications
share the same code for a number of tasks, such as the toolbar functionality
and the menu system. Such code reuse provides the end user with a consistent
look and feel, flattening the learning curve associated with the various Office
applications.
Code Reuse in ASP
ASP provides extensive opportunities for code use. Server-side includes and the
Server.Execute method, which we discussed in Chapter 1, Introduction, allow
developers to modularize their code; COM components, which we’ll discuss in
Chapter 7, Using Components, aid in business logic; as we discussed in Chapter 4,
Regular Expressions, Classes, and Dynamic Evaluation and Execution, VBScript
provides classes that enable the encapsulation of complexity and the ability to
hide implementation details. Despite these features, many ASP developers fail to
make a serious effort at code reuse.
I think part of the problem is that ASP pages are created using scripting languages.
Generic scripts, by their nature, are often small and simple, performing a
discrete task. Furthermore, scripts often each have a unique task to complete. For
this reason, rarely do ideas of code reuse or modularization come into mind when
developing a script.
ASP projects would benefit enormously if ASP developers paid greater attention to
issues of code reuse. More often than not, a vast number of ASP scripts on your
site perform nearly or even completely identical tasks. In my experience with ASP,
I’ve identified two distinct areas in which developers usually create several common
scripts: form use and database access. Creating multiple scripts to do nearly
the same thing is not only cumbersome, but also error-prone and far from maintainable.
These two areas would benefit the most from extensive and intelligent
code reuse.
This chapter, as its name suggests, deals with form reuse. Chapter 6, Database
Reuse, discusses techniques for code reuse in ASP scripts that utilize databases.
Examining form usage
On a data-driven web site, forms are extremely useful tools that allow visitors to
query, insert, delete, or update particular bits of data from a database. Without
forms, data-driven web sites are essentially useless.
106 Chapter 5: Form Reuse
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
When creating forms for an ASP web site, the developer will usually follow these
steps:
1. Create an HTML form with the proper form fields, complete with client-side
form validation.
2. Create a form-processing script and add the needed code.
3. Add server-side form validation code to the ASP page created in Step 2.
In this chapter, we’ll look at ways of automating Steps 1 and 3, which are common
to all forms throughout a web site. Step 2, of course, will be different
depending on the form’s intended purpose. However, if we are creating administration
pages for use in updating, deleting, or inserting records into a database, we
may be able to generalize Step 2. Chapter 6 focuses on reuse of this kind.
Before we jump into a thorough discussion of form reuse for Steps 1 and 3, it is
important you have a solid understanding of forms. Specifically, it is important you
understand what forms are, the HTML needed to create them, and the ASP code
used to process their results. In case you are a bit rusty, the next section provides
a quick review of form use.
A Primer on Form Use
Forms exist for one reason—to allow the user to send input to the web server.
This input can be used in a vast variety of ways. The input may be web site feedback
that is packaged into an email and sent to the webmaster; the input may be
search terms for viewing a particular subset of results from a database; the input
could be shipping and billing information for an item from an e-commerce web
site. Therefore, to collect user input, you must use a form.
Forms play a large role in making the Web truly dynamic. A web site
without forms is, most likely, a static web site. Forms have assisted
web development by allowing users to visit dynamic web pages
whose contents depend upon the user’s input.
When using forms, often two pages are used. The first page is the form creation
web page, and contains the HTML code needed to display the form and its elements
(text boxes, list boxes, checkboxes, etc.). The second page is the form processing
script. This is the ASP page that retrieves the form field values entered by
the user and performs whatever processing needs to be done.
The next two subsections detail these two pages. The “Creating HTML Forms” section
discusses how to generate a form in a web page using the proper HTML
A Primer on Form Use 107
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
code, while the “Processing Forms Through an ASP Page” section details how to
read in the form field values entered by a user and make programmatic decisions
based upon these values. If you are comfortable with creating forms and reading
their values through an ASP page, feel free to skip the following two sections.
Creating HTML Forms
Forms are created in HTML using the FORM tag. The FORM tag has a number of
available properties or attributes, the three most important ones being NAME,
ACTION, and METHOD. The NAME property gives a name to the form, which is useful
when referring to a form through client-side scripting code. The ACTION property
specifies the URL of the form processing script. This is the URL that is called
when the user submits the form. If omitted, the ACTION property defaults to the
URL of the form.
The METHOD property determines how the form data is sent to the web server
when the form is submitted. The METHOD property can take one of two values:
GET or POST. GET sends the form field names and values through the query string,
while POST sends the values through an HTTP header. Nine times out of ten,
you’ll find that using POST is preferable to using GET. With POST, the form field
values are not sent through the query string, which is useful under the following
two circumstances:
• The user has entered sensitive information, such as a password, into the form.
By displaying the form field information through the querystring, the password
could be seen by an onlooker.
• Lengthy form values are expected. Some older browsers limited the total
length of the querystring. When dealing with large forms, or forms that expect
lengthy inputs from the user, you should use the POST method to prevent
overloading of the querystring.
Using GET, though, has its advantages. One major advantage of passing the form
field values through the querystring is it allows the visitor to bookmark the results
of a particular set of form field values. Since a bookmark saves the entire URL, the
form field values that were passed through the querystring are part of the bookmark.
The start of a form is represented with the FORM tag, and the end by a closing
FORM tag (</FORM>). Here is an example of a form with no form elements:
<FORM NAME=frmMyForm METHOD=POST ACTION="/scripts/FormProcess.asp">
</FORM>
The building blocks of forms
A form without any form elements is pretty useless. Form elements provide for the
collection of user input. Each text box, list box, checkbox, radio button, and image
108 Chapter 5: Form Reuse
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
is considered a form element (sometimes referred to as a form field). Text boxes,
checkboxes, buttons, and radio buttons are created using the INPUT tag; list boxes
are created using the SELECT tag with zero to many OPTION tags; and multi-row
text boxes are created using the TEXTAREA tag.
Each of these three tags has its own set of available properties. However, all have
one very important property in common: the NAME property. The NAME property
uniquely identifies each form field within the form. This NAME property is also
what is used to retrieve the value entered for a particular form field in the formprocessing
script.
Since text boxes, checkboxes, and radio buttons are all created using the same
HTML INPUT tag, a mechanism to specify what type of form element to create
with the INPUT tag must exist. The TYPE property of the INPUT tag performs this
task. The TYPE property can have one of three values: TEXT, CHECKBOX, or RADIO,
which create a text box, checkbox, or radio button, respectively.
To learn more about the various properties of the form element tags, be sure to
pick up a book on HTML design. Throughout the examples in this chapter, the
form elements usually contain only the essential properties: NAME and TYPE.
Processing Forms Through an ASP Page
Recall that form information is sent from the client to the server via an HTTP
request. It’s not surprising, then, that the ASP Request object is used to obtain the
form information submitted by the web visitor. The Request object contains two
collections, Form and QueryString, that are useful for collecting form information.
When a user submits information through a form, the browser compacts the user’s
form field entries into a single string. This string contains each form element’s
name and value. Each form element name/value is delimited by an ampersand (&);
in each name/value pair, an equals sign delimits the name and value. A raw string
sent by the browser to the web server looks like:
FormFieldName1=FormFieldValue1&FormFieldName2=FormFieldValue2...
When the ASP form-processing script receives this string, it parses it into the appropriate
Request object collection. If the form’s METHOD parameter was set to GET, the
above string is passed through the query string and the Request.QueryString collection
is populated. If the form’s METHOD parameter was set to POST instead, the
above string is populated in the Request.Form collection. Even though ASP
unpacks this string into the various collections, the string in its entirety can easily
be accessed with the following code:
'Display the packed string that has been populated in
'the QueryString collection
Response.Write Request.QueryString
A Primer on Form Use 109
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
'Display the packed string that has been populated in
'the Form collection
Response.Write Request.Form
To access a particular element in the form field, simply specify the name of the
form field like so:
'Return the value of the form field named FormField1
Request.QueryString("FormFieldName1")
If you attempt to retrieve the value of the QueryString or the Form property when
the property contains no data, or if you attempt to retrieve the value of a nonexistent
field from either collection, an empty string will be returned; ASP does not
raise an error.
Working with the QueryString and Form collections
Since Request.QueryString and Request.Form are collections, you can use the standard
methods and properties of collections. For example, to return the number of
elements in the Request.Form collection, use the Count property like so:
'Output the number of elements in the Request.Form collection
Response.Write "There are " & Request.Form.Count & " element(s) " & _
"in the Request.Form collection"
To iterate through either one of these collections, use either a For Each ... Next
loop or a counting loop. An example of using a For Each ... Next loop can be
seen in Example 5-1. Note that when iterating through the collection using a For
Each ... Next loop, each name in the name/value pairs is returned. In
Example 5-1, strName is assigned the value of the name in the name/value pair
through the For Each ... Next loop. To obtain the value in a name/value pair,
use Request.QueryString(strName).
Example 5-1. Iterating with a For Each … Next Loop
<%@ LANGUAGE = "VBSCRIPT" %>
<% Option Explicit %>
<HTML>
<BODY>
<TABLE ALIGN=CENTER BORDER=1>
<TR>
<TH>Name</TH>
<TH>Value</TH>
</TR>
<%
'Use a For Each ... Next to loop through the QueryString collection
Dim strName
For Each strName in Request.QueryString
Response.Write "<TR><TD>"
Response.Write strName
Response.Write "</TD><TD>"
Response.Write Request.QueryString(strName)
Response.Write "</TD></TR>"
110 Chapter 5: Form Reuse
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
The QueryString or Form collections can also be iterated by using a counting loop.
Since we can obtain the number of elements in the collection using the Count
property, and since the elements of a collection can be visited via an index, we
can use the following code to visit each element in a collection:
<%@ LANGUAGE = "VBSCRIPT" %>
<% Option Explicit %>
<%
'Use a counting loop to visit each element
Dim iLoop, iCount
iCount = Request.Form.Count
For iLoop = 1 to iCount
Response.Write "Value at " & iLoop & ": " & Request.Form(iLoop)
Response.Write "<P><HR><P>"
Next
%>
The QueryString and Form collections are indexed starting at one.
Often you’ll see code that doesn’t specify whether to read from the QueryString or
Form collection. For example, the code might read:
'Store the user's name into strName
Dim strName
strName = Request("UserName")
When a collection is not specified, the QueryString collection is scanned first. If
the particular form field name/value pair does not exist within the QueryString collection,
the Form collection is searched. If the name/value pair is not found in the
Form collection either, a blank string (" ") is returned. This method’s main advantage
is that as a developer creating a form-processing script, you don’t have to
worry about the method used by the developer creating the HTML form. The disadvantage
of using this method, though, is that by not explicitly specifying what
collection to search for, the code becomes a bit ambiguous. As we begin to practice
form reuse, you’ll see that the advantage of not having to specify the correct
collection greatly outweighs the disadvantage.
Next
%>
</TABLE>
</BODY>
</HTML>
Example 5-1. Iterating with a For Each … Next Loop (continued)
Form Validation 111
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Form Validation
With ASP, information collected through a form interface is often used to modify a
database in some way. Databases are only as good as the data within them. A
database saturated with meaningless or improperly formatted data is worthless.
Therefore, when creating ASP pages that will directly modify databases based
upon input gathered through a form, it is vital that the user entering the form
information enters valid data.
The process of ensuring that form fields contain valid input is referred as form validation.
Form validation can occur on both the client side and server side. In the
next two sections, “Client-Side Form Validation” and “Server-Side Form Validation,”
we’ll discuss the advantages and disadvantages of each method.
Client-Side Form Validation
Client-side form validation, as the name implies, is form validation that happens
on the client. A web page containing a form can optionally contain client-side
JavaScript code that will execute when the user attempts to submit the form. This
JavaScript code can then scan the user’s form entries, making sure they have a particular
format.
The main advantage of client-side form validation is the fact that the validation
occurs completely on the client’s computer. Therefore, if the user has entered
invalid form data, they don’t have to wait for a round trip to the web server before
they know whether or not they’ve entered invalid data. Many sites employ this
type of form validation. For example, an e-commerce site might provide form validation
for the form fields supplied for the customer’s address.
Client-side form validation is not without its disadvantages, though. If the browser
being used does not support JavaScript, or the user has JavaScript disabled, then
client-side form validation will not execute. Furthermore, any visitor can easily
view the client-side form validation code by viewing the HTML source sent to her
browser. If this user wishes to send invalid data, they could take the time to determine
what types of data would “slip by” the client-side form validation. Additionally,
a visitor entering the URL of the form-processing script into his or her
browser and passing illegal form variable values through the querystring can circumvent
client-side form validation.
112 Chapter 5: Form Reuse
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Also, a malicious user can sidestep client-side form validation
through the following sequence of steps:
1. Save the HTML/client-side validation source code.
2. Remove the client-side validation source code.
3. Reload the modified HTML through a browser and submit the
form with invalid data.
For this reason, you should not rely on just client-side form validation; it should
always be used in conjunction with server-side form validation. Client-side form
validation is a nice tool, though, and can be used in addition to server-side form
validation.
An example of client-side form validation can be seen in Example 5-2.
Example 5-2. Client-Side Form Validation
<HTML>
<HEAD>
<SCRIPT LANGUAGE="JavaScript">

</SCRIPT>
</HEAD>
<BODY>
<FORM NAME=frmInfo METHOD=POST ACTION="somePage.asp"
ONSUBMIT="return ValidateData();">
Enter your zip code:<BR>
<INPUT TYPE=TEXT NAME=Zip SIZE=10>
<P>
<INPUT TYPE=SUBMIT>
</FORM>
</BODY>
</HTML>
Example 5-2. Client-Side Form Validation (continued)
114 Chapter 5: Form Reuse
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Of course, this disadvantage is becoming less and less of an issue as Internet connection
speeds become faster and faster. However, with a large number of users
still connecting via a modem, this disadvantage is still a legitimate concern. This
drawback of server-side validation, though, can be compensated with adequate client-
side form validation. When both client-side and server-side form validation are
used, the server-side form validation should duplicate the validation performed on
the client side. In such a scenario, the server-side form validation serves only as a
safety catch in case the user either mistakenly steps around the client-side form
validation (perhaps they have JavaScript disabled), or purposely avoids client-side
validation (a malicious user).
In Example 5-2, we used client-side form validation to validate a zip code. In
Example 5-3, the same data is validated, but this time using server-side validation.
It is recommended you use both client-side and server-side form validation when
validating user form input.
Example 5-3. Classical Server-Side Form Validation
<% @LANGUAGE="VBSCRIPT" %>
<% Option Explicit %>
<%
'Read in the form variable
Dim strZip, iLoop
strZip = Request("Zip")
'Check to make sure the zip contains either ##### or #####-####
If Len(strZip) = 5 or Len(strZip) = 10 then
' ok, the zip code is the correct length, make sure
' it contains all numeric characters if length is
' 5, or, if length is 10, contains 5 numeric, a dash,
' and 4 numeric
If Len(strZip) = 5 then
For iLoop = 1 to Len(strZip)
If Asc(Mid(strZip, iLoop, 1)) < Asc("0") or _
Asc(Mid(strZip, iLoop, 1)) > Asc("9") then
Response.Write "You entered an invalid zip code."
Response.End
End If
Next
Else
' the string is 10 characters long
For iLoop=1 to 5
If Asc(Mid(strZip, iLoop, 1)) < Asc("0") or _
Asc(Mid(strZip, iLoop, 1)) > Asc("9") then
Response.Write "You entered an invalid zip code."
Response.End
End If
Next
If Asc(Mid(strZip, iLoop, 6)) <> Asc("-") then
Response.Write "You entered an invalid zip code."
Response.End
Form Validation 115
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
In Example 5-3, classical VBScript methods were used to validate a particular
input. While these methods are sufficient, they leave a lot to be desired. With the
VBScript 5.0 scripting engine, a new regular expression object has been added,
allowing regular expression pattern matching in VBScript!
Regular expressions offer more robust form-validation techniques than simply
using the techniques in Example 5-3. Example 5-4 uses regular expression matching
to perform the same form validation as in Example 5-3.
End If
For iLoop=7 to Len(strZip)
If Asc(Mid(strZip, iLoop, 1)) < Asc("0") or _
Asc(Mid(strZip, iLoop, 1)) > Asc("9") then
Response.Write "You entered an invalid zip code."
Response.End
End If
Next
End If
Else
Response.Write "You entered an invalid zip code."
Response.End
End If
' If we reached here, we had valid input, continue
' with form processing...
Response.Write "You entered valid input, thanks!"
%>
Example 5-4. Server-Side Form Validation via Regular Expressions
<% @LANGUAGE="VBSCRIPT" %>
<% Option Explicit %>
<%
'Read in the form variable
Dim strZip, iLoop
strZip = Request("Zip")
'Check to make sure the zip contains either ##### or #####-####
Dim objRegExp
Set objRegExp = New regexp
objRegExp.Pattern = "^\d{5}(-\d{4})?$"
If Not objRegExp.Test(strZip) then
Response.Write "Invalid input - zip code has an invalid value!"
Response.End
End If
'Otherwise, we have valid, so do stuff!
' ...
Response.Write "Valid data!!!"
%>
Example 5-3. Classical Server-Side Form Validation (continued)
116 Chapter 5: Form Reuse
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
For more information on form validation, be sure to check out the following online
resources:
• “Form Validation Using JavaScript” at http://www.4guysfromrolla.com/webtech/
091998-1.shtml
• “Server-side Form Validation” found at http://www.4guysfromrolla.com/
webtech/120199-1.shtml
• “Advanced Form Validation” found at http://www.asptoday.com/articles/
19990708.htm
• “Validate What? Using Regular Expressions to Validate Input” at http://www.
4guysfromrolla.com/webtech/050399-2.shtml
Creating Reusable Server-Side Form
Validation Routines
Earlier in this chapter, in the section “Code Reuse in ASP,” we discussed the three
common steps taken to create a form:
1. Create an HTML form with the proper form fields, complete with client-side
form validation.
2. Create a form-processing script and add the needed code.
3. Add server-side form validation code to the ASP page created in Step 2.
At that time, I mentioned we would look at ways to create robust, reusable code
for Steps 1 and 3. Since server-side form validation is paramount to client-side
form validation, it makes sense to start by creating code for Step 1. In this section,
we will look at how to create a single, reusable ASP page that uses regular expressions
to validate form data. In “Developing Reusable Form Creation Routines,” the
discussion will switch to creating a class interface to aid in accomplishing Step 1.
Our reusable server-side form validation routine will exist on the /CODEREUSE/
ValidateForm.asp ASP page. (That means every form on our web site will have its
ACTION property set to /CODEREUSE/ValidateForm.asp.) This page will contain
code that will iterate through the proper collection of the Request object, reading
in the value of each form field and determining whether or not the value is valid.
Our validation page, /CODEREUSE/ValidateForm.asp, needs to know what constitutes
a valid entry for each and every form field. We’ll supply this information in
the NAME property of the form field. Since regular expressions are used to validate
form field input, each form field that needs to be validated will contain the proper
regular expression appended to the NAME property. For example, if we had a form
Creating Reusable Server-Side Form Validation Routines 117
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
field for the user to enter his telephone number, we would use the following
HTML to create the text box:
<FORM METHOD="POST" ACTION="/CODEREUSE/ValidateForm.asp">
<INPUT TYPE="HIDDEN" NAME="Collection" VALUE="Form">
<INPUT TYPE="HIDDEN" NAME="Redirect" VALUE="/SomePage.asp">

Enter your phone number in the following format: ###-###-####<BR>
<INPUT TYPE="TEXT" NAME="Phone Number!^\d{3}-\d{3}-\d{4}$">
<P>

Enter your first name:<BR>
<INPUT TYPE="TEXT" NAME="First Name">
<P>
<INPUT TYPE="SUBMIT" VALUE="Submit!">
</FORM>
Note that an exclamation point separates the English-like name of the form field
(txtPhone) and the regular expression (\d{3}-\d{3}-\d{4}). The regular
expression will be used to validate the user’s input in /CODEREUSE/
ValidateForm.asp. Also note that the form’s ACTION property directs the submitted
form to /CODEREUSE/ValidateForm.asp.
The above form also contains two HIDDEN form fields. The first HIDDEN form field,
Collection, specifies what Request object collection to use. The Value for this
HIDDEN form field would have been QueryString, had the form’s METHOD been
set to GET. The second HIDDEN form field, Redirect, specifies the ASP page that
is responsible for processing the form data. This page (via a Server.Transfer) is
called only if the submitted form data is valid.
What Format Do You Want That In?
If you are going to require that your users enter data in a particular format
(such as a phone number in ###-###-#### format), make sure you let your
users know what format they need to enter the information in!
A common mistake web developers make is not letting their users know what
format they are expected to enter the data in. Therefore, if the user enters the
data in an invalid format, she may become agitated, especially if she is continually
asked to re-enter “valid” data when she doesn’t know what the “valid”
format is!
118 Chapter 5: Form Reuse
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Be absolutely sure to place the quotes around the form field NAME
value! If not, any spaces in the form field NAME will cut off the rest of
the NAME when passed to the server-side validation script. If you
need to have double quotes in the regular expression, you must represent
the double quote with some other character and have it transformed
back into a double quote in the server-side validation script.
The code for /CODEREUSE/ValidateForm.asp will perform the following steps:
1. Determine what Request collection to use. (If this isn’t specified via the
Collection HIDDEN variable, the Request.Form collection is used by default.)
2. Iterate through the proper collection, checking to see what form fields contain
an exclamation point.
3. For those elements that do contain an exclamation point, read in the regular
expression pattern and perform a regular expression match using the Test
method of the RegExp object.
4. If the Test method returns False, then the input is invalid. Note this invalid
entry in an error log.
5. Once the entire collection has been iterated, determine whether any validation
errors occurred. If so, output these errors. If not, use Server.Transfer to
forward the user onto the form-processing script specified by the HIDDEN variable
Redirect.
Quite a tall order for /CODEREUSE/ValidateForm.asp! Here is the code, with the
regular expression validation emphasized:
<%@ LANGUAGE = "VBSCRIPT" %>
<% Option Explicit %>
<%
Function EnglishName(str)
'Check to see if there is an exclamation point: if so, hack off
'all contents of the string to the right of the exclamation point
If InStr(1,str,"!") > 0 then
EnglishName = Left(str,InStr(1,str,"!") - 1)
End If
End Function
'Determine what collection to use
Dim strCollection
strCollection = Request("Collection")
'Create a reference to the property Request collection
Dim colRequestCol
If Ucase(strCollection) = "QUERYSTRING" then
Set colRequestCol = Request.QueryString
Else
Creating Reusable Server-Side Form Validation Routines 119
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Set colRequestCol = Request.Form
End If
Dim strItem, strErrors, objRegExp, strErrorLog
Set objRegExp = New regexp ' Create a regexp instance
objRegExp.IgnoreCase = True
objRegExp.Global = True
'Iterate through each of the form field elements
For Each strItem in colRequestCol
'See if there is an exclamation point. If there is,
'then we need to perform form validation
If InStr(1,strItem,"!") > 0 then
'Grab the regular expression pattern
objRegExp.Pattern = Mid(strItem, InStr(1,strItem,"!") + 1, Len(strItem))
If Not objRegExp.Test(colRequestCol(strItem)) then
'Input invalid! Append to the error log
strErrorLog = strErrorLog & "<BR>Invalid Input for " & _
EnglishName(strItem)
End If
End If
Next
'Are there any errors?
If Len(strErrorLog) > 0 then
Response.Write "<B>The following validation errors occurred:</B>"
Response.Write strErrorLog
Else
'No form validation errors occurred!
'Use Server.Transfer to send the user to the proper
'form validation script
'If the user didn't specify a redirect, raise an error
If Len(Request("Redirect")) = 0 then
Raise vbObjectError + 1010, "Validation Error", "Redirect not specified"
Else
Server.Transfer Request("Redirect")
End If
End If
Set colRequestCol = Nothing ' Clean up
Set objRegExp = Nothing
%>
Keep in mind that /CODEREUSE/ValidateForm.asp serves only one purpose—to
perform server-side validation. If there are no validation errors, the script transfers
control to the form-processing page specified by the HIDDEN form field Redirect.
Due to the fact that /CODEREUSE/ValidateForm.asp only performs server-side validation
and transfers the user to the correct form-processing script, all of our HTML
pages can call this single validation page. This means once this system is implemented,
when creating a form, you will not have to code its server-side validation!
Figure 5-1 further illustrates the page flow that occurs when a form is
submitted.
120 Chapter 5: Form Reuse
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Now, let’s look at how to encapsulate the logic in /CODEREUSE/ValidateForm.asp
through the use of classes!
Figure 5-1. /CODEREUSE/ValidateForm.asp validates all forms on the site
Beware of Malicious Users
As discussed in the previous section “Server-Side Form Validation,” one of the
reasons to use server-side form validation is to prevent malicious users from
circumventing form validation. With the approach shown in /CODEREUSE/
ValidateForm.asp, a malicious user could easily bypass the server-side form
validation. By simply stripping the exclamation point and the regular expression
that follows it from the form element NAMEs, a malicious user could send
invalid data to the form-processing script that wouldn’t be caught by the serverside
form-validation routine.
Due to this issue, this server-side validation approach should not be used
where malicious users might attempt to bypass validation. For example, on a
large, publicly accessible e-commerce site, such a form-validation technique
should not be applied to vital forms, such as billing information forms. This is
a trade off that must be made for this application. When building a generic,
reusable script, often other sacrifices must be made.
Note that this server-side form-validation approach will still work for those
users who have JavaScript disabled. Since the form field validation information
is stored in the form element NAMEs, and since the actual validation is performed
on the server, it does not matter if the user has JavaScript enabled.
ASP PAGE
WITH FORM
ASP PAGE
WITH FORM
ASP PAGE
WITH FORM
/CODEREUSE/ValidateForm.asp
This page is called by all of
the forms on the web site.
FORM-PROCESSING
SCRIPT
FORM-PROCESSING
SCRIPT
Every ASP page that creates a form
sets the form's ACTION property to
redirect to the global form validation
script. Passed along is the ASP page
URL that needs to be called once the
data is validated.
If the form data is valid,
the suitable formprocessing
script is called.
Creating Reusable Server-Side Form Validation Routines 121
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Using Classes to Enhance Server-Side Validation
As discussed in Chapter 4, classes provide a means to encapsulate complexity. Our
first implementation of /CODEREUSE/ValidateForm.asp is rich in complexity. The
developer who wishes to make any modifications to this page must know that the
names of form variables may contain an exclamation point and regular expression
code. It would be nice to hide such complexity.
There is also a lot of complexity in creating the HTML forms. We’ll
look at ways to reduce this complexity using classes later on!
When encapsulating the server-side form-validation routines into a class, the functionality
remains constant—only the implementation changes. For that reason, creating
the class is mostly cut and paste work! A few enhancements have been made
to the new class version, though. The major enhancement is the use of a Dictionary
object to store the validation error information (objErrorDict). The changes
made due to this enhancement are emphasized in the following code snippet:
<%
Class FormData
'****************** PROPERTIES ********************
Private colRequestCol
Private objRegExp
Private objErrorDict
'**************************************************
'************** INITIALIZE/TERMINATE **************
Private Sub Class_Initialize()
'Determine what Request collection to use
If Ucase(Request("Collection")) = "QUERYSTRING" then
Set colRequestCol = Request.QueryString
Else
Set colRequestCol = Request.Form
End If
'Instantiate a regexp object
Set objRegExp = New regexp
objRegExp.IgnoreCase = True
objRegExp.Global = True
'Instantiate the error log
Set objErrorDict = CreateObject("Scripting.Dictionary")
End Sub
Private Sub Class_Terminate()
Set colRequestCol = Nothing
122 Chapter 5: Form Reuse
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Set objRegExp = Nothing
Set objErrorDict = Nothing
End Sub
'**************************************************
'************* PROPERTY GET STATMENT **************
Public Property Get ErrorLog()
'Return the Error Log Dictionary Object
Set ErrorLog = objErrorDict
End Property
'**************************************************
'********************* METHODS ********************
Public Function ValidInputs()
'Checks to see if data is valid. If it is, returns
'True, else returns False. A list of errors can be
'obtained through PrintErrors
Dim strItem
'Iterate through each of the form field elements
For Each strItem in colRequestCol
'See if there is an exclamation point. If there is,
'then we need to perform form validation
If InStr(1,strItem,"!") > 0 then
'Grab the regular expression pattern
objRegExp.Pattern = Mid(strItem, InStr(1,strItem,"!") + 1, _
Len(strItem))
If Not objRegExp.Test(colRequestCol(strItem)) then
'Input invalid! Append to the error log
objErrorDict.Add EnglishName(strItem), _
colRequestCol(strItem).Item
End If
End If
Next
'Did we encounter any errors?
If objErrorDict.Count > 0 then
ValidInputs = False
Else
ValidInputs = True
End If
End Function
Private Function EnglishName(ByVal str)
'Check to see if there is an exclamation point: if so, hack off
'all contents of the string to the right of the exclamation point
If InStr(1,str,"!") > 0 then
str = Left(str,InStr(1,str,"!") - 1)
End If
Creating Reusable Server-Side Form Validation Routines 123
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
EnglishName = Replace(str, "_", " ")
End Function
Public Function ErrorMessage()
'Returns a string containing an error message
If objErrorDict.Count = 0 then
ErrorMessage = "There were no validation errors."
Else
Dim strName
For Each strName in objErrorDict
ErrorMessage = ErrorMessage & "Error in " & strName & _
" - the entry " & objErrorDict(strName) & _
" is invalid." & vbCrLf
Next
End If
End Function
'**************************************************
End Class
%>
If we place the above class definition into a file, then we can use a server-side
include to import the class definition into the ASP pages that are interested in
using the class. Classes reduce complexity by hiding implementation details, and
as they say, the proof is in the pudding. Previously, /CODEREUSE/ValidateForm.
asp consisted of 57 lines of code. Using classes and include files, the line count
has been reduced drastically to 18 lines!
<%@ LANGUAGE = "VBSCRIPT" %>
<% Option Explicit %>

<%
'Instantiate our class defined in FormData.Class.asp
Dim objFormData
Set objFormData = New FormData
'Test to see if form inputs are valid
If objFormData.ValidInputs() then
'No form validation errors occurred!
'Use Server.Transfer to send the user to the proper
'form validation script
'If the user didn't specify a redirect, raise an error
If Len(Request("Redirect")) = 0 then
Raise vbObjectError + 1010, "Validation Error", "Redirect not specified"
Else
Server.Transfer Request("Redirect")
End If
Else
'Display an error message
Response.Write objFormData.ErrorMessage()
End if
Set objFormData = Nothing
%>
124 Chapter 5: Form Reuse
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
If you are a little rusty on using include files within an ASP page, be
sure to review Chapter 1 and check out the following online article
“The low-down on #includes” at http://www.4GuysFromRolla.com/
webtech/080199-1.shtml.
Further enhancements to the server-side validation class
There are, of course, numerous improvements that could be made to the serverside
validation class. One major improvement would be to have a method that not
only displays the invalid form entries, but also displays the form the user just filled
out, so they can quickly make the changes they need to.
One possible way to accomplish this would be to have the form supply yet
another HIDDEN variable that would contain the URL to the HTML form. Then a
method could be added that would perform a Server.Transfer back to the form
after displaying the invalid form entries. That way, the user would see a list of the
invalid form fields at the top of the HTML page, and then would be shown the
form again. (This would save the user from having to click the back button when
he or she has entered invalid form data.)
Developing Reusable Form Creation
Routines
As promised, in this section we will examine reusable methods to create HTML
forms with the proper form fields, complete with client-side form validation. These
methods will hide the specific details of properly naming form fields and will not
bother the developer with creating client-side form validation routines.
Form Implementation Complexity
Usually, creating a form is a trivial task; it’s just a matter of HTML coding. However,
due to our reusable server-side validation code, creating a form has become a much
more complex issue. When using a global server-side form validation page that all
forms initially submit to, all the forms need to follow a certain standard. Specifically,
the NAME of each form element that requires form validation must contain an
exclamation point followed by the validation regular expression. Furthermore, the
form’s ACTION property must direct to the validation page (/CODEREUSE/
ValidateForm.asp) and pass two HIDDEN variables: Collection and Redirect.
Remembering these specific details is cumbersome and serves as an easy place to
make a mistake, resulting in a bug. Also, if other developers are to be working on
Developing Reusable Form Creation Routines 125
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
creating the HTML forms, they will be less familiar with the conventions specified
by the server-side form validation script, /CODEREUSE/ValidateForm.asp.
As discussed earlier, client-side form validation is a nice option, although not as
critical as server-side form validation. As with server-side form validation, clientside
form validation lends itself to reuse.
Masking Implementation Complexity Using Classes
To reduce implementation complexity and provide a black box-like interface, we’ll
use classes! Two classes can be created that will provide the following benefits:
• Provide the end developer with an easy interface to create forms that contain
complex implementation details.
• Provide automatic code generation for client-side form validation.
The first class defines a form field element as an object. Each form field object has
a certain set of properties: NAME, TYPE, SIZE, and VALUE are four standard ones.
For our custom server-side validation routines, an optional validation regular
expression is yet another property. Our rather simplistic form field class,
FormElement, contains the following definition, which is stored in the file
CODEREUSE\FormCreation.Class.asp:
<%
Class FormElement
'****************** PROPERTIES ********************
Private strName
Private strRegExp
Private strType
Private strPreHTML
Private strPostHTML
Private strSize
Private strValue
'**************************************************
'************* PROPERTY LET STATMENTS *************
Public Property Let Name(str)
strName = str
End Property
Public Property Let RegularExpression(str)
strRegExp = str
End Property
Public Property Let ElementType(str)
'Only one of six types possible: SELECT, TEXTAREA, TEXT,
' RADIO, HIDDEN, or CHECKBOX (TEXT is the default)
If Ucase(str) = "TEXT" OR Ucase(str) = "SELECT" _
OR Ucase(str) = "TEXTAREA" OR Ucase(str) = "RADIO" _
OR Ucase(str) = "CHECKBOX" OR Ucase(str) = "HIDDEN" then
strType = str
126 Chapter 5: Form Reuse
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Else
strType = "TEXT" 'TEXT is the default type
End If
End Property
Public Property Let Size(str)
strSize = str
End Property
Public Property Let PreHTML(str)
strPreHTML = str
End Property
Public Property Let PostHTML(str)
strPostHTML = str
End Property
Public Property Let Value(str)
strValue = str
End Property
'**************************************************
'************* PROPERTY GET STATMENTS *************
Public Property Get Name()
Name = strName
End Property
Public Property Get RegularExpression()
RegularExpression = strRegExp
End Property
Public Property Get ElementType()
ElementType = strType
End Property
Public Property Get Size()
Size = strSize
End Property
Public Property Get PreHTML()
PreHTML = strPreHTML
End Property
Public Property Get PostHTML()
PostHTML = strPostHTML
End Property
Public Property Get Value()
Value = strValue
End Property
'**************************************************
'********************* METHODS ********************
Public Sub Clear()
strName = ""
Developing Reusable Form Creation Routines 127
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
strRegExp = ""
strType = ""
strSize = ""
strPreHTML = ""
strPostHTML = ""
strValue = ""
End Sub
'**************************************************
End Class
%>
A FormElement class instance represents a discrete form field element in a form.
This implementation of FormElement contains only a small subset of the properties
a form field element can contain. If your web site requires more elegant form
fields, add the needed form field properties in the above class. The following list
shows the various properties of the FormElement class:
Name (Read/Write)
Specifies the NAME property of the form element. It is essential this property is
specified. Furthermore, it is recommended that you give the NAME property an
English-like name, since if there is a validation error, this property is reported
in the error message.
RegularExpression (Read/Write)
Specifies the regular expression validation code. Remember the validation routine
will search to see if the user’s input matches this regular expression.
ElementType (Read/Write)
Specifies the TYPE for the form element. This can have one of six possible values:
SELECT, TEXTAREA, TEXT, RADIO, CHECKBOX, or HIDDEN, which create a
list box, a multi-row text box, a single-row text box, a radio button, a checkbox,
or a hidden form field, respectively.
PreHTML (Read/Write)
This property specifies any HTML that should occur before the actual INPUT tag.
PostHTML (Read/Write)
This property specifies any HTML that should occur after the actual INPUT tag.
Size (Read/Write)
This property specifies the SIZE property for the form element. In a multi-row
text box, there are as many columns as specified by the Size property and
one-fourth as many rows.
Value (Read/Write)
This property specifies the VALUE property of the INPUT tag. It is meaningless
if you are creating a list box, multi-row text box, radio button, or checkbox.
The Value property is very useful when creating HIDDEN form fields.
128 Chapter 5: Form Reuse
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
The FormElement class uses the Clear method, as described in the following
definition:
Clear
The Clear method erases all of the property values. Normally such a method
wouldn’t be needed. The sidebar “Adding Form Elements” later in this chapter
explains why I chose to supply such a method.
We’ll look at how to use the FormElement class and its properties and methods
once we examine the GenerateForm class.
GenerateForm is the class used to create the HTML that will physically construct
the form on the users’ browsers. Each function of GenerateForm’s properties is
outlined here:
Collection (Read/Write)
Specifies whether to submit the form using an ACTION of POST (the default) or
GET. To use GET, Collection must be set equal to QueryString. If it equals any
other value, the form will submit with an ACTION of POST.
Redirect (Read/Write)
The URL of the actual form-processing script. Once the data is validated by
/CODEREUSE/ValidateForm.asp, this is the page the user is passed to.
FormName (Read/Write)
Specifies the NAME property of the form. If not specified, frmForm1 is used. If
you plan to use the GenerateForm class to create more than one form on an
HTML page, it is essential that each form have its own, unique FormName for
client-side form-validation purposes.
FormAction (Read/Write)
Specifies the URL of the server-side form validation script. Set to /CODEREUSE/
ValidateForm.asp in the Initialize event, but can be modified.
SubmitTitle (Read/Write)
The title for the form’s submit button. The title, if not specified, defaults to
“Submit Query!”
objFormElementDict (Write (indirectly))
This property is indirectly write-only. This property is a Dictionary object that
contains zero to many HTML strings that correspond to form field elements.
To add a new form field element, use the AddElement method.
The following list discusses the GenerateForm class’s methods:
AddElement(objFormElement)
Adds a form field element to objFormElementDict. objFormElement is
expected to be an instance of the FormElement class. A more thorough
Developing Reusable Form Creation Routines 129
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
discussion of why this technique is used to add a form element appears in a
later sidebar.
GenerateForm( )
Returns the HTML that will generate a form specified by the form field elements
in objFormElementDict.
GenerateValidation( )
Returns the client-side JavaScript code that will validate the form field inputs.
The client-side JavaScript takes advantage of regular expressions; hence the
client-side validation code will only work properly in browsers that support
JavaScript 1.2 or greater: Internet Explorer or Netscape Navigator 4.0 and up.
GenerateHTMLDocument(strTitle)
Returns the entire HTML document in one string. Only useful if the HTML
page will contain one form and one form only. The strTitle parameter displays
a title at the beginning of the HTML page (right after the <BODY> tag).
GenerateHTMLDocument(X) is synonymous with:
<HTML><HEAD>
GenerateValidation()
</HEAD><BODY>
X
GenerateForm()
</BODY></HTML>
The GenerateForm class definition, which is listed in Example 5-5 and is also
stored in the file \CODEREUSE\FormCreation.Class.asp along with the definition
of the FormElement class, is fairly straightforward and easy to use. To create a
form, two object instances are needed: an instance of GenerateForm and an
instance of FormElement. Start by creating the GenerateForm instance and set its
properties. Next, create a single FormElement instance, set the needed properties
in the FormElement instance, and call the AddElement method of the
GenerateForm instance, passing in the FormElement instance. Finally, call the
FormElement instance’s Clear method and repeat the previous three steps for
each element in the form.
Example 5-5. The GenerateForm Class Definition
<%
Class GenerateForm
'****************** PROPERTIES ********************
Private strRedirect
Private strCollection
Private strFormAction
Private objFormElementDict
Private strFormName
Private strSubmitTitle
'**************************************************
130 Chapter 5: Form Reuse
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
'************** INITIALIZE/TERMINATE **************
Private Sub Class_Initialize()
'Set the default property values
strCollection = "FORM"
strFormAction = "/CODEREUSE/ValidateForm.asp"
strFormName = "frmForm1"
strRedirect = Request.ServerVariables("SCRIPT_NAME")
strSubmitTitle = "Submit Query!"
Set objFormElementDict = CreateObject("Scripting.Dictionary")
End Sub
Private Sub Class_Terminate()
Set objFormElementDict = Nothing ' Clean up
End Sub
'**************************************************
'************* PROPERTY LET STATMENTS *************
Public Property Let Collection(str)
'Set the strCollection private property, making sure it is
'set to either QueryString or Form (default to Form)
If Ucase(str) = "QUERYSTRING" then
strCollection = "QueryString"
Else
strCollection = "Form"
End If
End Property
Public Property Let Redirect(str)
'Set the strRedirect private property; if trying to assign it
'a blank string, assign it the value of the current ASP page
If Len(str) = 0 then
strRedirect = Request.ServerVariables("SCRIPT_NAME")
Else
strRedirect = str
End If
End Property
Public Property Let FormAction(str)
'Set the strFormAction private property
strFormAction = str
End Property
Public Property Let FormName(str)
'Set the strFormName private property
strFormName = str
End Property
Public Property Let SubmitTitle(str)
'Set the strSubmitTitle private property
If Len(str) = 0 then
strSubmitTitle = "Submit Query!"
Example 5-5. The GenerateForm Class Definition (continued)
Developing Reusable Form Creation Routines 131
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Else
strSubmitTitle = str
End If
End Property
'**************************************************
'************* PROPERTY GET STATMENTS *************
Public Property Get Collection()
Collection = strCollection
End Property
Public Property Get Redirect()
Redirect = strRedirect
End Property
Public Property Get FormAction()
FormAction = strFormAction
End Property
Public Property Get FormName()
FormName = strFormName
End Property
Public Property Get SubmitTitle()
SubmitTitle = strSubmitTitle
End Property
'**************************************************
'********************* METHODS ********************
Public Function AddElement(objFormElement)
'Adds a form field element to the objFormElementDict object
'Expects to be passed a valid objFormElement instance
Dim strHTML, strTechnicalName
'Determine if this form field needs to be validated
If Len(objFormElement.RegularExpression) > 0 then
strTechnicalName = objFormElement.Name & "!" & _
objFormElement.RegularExpression
Else
strTechnicalName = objFormElement.Name
End If
'Determine what form field type we are dealing with
If objFormElement.ElementType = "SELECT" then
strHTML = objFormElement.PreHTML & vbCrLf & _
"<SELECT NAME=""" & objFormElement.Name & """"
If Len(objFormElement.Size) > 0 then
strHTML = strHTML & " SIZE=" & objFormElement.Size
End If
strHTML = strHTML & ">" & vbCrLf & _
objFormElement.PostHTML & _
vbCrLf & "</SELECT>" & vbCrLf & vbCrLf
Example 5-5. The GenerateForm Class Definition (continued)
132 Chapter 5: Form Reuse
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Elseif objFormElement.ElementType = "TEXTAREA" then
strHTML = objFormElement.PreHTML & vbCrLf & _
"<TEXTAREA NAME=""" & strTechnicalName & """"
If Len(objFormElement.Size) > 0 then
strHTML = strHTML & " COLS=" & objFormElement.Size & _
" ROWS=" & objFormElement.Size / 4
End If
strHTML = strHTML & "></TEXTAREA>" & _
vbCrLf & objFormElement.PostHTML & _
vbCrLf & vbCrLf
Else 'must be one of the other types
strHTML = objFormElement.PreHTML & vbCrLf & _
"<INPUT NAME=""" & strTechnicalName & """"
If Len(objFormElement.Size) > 0 then
strHTML = strHTML & " SIZE=" & objFormElement.Size
End If
If Len(objFormElement.Value) > 0 then
strHTML = strHTML & " VALUE=""" & objFormElement.Value & """"
End If
strHTML = strHTML & " TYPE=" & objFormElement.ElementType & _
">" & vbCrLf & objFormElement.PostHTML & _
vbCrLf & vbCrLf
End If
'Add the HTML to the FormElement dictionary
objFormElementDict.Add strTechnicalName, strHTML
End Function
Public Function GenerateForm()
'Iterates through the objFormElementDict collection and
'generates the resulting form
Dim strResultingForm
strResultingForm = "<FORM NAME=""" & strFormName & _
""" METHOD="
If strCollection = "QueryString" then
strResultingForm = strResultingForm & "GET"
Else
strResultingForm = strResultingForm & "POST"
End If
strResultingForm = strResultingForm & " ACTION=""" & _
strFormAction & """ ONSUBMIT=""return ValidateData();"">" & _
vbCrLf
'Add the HIDDEN form variables
strResultingForm = strResultingForm & vbTab & "<INPUT TYPE=HIDDEN " & _
"NAME=Collection VALUE=""" & strCollection & """>" & vbCrLf & _
Example 5-5. The GenerateForm Class Definition (continued)
Developing Reusable Form Creation Routines 133
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
vbTab & "<INPUT TYPE=HIDDEN NAME=Redirect VALUE=""" & _
strRedirect & """>" & vbCrLf
'Iterate through the form element dictionary, outputting the
'form field elements
Dim strName
For Each strName in objFormElementDict
strResultingForm = strResultingForm & vbTab & objFormElementDict(strName)
Next
strResultingForm = strResultingForm & "<P><INPUT TYPE=SUBMIT VALUE=""" & _
strSubmitTitle & """>" & vbCrLf & vbCrLf
strResultingForm = strResultingForm & "</FORM>" & vbCrLf & vbCrLf
GenerateForm = strResultingForm
End Function
Public Function GenerateValidation()
'Creates the client-side validation code
Dim strCode
strCode = "<SCRIPT LANGUAGE=""JavaScript"">" & vbCrLf & "" & vbCrLf & "</SCRIPT>" & vbCrLf
GenerateValidation = strCode
End Function
Public Function GenerateHTMLDocument(strTitleHTML)
'This method generates the HTML/BODY tags, the form and client-side
'form validation all in one call
Dim strResultHTML
strResultHTML = "<HTML><HEAD>" & vbCrLf & GenerateValidation() & _
Example 5-5. The GenerateForm Class Definition (continued)
134 Chapter 5: Form Reuse
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
For example, the following code will create a form with four form fields (two text
boxes, a text area, and a list box). The first two text boxes are validated on the
client side when the form is submitted (and will also be subject to validation on
the server side when submitted to /CODEREUSE/ValidateForm.asp). For the
example, the two class definitions were placed in an separate file and included in
the \05FormCreationDemo.asp file, whose contents are as follows:
<%@ LANGUAGE = "VBSCRIPT" %>
<% Option Explicit %>

<%
'Create an instance of the GenerateForm class
Dim objGenerateForm
Set objGenerateForm = New GenerateForm
'Set the objGenerateForm properties
objGenerateForm.Collection = "Form"
objGenerateForm.Redirect = "/scripts/Something.asp"
objGenerateForm.SubmitTitle = "--- GO! ---"
'Create an instance of the FormElement class
Dim objFormElement
Set objFormElement = New FormElement
'*** Create the form field elements ***
'Create an age form field element
objFormElement.Name = "Age"
objFormElement.Size = 4
objFormElement.ElementType = "TEXT"
objFormElement.RegularExpression = "\d{1,2}"
objFormElement.PreHTML = "What is your Age?<BR>"
objGenerateForm.AddElement objFormElement 'Add it to the form
'Create an email form field element
objFormElement.Clear 'clear out the form element values before reuse
objFormElement.Name = "Email"
objFormElement.ElementType = "TEXT"
objFormElement.RegularExpression = "^[a-z_0-9\-\.]+@\w+\.\w+$"
objFormElement.PreHTML = "<P>What is your email address?<BR>"
objGenerateForm.AddElement objFormElement 'Add it to the form
"</HEAD>" & vbCrLf
strResultHTML = strResultHTML & "<BODY>" & vbCrLf & strTitleHTML & _
vbCrLf & GenerateForm() & vbCrLf & "</BODY></HTML>"
GenerateHTMLDocument = strResultHTML
End Function
'**************************************************
End Class
%>
Example 5-5. The GenerateForm Class Definition (continued)
Developing Reusable Form Creation Routines 135
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
'Create a comment form field with no validation
objFormElement.Clear 'clear out the form element values before reuse
objFormElement.Name = "User Comments"
objFormElement.ElementType = "TEXTAREA"
objFormElement.PreHTML = "<P>Any comments?<BR>"
objFormElement.Size = 40
objGenerateForm.AddElement objFormElement 'Add it to the form
'Create a list box
objFormElement.Clear 'clear out the form element values before reuse
objFormElement.Name = "Experience"
objFormElement.ElementType = "SELECT"
objFormElement.PreHTML = "<P>How many years have you been on the Net?<BR>"
objFormElement.Size = 1
objFormElement.PostHTML = "<OPTION VALUE=1>One Year</OPTION>" & vbCrLf & _
"<OPTION VALUE=2>Two Years</OPTION>"
objGenerateForm.AddElement objFormElement 'Add it to the form
'Generate the entire HTML document
Response.Write objGenerateForm.GenerateHTMLDocument("<H1>Information</H1>")
'(synonymous to doing:
' Response.Write "<HTML><HEAD>"
' Response.Write objGenerateForm.GenerateValidation()
' Response.Write "</HEAD><BODY>"
' Response.Write objGenerateForm.GenerateForm()
' Response.Write "</BODY></HTML>")
Set objFormElement = Nothing 'Clean up!
Set objGenerateForm = Nothing 'Clean up!
%>
Note that the GenerateHTMLDocument method was used to generate the HTML
output. A simple Response.Write sends the form HTML to the client. As with any
generalized implementation, there are a few caveats when using the
GenerateForm class:
• The value of the Name property of the FormElement class should not have a
double quote. Including one will result in unexpected behavior.
• When creating a list box, it is imperative the series of HTML tags used to create
the list box items be inserted in the PostHTML property. (The Experience
form field properties provide a good example for creating a select box.)
• These classes are not designed to create complex forms. The HTML they output
is not pretty, and is not intended to be so. However, refinements of the
GenerateForm and FormElement classes could be made to accommodate
more complex forms and more detailed form fields.
136 Chapter 5: Form Reuse
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
The Practicality of Reuse
Earlier in the chapter, the three general steps of form creation were outlined as
follows:
1. Create an HTML form with the proper form fields, complete with client-side
form validation.
2. Create a form-processing script and add the needed code.
3. Add server-side form-validation code to the ASP page created in Step 2.
In this chapter we examined code that took advantage of the repetitive nature of
Steps 1 and 3. How practical, though, is code reuse? The answer depends on how
many times we are going to repeat Steps 1 and 3. Since there is such a high initial
cost in developing robust, generic, reusable code, it may not seem sensible to proceed
with such code development unless you plan on reusing the code
extensively.
I strongly suggest that you strive to make all of your code as reusable as possible.
While this may seem time-consuming and a bit overzealous, it will be beneficial in
the long run. For example, creating a reusable form class, as we did in this chapter,
will save time when creating a form in any project! Once the reusable formgeneration
and validation code has been written, the timesaving benefits will be
automatically included for all future projects using forms.
Adding Form Elements
The AddElement method expects an instance of the FormElement class as its
single parameter. Once it receives the object instance, it refers to its various
properties to create an HTML string, which is then stored in the
objFormElementDict object.
A more sound and straightforward approach, in my opinion, would have been
to have an actual instance of the FormElement class added to the dictionary.
Then in the GenerateForm and GenerateValidation methods, the object
instances in the objFormElementDict object would be dissected and converted
into valid HTML strings.
I chose not to use this approach, though, because I didn’t want to have to create
an instance of the FormElement class for every form element. Therefore, I
decided to use the current approach, so that the developer using these classes
could simply reuse a single FormElement object instance. Hence, I included a
Clear method in the FormElement class.
Further Reading 137
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
To summarize: although writing reusable code may seem like a great deal of initial
overhead—and therefore overkill for small projects—the benefits of code reuse
almost always extend far beyond the current project. For this reason, strive to
apply code reuse methodologies to all of your projects.
Further Reading
To learn more about the topics covered in this chapter, be sure to read the following
online articles:
• A good article on the hidden cost of code reuse. As mentioned at the end of
the chapter, it may at times seem like creating reusable code has too high a
cost initially. Read more at http://www.iweek.com/708/08iuhid.htm.
• In this chapter, we looked at using regular expressions on the client-side for
form validation. For more information on this, be sure to read this article, http:
//www.asptoday.com/articles/19990629.htm, which presents a library of regular
expressions for form validation.
• Another good article focusing on form validation using regular expressions can
be found at http://www.4guysfromrolla.com/webtech/050399-2.shtml.
• When creating client-side form-validation routines, it is important you use
JavaScript as the client-side scripting language, since it is the only scripting
language supported by both Internet Explorer and Netscape Navigator.
Detailed technical documentation on JavaScript’s objects, methods, and properties
can be found at: http://developer.netscape.com/docs/manuals/index.
html?content=javascript.html.
138
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Chapter 6 6
Database Reuse
In Chapter 5, Form Reuse, we discussed how to use code reuse mechanisms to
build versatile form-creation and form-validation classes. In this chapter, we will
look at how to build reusable database classes. As with the previous chapter, we
will begin by examining database usage from a high level and look at how to
accomplish various database-related tasks using standard, procedural code. Finally,
we’ll convert these various procedural scripts into a robust set of classes, prime for
reuse!
I think you’ll find the code presented and the topics discussed in this chapter to be
extremely applicable. Database access is common among ASP web sites, and any
steps that can be taken to reduce the time needed to create new database interface
pages are steps in the right direction. However, presenting such a large and
advanced application requires a great deal of time and explanation. This chapter
is, by far, the longest in this book. It will most likely take you a long while to
work through this chapter. If you stick with it, though, your fortitude will pay dividends,
since the application presented in this chapter is very useful in the real
world.
It is assumed that you are knowledgeable in relational database design and theory.
If you are unfamiliar with the SQL syntax, or have never created a relational
data model, it is strongly suggested that you become more familiar with these topics
before beginning on this chapter.
Examining Database Usage
As discussed in Chapter 5, there are a number of typical form interfaces. One
interface used often is the administration pages interface. In this interface, a
Examining Database Usage 139
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
superuser is able to add, edit, and remove records from a particular database table.
With this interface, there is great room for reuse, since all of the administration
pages follow a particular formula:
• Provide a list of all the editable database tables. Each database table name
should be a hyperlink to an administration page for that particular table.
• On each table’s specific administration page, there should be three hyperlinked
options: “Insert a New Record,” “Edit an Existing Record,” and “Delete
an Existing Record.”
— In the “Insert a New Record” page, the user should be presented with a
form that contains form elements for each of the columns in the table.
Once the form is filled out and submitted, a new record should be added
to the table with the values entered by the user.
— In the “Edit an Existing Record” page, the user should be able to select an
existing record in the table. Upon selecting an existing record, the user
should be taken to a page that contains a form similar to the form in the
“Insert a New Record” page. The only difference is the form fields should
have the existing record’s information entered into the various form fields.
When the user has made changes to the record’s data and has submitted
the form, the specified record should be updated with the new values.
— In the “Delete an Existing Record” page, the user should be able to select
an existing record to delete.
The administration page interface lends itself well to reuse since a number of similar
steps need to be repeated for each database table’s administration page. The
steps for creating an administration page for a specific table include:
1. Create a web page with a form that contains a form element for each of the
table’s columns. This is the form the user will visit when they choose “Insert a
New Record.”
2. Create a form-processing script for the form in Step 1 that will add the new
record to the table.
3. Create a web page that lists all of the existing records for the user to choose
from when they want to edit an existing record.
4. Create a web page that displays a form that contains a form element for each
of the table’s columns. The values of these form elements are identical to the
values of the record selected by the user in Step 3.
5. Create a form-processing script for the form created in Step 4. This formprocessing
script is responsible for updating an existing database record
(selected through the form in Step 3) with the values entered into the form in
Step 4.
140 Chapter 6: Database Reuse
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
6. Create a web page that lists all of the existing records for the user to choose
from when they want to delete an existing record.
7. Create a form-processing script for the form in Step 6 that will delete the
record selected by the user.
These seven steps must be repeated for every table in the database that requires an
administration page! Clearly, if any of these steps can be reused for each table’s
administration page, we will save ourselves considerable time. In this chapter, we
will examine how to turn each of the above steps into reusable code. For some of
the steps, a brief discussion will be presented on potential methods that could be
used to make the step highly reusable. For other steps, a more in-depth analysis
will occur, accompanied by code examples. These topics are discussed in detail in
the next section.
You can probably see that some of these steps can be combined,
reducing the overall number of steps needed for each table’s administration
page. For example, the forms in Step 3 and Step 6 could
share a single web page, as could the forms in Step 1 and Step 4.
The Building Blocks for Creating
Reusable Administration Pages
The biggest challenge in creating truly reusable administration pages arrives when
dealing with complex database tables. Imagine, for a moment, that all our tables
were simple database tables that don’t impose foreign key constraints upon other
tables, and have no foreign key constraints imposed upon them (that is, a simple
database table is one that has no explicit relationships with other tables). Rarely
are there many simple database tables in a database. Since relational databases
inherently encourage the developer to create several small, related tables, simple
database tables are found only in simple data models. Therefore, we need a system
that gracefully handles complex tables as well as simple tables.
In this section we’ll examine the theory behind creating powerful, reusable database
administration pages that can be used for complex tables. We’ll begin with a
discussion of what the ideal reusable administration page should contain. Since the
reusable administration pages we’ll later create will make heavy use of ADO schemas,
an entire section is dedicated to this topic.
The Ideal Reusable Administration Page Script
Without code reuse, a developer needs to perform seven steps for each administration
page. With code reuse, however, the need to repeat each of the seven
The Building Blocks for Creating Reusable Administration Pages 141
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
steps for each administration page can be eliminated. Can code reuse help to eliminate
the fact that seven steps are needed, though? The ideal situation would be to
have only one step needed to create an entire set of administration pages.
Imagine that we had a page called DatabaseAdministration.asp that expected one
parameter through the querystring, the table name. This page would then display
the appropriate options for adding, deleting, and updating records, gracefully handling
one-to-many and one-to-one relationships. Furthermore, the page would
know what form elements would be best suited for collecting input from the user.
For example, if a table had a date/time column, it might be wiser in some
instances to use select boxes to represent the month, day, and year; other times, it
might be wiser to use a simple text box.
This single administration page would also need to be able to determine what columns
in a table had foreign key constraints and provide a mechanism to allow the
user to only select applicable values from the related table. For example, imagine
that we had an Employee table with the structure shown in Figure 6-1.
Note that Title is a lookup table, and its Name column contains values like “President,”
“CEO,” “Manager,” “Secretary,” and “Grunt.” Through referential integrity,
TitleID in Employee can contain only integer values that are also present in
TitleID in Title.
How should our administration page handle such foreign keys when adding a new
record or updating an existing record? A simple text box could be used, in which
the user could enter an integer value corresponding to a value in the Title table.
Of course, this would require the user to know that the title “CEO” had a TitleID
of, say, 5.
This is not an acceptable solution, since the user cannot be held responsible for
knowing what titles correspond to what TitleIDs and what TitleIDs are valid.
Rather than seeing a text box, the user should be presented with a list box containing
the acceptable title names. To accomplish this feat, our administration page
would not only have to identify what columns are foreign keys, but would also
have to decide what column from the lookup table to display in the list box!
Figure 6-1. The Title table serves as a lookup table to Employee
Employee
EmployeeID int PK IDENTITY(1,1)
FirstName varchar(50)
LastName varchar(50)
TitleID int FK
...
Title
TitleID int PK IDENTITY(1,1)
Name varchar(50)
There is a one-to-many relationship between Title
and Employee. Title serves as a lookup table for
Employee.
142 Chapter 6: Database Reuse
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Such a single, robust administration page is quite impossible to create, since it
would require that the page be autonomous and know the best way to display
form fields and lookup table lists. We can’t expect the script to be able to make
the best decisions, but what if we could give “hints” to the script to help it decide
how to handle form elements and foreign keys? In this chapter, we will strive to
build a collection of ASP pages that allow the end user to update, insert, and
delete database records with some input from the end developer to help the
administration page decide what methods to use to collect user input.
In creating this set of reusable administration pages, some mechanism for determining
a particular table’s columns and column properties needs to be used. Furthermore,
since these pages strive to be as generic as possible, they shouldn’t be
dependent upon the database used; that is, these administration pages should
work if we use Access, Microsoft SQL Server, or some other database system.
Thankfully, ADO provides a means to collect table and column information
regardless of the database provider.
Database Schemas
One of ADO’s most useful features is schemas. Schemas provide low-level database
information, such as table information, column information, foreign key constraints,
and primary key information, in a high-level format. Using schemas, this
system information can be collected in the same manner for a number of different
database providers.
Most database systems provide their own methods for collecting such low-level
information. For example, if you want to access table and column information in
Microsoft SQL Server without using schemas, you would have to query system
tables, like sysobjects or syscolumns. Example 6-1 illustrates how to obtain all of
the tables in a SQL Server database by querying sysobjects.
Example 6-1. List the Tables in a SQL Server Database Using the sysobjects Table
<% @LANGUAGE="VBSCRIPT" %>
<% Option Explicit %>
<%
'Open up a connection to the database
Dim objConn
Set objConn = Server.CreateObject("ADODB.Connection")
objConn.ConnectionString = "DSN=MyDatabase"
objConn.Open
'We want to query the sysobjects table where type = 'U'
Dim strSQL
strSQL = "SELECT * FROM sysobjects WHERE type='U'"
'Execute the query
Dim objRS
The Building Blocks for Creating Reusable Administration Pages 143
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
The sysobjects table is fairly cryptic, containing several columns with single character
values that denote their purpose. Rows that have a “U” in the Type column are
tables. I think you’ll agree that the code in Example 6-1 is fairly cryptic. Furthermore,
the code will only work for Microsoft SQL Server databases.
Schemas make listing database information much easier and more portable. Since
the code in Example 6-1 will only work for Microsoft SQL Server databases, it is
anything but portable, and portability should be a key concern when developing
reusable scripts.
Importing enumerations with adovbs.inc
Before you start using schemas, it is important that you have access to the enumerations
defined in ADO. There are currently two ways to do this:
• Use a server-side include to import the contents of adovbs.inc into each ASP
page that needs to reference these database enumerations.
• Use the METADATA tag in Global.asa to import the ADO constants.
adovbs.inc is a text file that contains all of the ADO enumerations in the form of
VBScript constants. To use adovbs.inc, copy it into the directory in which you
place your include files (I recommend creating a directory named /inc and turning
off Read permissions.) adovbs.inc can be found in Program Files\Common Files\
System\ado.
Once adovbs.inc is in a web-accessible directory, you can use a server-side
include to import the contents of adovbs.inc into any ASP page that needs to
access the ADO enumerations. Simply use the following line of code at the top of
your document:
<% @LANGUAGE="VBSCRIPT" %>
<% Option Explicit %>

Set objRS = objConn.Execute(strSQL)
'Display the table names
Do While not objRS.EOF
Response.Write objRS("Name") & "<BR>"
objRS.MoveNext
Loop
'Clean up...
objConn.Close
Set objConn = Nothing
%>
Example 6-1. List the Tables in a SQL Server Database Using the sysobjects Table (continued)
144 Chapter 6: Database Reuse
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Take a moment to look at the contents of adovbs.inc. Note that it is nothing more
than a lengthy list of VBScript constants assigned appropriate hexadecimal values.
Below is a short excerpt from the schema section of adovbs.inc:
'---- SchemaEnum Values ----
Const adSchemaProviderSpecific = -1
Const adSchemaAsserts = 0
Const adSchemaCatalogs = 1
Const adSchemaCharacterSets = 2
Const adSchemaCollations = 3
Const adSchemaColumns = 4
For more information on server-side includes, be sure to read
Chapter 1, Introduction, or check out “The low-down on #includes,”
available at http://www.4guysfromrolla.com/webtech/080199-1.shtml.
Importing enumerations with the METADATA tag
The METADATA tag imports enumerations from a type library. For example, in each
ASP page that uses a server-side include to import the contents of adovbs.inc, we
could replace:

with the following line of code:

The METADATA tag has the following syntax (the line breaks are for enhanced
readability and are not required):

You only need to specify either the UUID or the FILE when importing constants
from a type library. The UUID expects the type library’s GUID, while the FILE
expects the full, physical path to the type library. The GUID for the ActiveX Data
Objects type library is: 00000201-0000-0010-8000-00AA006D2EA4. I find that
specifying the FilePath instead of the GUID makes for much easier-to-read code.
However, there is always the chance that the file may be renamed or moved,
whereas the GUID won’t change.
The Building Blocks for Creating Reusable Administration Pages 145
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Now, you may be wondering why anyone would want to use the METADATA tag in
place of adovbs.inc. Typing:

instead of:

is quicker and less prone to mistake. METADATA’s advantage is that you can specify
it once in Global.asa and have access to all of the ADO enumerations in all of
your ASP pages! Also, according to Microsoft’s ASP Performance Tips (http://msdn.
microsoft.com/library/psdk/bdg/bdgapp03_3rhv.htm), using a single METADATA tag
in Global.asa provides for better performance than using server-side includes on
each page.
If you place the METADATA tag in Global.asa, remember that every
ASP page will have access to the ADO constants. If you attempt to
include adovbs.inc in a page, you will receive a Name redefined:
'adOpenForwardOnly' error. This error occurs because you are
attempting to create a constant in adovbs.inc that already exists from
the METADATA import. adOpenForwardOnly is listed because it is the
first constant defined in adovbs.inc.
Opening schemas
To open a schema, use the OpenSchema method of the Connection object. The
OpenSchema method has the following definition:
Set recordset = connection.OpenSchema(QueryType[, Criteria[, SchemaID]])
QueryType is a required parameter that specifies the type of schema to open, and
must be set to a valid SchemaEnum type. There are a number of possible schema
types that can return a vast array of database information. There are schemas for
listing tables, columns, column privileges, foreign keys, indexes, primary keys, referential
constraints, and other miscellaneous information. The valid SchemaEnums
are listed in adovbs.inc under the heading “SchemaEnum Values.” Each
SchemaEnum constant defined in adovbs.inc is prefixed with an adSchema; some
of the more commonly used SchemaEnum values (including those that will be used
in code presented in this chapter) are listed in Table 6-1.
146 Chapter 6: Database Reuse
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
The available SchemaEnums that can be specified by the QueryType parameter differ
among the various database providers. Many database providers support a
common set of SchemaEnums (such as adSchemaColumns and adSchemaTables),
but some database providers may contain their own unique schemas that aren’t
represented by a value in the SchemaEnum list. If you need to use such a schema,
set the QueryType to adSchemaProviderSpecific and specify the unique
schema in the SchemaID parameter.
The various schemas supported depend upon the database provider.
For example, the OLE DB provider for AS/400 and VSAM only
supports four schemas, while the OLE DB provider for SQL Server
supports several more.
Example 6-2 contains code that is functionally identical to the code presented in
Example 6-1; both scripts list the tables in a database. Example 6-2, however, uses
the adSchemaTables QueryType rather than sysobjects, which makes
Example 6-2 easier to read and more portable.
Table 6-1. Several Common SchemaEnums Are Defined in adovbs.inc
SchemaEnum Description
adSchemaColumns Contains detailed information about each of the columns in a
database
adSchemaTables Contains detailed information about each table in a database
adSchemaForeignKeys Contains detailed information about all of the foreign key
constraints in a database
Example 6-2. Listing the Tables in a Database with Schemas
<% @LANGUAGE="VBSCRIPT" %>
<% Option Explicit %>

<%
'Open a connection to the database
Dim objConn
Set objConn = Server.CreateObject("ADODB.Connection")
objConn.ConnectionString = "DSN=MyDatabase"
objConn.Open
'Use the OpenSchema method to grab the table schema
Dim objRS
Set objRS = objConn.OpenSchema(adSchemaTables)
'Loop through the contents of the schema.
Do While Not objRS.EOF
'Only display TABLES – not SYSTEM TABLES, not VIEWS
If objRS("TABLE_TYPE") = "TABLE" then
Response.Write objRS("TABLE_NAME") & "<BR>"
The Building Blocks for Creating Reusable Administration Pages 147
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Since a schema is returned as a Recordset object, we can use the Recordset
object’s properties and methods to iterate through the schema. A schema, like an
ordinary database table, contains columns, which describe the properties of an
abstract object, and rows, which serve as an instantiation of an object. The
adSchemaTables schema we used in Example 6-2 contains several columns
describing each table, including TABLE_NAME, TABLE_TYPE, DESCRIPTION, DATE_
MODIFIED, and DATE_CREATED.
The adSchemaTables schema returns the views and system tables along with the
user tables. The TABLE_TYPE column specifies what type of table is being
returned, and in Example 6-2 we list only user tables by assuring that TABLE_TYPE
equals “TABLE” before displaying the TABLE_NAME column.
Note that we explicitly included adovbs.inc in Example 6-2. This
explicit include would not be needed if the METADATA tag were used
in Global.asa. It is strongly recommended that you use the
METADATA approach. All the examples in this book that need to use
ADO enumerations, however, will explicitly include adovbs.inc to
assist with readability.
The Criteria parameter
The OpenSchema method has an optional parameter named Criteria. This
parameter can be specified to limit the resulting schema. For example, in
Example 6-2, we didn’t specify a Criteria, and all tables were returned, including
views and system tables. We could pass in a Criteria parameter, however,
that would inform ADO to return only those tables that were user tables. Using a
Criteria to limit the contents of a schema is similar to using a WHERE clause to
limit the results of a SQL query.
Each SchemaEnum has its own predefined set of potential Criteria that can be
used to limit the results of a schema. Some SchemaEnums have no limiting
End If
objRS.MoveNext
Loop
'Clean up...
objRS.Close
Set objRS = Nothing
objConn.Close
Set objConn = Nothing
%>
Example 6-2. Listing the Tables in a Database with Schemas (continued)
148 Chapter 6: Database Reuse
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Criteria, while others can have several columns to which Criteria can be
applied. To accommodate the differing number of Criteria accepted by various
SchemaEnums, the Criteria parameter should be passed in as an array. For
example, if a particular SchemaEnum expects four Criteria, the array should contain
four elements.
The adSchemaTables schema expects four Criteria: TABLE_CATALOG, TABLE_
SCHEMA, TABLE_NAME, and TABLE_TYPE. To retrieve only user tables when opening
the adSchemaTables schema, we need to supply a Criteria parameter that
was a four-element array, defined as:
Array(Empty, Empty, Empty, "TABLE")
Since we do not wish to filter on the first three Criteria—TABLE_CATALOG,
TABLE_SCHEMA, and TABLE_NAME—we simply leave these blank by specifying the
resulting array element as Empty. By using the above Criteria, we can alter
Example 6-2 so the check for objRS("TABLE_TYPE") = "TABLE" can be avoided.
Example 6-3 displays this new code using the Criteria parameter of the Open-
Schema method.
Example 6-3. Using the Criteria Parameter to Selectively List Tables of a Database
<% @LANGUAGE="VBSCRIPT" %>
<% Option Explicit %>

<%
'Open a connection to the database
Dim objConn
Set objConn = Server.CreateObject("ADODB.Connection")
objConn.ConnectionString = "DSN=MyDatabase"
objConn.Open
'Use the OpenSchema method to grab the table schame
'In this example, we specify the Criteria parameter of the
'OpenSchema method
Dim objRS
Set objRS = objConn.OpenSchema(adSchemaTables, _
Array(Empty, Empty, Empty, "TABLE"))
'Loop through the contents of the schema.
Do While Not objRS.EOF
'Since the Criteria parameter returned only user table,
'we don't need to check TABLE_TYPE here, like we did in Example 6-2
Response.Write objRS("TABLE_NAME") & "<BR>"
objRS.MoveNext
Loop
'Clean up...
objRS.Close
Set objRS = Nothing
objConn.Close
Creating Reusable Administration Pages 149
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Keep in mind that Example 6-1, Example 6-2, and Example 6-3 all are functionally
equivalent.
There are several possible SchemaEnum values, each with their own
columns and possible criteria values—far too many to list in a nonreference
book like this one. You can obtain a full SchemaEnum listing
at http://msdn.microsoft.com/library/psdk/dasdk/mdae0wfh.htm
or through a technical reference book, like Wrox’s ADO 2.0 Programmer’s
Reference (Wrox Press Inc.).
Schemas will greatly assist in the quest to create a generic database access script.
Since schemas provide portable, low-level database information through an easyto-
use interface, they make an ideal solution for determining how to display particular
database fields and database input mechanisms on a generic database page.
For more information on ADO schemas, be sure to read the following tutorials and
articles:
• “Listing the Tables and Columns in a Database,” found at http://www.
4guysfromrolla.com/webtech/101799-1.shtml.
• Technical documentation for OpenSchema, found at http://msdn.microsoft.
com/library/psdk/dasdk/mdam2ppd.htm.
• For general ADO information, check out Microsoft’s ADO site at http://www.
microsoft.com/data/ado/.
Creating Reusable Administration Pages
In our quest to build a set of truly reusable administration page generation scripts,
there are three general problems we’ll encounter and have to solve:
• How does one gather table and column information, and what specific table
and column information is worth noting?
• How does one create easy-to-use forms for inserting, editing, and deleting
records in a database table? How can the information collected about the database
tables and columns be used to create easier-to-use administration page
forms?
• Once the user submits an administration page form, how will the changes find
their way into the database?
Set objConn = Nothing
%>
Example 6-3. Using the Criteria Parameter to Selectively List Tables of a Database (continued)
150 Chapter 6: Database Reuse
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
These questions are fairly easy to answer when one is given the liberty to create a
set of administration pages for each and every database table, but when we limit
ourselves to only a small set of generic administration pages, regardless of the
number of database tables, these questions quickly become much more difficult to
answer.
The remainder of this chapter is dedicated to answering these questions. The following
sections each focus on answering a particular question:
• “Gathering Column Information” answers the question, “How does one gather
column information?”
• “Gathering Foreign Key Information” answers the question, “How does one
gather table information?”
• “Deciding How to Display the Table Columns in a Form” answers the question,
“How can the information collected about the database tables and columns
be used to create easier-to-use administration page forms?”
• “Creating the Administration Page Forms” answers the question, “How does
one create easy-to-use forms for inserting, editing, and deleting records in a
database table?”
• “Inserting, Updating, and Deleting Database Records” answers the question,
“Once the user submits an administration page form, how will the changes
find their way into the database?”
The journey to answer these questions is a long one, including multitudes of
classes and over a thousand lines of source code. Each section focuses on developing
a class (or set of classes) to accomplish a certain task. At the conclusion of
each of these sections, we’ll pause for a moment to test the class or classes we just
created, and to look at how they fit into the big picture.
As classes essential to the generic administration page application are presented
throughout the chapter, they will be tested against an example database. This database
contains information on products and catalogs. There can be many products
in a given catalog, and one product can appear in many catalogs. Each product
has a specific product type, which is indicated by the ProductType look-up table.
There are two relationships in this data model: a many-to-many relationship
between Catalog and Product, and a one-to-many relationship between
ProductType and Product.
The structure for this example database can be seen in Table 6-2. Some example
data for the Catalog, ProductType, Product, and CatalogProduct tables can
be seen in Table 6-3 through Table 6-6.
Creating Reusable Administration Pages 151
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Table 6-2. The ProductInfo Database Table Structures
Table Name Column Name Description
Catalog CatalogID Primary key; int.
Name The name of the catalog; varchar(50).
ReleasedDate The date the catalog was released to the
public; date/time.
ContainsPictures Does the catalog contain pictures? Bit.
Product ProductID Primary key; int.
ProductTypeID Foreign key; int.
Price The retail price; money.
DateEntered The date the product was entered into
inventory; date/time.
Quantity The current, in-stock quantity; int.
ProductType ProductTypeID Primary key; int.
Name The name of the product type; varchar(50).
Description A description of the product type; varchar(
255).
CatalogProduct CatalogID Primary key, foreign key; int.
ProductID Primary key, foreign key; int. The
CatalogID and ProductID form a composite
primary key.
Table 6-3. Catalog Table Example Data
CatalogID Name Released Date Contains Pictures
2 Spring 2000 4/1/2000 1
3 Fall 2000 9/1/2000 0
4 2000 Year in Review 12/15/2000 0
Table 6-4. ProductType Table Example Data
ProductTypeID Name Description
1 Retail For sale
2 Defective These are defective items
Table 6-5. Product Table Example Data
ProductID ProductTypeID Price Date Entered Quantity
1 1 45 4/1/2000 555
2 1 155 4/2/2000 15
3 2 5 4/2/2000 10
152 Chapter 6: Database Reuse
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Before we start presenting the classes needed for the generic administration page
generation application, however, it would be prudent to take a moment and discuss
some important database terms.
A Review of Database Terms
There are three terms used extensively throughout this chapter. It is imperative
that I, the author, and you, the reader, speak the same lingo. So let’s take a
moment to make sure we are both on the same page when it comes to the following
three database terms.
Primary key constraints
A primary key constraint can be placed on one or more columns in a table, and
ensures that the column or columns selected contain unique values. Primary keys
play an important role in relational databases, providing entity integrity. Primary
keys are synonymous UNIQUE constraints that do not allow NULL values. Be sure
to place primary key constraints on the column or columns that uniquely identity a
particular row.
AutoNumber columns
An AutoNumber column is a numeric column that is automatically incremented
each time a new record is added to the table. AutoNumber columns are created
differently for each database system; in Access, an AutoNumber column is created
by specifying the column’s datatype as AutoNumber; in Microsoft SQL Server, the
IDENTITY property is used.
An AutoNumber is not a datatype, despite the fact that in Access you set the column’s
datatype to AutoNumber. Rather, AutoNumber is a column property on a
column that contains a numeric datatype. The AutoNumber property should only
be applied to a column that is a primary key.
AutoNumbers are the easiest way to uniquely identity each row. When a new
record is inserted into a table with an AutoNumber column, the developer doesn’t
need to specify the new AutoNumber value in the INSERT statement; the database
system does this automatically, picking the next sequential AutoNumber.
Table 6-6. CatalogProduct Table Example Data
CatalogID ProductID
3 2
3 3
Creating Reusable Administration Pages 153
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
A primary key column does not need to have the AutoNumber property
assigned to it. For example, in a catalog of widgets, the primary
key column might be the widget’s serial number, which is unique
among all widgets. However, the AutoNumber property should only
be applied to primary key columns.
Foreign key constraints
In a relational database, tables share relationships with one another. When creating
a relationship between two database tables, a foreign key constraint is used to
ensure referential integrity.
For example, assume that you have two database tables, Employee and
EmployeeTitle. The EmployeeTitle table represents the different titles available
at your company (President, Manager, Engineer, etc.). The Employee table
contains employee information, with a row for every employee at your company.
The Employee table could have many columns storing bits of information about
each employee, such as social security number, birth date, name, address, and
title. A relationship could be set up between the Employee and EmployeeTitle
tables. The TitleID column in the Employee table would be an integer value,
mapping to a specific title in the EmployeeTitle table. Figure 6-2 shows some
values that might be found in the Employee and EmployeeTitle tables.
Figure 6-2. The Employee and EmployeeTitle tables are related
FirstName
Employee Table
LastName SocialSecurityNumber TitleID
Scott Mitchell 123-45-6789 10
Kevin Spacey 098-76-5432 2
Joshua Joy 321-54-9876 7
James Ransom 090-09-0909 10
Al Hornets 456-12-3456 10
Skippy Tourna 555-55-5555 2
TitleID
EmployeeTitle Table
Title
1 Engineering Systems Engineer
2 Flight Simulation Director
5 Testing Engineer
7 Secretary
9 President
10 Senior Manager
The TitleID in the Employee table maps to a Title in the
EmployeeTitle table. Therefore, to determine an employee's
title, simply perform a lookup in the EmployeeTitle table.
154 Chapter 6: Database Reuse
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Note that each value in the TitleID in the Employee table maps to a value in the
TitleID column in the EmployeeTitle table. Through this relationship, one can
quickly determine the title of an employee. The data shown in Figure 6-2 exhibits
referential integrity; that is, there are no logical discrepancies between the two
related columns.
Imagine for a moment that the TitleID column for Scott Mitchell was changed to
a value of 800, which is clearly not a valid value in the EmployeeTitle table. Or,
imagine that a row in the EmployeeTitle table that maps to a row in the
Employee table (such as the Secretary table, which maps to employee Joshua Joy)
was deleted. Either of these events would violate referential integrity since, in
either case, an Employee would report a nonexistent employee title.
Foreign key constraints assist by ensuring referential integrity. If a foreign key constraint
was created between the TitleID columns of the Employee and
EmployeeTitle tables, the hypothetical actions discussed in the preceding paragraph
could not happen. If an attempt was made to delete an employee title that
mapped to a row in the Employee table, a referential integrity error would be displayed
and the delete command would not succeed. Similarly, if there was an
attempt to alter an employee’s TitleID to a value not present in the
EmployeeTitle table, an error would occur and the update would not be
committed.
All relational databases provide the capability to create foreign key
constraints. To learn how to establish foreign key constraints for
your development database, consult the database’s documentation.
Since foreign key columns are usually stored as numeric data, they are a bit tricky
to handle in a generic administration page. When a user creates or updates an
existing record that contains foreign keys, asking the user for a numeric ID is preposterous.
Instead, the user should be presented with a list box that displays a
readable list of options.
Part of the challenge in creating a generic administration page generation script is
determining what foreign keys exist in what tables. With the aid of schemas, however,
this process is simplified. In the later section “Gathering Foreign Key Information,”
we’ll examine how to determine the foreign key constraints in a particular
database table. Another challenge is being able to map a foreign key to a set of
readable options in the related table. To overcome this challenge, we’ll allow the
end developer to give some “hints” to the administration page generation class on
how related tables should have their contents listed.
Creating Reusable Administration Pages 155
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Gathering Column Information
For all aspects of the administration page interface, being able to determine the
columns and column types of a given record is important. As discussed in “Database
Schemas,” we will use ADO schemas to ascertain column information for a
given table. Example 6-2 and Example 6-3 demonstrated how to collect column
information for a particular table. While this procedure is not especially difficult, it
would be nice to have a class that hid the steps needed to acquire such information.
An end developer using the code in Example 6-2 or Example 6-3 needs to be
aware of what schemas are and how they are used. We can remove this constraint
by packaging the functionality of Example 6-3 into an easy-to-use class.
Before designing the class to gather column information from a given table, it is
important to know what information will need to be stored for a given database
column. Since the administration pages must know a great deal of information
about each column—including the name, the datatype, the default value, the column
description, and whether or not the field is an AutoNumber field—it is essential
that extensive information be recorded for each column in a database table.
In this section, two classes are presented: ColumnInformation and Columns.
ColumnInformation and Columns are used in the generic administration page
generation application to represent a generic table column and a collection of
table columns, respectively. Both the ColumnInformation and the Columns
classes should be placed in the same file, /CODEREUSE/Column.Class.asp.
The ColumnInformation class
To assist with the collection of column information for a given table, a
ColumnInformation class is presented. Each instance of this class represents a
single column in a database. This class will not be used directly by the end developer,
but by the Columns class. The Columns class, which we’ll discuss shortly in
the section “The Columns class,” is used by other classes to inspect a database
table’s various columns. The Columns class contains a collection of
ColumnInformation class instances, one for each column of a particular database
table.
Since the ColumnInformation class is used solely for holding the needed information
for a particular table column, the class contains only a number of read/write
properties. Example 6-4 contains the class definition for ColumnInformation.
Example 6-4. The ColumnInformation Class Stores Information for a Particular Database
Table Column
<%
Class ColumnInformation
'************** MEMBER VARIABLES *********************
Private strName
156 Chapter 6: Database Reuse
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Private iDataType
Private bolDefault
Private strDefault
Private bolNullable
Private strDescription
Private varValue
Private bolIdentity
Private bolPrimaryKey
'*****************************************************
'****************** GET PROPERTIES *******************
Public Property Get Identity()
Identity = bolIdentity
End Property
Public Property Get PrimaryKey()
PrimaryKey = bolPrimaryKey
End Property
Public Property Get Name()
Name = strName
End Property
Public Property Get DataType()
DataType = iDataType
End Property
Public Property Get HasDefault()
HasDefault = bolDefault
End Property
Public Property Get DefaultValue()
DefaultValue = strDefault
End Property
Public Property Get Nullable()
Nullable = bolNullable
End Property
Public Property Get Description()
Description = strDescription
End Property
Public Property Get Value()
Value = varValue
End Property
'*****************************************************
'****************** LET PROPERTIES *******************
Public Property Let Identity(bolIdent)
bolIdentity = bolIdent
End Property
Example 6-4. The ColumnInformation Class Stores Information for a Particular Database
Table Column (continued)
Creating Reusable Administration Pages 157
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
The class presented in Example 6-4 is very simple, containing nothing but properties.
The purpose of this class is to serve as a black box for describing what,
exactly, a column is.
The Columns class
Let’s examine a class that collects all of the column information from a given table.
Rather than having this class, Columns, be responsible for maintaining each discrete
bit of information that pertains to a column, it simply contains a variable
number of ColumnInformation instances. The Columns class takes advantage of
the black box interface to a table column the ColumnInformation class provides.
Since the Columns class is quite hefty, we’ll examine each part of the class separately,
as opposed to listing pages of raw source code.
Public Property Let PrimaryKey(bolPK)
bolPrimaryKey = bolPK
End Property
Public Property Let Name(str)
strName = str
End Property
Public Property Let DataType(iValue)
iDataType = iValue
End Property
Public Property Let HasDefault(bolHasDefault)
bolDefault = bolHasDefault
End Property
Public Property Let DefaultValue(str)
strDefault = str
End Property
Public Property Let Nullable(bolIsNullable)
bolNullable = bolIsNullable
End Property
Public Property Let Description(str)
strDescription = str
End Property
Public Property Let Value(var)
varValue = var
End Property
'*****************************************************
End Class
%>
Example 6-4. The ColumnInformation Class Stores Information for a Particular Database
Table Column (continued)
158 Chapter 6: Database Reuse
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Columns’s class properties and member variables
The Columns class contains a single member variable, objColumnDict, which is a
Dictionary object, storing zero to many ColumnInformation instances. This collection
is available to the end developer through a Property Get statement; furthermore,
the number of elements in the objColumnDict Dictionary object is also
available to the end developer.
Of course, the end developer won’t always want to obtain the entire
objColumnDict Dictionary object. There will be times when a developer is interested
in retrieving just a specific ColumnInformation instance from the
objColumnDict collection. To accommodate this, a method (GetColumnInformation)
is provided to return a specific element from the objColumnDict Dictionary
object.
Example 6-5 shows the Columns class definition’s event handlers, member variables,
and properties. The Initialize and Terminate event handlers are responsible
for managing the creation and destruction of objColumnDict. Note the two
Property Get statements—Count and ColumnList—that, allow for the retrieval of
the number of elements in objColumnDict and the retrieval of objColumnDict
itself, respectively.
Example 6-5. The Properties and Member Variable of the Columns Class
<%
Class Columns
'************** MEMBER VARIABLES *********************
Private objColumnDict 'store zero to many columns
'*****************************************************
'******************* EVENT HANDLERS ******************
Private Sub Class_Initialize()
'Create an instance of the Dictionary object
Set objColumnDict = Server.CreateObject("Scripting.Dictionary")
End Sub
Private Sub Class_Terminate()
Set objColumnDict = Nothing 'Clean up...
End Sub
'*****************************************************
'********************** METHODS **********************
Public Function GetColumnInformation(strColumnName)
'Returns a specific column's information... that is, returns a
'reference to an instance of a ColumnInformation object
Set GetColumnInformation = objColumnDict(strColumnName)
End Function
' ... Numerous methods removed here, but are presented in
' Example 6-6 and Example 6-7 ...
'*****************************************************
Creating Reusable Administration Pages 159
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Columns’s class methods
Columns contains a number of methods. The first method we’ll look at, Populate-
Columns, is responsible for reading column information from a particular database
table and populating the objColumnDict Dictionary object with the results.
The PopulateColumns method has the following definition:
Public Sub PopulateColumns(strTableName, objConn, strWhereClause)
Since Columns serves as a collection of database table columns, the specific database
table to use must be specified by the end developer through the
strTableName parameter. The objConn parameter should contain an opened
connection to the database that contains the table specified by strTableName.
PopulateColumns will use the OpenSchema method of the objConn Connection
object passed in by the end developer. The strWhereClause is used only when a
record is being updated and the column values for a particular row must be
known. When inserting a new row, the strWhereClause parameter should be an
empty string ("").
The end developer won’t have to worry about passing the proper
strWhereClause to PopulateColumns. Another class, Admin-
PageGenerator, will be used to construct the various forms for
updating, deleting, and inserting records. This class will also correctly
call PopulateColumns. In fact, the end developer does not
even need to know that the ColumnInformation and Columns
classes exist! The AdminPageGenerator class is discussed in greater
detail in the section “The AdminPageGenerator class.”
The PopulateColumns method, which appears in Example 6-6, uses the
adSchemaColumns schema to populate the objColumnDict Dictionary object. For
each column in the database table specified by strTableName, there is a
ColumnInformation instance created and added to objColumnDict. As the column
information is collected, a SQL query is built up that will later be used to
'****************** GET PROPERTIES *******************
Public Property Get Count()
Count = objColumnDict.Count
End Property
Public Property Get ColumnList()
Set ColumnList = objColumnDict
End Property
'*****************************************************
End Class
%>
Example 6-5. The Properties and Member Variable of the Columns Class (continued)
160 Chapter 6: Database Reuse
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
determine what columns are AutoNumber columns, and if the strWhereClause
was specified, the values for a particular row.
Example 6-6. The PopulateColumns Method Obtains Detailed Information About Each
Column in a Particular Database Table
<%
Class Columns
'... Member variables and Class Event Handlers not shown here;
' These were both shown in Example 6-5 ...
'********************** METHODS **********************
Public Sub PopulateColumns(strTableName, objConn, strWhereClause)
'This sub is responsible for populating the objColumnDict
'dictionary object with the columns from the table strTableName.
'objConn is expected to be an opened connection to the
'proper database
'Make sure a proper table name was passed in
If Len(strTableName) = 0 then
Err.Raise vbObjectError + 1280, "Columns Class", _
"In PopulateColumns method: TableName not supplied."
End If
Dim objRS 'Open up the column schema
Set objRS = objConn.OpenSchema(adSchemaColumns, _
Array(Empty, Empty, strTableName, Empty))
'Add a ColumnInformation object instance for each column that
'exists within the table specified by strTableName. Also,
'build up a SQL statement
Dim objColumn, strSQL
strSQL = "SELECT "
Do While Not objRS.EOF
Set objColumn = New ColumnInformation
objColumn.Name = objRS("COLUMN_NAME")
objColumn.DataType = objRS("DATA_TYPE")
objColumn.HasDefault = objRS("COLUMN_HASDEFAULT")
objColumn.DefaultValue = objRS("COLUMN_DEFAULT")
objColumn.Nullable = objRS("IS_NULLABLE")
objColumn.Description = objRS("DESCRIPTION")
objColumn.Identity = False
objColumn.PrimaryKey = False
objColumnDict.Add objColumn.Name, objColumn
strSQL = strSQL & "[" & objColumn.Name & "]"
Set objColumn = Nothing
objRS.MoveNext
If Not objRS.EOF then strSQL = RTrim(strSQL) & ","
Loop
'Now we need to grab the PrimaryKey information
Creating Reusable Administration Pages 161
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
The tasks PopulateColumns needs to complete are pretty straightforward, but
some convoluted code is needed to accomplish them. To populate the
objColumnDict object, the adSchemaColumns schema is used with the Criteria
set to only return columns for the database table specified by strTableName. As
the schema contents are iterated, a ColumnInformation object instance is creobjRS.
Close
Set objRS = Nothing
Set objRS = objConn.OpenSchema(adSchemaPrimaryKeys, _
Array(Empty, Empty, strTableName))
Do While Not objRS.EOF
objColumnDict(objRS("COLUMN_NAME").Value).PrimaryKey = True
objRS.MoveNext
Loop
'Now we need to obtain the identity columns/column values
strSQL = strSQL & " FROM [" & strTableName & "]"
If Len(strWhereClause) > 0 then
strSQL = strSQL & " WHERE " & strWhereClause
End If
'We need to grab a row from the database to determine if we have any
'identity columns, and to populate our row values, if needed
objRS.Close
Set objRS = Nothing
Set objRS = Server.CreateObject("ADODB.Recordset")
objRS.MaxRecords = 1 'Get only one record
objRS.Open strSQL, objConn
'If there are no records in this table, we can't perform this loop
If Not objRS.EOF then
Dim objField
For Each objField in objRS.Fields
objColumnDict(objField.Name).Value = objField.Value
If objField.Properties("ISAUTOINCREMENT") then
objColumnDict(objField.Name).Identity = True
End If
Next
End If
'Clean up...
objRS.Close
Set objRS = Nothing
End Sub
'... Some methods and properties not shown; these methods are shown
' in Example 6-7; the properties were presented in Example 6-5 ...
End Class
%>
Example 6-6. The PopulateColumns Method Obtains Detailed Information About Each
Column in a Particular Database Table (continued)
162 Chapter 6: Database Reuse
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
ated at each iteration, and its properties are set according to the values for the current
column in the schema. Note that the column names are also added to a SQL
query as this loop progresses.
Once this loop has finished, PopulateColumns knows a great deal about each column
in the database table specified by strTableName. However, it is still
unknown what columns (if any) are AutoNumber columns and what columns (if
any) are primary key columns.
To determine what columns are primary keys, we use the OpenSchema method to
open the adSchemaPrimaryKeys schema, restricting the primary keys returned to
the primary keys in the table strTableName. Next, the schema is iterated, and the
columns in objColumnDict that match the COLUMN_NAME returned by the
adSchemaPrimaryKeys have their PrimaryKey property set to True.
Determining what columns are AutoNumber columns is a tad more difficult. The
AutoNumber property is not available in any schema. Part of the reason is
AutoNumber is not a column datatype or column constraint; rather, it is a column
property, making it a bit of an anomaly. Since AutoNumber columns are not a
required standard in relational databases, ADO does not provide AutoNumber
information in schemas, which is really annoying, in my opinion.
To determine whether a particular column is an AutoNumber or not, we must consult
the dynamic Properties collection of the ADO Recordset. The Properties collection
is available through the Field object, and can contain a varying number of
properties. What we are looking for is a property named IS_AUTONUMBER, which,
if the property exists, contains a True or False value.
At this point, PopulateColumns might also need to determine the values of the columns
for a particular row (this would be the case if the user is updating an existing
record, in which case the update form needs to know the current values for
the column to be edited). Therefore, we’ll kill two birds with one stone, gathering
both the values for a particular record, if needed, and determining what columns
are AutoNumbers.
To accomplish this, we need to retrieve the appropriate query from the table—this
is where the dynamic SQL query that was built comes in. This SELECT statement
query is used to return the particular row of interest, as specified by
strWhereClause. If strWhereClause is not specified, the entire table might be
returned, which is fine, but a bit of a performance drag if the table contains a large
number of rows. Therefore, the MaxRecords property is set to 1, which, for database
providers that support the MaxRecords property, will bring back only one row.
The Columns class contains four other methods that return information about primary
key columns; these are shown in Example 6-7. When the “Edit Existing
Records” administration page is run, information uniquely identifying the record
Creating Reusable Administration Pages 163
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
being edited from all other records must be known. The GetPrimaryKeys method
returns a comma-delimited list of primary key column names and values. This list
of names and values will come in handy when updating a record in the database.
The QuotedValue private member function surrounds a primary key value in single
quotes if the column datatype is a textual type or a date/time type. This is a
requirement for SQL statements; that is, the value of a textual or date/time column
must be surrounded by quotes in the WHERE clause. The Replace method is used
on the column’s value, replacing all instances of apostrophes with two concurrent
apostrophes. This measure prevents SQL from choking on single apostrophes in
textual data. To learn more about this, be sure to read “How to Deal with Apostrophes
in your SQL String,” available at: http://www.4GuysFromRolla.com/webtech/
051899-1.shtml.
To delete a record or select a particular record to edit, the contents of a database
table must be listed, and a checkbox or radio button must be present to allow the
user to either select a number of records (in the case of deleting) or select a particular
record (in the case of choosing which record to edit). In both of these cases,
the primary key(s) and their respective value(s) must be returned for each row
listed, since the primary key(s) are what uniquely identify each record in the table.
To delete one or more records, it is essential that the right primary key values be
known!
The GetCurrentPrimaryKeys method is nearly identical to the GetPrimaryKeys
method; it differs in two ways: it expects an opened Recordset object to be passed
in, and it calls the CurrentQuotedValue helper function instead of QuotedValue.
The CurrentQuotedValue helper function operates just like QuotedValue, except,
like GetCurrentPrimaryKeys, it expects an opened Recordset object to be passed
in. Also, CurrentQuotedValue returns the value of the current row of the recordset,
as opposed to the Value property of the ColumnInformation instance.
When we examine our generic administration page generation script later in the
chapter, we’ll see how the four methods listed in Example 6-7 aid in the administration
page generation process.
Example 6-7. The Remaining Columns Class Methods
<%
Class Columns
'... Member variables and Class Event Handlers not shown here;
' These were both shown in Example 6-5 ...
'********************** METHODS **********************
Public Function GetPrimaryKeys()
Dim strColumnName, strResult
For Each strColumnName in objColumnDict
If objColumnDict(strColumnName).PrimaryKey then
If Len(strResult) > 0 then strResult = strResult & " AND "
164 Chapter 6: Database Reuse
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
strResult = strResult & "[" & _
objColumnDict(strColumnName).Name & _
"] = " & QuotedValue(strColumnName)
'Make sure we got a valid QuotedValue
If Len(QuotedValue(strColumnName)) = 0 then
Err.Raise vbObjectError + 1281, "Columns Class", _
"In method GetPrimaryKeys: Unable to Retrieve Value"
End If
End If
Next
GetPrimaryKeys = strResult
End Function
Private Function QuotedValue(strColumnName)
'Returns the value of a column, surrounding it with
'single tick marks if it is a text or date field
Dim iDataType, strResult
iDataType = objColumnDict(strColumnName).DataType
If iDataType = adDate OR iDataType = adDBTimeStamp _
OR iDataType = adChar OR iDataType = adVarChar _
OR iDataType = adLongVarChar OR iDataType = adWChar _
OR iDataType = adVarWChar OR _
iDataType = adLongVarWChar then
strResult = "'" & _
Replace(objColumnDict(strColumnName).Value, "'", "''") & "'"
Else
strResult = objColumnDict(strColumnName).Value
End If
QuotedValue = strResult
End Function
Public Function GetCurrentPrimaryKeys(objRS)
Dim strColumnName, strResult
For Each strColumnName in objColumnDict
If objColumnDict(strColumnName).PrimaryKey then
If Len(strResult) > 0 then strResult = strResult & " AND "
strResult = strResult & "[" & _
objColumnDict(strColumnName).Name & _
"] = " & CurrentQuotedValue(strColumnName, objRS)
'Make sure we got a valid QuotedValue
If Len(CurrentQuotedValue(strColumnName, objRS)) = 0 then
Err.Raise vbObjectError + 1281, "Columns Class", _
"In method GetCurrentPrimaryKeys: Unable to Retrieve Value"
End If
End If
Next
Example 6-7. The Remaining Columns Class Methods (continued)
Creating Reusable Administration Pages 165
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Testing the Columns and ColumnInformation classes
Now that we have created two classes to assist with storing database column information,
let’s give them a whirl!
A “testing” section will immediately follow each major section that
introduces a new class or set of classes. This testing section will
show the class in use.
Since the ColumnInformation class contains only a set of read/write properties, a
thorough example is not possible. However, Example 6-8 illustrates a simple test
driver for the class.
GetCurrentPrimaryKeys = strResult
End Function
Private Function CurrentQuotedValue(strColumnName, objRS)
'Returns the value of a column, surrounding it with
'single tick marks if it is a text or date field
Dim iDataType, strResult
iDataType = objColumnDict(strColumnName).DataType
If iDataType = adDate OR iDataType = adDBTimeStamp _
OR iDataType = adChar OR iDataType = adVarChar _
OR iDataType = adLongVarChar OR iDataType = adWChar _
OR iDataType = adVarWChar OR _
iDataType = adLongVarWChar then
strResult = "'" & _
Replace(objRS(strColumnName).Value, "'", "''") & "'"
Else
strResult = objRS(strColumnName).Value
End If
CurrentQuotedValue = strResult
End Function
'*****************************************************
'... The PopulateColumns method and some properties are not shown;
' the PopulateColumns method was presented in Example 6-6, and
' the Properties were presented in Example 6-5 ...
End Class
%>
Example 6-8. ColumnInformation Stores Information About a Table Column
<% @LANGUAGE="VBSCRIPT" %>
<% Option Explicit %>

Example 6-7. The Remaining Columns Class Methods (continued)
166 Chapter 6: Database Reuse
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
To use the class, we start by importing the ColumnInformation class definition
via a server-side include. Next, an instance of the class is created and some of its
properties are assigned values. After the properties have been assigned values, feel
free to do whatever testing you want. With this class you are limited to reassigning
property values or reading the property values. For a simple test, Example 6-8
simply outputs the value of the Name property.
Note that the DataType property is assigned the value of an ADO
constant. adVarChar (which has a value of 200) is a constant
defined in adovbs.inc. The previous script does not explicitly include
adovbs.inc since it assumes you have implemented the METADATA
alternative discussed in the section “Importing enumerations with the
METADATA tag.” If this is not the case, however, you need to
include adovbs.inc in each ASP page before you include any of the
generic administration page generation classes.
The Columns class, which uses the ColumnInformation class internally, maintains
column information for a particular database table. As discussed earlier, the
PopulateColumns method, which uses the adSchemaColumns schema to acquire
column information about a specified database table, takes three parameters, having
the following definition:
Public Sub PopulateColumns(strTableName, objConn, strWhereClause)
For our test examples, we’ll examine the Products table of our example database
(see Table 6-2). Before calling the PopulateColumns method, we’ll need to
have successfully opened a connection to the database. The opened Connection
object is the second parameter.
<%
'Create an instance of the ColumnInformation class
Dim objColumnInfo
Set objColumnInfo = New ColumnInformation
'Set the properties of the class
objColumnInfo.Name = "FirstName"
objColumnInfo.DataType = adVarChar
objColumnInfo.Nullable = False
objColumnInfo.Identity = False
objColumnInfo.PrimaryKey = False
'Use the class somehow...
Response.Write "COLUMN NAME: " & objColumnInfo.Name
Set objColumnInfo = Nothing 'Clean up!
%>
Example 6-8. ColumnInformation Stores Information About a Table Column (continued)
Creating Reusable Administration Pages 167
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
For our first test, we will send a blank string for the strWhereClause parameter.
This tells PopulateColumns that we are just interested in knowing the column
structure of the table. In the “Insert a New Record” administration page, where we
are just interested in knowing the topology of the table, we will not supply a
strWhereClause parameter.
If we were interested in knowing the values for a specific row in the table, we’d
pass in a legal SQL WHERE clause statement as the strWhereClause parameter
that would return exactly one row from the table. In the “Edit an Existing Record”
administration page, where we’d like to know the existing values for the record
being edited (so the user can see what the current values are), we’ll supply a
strWhereClause parameter.
Example 6-9 contains a test driver for the Columns class that obtains the column
information for the Product table. Note that in this example the PopulateColumns
method is called without specifying a strWhereClause parameter.
Example 6-9. The Generic Topology of the Product Table Is Obtained
<% @LANGUAGE="VBSCRIPT" %>
<% Option Explicit %>

<%
'Open a database connection
Dim objConn
Set objConn = Server.createObject("ADODB.Connection")
objConn.ConnectionString = "Provider=SQLOLEDB;Data Source=mitchell;" & _
"Initial Catalog=ProductInfo;User ID=sa;Password="
objConn.Open
'Create an instance of the Columns class
Dim objColumns
Set objColumns = New Columns
'Populate the columns collection with the Product table information
objColumns.PopulateColumns "Product", objConn, ""
'At this point, we have access to a list of the column information in the
'Product table.
Dim strColumnName, objColumnInfo
'We wish to loop through all of the columns in objColumns ColumnList
For Each strColumnName in objColumns.ColumnList
'Assign the specific ColumnInformation instance to a local variable
Set objColumnInfo = objColumns.GetColumnInformation(strColumnName)
'Output the ColumnInformation instance's properties
Response.Write "<B>" & objColumnInfo.Name & "</B><BR>"
Response.Write "Value: " & objColumnInfo.Value & ", "
Response.Write "DataType: " & objColumnInfo.DataType & ", "
Response.Write "<P><HR><P>"
168 Chapter 6: Database Reuse
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Example 6-9 shows how easy it is to acquire detailed column information about a
database table using just a few lines of code. The Columns class masks all the
complexity in opening the proper schema and iterating through it. Since we did
not specify a strWhereClause in our call to PopulateColumns, the Value property
for each of the columns is meaningless, grabbed from a random record in the
table.
Figure 6-3 contains a screenshot of Example 6-9 when viewed through a browser.
When generating the “Insert a New Record” administration page, code similar to
that in Example 6-9 will be used to ascertain the topology of the administration
page table. When displaying the “Edit an Existing Record” administration page,
however, we will need to be able to grab information about a particular record in
the database; specifically, we’re interested in the record the user selected to edit.
If the strWhereClause parameter is specified in the call to PopulateColumns, the
column values for a particular row are entered into the Value property of each
ColumnInformation instance. Therefore, if we want to obtain column values for
a particular row along with the general topology of the Product table, we could
adjust our call to PopulateColumns to contain an strWhereClause parameter. For
example, if we want our objColumns class instance in Example 6-9 to have the
column values for the column where ProductID equals 2, we would only need to
adjust our call to PopulateColumns like so:
objColumns.PopulateColumns "Product", objConn, "ProductID = 2"
Now that we’ve examined how to collect column information, we’ll look at how to
use similar techniques to obtain foreign key constraining information.
Gathering Foreign Key Information
In “A Review of Database Terms,” we discussed the importance of properly displaying
foreign key columns in a database table. Remember that foreign key columns
in a table will contain data that maps back to a primary key in another table.
Since foreign key columns are usually stored as numeric data, foreign key columns
are a bit tricky to handle in a generic administration page. When adding a
Set objColumnInfo = Nothing
Next
Set objColumns = Nothing 'Clean up
'Close the database
objConn.Close
Set objConn = Nothing
%>
Example 6-9. The Generic Topology of the Product Table Is Obtained (continued)
Creating Reusable Administration Pages 169
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
new record or deleting an existing record, all foreign key constraints should be
displayed as a list box of easy-to-read options.
Of course, to be able to accomplish this goal, all of the foreign key constraints in a
table must be identified. The adSchemaForeignKeys schema, as its name suggests,
provides a list of all of the foreign key constraints in a database.
Unfortunately, the adSchemaForeignKeys schema somewhat limits our choices of
database providers. At the time of this writing, only the SQL OLE-DB and the Jet 4.0
providers support this schema. For instance, trying to access the
adSchemaForeignKeys schema using an ODBC provider, will result in an error.
This is a tradeoff worth making, though, since an generic administration page generation
application that cannot properly handle tables with foreign key constraints
is severely limited in its usefulness.
Figure 6-3. The topology of the Product table is revealed through the Columns class
170 Chapter 6: Database Reuse
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
The ForeignKeyInformation class
To collect information about the columns in a database table, two classes—
ColumnInformation and Columns—were used. The ColumnInformation class
was responsible for holding the information needed to describe a particular column.
The Columns class served as a container for zero to many ColumnInformation
instances, representing all of the columns in a particular database table.
A very similar approach is used to gather foreign key information. A
ForeignKeyInformation class is used to store specific information on each foreign
key constraint, while a ForeignKeys class serves as a container, holding zero
or more ForeignKeyInformation instances. Both of these classes should be created
in a single file, /CODEREUSE/ForeignKey.Class.asp.
As with ColumnInformation, ForeignKeyInformation simply contains read/
write properties needed to describe a foreign key constraint. Only three member
variables are needed: the column name that has the foreign key constraint placed
upon it (strColumnName), the name of the related table that issued the foreign
key constraint (strPKTableName), and the name of the primary key column in the
related table (strPKColumn).
With our example database (described in Table 6-2), if we wanted to identify all
foreign key constraints for the Product table, we would find that there was
exactly one foreign key constraint, which would be represented by a single
instance of the ForeignKeyInformation class. In this instance of the
ForeignKeyInformation class, the strColumnName member variable would be
set to the Product table’s ProductTypeID column; the strPKTableName member
variable would be set to ProductType; finally, the strPKColumnName
member variable would be set to ProductType’s primary key, ProductTypeID.
Example 6-10 contains the ForeignKeyInformation class definition. Like the
ColumnInformation class, there are no methods or event handlers, just property
statements for the three member variables.
Example 6-10. A ForeignKeyInformation Instance Represents a Single Foreign
Key Constraint
<%
Class ForeignKeyInformation
'************** MEMBER VARIABLES *********************
Private strColumnName
Private strPKTableName
Private strPKColumnName
'*****************************************************
'****************** GET PROPERTIES *******************
Public Property Get ColumnName()
ColumnName = strColumnName
End Property
Creating Reusable Administration Pages 171
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
The ForeignKeys class
The ForeignKeys class serves as a container for zero to many
ForeignKeyInformation instances. Like the Columns class, the ForeignKeys
class only has a single member variable, objFKDict, which is a Dictionary object
instance that holds information about the foreign key constraints for a particular
table.
The ForeignKeys class contains three methods. The first method, PopulateForeignKeys,
simply populates the objFKDict Dictionary object with the foreign key
constraints from a particular table. The second method, GetForeignKeyInformation,
returns a specific foreign key constraint instance from objFKDict. The final
method, ForeignKeyExists, allows the developer to determine whether or not a
particular foreign key constraint exists within the objFKDict Dictionary object.
The methods, properties, and event handlers for the ForeignKeys class are presented
in Example 6-11.
Public Property Get PrimaryKeyTable()
PrimaryKeyTable = strPKTableName
End Property
Public Property Get PrimaryKeyColumn()
PrimaryKeyColumn = strPKColumnName
End Property
'*****************************************************
'****************** LET PROPERTIES *******************
Public Property Let ColumnName(str)
strColumnName = str
End Property
Public Property Let PrimaryKeyTable(strPKTable)
strPKTableName = strPKTable
End Property
Public Property Let PrimaryKeyColumn(strPKColumn)
strPKColumnName = strPKColumn
End Property
'*****************************************************
End Class
%>
Example 6-11. The ForeignKeys Class Serves as a Container for Zero to Many
ForeignKeyInformation Instances
<%
Class ForeignKeys
'************** MEMBER VARIABLES *********************
Example 6-10. A ForeignKeyInformation Instance Represents a Single Foreign
Key Constraint (continued)
172 Chapter 6: Database Reuse
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Private objFKDict
'*****************************************************
'******************* EVENT HANDLERS ******************
Private Sub Class_Initialize()
Set objFKDict = Server.CreateObject("Scripting.Dictionary")
End Sub
Private Sub Class_Terminate()
Set objFKDict = Nothing 'Clean up...
End Sub
'*****************************************************
'********************** METHODS **********************
Public Sub PopulateForeignKeys(strTableName, objConn)
'This sub is responsible for populating the objFKDict
'dictionary object with the foreign keys from the table
'strTableName. objConn is expected to be an opened connection
'to the proper database.
Dim objRS 'Open up the column schema
Set objRS = objConn.OpenSchema(adSchemaForeignKeys, _
Array(Empty, Empty, Empty, Empty, Empty, strTableName))
'Add a ForeignKeyInformation object instance for each column that
'exists within the table specified by strTableName
Dim objFK
Do While Not objRS.EOF
Set objFK = New ForeignKeyInformation
objFK.ColumnName = objRS("FK_COLUMN_NAME")
objFK.PrimaryKeyTable = objRS("PK_TABLE_NAME")
objFK.PrimaryKeyColumn = objRS("PK_COLUMN_NAME")
objFKDict.Add objFK.ColumnName, objFK
Set objFK = Nothing
objRS.MoveNext
Loop
'Clean up...
objRS.Close
Set objRS = Nothing
End Sub
Public Function GetForeignKeyInformation(strFKName)
'Return a specific FK's information... that is, return a
'reference to an instance of a ForeignKeyInformation object
If objFKDict.Exists(strFKName) then
Set GetForeignKeyInformation = objFKDict(strFKName)
Else
Set GetForeignKeyInformation = Nothing
End If
Example 6-11. The ForeignKeys Class Serves as a Container for Zero to Many
ForeignKeyInformation Instances (continued)
Creating Reusable Administration Pages 173
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Take a moment to compare the coding styles used to create the ForeignKeys and
Columns classes. Although these two classes are nearly functionally identical, they
have differing methods. For example, the Columns class contains a Property Get
function that returns the entire objColumnDict Dictionary object. The
ForeignKeys class, however, uses a method to return the objFKDict collection.
The reason for these differences is to demonstrate that there is more than one way
to do things. Of course, when using several related classes to accomplish a common
goal, it is usually best to have those classes share similar interfaces. That is, if
two classes are as functionally similar as Columns and ForeignKeys, it probably
only confuses an end developer who has to use different methods and properties
for each class to accomplish nearly the same thing.
A good example is the ADO objects, which share common method names. For
example, the Connection and Recordset objects both have an Open method,
which, from a high-level view, does the same thing. Since the classes presented in
this chapter are not going to be directly used in practice, I think showing multiple
ways to accomplish the same thing is useful for the reader; hence the differences
between the Columns class and the ForeignKeys class.
Additionally, ForeignKeys differs from the Columns class in that it lacks a property
to return the entire collection of ForeignKeyInformation instances. Rather,
the ForeignKeys class contains only a single property that returns the number of
ForeignKeyInformation instances in the objFKDict collection, and a collection
to pick out a specific element from objFKDict if the developer knows the
name beforehand. This makes it difficult for an end developer to iterate the
objFKDict collection, since the only way an end developer can access a
ForeignKeyInformation instance is if he knows the column name of the
ForeignKeyInformation instance.
This is not a problem, though, because in the generic administration page generation
application we’re creating, the end developer will have to specify how to
End Function
Public Function ForeignKeyExists(strFKName)
ForeignKeyExists = objFKDict.Exists(strFKName)
End Function
'*****************************************************
'****************** GET PROPERTIES *******************
Public Property Get Count()
Count = objFKDict.Count
End Property
'*****************************************************
End Class
%>
Example 6-11. The ForeignKeys Class Serves as a Container for Zero to Many
ForeignKeyInformation Instances (continued)
174 Chapter 6: Database Reuse
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
handle each foreign key column. Since the end developer provides this information,
we know exactly what the column names of each ForeignKeyInformation
instance will be. This may seem a bit confusing now, but it should become clearer
when we discuss the AdminPageGenerator class in the section “Creating the
Administration Page Forms.”
Testing the ForeignKeys and ForeignKeyInformation classes
Like the ColumnInformation class, the ForeignKeyInformation class contains
only a set of read/write properties; therefore, a detailed example is not possible.
Regardless, Example 6-12 illustrates a simple test driver for the class. As with
Example 6-8, in which we created a simple test driver for the Column-
Information class, you must use a server-side include to import the class definition
of ForeignKeyInformation in /CODEREUSE/ForeignKey.Class.asp when
using the class.
Since we cannot iterate through all the items in the objFKDict collection in the
ForeignKeys class, we’ll have to use a specific table that we know has a foreign
key constraint for our test example of the ForeignKeys class. We know the
Product table contains one foreign key constraint, from its ProductTypeID column
to the ProductTypeID column of the ProductType table. Therefore, let’s
use the ForeignKeys class to obtain the foreign key constraint information for
Product.
Like the Columns class, the ForeignKeys class contains a method that does all of
the low-level work needed to obtain the foreign key constraint information. This
Example 6-12. The ForeignKeyInformation Class Contains Three Read/Write Properties
<% @LANGUAGE="VBSCRIPT" %>
<% Option Explicit %>

<%
'Create an instance of the ForeignKeyInformation class
Dim objFKInfo
Set objFKInfo = New ForeignKeyInformation
'Set the properties
objFKInfo.PrimaryKeyTable = "ProductType"
objFKInfo.PrimaryKeyColumn = "ProductTypeID"
objFKInfo.ColumnName = "ProductTypeID"
'Now that we have set the ForeignKeyInformation properties, do something!
Response.Write "Primary Key Table = " & objFKInfo.PrimaryKeyTable
Set objFKInfo = Nothing 'Clean up
%>
Creating Reusable Administration Pages 175
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
method, PopulateForeignKeys, expects two parameters: a table name and an
opened Connection object. The method’s task is to populate the objFKDict Dictionary
object with the foreign key constraint information for the specified database
table.
Example 6-13 contains a test driver for the ForeignKeys class. This test driver
starts by opening a connection to the database. Next, an instance of the
ForeignKeys class is created and the PopulateForeignKeys method is called.
Finally, the foreign key constraint information for the Product table’s Product-
TypeID column is output. The output of the code presented in Example 6-13,
when viewed through a browser, can be seen in Figure 6-4.
Example 6-13. The ForeignKeys Class Canr Obtain the Foreign Key Constraints for a
Particular Table
<% @LANGUAGE="VBSCRIPT" %>
<% Option Explicit %>

<%
'Create a Connection object
Dim objConn
Set objConn = Server.CreateObject("ADODB.Connection")
objConn.ConnectionString = "Provider=SQLOLEDB;Data Source=mitchell;" & _
"Initial Catalog=ProductInfo;User ID=sa;Password="
objConn.Open
'Create an instance of the ForeignKeyInformation class
Dim objFKs
Set objFKs = New ForeignKeys
'Load the foreign key constraints for the Product table
objFKs.PopulateForeignKeys "Product", objConn
'Display the ForeignKeyInformation info for the ProductTypeID column
Dim objFKInfo
Set objFKInfo = objFKs.GetForeignKeyInformation("ProductTypeID")
Response.Write "The Product table contains " & objFKs.Count
Response.Write " foreign key constraints. Specifically, the column "
Response.Write objFKInfo.ColumnName & " of the Product table contains "
Response.Write "a foreign key constraint to the " & objFKInfo.PrimaryKeyColumn
Response.Write " column of the " & objFKInfo.PrimaryKeyTable & " table."
Set objFKs = Nothing 'Clean up
'Close the Connection to the DB
objConn.Close
Set objConn = Nothing
%>
176 Chapter 6: Database Reuse
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Deciding How to Display the Table Columns
in a Form
Part of the challenge of an administration page is deciding how to display the
information from a database in a nice-looking HTML form. When creating an
administration page for each database table by hand, once the developer decides
how to display each table column in a form, all she has to do is create the needed
HTML. With a generic, reusable administration page generation script, however,
the situation is a bit more difficult, since the burden of deciding what form elements
to use for what database table columns is placed squarely on the ASP script.
Needless to say, this makes creating nice-looking administration pages anything
but easy. For example, imagine that the generic administration page generation
script decided to display all numeric database types using a text box. While this
will work, there are times when using a list box is easier for the end user. It’s
unfair to expect the generic administration page generation script to be intelligent
enough to determine the best form field for a particular table column.
Since we can’t create an all-knowing administration page generation script, we’ll
create the next best thing: an administration page generation script that doesn’t
mind receiving advice. The generic administration page generation script will have
its own ideas on what the best form fields are for a given database column type,
but if the end developer kindly asks the script to use a different form field for a
particular database column, the script will dutifully oblige.
The DataTypeFormElementInformation class
Not surprisingly, to provide the functionality needed to determine what form element
to use with what database table column, we’ll employ a class. This class,
DataTypeFormElementInformation, does exactly what its verbose name
Figure 6-4. The foreign key constraint information on the Product table’s ProductTypeID is
displayed via the ForeignKeys class
Creating Reusable Administration Pages 177
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
implies: it maps a particular datatype to a particular form element. A member variable,
objDataTypeDict, is used to store a form element type for each possible
datatype. When the form for updating an existing record or inserting a new record
is generated, an instance of this class is consulted to aid in determining how to
display the table’s particular columns.
Since the objDataTypeDict member variable proposes to have an
entry for each possible datatype, it is important that we define the
set of possible datatypes. For our application, the set of possible
datatypes are those datatypes that have constants defined in
adovbs.inc.
Example 6-14 holds the class definition for DataTypeFormElementInformation,
which should be stored in the file /CODEREUSE/SchemaInfo.Class.asp. Two other
member variables used in the DataTypeFormElementInformation class are
objFormNameDict and objFormDescriptionDict; we’ll look at these momentarily.
For the time being, focus on the emphasized text in Example 6-14, which
contains the objDataTypeDict member variable declaration, its initialization, and
its Property Let and Property Get statements.
Example 6-14. The DataTypeFormElementInformation Class Provides Form Display
Information for Specific Database Table Columns
<%
Class DataTypeFormElementInformation
'************** MEMBER VARIABLES *********************
Private objDataTypeDict
Private objFormNameDict
Private objFormDescriptionDict
'*****************************************************
'******************* EVENT HANDLERS ******************
Private Sub Class_Initialize()
Set objDataTypeDict = Server.CreateObject("Scripting.Dictionary")
Set objFormNameDict = Server.CreateObject("Scripting.Dictionary")
Set objFormDescriptionDict = Server.CreateObject("Scripting.Dictionary")
'Set the default datatype form element values
objDataTypeDict.Add CStr(adInteger), _
"<INPUT TYPE=TEXT SIZE=10 NAME=""col_~COLUMNNAME~""" & _
" VALUE=""~COLUMNVALUE~"">"
objDataTypeDict.Add CStr(adCurrency), _
"$<INPUT TYPE=TEXT SIZE=10 NAME=""col_~COLUMNNAME~""" & _
" VALUE=""~COLUMNVALUE~"">"
objDataTypeDict.Add CStr(adNumeric), _
"<INPUT TYPE=TEXT SIZE=15 NAME=""col_~COLUMNNAME~""" & _
" VALUE=""~COLUMNVALUE~"">"
objDataTypeDict.Add CStr(adBoolean), _
"<SELECT NAME=""col_~COLUMNNAME~"">" & vbCrLf & _
178 Chapter 6: Database Reuse
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
"<OPTION VALUE=""1"" ~COLUMNVALUE~>Yes</OPTION>" & vbCrLf & _
"<OPTION VALUE=""0"" ~COLUMNVALUE~>No</OPTION>" & vbCrLf & _
"</SELECT>"
objDataTypeDict.Add CStr(adDate), _
"<INPUT TYPE=TEXT SIZE=10 NAME=""col_~COLUMNNAME~""" & _
" VALUE=""~COLUMNVALUE~"">"
objDataTypeDict.Add CStr(adChar), _
"<INPUT TYPE=TEXT SIZE=25 NAME=""col_~COLUMNNAME~""" & _
" VALUE=""~COLUMNVALUE~"">"
objDataTypeDict.Add CStr(adWChar), _
"<INPUT TYPE=TEXT SIZE=25 NAME=""col_~COLUMNNAME~""" & _
" VALUE=""~COLUMNVALUE~"">"
objDataTypeDict.Add CStr(adVarChar), _
"<INPUT TYPE=TEXT SIZE=25 NAME=""col_~COLUMNNAME~""" & _
" VALUE=""~COLUMNVALUE~"">"
objDataTypeDict.Add CStr(adVarWChar), _
"<INPUT TYPE=TEXT SIZE=25 NAME=""col_~COLUMNNAME~""" & _
" VALUE=""~COLUMNVALUE~"">"
' ... Several datatypes were excluded here for brevity. Consult
' adovbs.inc for a full listing of possible datatypes ...
End Sub
Private Sub Class_Terminate()
'Cleanup time!
Set objDataTypeDict = Nothing
Set objFormNameDict = Nothing
Set objFormDescriptionDict = Nothing
End Sub
'*****************************************************
'****************** GET PROPERTIES *******************
Public Property Get DataTypeFormElement(iDataType)
DataTypeFormElement = objDataTypeDict(CStr(iDataType))
End Property
Public Property Get FormElementName(strFormName)
FormElementName = objFormNameDict(strFormName)
End Property
Public Property Get FormElementDescription(strFormName)
FormElementDescription = objFormDescriptionDict(strFormName)
End Property
'*****************************************************
'****************** LET PROPERTIES *******************
Public Property Let DataTypeFormElement(iDataType, strFormValue)
'Check to see if the element exists in the dictionary object.
'If it does, update the current value, else add a new value
If objDataTypeDict.Exists(CStr(iDataType)) then
'Alter the current value
Example 6-14. The DataTypeFormElementInformation Class Provides Form Display
Information for Specific Database Table Columns (continued)
Creating Reusable Administration Pages 179
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
In the Initialize event handler, objDataTypeDict is assigned to a new Dictionary
object instance and is populated with the various table column datatypes and their
corresponding form element values. For example, integer (adInteger) data is to
be displayed using an HTML text box control (<INPUT TYPE=TEXT...>). Note
that the form elements contain strings like ~COLUMNNAME~ and ~COLUMNVALUE~.
These tilde-delimited strings are referred to as placeholders. When generating the
administration page, before the HTML to display the form element is sent to the
browser, these placeholders are replaced with specific values. The ~COLUMNAME~
placeholder is replaced with the name of the database field that the form element
represents. At the “Insert a New Record” administration page, the ~COLUMNVALUE~
placeholder is replaced with the column’s default value, if it exists. If the column
has no default value, the ~COLUMNVALUE~ placeholder is simply removed before
the HTML is sent to the browser. In the “Edit an Existing Record” administration
page, the ~COLUMNVALUE~ placeholder is replaced with the value of the column of
the record being edited.
objDataTypeDict(CStr(iDataType)) = strFormValue
Else
'A new element, so add it!
objDataTypeDict.Add CStr(iDataType), strFormValue
End If
End Property
Public Property Let FormElementName(strFormNameValue, strFormName)
'Check to see if the element exists in the dictionary object.
'If it does, update the current value, else add a new value
If objFormNameDict.Exists(CStr(iDataType)) then
'Alter the current value
objFormNameDict(strFormName) = strFormNameValue
Else
'A new element, so add it!
objDataTypeDict.Add strFormName, strFormNameValue
End If
End Property
Public Property Let FormElementDescription(strFormValue, strFormName)
'Check to see if the element exists in the dictionary object.
'If it does, update the current value, else add a new value
If objDataTypeDict.Exists(CStr(iDataType)) then
'Alter the current value
objDataTypeDict(CStr(iDataType)) = strFormValue
Else
'A new element, so add it!
objDataTypeDict.Add CStr(iDataType), strFormValue
End If
End Property
'*****************************************************
End Class
%>
Example 6-14. The DataTypeFormElementInformation Class Provides Form Display
Information for Specific Database Table Columns (continued)
180 Chapter 6: Database Reuse
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Note that objDataTypeDict is populated with a default form element
string for each datatype in the Initialize event handler. For
brevity, only a subset of available datatypes is shown in the Initialize
event handler. When using this class in practice, it is important
that all possible datatypes have an entry in objDataTypeDict. Consult
adovbs.inc to see a complete listing of possible datatypes.
To retrieve the form element string that’s used to display a particular datatype from
the objDataTypeDict Dictionary object, use the DataTypeFormElement property.
This Property Get statement returns a single value from the
objDataTypeDict Dictionary object. For example, the form element string for an
Integer datatype can be obtained using the following code:
objDataTypeFormElementInfoInstance.DataTypeFormElement(adInteger)
As mentioned earlier, the end developer has an opportunity to provide the administration
page generation script with some hints on how to display particular
datatypes. The DataTypeFormElement’s Property Let statement adds a new key/
value pair to objDataTypeDict if the key does not already exist; otherwise, the
Property Let statement modifies the existing entry in objDataTypeDict, and is
used in the following manner:
ObjInstance.DataTypeFormElement(DataType) = strNewFormValue
The end developer will never call this property directly. Rather, if the
DataTypeFormElement property needs to be modified, the
AdminPageGenerator class will call it. The AdminPageGenerator
class, which is discussed in detail later in this chapter in “Creating
the Administration Page Forms,” is responsible for creating the actual
forms needed for the various administration pages.
DataTypeFormElementInformation’s other methods
The DataTypeFormElementInformation class has two other member variables
that have similar Property Get and Property Let statements. These two member
variables, objFormNameDict and objFormDescriptionDict, are both Dictionary
object instances, like objDataTypeDict. Like objDataTypeDict, these
two member variables aid in the display of a form. As the code in Example 6-14
shows, they are accessible through the read/write FormElementName and
FormElementDescription properties.
objFormNameDict can be used to give an English-like name to a form field in the
administration form, while objFormDescriptionDict can be used to provide a
Creating Reusable Administration Pages 181
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
descriptive sentence or two for each form field. Note that both of these Dictionary
objects are initially empty—it is the end developer’s responsibility to populate
these values. If the end developer fails to do so, the literal column names are presented
as the form field names, and the column’s description, if it exists, is used as
the form field’s description.
Figure 6-5 shows the “Insert a New Record” administration page with English-like
form name and description values entered by the end developer. Figure 6-6 shows
the same administration page’s output if the end developer did not specify these
helpful names in objFormNameDict and objFormDescriptionDict.
From the end developer’s standpoint, providing labels and detailed explanations
for any given number of form fields is quite simple. The end developer simply
hints at what particular form fields she’d like to give a specific label or a specific
description. The rest default to the column name and the column description, if it
exists. An example of assigning customized names and descriptive labels to
Figure 6-5. The “Insert a New Record” administration page with helpful form element names
and descriptions
182 Chapter 6: Database Reuse
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
database fields is shown in Example 6-15, which contains a code fragment that
might be written by an end developer to take advantage of the reusable database
coding framework we’re developing in this chapter. The code in Example 6-15
may seem a bit Greek, since we’ve yet to discuss the AdminPageGenerator class,
but it demonstrates how simple it is to label or give a description for a number of
form elements.
Figure 6-6. The “Insert a New Record” administration page without the helpful form element
names and descriptions
Example 6-15. Providing Form Element Labels and Descriptions Is an Easy Task for the End
Developer
'Create an AdminPageGenerator instance
Dim objAdmin
Set objAdmin = new AdminPageGenerator
'Add some form element label hints for specific columns
objAdmin.FormNameHint "ProductTypeID", "Product Type"
objAdmin.FormNameHint "DateEntered", "Date Added"
objAdmin.FormNameHint "Quantity", "Available Quantity"
'Add some form element description hints for specific columns
Creating Reusable Administration Pages 183
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Testing the DataTypeFormElementInformation class
The DataTypeFormElementInformation class, like the ColumnInformation
and ForeignKeyInformation classes, is not well-suited to a detailed test case. A
simple test driver can be seen in Example 6-16. This test driver outputs the default
value of the form element HTML for the adInteger datatype. Next, it assigns a
new form element HTML string for adInteger and outputs the new value.
Note that Example 6-16 starts with a server-side include to import the contents of
/CODEREUSE/SchemaInfo.Class.asp, which contains the definition of the
DataTypeFormElementInformation class. Next, an instance of the DataType-
FormElementInformation class is created, and its default DataTypeForm-
Element value for the adInteger datatype is output.
objAdmin.FormDescriptionHint "DateEntered", _
"Enter the first date the product was inventoried."
objAdmin.FormDescriptionHint "Quantity", _
"Do not count pending orders as available quantity."
objAdmin.FormDescriptionHint "Price", _
"Enter the suggested retail price."
Example 6-16. The DataTypeFormElementInformation Class Provides Display Information for
the Various Datatypes
<% @LANGUAGE="VBSCRIPT" %>
<% Option Explicit %>

<%
'Create an instance of the DataTypeFormElementInformation class
Dim objDTInfo
Set objDTInfo = New DataTypeFormElementInformation
'Use the class somehow... (display the form element for adInteger)
Response.Write "The default form element HTML for an Integer datatype is:<BR>"
Response.Write "<XMP>" & objDTInfo.DataTypeFormElement(adInteger) & "</XMP><BR>"
Response.Write "<FORM>" & objDTInfo.DataTypeFormElement(adInteger) & "</FORM>"
'Now, adjust the default form element HTML for adInteger
objDTInfo.DataTypeFormElement(adInteger) = "<SELECT NAME=""~COLUMNNAME~"">" & _
"<OPTION VALUE=1>1</OPTION><OPTION VALUE=2>2</OPTION></SELECT>"
'Display the new value
Response.Write "<P>The new form element HTML for an Integer datatype is:<BR>"
Response.Write "<XMP>" & objDTInfo.DataTypeFormElement(adInteger) & "</XMP>"
Response.Write "<FORM>" & objDTInfo.DataTypeFormElement(adInteger) & "</FORM>"
Set objDTInfo = Nothing 'Clean up
%>
Example 6-15. Providing Form Element Labels and Descriptions Is an Easy Task for the End
Developer (continued)
184 Chapter 6: Database Reuse
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Since there may be times when the end developer wishes to have a particular
datatype displayed a different way, he can set a particular datatype to a particular
HTML string in DataTypeFormElement. In Example 6-16, the adInteger
datatype is adjusted to display a list box containing two options: 1 and 2. (This, of
course, is not very practical in a real-world application, since there are far more
integers than just 1 and 2!) This new form element HTML string is then output.
Figure 6-7 contains a screenshot of the code in Example 6-16 when viewed
through a browser.
The XMP tag used in Example 6-16 displays characters in a fixedwidth
font, ignoring HTML tags. For more information on the XMP
tag, check out: http://developer.netscape.com/docs/manuals/htmlguid/
tags4.htm#1424741.
Before moving on to generating the needed HTML to create the various administration
page forms, let’s step back and take a look at what information we currently
have at our disposal:
Column information
The Columns class contains a collection of ColumnInformation classes, storing
important information for each column in a particular database table.
Figure 6-7. The DataTypeFormElementInformation class determines how each column in a
table will be displayed in an administration page
Creating Reusable Administration Pages 185
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Foreign key constraint information
The ForeignKeys class contains a collection of ForeignKeyInformation
classes, storing important information for each foreign key constraint in a particular
database table.
Form display information
DataTypeFormElementInformation provides detailed information about
how each potential column datatype should be represented in an HTML form.
This class also permits the end developer to selectively label and provide
detailed information for each form field element in a form.
With this information, we can generate a form to update, insert, or delete records
from a database with some ease. In the next section, “Creating the Administration
Page Forms,” the AdminPageGenerator class is dissected. This class will be
responsible for creating the needed forms.
Creating the Administration Page Forms
Recall that there are three types of administration pages that need to be created:
Insert a new record
Adding a new record requires a form that presents a form field for most columns
in the database table. (AutoNumber columns, for example, should not
have a form field presented.) Once the user enters values into all of these
form fields, a submit button should then insert a new record into the table,
using the information just provided by the user.
Delete an existing record
This administration page needs to provide a listing of all of the records currently
in the database table. Each record listing should have a checkbox,
allowing the user to select multiple records. Once the form is submitted, those
selected records should be deleted.
Edit an existing record
This administration page is a hybrid of both the “Insert a New Record” and
“Delete an Existing Record” administration pages. Initially, a user needs to
select a particular record to edit. This can be accomplished in a fashion similar
to the “Delete an Existing Record” interface. Rather than having a checkbox
beside each record, though, each record should have a radio button, so
the user can only select a single record to edit.
Once the record has been selected, the user should be taken to a form similar
to the one in the “Insert a New Record” interface like the ones shown earlier
in Figure 6-5 and Figure 6-6. In the “Edit an Existing Record” version of this
form, though, each form field should have the value of the record we’re editing.
The user can then make changes to the existing values and have these
changes saved to the database once the form is selected.
186 Chapter 6: Database Reuse
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Once we display the proper administration page form, our job is only half done;
we must still alter the database to reflect this change! This section focuses just on
the AdminPageGenerator class, which, as its name suggests, is used for creating
the administration page forms. In the section “Inserting, Updating, and Deleting
Database Records,” we’ll examine how to commit the users’ changes to the
database.
The AdminPageGenerator class
Creating a particular administration page form should be a simple task for the end
developer using the generic administration page generator presented in this chapter.
Ideally, the end developer would only have to use the following code:
'Create an instance of the AdminPageGenerator class
Dim objAdmin
Set objAdmin = New AdminPageGenerator
'Display the "Insert a New Record" form for the table Users
Response.Write objAdmin.DisplayInsertRecordForm("Users")
Set objAdmin = Nothing 'Clean up
With the AdminPageGenerator class, it is nearly that simple. Before we delve into
the syntax for the methods that generate the various administration page forms, let’s
look at the member variables and properties of the AdminPageGenerator class.
The AdminPageGenerator member variables and properties
Since each administration page is responsible for allowing the user to insert,
update, and delete records from a particular database table, the Admin-
PageGenerator class is only concerned with the columns and foreign key constraints
from a single table. This class should be placed in /CODEREUSE/
SchemaInfo.Class.asp, which should also contain the class definition for
DataTypeFormElementInformation. In all the examples, we’ll look at showing
methods, properties, or event handlers for the AdminPageGenerator class.
Remember that these are all in one class, which should be located in /CODEREUSE/
SchemaInfo.Class.asp.
Classes to store column and foreign key constraint information for a given database
table have already been examined. As its source code in Example 6-17
shows, AdminPageGenerator contains two member variables—objColumns and
objForeignKeys—that are instances of the Columns and ForeignKeys classes.
To generate an administration page form, we need to know not only the column
and foreign key constraint information for a particular database table, but also
how to display the form fields for each of the various database columns. Since
Creating Reusable Administration Pages 187
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
we already have a class that provides information on how each potential database
column datatype should be displayed as a form element, why not use an
instance of this class in AdminPageGenerator? The objDataTypeFormElements
member variable does exactly this, serving as an instance of the
DataTypeFormElementInformation class.
Recall that DataTypeFormElementInformation contains suggestions on how to
display each datatype. Although the AdminPageGenerator does not have any
methods or properties that allow the end developer to alter these set defaults, adding
such functionality is trivial. However, even with this added functionality, the
DataTypeFormElementInformation class does not allow for total flexibility. For
example, imagine the developer wanted to use two different form elements for
two database columns that had the same datatype. Ugh.
To permit such flexibility, we return to a technique used previously in the chapter:
hinting. If the end developer wishes to have a particular table column use a
form element that is different from the default form element for the given datatype,
he can give the AdminPageGenerator class a friendly hint. To accommodate this
feature, the AdminPageGenerator class contains a member variable,
objFormHints, which is a Dictionary object. For each form field hint the end
developer needs to supply, a new key/value pair will be placed in this Dictionary
object. The FormHint method (discussed in “The AdminPageGenerator methods”
section) is available for the end developer to add hints specifying how to display a
particular column as a form element.
Since the Columns and ForeignKeys classes require an opened Connection
object to obtain schema information for a particular database table, the
AdminPageGenerator class also contains private member variables that store
information on how to connect to the correct database. These member variables—
strConnectionString, strUser, and strPassword—store the connection
string, user name, and password, respectively. When the end developer wishes to
display a particular administration page form for a particular database table, she
will need to supply the database connection information. Details on providing this
information are covered in more detail in the next section, “The AdminPageGenerator
methods.”
Another member variable of the AdminPageGenerator class is
strIdentityColumn. When adding a new record to a table in the “Insert a New
Record” interface, the user should not be shown form fields for AutoNumber columns.
Recall that when the Columns class’s PopulateColumns method is called,
extensive information about each column in the database table is collected. One
188 Chapter 6: Database Reuse
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
tidbit of information collected about each column is whether it is an AutoNumber
column or not. In PopulateColumns, to determine which column, if any, is an
AutoNumber column, a single row is grabbed from the database table in question.
Then the dynamic Properties collection of the Recordset object is examined to
determine if any columns are AutoNumber columns. (The PopulateColumns
method is listed in Example 6-6.)
A problem arises with this approach if there are no records in the database table
to begin with, because then the Properties collection of the Recordset object isn’t
available for inspection. To handle this, the end developer can hint at the proper
AutoNumber column. It is important that the end developer provide this information
if the database table for the administration page has or always will have zero
records in it.
Keep in mind that the AdminPageGenerator class is responsible for only one
thing: creating the various administration page forms. This class is not too
interested in what happens once the form is submitted, and the actual work of
inserting, updating, or deleting a database record is left to another class. By
default, the ASP page /CODEREUSE/AdminPageAction.asp is called when an
administration page form is submitted.
Once the administration page form is submitted and a database record has been
inserted, deleted, or updated, what should happen next? The most flexible option
is to have /CODEREUSE/AdminPageAction.asp redirect the user to some other ASP
page. Figure 6-8 depicts the flow between ASP pages that occurs when a user visits
and submits an administration page form.
The AdminPageGenerator class provides the Redirect property for the end developer
to specify where /CODEREUSE/AdminPageAction.asp should send the user
once the database has been modified. The Redirect property is shown in
Example 6-17.
Despite the AdminPageGenerator class’s complexity, it has only two properties,
IdentityColumnHint and Redirect. (As we’ll see in the next section, the meat of the
AdminPageGenerator class is in its methods!) Example 6-17 contains the Admin-
PageGenerator’s member variables, property, and Initialize and Terminate
event handlers. Note that the Initialize event handler instantiates the five
object member variables, while the Terminate event handler simply cleans up.
Creating Reusable Administration Pages 189
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Figure 6-8. When a user submits an administration page form, the database is modified first,
and then the user is redirected to some confirmation page
Example 6-17. The AdminPageGenerator Member Variables, Property, and
Event Handlers
<%
Const strAction ="/CODEREUSE/AdminPageAction.asp"
%>


<%
Class AdminPageGenerator
'************** MEMBER VARIABLES *********************
Private objColumns
Private objForeignKeys
Private objDataTypeFormElements
Private objFormHints
Private objFKHints
Private strIdentityColumn
Private strConnectionString
Private strUserName
Private strPassword
Private strRedirect
Administration Page Form
The user is presented with a form to insert,
update, or delete a record. Once the user
submits the form, they are taken to the next
page, /CODEREUSE/AdminPageAction.asp.
AdminPageAction.asp
The page does the actual work of inserting,
updating, or deleting a record from the
database. Once the database table has been
modified, the user is forwarded on to a page
specified by the end developer.
SomePage.asp
This can be a confirmation page of sorts.
Alternatively, you could redirect the user back
to the Administration Page Menu.
When the administration page form is submitted,
AdminPageAction.asp is visited, which does the
work of modifying the database.
After the database is modified, a Server.Transfer sends
the user to a page specified by the end developer.
190 Chapter 6: Database Reuse
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
The AdminPageGenerator methods
The AdminPageGenerator class is responsible for creating the “Insert a New
Record,” “Edit an Existing Record,” and “Delete an Existing Record” forms. The
AdminPageGenerator class contains two methods to create the “Insert a New
Record” and “Delete an Existing Record” administration page forms: CreateRecord-
Form and DeleteRecordForm, respectively. The “Edit an Existing Record” administration
page requires two methods, since two forms are used. The first form,
created by the UpdateRecordList method, generates a list of potential records to
edit, from which the user can select a single record; the second form, created by
the UpdateRecordForm method, contains the form fields for the user to edit the
particular record. These four administration page form generation methods have
the following definitions:
'*****************************************************
'****************** LET PROPERTIES *******************
Public Property Let IdentityColumnHint(strIdentity)
strIdentityColumn = strIdentity
End Property
Public Property Let Redirect(str)
strRedirect = str
End Property
'*****************************************************
'******************* EVENT HANDLERS ******************
Private Sub Class_Initialize()
Set objColumns = New Columns
Set objForeignKeys = New ForeignKeys
Set objDataTypeFormElements = New DataTypeFormElementInformation
Set objFormHints = Server.CreateObject("Scripting.Dictionary")
Set objFKHints = Server.CreateObject("Scripting.Dictionary")
End Sub
Private Sub Class_Terminate()
Set objFormHints = Nothing
Set objFKHints = Nothing
Set objColumns = Nothing
Set objForeignKeys = Nothing
Set objDataTypeFormElements = Nothing
End Sub
'*****************************************************
'The AdminPageGenerator methods have been omitted. They are discussed
'in the next section, "The AdminPageGenerator Methods."
End Class
%>
Example 6-17. The AdminPageGenerator Member Variables, Property, and
Event Handlers (continued)
Creating Reusable Administration Pages 191
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Public Function CreateRecordForm(strTableName, strConnString, _
strUser, strPass)
Public Function UpdateRecordList(strTableName, strConnString, _
strUser, strPass)
Public Function UpdateRecordForm(strTableName, strWhereClause, _
strConnString, strUser, strPass)
Public Function DeleteRecordForm(strTableName, strConnString, _
strUser, strPass)
The parameters that need to be passed into these methods are fairly straightforward.
strTableName specifies what database table the administration page is for.
strConnString, strUser, and strPass contain database connection information.
The strWhereClause in the UpdateRecordForm method specifies a valid
SQL WHERE clause that will grab exactly one record from the table strTableName.
Each of these methods returns a string that contains the HTML needed to generate
the proper form.
For example, to display the “Insert a New Record” form for the ProductType
database table, the following code could be used:
'Create an instance of AdminPageGenerator
Dim objAdmin
Set objAdmin = New AdminPageGenerator
Dim strConnString
strConnString = "Provider=SQLOLEDB;Data Source=mitchell;" & _
"Initial Catalog=ProductInfo;User ID=sa;Password="
'Send the HTML to the client.
Response.Write objAdmin.CreateRecordForm("ProductType", strConnString, "", "")
Set objAdmin = Nothing 'Clean up!
Note that the connection string in the previous code snippet uses the
SQLOLEDB database provider. The adSchemaForeignKeys schema is
supported only by the SQLOLEDB and Jet 4.0 database providers.
You can obtain the latest OLE DB providers at http://www.microsoft.
com/data/.
These three methods will be examined in great detail over the next three sections.
Before that, though, let’s look at the four remaining AdminPageGenerator methods.
These four methods enable the end developer to give “hints” to
AdminPageGenerator. Recall that in “The DataTypeFormElementInformation
class” section we discussed how DataTypeFormElementInformation provided a
form field template for each possible column datatype. However, problems could
arise if the developer wanted to have two columns that had the same datatype to
have differing form fields.
192 Chapter 6: Database Reuse
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
To accommodate this, the FormHint method can be used. The FormHint method
expects two parameters: the name of a particular column, and the form element
code to use in place of the form element code “recommended” by
DataTypeFormElementInformation. The form element code should be valid
and complete HTML code that generates a particular form element. The placeholders
~COLUMNNAME~ and ~COLUMNVALUE~ can be used in the form element code,
just as they were in the DataTypeFormElementInformation class. For example,
if you had a column named Price that you wanted to create as a text box with its
SIZE property set to 2, you could use the FormHint method like so:
objAdmin.FormHint "Price", "$<INPUT TYPE=TEXT SIZE=2 " & _
"NAME=""col_~COLUMNNAME~""" & _
" VALUE=""~COLUMNVALUE~"">"
The FormNameHint and FormDescriptionHint methods allow the end developer to
provide English-like names and descriptions for each of the form elements. Both
of these methods take two parameters: the name of the column, and the name or
description to use. In “DataTypeFormElementInformation’s other methods,” we
examined a code snippet that used these two methods.
Recall from our earlier discussions in this chapter that displaying tables that have
foreign key constraints is a bit of a sticky matter. It is important that a list box be
presented that contains all of the valid options. For example, imagine that we had
a Portfolio table that had a column named InvestmentTypeID, which was a
foreign key to an InvestmentType table. For this example, InvestmentType
contains two columns, ID and Name, serving as a look-up table pairing a numeric
ID to a particular type of investment (bond, stock, mutual fund, etc.). When a user
is presented the “Edit an Existing Record” administration page for the Portfolio
table, the last thing we want to do is show them a numeric field in the
InvestmentTypeID form field. Rather, we’d like to show them a list box containing
the various types of investments.
Unfortunately, AdminPageGenerator has absolutely no idea what columns in
InvestmentType to display in the list box in the “Edit an Existing Record” administration
page form. Therefore, it is the responsibility of the end developer to hint
at what fields she’d like to see listed. The ForeignKeyHint method of the
AdminPageGenerator class is the mechanism the end developer uses to provide
foreign key constraint hints.
In our Product/Catalog example, there are a total of three foreign
key constraints. The Product table has a single foreign key constraint
with the ProductType table. The CatalogProduct table has
two foreign key constraints, one to the Catalog table’s primary key,
and the other to the Product table’s primary key.
Creating Reusable Administration Pages 193
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
The ForeignKeyHint method takes two parameters: the name of the column that is
a foreign key, and the text to display in the list box, having the following
definition:
Public Sub ForeignKeyHint(strColumnName, strListBoxValue)
In our Portfolio/InvestmentType example, the end developer might want to
have the text: “INVESTMENT NAME Investment” appear in the list box, where
INVESTMENT NAME is the value in the InvestmentType table’s Name column. To
accomplish this, the end developer could set strListBoxValue to: “~Name~
Investment.” To insert a column value, simply enter the column name surrounded
by tildes (~).
objAdmin.ForeignKeyHint "InvestmentTypeID", "~Name~ Investment"
Example 6-18 contains the code for these four methods of the Admin-
PageGenerator class: ForeignKeyHint, FormHint, FormNameHint, and FormDescriptionHint.
Example 6-18. AdminPageGenerator Provides Four Methods
Class AdminPageGenerator
' ... Properties, event handlers, and other methods omitted for
' brevity. The properties and event handlers are shown in
' Example 6-12 ...
'********************** METHODS **********************
Public Sub ForeignKeyHint(strColumnName, strListBoxValue)
If objFKHints.Exists(UCase(strColumnName)) then
objFKHints(UCase(strColumnName)) = strListBoxValue
Else
objFKHints.Add UCase(strColumnName), strListBoxValue
End If
End Sub
Public Sub FormHint(strFormName, strFormValue)
If objFormHints.Exists(strFormName) then
objFormHints(UCase(strFormName)) = strFormValue
Else
objFormHints.Add UCase(strFormName), strFormValue
End If
End Sub
Public Sub FormNameHint(strFormName, strName)
objDataTypeFormElements.FormElementName(UCase(strFormName)) = strName
End Sub
Public Sub FormDescriptionHint(strFormName, strFormDesc)
objDataTypeFormElements.FormElementDescription(UCase(strFormName)) = _
strFormDesc
End Sub
'*****************************************************
End Class
194 Chapter 6: Database Reuse
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
The “Insert a New Record” administration page
Example 6-19 contains the code for InsertRecord.asp, which displays the “Insert a
New Record” administration page form shown in Figure 6-9. The CreateRecord-
Form method of the AdminPageGenerator class is responsible for generating an
administration page form that permits the user to add a new record to a table. In
Figure 6-9, you can see a screenshot of the “Insert a New Record” administration
page form for the Product table. Note that the ProductTypeID column (which is
a foreign key to the ProductType table) is displayed as a list box with the available
product type names, rather than their corresponding product type codes.
Figure 6-9. The “Insert a New Record” administration page for the Product table permits the
user to create a new record
Example 6-19. The “Insert a New Record” Administration Page Can Be Created with Just a Few
Lines of Code
<% @LANGUAGE="VBSCRIPT" %>
<% Option Explicit %>

<%
Dim strTableName
strTableName = "Product"
Creating Reusable Administration Pages 195
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Since the administration page table name is hardcoded in
Example 6-19 (strTableName = "Product"), InsertRecord.asp will
only display a very specific administration page. Later in this chapter
we’ll look at how to make a more dynamic ASP page, able to display
any administration page for any database table.
Creating an “Insert a New Record” administration page form consists of a few
steps:
1. Create an instance of the AdminPageGenerator class.
2. Set the IdentityColumnHint property, if necessary.
3. For each foreign key constraint in the specified table, issue a call to Foreign-
KeyHint.
4. Use the CreateRecordForm method, passing it the table name and database
connection information, to generate the proper form.
It’s that simple! In fact, later in this chapter we’ll examine how to add yet another
level of encapsulation to make it even easier to create a generic administration
page form!
Now that we’ve looked at how an end developer uses the AdminPageGenerator
class to create an “Insert a New Record” administration page form, it’s time we
Dim objAdmin
Set objAdmin = new AdminPageGenerator
'Set the ForeignKeyHint and the IdentityColumnHint
objAdmin.IdentityColumnHint = "ProductID"
objAdmin.ForeignKeyHint "ProductTypeid", "Product Type ~Name~"
%>
<HTML><BODY>
<h1>Insert a <%=UCase(strTableName)%></h1><BR>
<%
Dim strConnString
strConnString = "Provider=SQLOLEDB;Data Source=mitchell;" & _
"Initial Catalog=ProductInfo;User ID=sa;Password="
'Display the form
Response.Write objAdmin.CreateRecordForm(strTableName, _
strConnString, "", "")
Set objAdmin = Nothing 'Clean up...
%>
</BODY></HTML>
Example 6-19. The “Insert a New Record” Administration Page Can Be Created with Just a Few
Lines of Code (continued)
196 Chapter 6: Database Reuse
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
examined the source code for its CreateRecordForm method, which can be seen in
Example 6-20.
Recall that the “Edit an Existing Record” administration page requires two forms;
the first form lists the records in the database table, allowing the user to select
one; the second form provides the user a chance to edit the values for that particular
record. Note that the second form in the “Edit an Existing Record” administration
page and the single form in the “Insert a New Record” administration page are
nearly identical. The only difference is the “Edit an Existing Record” form lists the
values of the record being edited in the form fields.
Due to the fact that the “Insert a New Record” and “Edit an Existing Record”
administration page forms are nearly identical, it’s not surprising that the source
code for CreateRecordForm and UpdateRecordForm would be nearly identical as
Example 6-20. When Creating the “Insert a New Record” Administration Page Form,
CreateRecordForm Calls the Reusable CreateForm Helper Function


<%
Class AdminPageGenerator
' ... Properties, event handlers, and other methods omitted for
' brevity. The properties and event handlers are shown in
' Example 6-12 ...
'********************** METHODS **********************
Public Function CreateRecordForm(strTableName, strConnString, _
strUser, strPass)
'Assign the private member variables the values
strConnectionString = strConnString
strUserName = strUser
strPassword = strPass
'Populate the objColumns and objForeignKeys objects
'Create a connection to the database
Dim objConn
Set objConn = Server.CreateObject("ADODB.Connection")
objConn.Open strConnectionString, strUserName, strPassword
'Populate the objColumn and objForeignKeys member variables
objColumns.PopulateColumns strTableName, objConn, ""
objForeignKeys.PopulateForeignKeys strTableName, objConn
objConn.Close
Set objConn = Nothing 'Clean up!
CreateRecordForm = CreateForm(True, strTableName) 'Call the helper function
End Function
'*****************************************************
End Class
%>
Creating Reusable Administration Pages 197
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
well. Rather than simply cut and paste code from these two methods, a common
helper function is called from both CreateRecordForm and UpdateRecordForm.
This helper function, CreateForm, which is shown in Example 6-21, expects two
parameters: bolNewRecord, a Boolean, indicating whether or not the function is
being called from CreateRecordForm or UpdateRecordForm, and strTableName,
the name of the table that the administration page form is being created for. This
helper function simplifies the code in CreateRecordForm to simply opening a connection
to the database, populating the objColumns and objForeignKeys member
variables of the AdminPageGenerator class, closing the connection, and
calling CreateForm, passing it a value of True.
The CreateForm helper function shown in Example 6-21 returns a string containing
the HTML to display the proper administration page form. Because Create-
Form is a helper function only to be called by AdminPageGenerator methods, it
is declared as Private. Since CreateForm is charged with the task of creating a
generic insert or edit form, it’s no surprise that its code is lengthy and painfully
unwieldy.
Before trying to dissect the code in Example 6-21, ask yourself what the code for
CreateForm needs to accomplish. Look at the problem from a high-level view.
Large tasks don’t seem so daunting if viewed in small, discrete chunks. The highlevel
tasks assigned to CreateForm include:
1. Determining if any foreign key list boxes need to be displayed. If so, a database
connection must be established.
2. Looping through each of the columns in the objColumns member variable. For
each column, display the correct form field. Displaying the correct form field
consists of checking the hint collections and objDataTypeFormElements, the
member variable instance of the DataTypeFormElementInformation class.
3. Creating the proper HTML tags to generate a properly functioning form,
including a Submit button.
Obviously, the meat of CreateForm comes from Step 2. Note the comments in
Example 6-21 that show where each of the above three steps roughly begins.
Example 6-21. CreateForm Is Responsible for Generating the Forms for the “Insert a New
Record” and “Edit an Existing Record” Administration Pages
<%
Const strAction ="/CODEREUSE/AdminPageAction.asp"
%>


<%
Class AdminPageGenerator
' ... Properties, event handlers, and other methods omitted for
' brevity. The properties and event handlers are shown in
' Example 6-12 ...
198 Chapter 6: Database Reuse
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
'********************** METHODS **********************
Private Function CreateForm(bolNewRecord, strTableName)
'This function creates an HTML form to add a new record to a database
Dim strResult, strElement, objRS, strSQL, objConn, _
strFKHint, objField, strTemp
'******** STEP 1: Determine if any foreign keys exist! ****************
'If we need to access foreign keys, create/open the connection and
'create the recordset
If objForeignKeys.Count > 0 AND objFKHints.Count > 0 then
Set objConn = Server.CreateObject("ADODB.Connection")
objConn.Open strConnectionString, strUserName, strPassword
Set objRS = Server.CreateObject("ADODB.Recordset")
End If
Dim strColName, objColumnInformation, strHeader, objFKInformation
strResult = "<FORM METHOD=POST ACTION=""" & strAction & """>" & vbCrLf
strResult = strResult & "<INPUT TYPE=HIDDEN NAME=ConnectionString " & _
"VALUE=""" & strConnectionString & """>" & vbCrLf
strResult = strResult & "<INPUT TYPE=HIDDEN NAME=UserName " & _
"VALUE=""" & strUserName & """>" & vbCrLf
strResult = strResult & "<INPUT TYPE=HIDDEN NAME=Password " & _
"VALUE=""" & strPassword & """>" & vbCrLf
'If we are updating a record, we need to record the value of the ID
'and record that the action is UPDATE
If Not bolNewRecord then
strResult = strResult & "<INPUT TYPE=HIDDEN NAME=""PrimaryKeyColumns""" & _
""" VALUE=""" & objColumns.GetPrimaryKeys() & """>" & vbCrLf
strResult = strResult & "<INPUT TYPE=HIDDEN NAME=""Action"" " & _
"VALUE = ""UPDATE"">" & vbCrLf
Else
'We are inserting a new record
strResult = strResult & "<INPUT TYPE=HIDDEN NAME=""Action"" " & _
"VALUE = ""INSERT"">" & vbCrLf
End If
'Pass along the table name
strResult = strResult & "<INPUT TYPE=HIDDEN NAME=""TableName"" " & _
"VALUE = """ & strTableName & """>" & vbCrLf
'Pass along the redirect URL (where the user is sent after the row is
'updated/inserted)
If Len(strRedirect) > 0 then
strResult = strResult & "<INPUT TYPE=HIDDEN NAME=""Redirect"" " & _
"VALUE = """ & strRedirect & "?TableName=" & strTableName & _
Example 6-21. CreateForm Is Responsible for Generating the Forms for the “Insert a New
Record” and “Edit an Existing Record” Administration Pages (continued)
Creating Reusable Administration Pages 199
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
""">" & vbCrLf
End If
'******** STEP 2: Loop through the Columns in objColumns! ************
For Each strColName in objColumns.ColumnList
Set objColumnInformation = objColumns.GetColumnInformation(strColName)
'Only display the form element if the column is not an IDENTITY field
If Not objColumnInformation.Identity And _
Not UCase(objColumnInformation.Name) = UCase(strIdentityColumn) then
'Determine whether this row is a foreign key or not...
If objForeignKeys.ForeignKeyExists(objColumnInformation.Name) _
And objFKHints.Exists(UCase(objColumnInformation.Name)) then
Set objFKInformation = objForeignKeys.GetForeignKeyInformation( _
objColumnInformation.Name)
strFKHint = objFKHints(UCase(objColumnInformation.Name))
If InStr(1, strFKHint, "<") then
strElement = strFKHint
Else
strElement = "<SELECT NAME=""col_~COLUMNNAME~"">" & vbCrLf
strSQL = "SELECT * FROM [" & _
objFKInformation.PrimaryKeyTable & "]"
objRS.Open strSQL, objConn
Do While Not objRS.EOF
strElement = strElement & "<OPTION VALUE=""" & _
objRS(objFKInformation.PrimaryKeyColumn) & """"
If objColumnInformation.Value = _
objRS(objFKInformation.PrimaryKeyColumn) then
strElement = strElement & " SELECTED "
End If
strElement = strElement & ">"
strTemp = strFKHint
For Each objField in objRS.Fields
strTemp = Replace(strTemp, "~" & objField.Name & "~", _
objField.Value, 1, -1, vbTextCompare)
Next
strElement = strElement & strTemp & "</OPTION>" & vbCrLf
objRS.MoveNext
Loop
objRS.Close 'Close the connection
strElement = strElement & "</SELECT><P>" & vbCrLf
End If
Else
'Check to see if a FormHint exists...
Example 6-21. CreateForm Is Responsible for Generating the Forms for the “Insert a New
Record” and “Edit an Existing Record” Administration Pages (continued)
200 Chapter 6: Database Reuse
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
If objFormHints.Exists(UCase(objColumnInformation.Name)) then
strElement = objFormHints(UCase(objColumnInformation.Name)) & _
"<P>" & vbCrLf
Else
strElement = objDataTypeFormElements.DataTypeFormElement(_
objColumnInformation.DataType) & _
"<P>" & vbCrLf
End If
End If
Set objFKInformation = Nothing 'Clean up...
strElement = Replace(strElement, "~COLUMNNAME~", _
objColumnInformation.Name)
If bolNewRecord then
If objColumnInformation.HasDefault then
strElement = Replace(strElement, "~COLUMNVALUE~", _
objColumnInformation.DefaultValue)
Else
strElement = Replace(strElement, "~COLUMNVALUE~", "")
End If
Else
If IsNull(objColumnInformation.Value) then
'A NULL value – this needs special care
strElement = Replace(strElement, "~COLUMNVALUE~", "")
Else
strElement = Replace(strElement, "~COLUMNVALUE~", _
objColumnInformation.Value)
End If
End If
If Len(objDataTypeFormElements.FormElementName(UCase(strColName))) > 0 _
then
strHeader = "<B>" & _
objDataTypeFormElements.FormElementName(UCase(strColName)) & _
"</B>" & vbCrLf
Else
strHeader = "<B>" & objColumnInformation.Name & _
"</B>" & vbCrLf
End If
If _ 'long line!
Len(objDataTypeFormElements.FormElementDescription(UCase(strColName))) _
> 0 then
strHeader = strHeader & ": " & _
objDataTypeFormElements.FormElementDescription(UCase(strColName)) & _
vbCrLf
Elseif Len(objColumnInformation.Description) > 0 then
strHeader = strHeader & ": " & objColumnInformation.Description & _
"" & vbCrLf
End If
strResult = strResult & strHeader & "<BR>" & strElement
End If
Example 6-21. CreateForm Is Responsible for Generating the Forms for the “Insert a New
Record” and “Edit an Existing Record” Administration Pages (continued)
Creating Reusable Administration Pages 201
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Near the top of CreateForm, an HTML FORM tag is created, with its ACTION property
set to the value in the variable strAction. strAction is a constant defined at
the top of the file in which the AdminPageGenerator class appears (/CODEREUSE/
SchemaInfo.Class.asp), specifying the URL of the form-processing script for the various
administration page forms. For these examples, strAction is set as follows:
Const strAction ="/CODEREUSE/AdminPageAction.asp"
Also, since AdminPageGenerator uses the ForeignKeys and Columns classes,
server-side includes must be used to import both /CODEREUSE/ForeignKey.Class.asp
as well as /CODEREUSE/Column.Class.asp into /CODEREUSE/SchemaInfo.Class.asp.
The generic administration page generation script presented throughout
this chapter is not suited to handle NULLable foreign key constraints.
This feature could be added, though, and is left as an
exercise to the interested reader.
Set objColumnInformation = Nothing 'Clean up!
Next
'******** STEP 3: Finish up the FORM tag ************
'Create submit button
strResult = strResult & "<P><INPUT TYPE=SUBMIT VALUE="""
If bolNewRecord then
strResult = strResult & "Add Record"
Else
strResult = strResult & "Update Record"
End If
strResult = strResult & """>"
strResult = strResult & vbCrLf & "</FORM>" & vbCrLf & vbCrLf
CreateForm = strResult
'Clean up if we created connection/recordset objects
If objForeignKeys.Count > 0 AND objFKHints.Count > 0 then
Set objRS = Nothing
objConn.Close
Set objConn = Nothing
End If
End Function
'*****************************************************
End Class
%>
Example 6-21. CreateForm Is Responsible for Generating the Forms for the “Insert a New
Record” and “Edit an Existing Record” Administration Pages (continued)
202 Chapter 6: Database Reuse
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
The “Delete an Existing Record” administration page
As with the “Insert a New Record” administration page, the “Delete an Existing
Record” administration page shares some uncanny similarities with the “Edit an
Existing Record” administration page. Namely, in both the “Delete an Existing
Record” administration page and the “Edit an Existing Record” administration page,
a form is needed to list the records in the table. In the “Delete an Existing Record”
administration page, the user can select one or more records to delete from the list
of current records; in the “Edit an Existing Record” administration page, the user
can select one and only one record to edit.
Before we explore this relationship further, let’s step back for a moment and take
a look at a screenshot of the “Delete an Existing Record” administration page,
which is shown in Figure 6-10.
Notice that each record in the table is listed with a checkbox next to it. Also note
that the ProductTypeID column contains the value from the Name column in the
ProductType table rather than the actual numeric id in the Product table. In the
“Insert a New Record” administration page form, the end developer could enter
foreign key hints to display foreign key constraint columns as list boxes with
applicable choices as opposed to a text box with a numeric id present. The same
mechanism is used in the “Delete an Existing Record” to provide for translation
from a numeric foreign key to a readable value.
Figure 6-10. The “Delete an Existing Record” administration page for the Product table
allows the user to delete one or more records
Creating Reusable Administration Pages 203
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
To display the “Delete an Existing Record” administration page form, the end
developer simply needs to output the returned HTML string from the
DeleteRecordForm method. DeleteRecordForm, like CreateRecordForm, takes four
parameters: the name of the database table whose administration page form it is to
display, the connection string, the username, and the password. Also, like
CreateRecordForm, DeleteRecordForm does very little in the way of actual work in
displaying the “Delete an Existing Record” administration page form. Another
helper function, CreateList, is responsible for actually doing the grunt work of creating
the form seen in Figure 6-10.
CreateList is a generic helper function that is also called from the UpdateRecordList
method. CreateList, as its name implies, creates a listing of all of the database
records in a given table, and has the following definition:
Private Function CreateList(bolDelete, objConn, strTableName)
The bolDelete function is a Boolean value, indicating whether the user is deleting
one or more records from the list of database records or is selecting a record to
edit. When bolDelete is True, a checkbox is placed beside each listed record;
when bolDelete is False, meaning the user is choosing a single record to edit, a
radio button is used instead. CreateList expects an opened database connection to
be passed in as the second parameter. The final parameter, strTableName, is the
name of the database table for which the administration page form is being created.
DeleteRecordForm, like CreateRecordForm, populates the objColumns and
objForeignKeys member variables and calls a helper function. The source code
for DeleteRecordForm can be seen in Example 6-22.
Example 6-22. DeleteRecordForm Obtains Column and Foreign Key Information and Hands
Display Responsibilities to CreateList


<%
Class AdminPageGenerator
' ... Properties, event handlers, and other methods omitted for
' brevity. The properties and event handlers are shown in
' Example 6-12 ...
'********************** METHODS **********************
Public Function DeleteRecordForm(strTableName, strConnString, _
strUser, strPass)
'Present a form that lists the database columns with a
'checkbox next to each column. The user can then select
'what columns he'd like to delete.
'Assign the private member variables the values
strConnectionString = strConnString
strUserName = strUser
strPassword = strPass
204 Chapter 6: Database Reuse
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
As with the CreateForm helper function, CreateList is a colossal function, responsible
for displaying a list of records for both the “Delete an Existing Record” and
“Edit an Existing Record” administration page forms. Before we delve into the
code of CreateList, let us step back and take a high-level view of the tasks Create-
List needs to complete:
1. Determine if any foreign key constraints exist. If any do, each time a row is
listed, a query will need to be run to grab the correct information to display
the readable name indicated by the specific foreign key hint.
2. Set up a series of HIDDEN form fields that pass along the needed information.
If CreateList is called from UpdateRecordList (that is, if bolDelete is False),
adequate information must be passed along so that once a record is selected
and the form submitted, UpdateRecordForm can be successfully called.
3. Obtain a snapshot of the database table and iterate through each record, displaying
the value for each column in tabular form.
The code for CreateList is presented in Example 6-23. Note the comments in
Example 6-23 that illustrate where each of the above three steps roughly begins.
'Create a connection to the database
Dim objConn
Set objConn = Server.CreateObject("ADODB.Connection")
objConn.Open strConnectionString, strUserName, strPassword
'Populate the objColumns and objForeignKeys objects
objColumns.PopulateColumns strTableName, objConn, ""
objForeignKeys.PopulateForeignKeys strTableName, objConn
'Call the helper function CreateList
DeleteRecordForm = CreateList(True, objConn, strTableName)
objConn.Close 'Close and Clean Up...
Set objConn = Nothing
End Function
'*****************************************************
End Class
%>
Example 6-23. The Generic Helper Function CreateList Creates a List of Records, Allowing the
User to Select One or More


<%
Class AdminPageGenerator
' ... Properties, event handlers, and other methods omitted for
' brevity. The properties and event handlers are shown in
Example 6-22. DeleteRecordForm Obtains Column and Foreign Key Information and Hands
Display Responsibilities to CreateList (continued)
Creating Reusable Administration Pages 205
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
' Example 6-12 ...
'********************** METHODS **********************
Private Function CreateList(bolDelete, objConn, strTableName)
'This function creates the HTML needed to select a particular
'record from a table
'******** STEP 1: Determine if any foreign key constraints exist ************
Dim objFKRS
'If we need to access foreign keys, gather the foreign key table rows
If objForeignKeys.Count > 0 AND objFKHints.Count > 0 then
Set objFKRS = Server.CreateObject("ADODB.Recordset")
End If
Dim strResult
strResult = "<FORM METHOD=POST ACTION=""" & strAction & """>" & vbCrLf
'******** STEP 2: Set up the needed HIDDEN form fields ************
strResult = strResult & "<INPUT TYPE=HIDDEN NAME=ConnectionString " & _
"VALUE=""" & strConnectionString & """>" & vbCrLf
strResult = strResult & "<INPUT TYPE=HIDDEN NAME=UserName " & _
"VALUE=""" & strUserName & """>" & vbCrLf
strResult = strResult & "<INPUT TYPE=HIDDEN NAME=Password " & _
"VALUE=""" & strPassword & """>" & vbCrLf
If bolDelete then
'Indicate that we are deleting a record
strResult = strResult & "<INPUT TYPE=HIDDEN NAME=""Action"" " & _
"VALUE = ""DELETE"">" & vbCrLf
Else
'Indicate that we have selected a record to be deleted.
strResult = strResult & "<INPUT TYPE=HIDDEN NAME=""Action"" " & _
"VALUE = ""UPDATESELECT"">" & vbCrLf
End If
strResult = strResult & "<INPUT TYPE=HIDDEN NAME=TableName " & _
"VALUE=""" & strTableName & """>" & vbCrLf
'Pass along the redirect URL (where the user is sent after the row is
'updated/inserted)
If Len(strRedirect) > 0 then
strResult = strResult & "<INPUT TYPE=HIDDEN NAME=""Redirect"" " & _
"VALUE = """ & strRedirect & "?TableName=" & _
strTableName & """>" & vbCrLf
End If
'See if we need to pass along ForeignKey hints. If we are updating a record
'we will need to pass along this information!
If Not bolDelete then
Dim objFKHintsInstance
Example 6-23. The Generic Helper Function CreateList Creates a List of Records, Allowing the
User to Select One or More (continued)
206 Chapter 6: Database Reuse
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
For Each objFKHintsInstance in objFKHints
strResult = strResult & "<INPUT TYPE=HIDDEN NAME=FKName " & _
"VALUE=""" & objFKHintsInstance & """>" & vbCrLf
strResult = strResult & "<INPUT TYPE=HIDDEN NAME=FKValue " & _
"VALUE=""" & objFKHints(UCase(objFKHintsInstance)) & _
""">" & vbCrLf
Next
End If
'We need to grab the entire contents of the database
Dim objRS
Set objRS = Server.CreateObject("ADODB.Recordset")
objRS.Open "[" & strTableName & "]", objConn, , , adCmdTable
Dim objField, objColumnList
Set objColumnList = objColumns.ColumnList 'Get the list of columns
strResult = strResult & "<TABLE BORDER=1 CELLSPACING=1>"
strResult = strResult & vbCrLf & "<TR>" & vbCrLf
strResult = strResult & "<TH> </TH>"
For Each objField in objRS.Fields
If Not objColumnList(objField.Name).Identity then
strResult = strResult & "<TH>" & objField.Name & "</TH>" & vbCrLf
End If
Next
strResult = strResult & "</TR>"
Dim strFKHint, objFKInformation, objColumnInformation, strElement, _
strSQL, objFKField, strTemp
'******* STEP 3: Iterate through the Snapshot Results of the Database *******
Do While Not objRS.EOF
strResult = strResult & "<TR>" & vbCrLf
'Do we need to display a checkbox or a radio button?
If bolDelete then
strResult = strResult & "<TD><INPUT TYPE=CHECKBOX NAME=ID VALUE=""" & _
objColumns.GetCurrentPrimaryKeys(objRS) & """>" & vbCrLf
Else
strResult = strResult & "<TD><INPUT TYPE=RADIO NAME=ID VALUE=""" & _
objColumns.GetCurrentPrimaryKeys(objRS) & """>" & vbCrLf
End If
For Each objField in objRS.Fields
If Not objColumnList(objField.Name).Identity then
Set objColumnInformation = objColumns.GetColumnInformation(objField.Name)
'Determine whether this row is a foreign key or not...
If objForeignKeys.ForeignKeyExists(objColumnInformation.Name) _
And objFKHints.Exists(UCase(objColumnInformation.Name)) then
Set objFKInformation = objForeignKeys.GetForeignKeyInformation(_
objColumnInformation.Name)
Example 6-23. The Generic Helper Function CreateList Creates a List of Records, Allowing the
User to Select One or More (continued)
Creating Reusable Administration Pages 207
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
strFKHint = objFKHints(UCase(objColumnInformation.Name))
strSQL = "SELECT * FROM [" & objFKInformation.PrimaryKeyTable & _
"] WHERE [" & _
objFKInformation.PrimaryKeyColumn & "] = " & _
objRS(objFKInformation.ColumnName).Value
objFKRS.Open strSQL, objConn
strTemp = strFKHint
For Each objFKField in objFKRS.Fields
strTemp = Replace(strTemp, "~" & objFKField.Name & "~", _
objFKField.Value, 1, -1, vbTextCompare)
Next
strResult = strResult & "<TD>" & strTemp & "</TD>" & vbCrLf
objFKRS.Close
Else
If IsNull(objField.Value) then
strResult = strResult & "<TD ALIGN=RIGHT> "
Else
Select Case objColumnList(objField.Name).DataType
Case adCurrency
strResult = strResult & "<TD ALIGN=RIGHT>" & _
FormatCurrency(objField.Value, 2)
Case adDate
strResult = strResult & "<TD ALIGN=RIGHT>" & _
FormatDateTime(objField.Value, 2)
Case adDBTimeStamp
strResult = strResult & "<TD ALIGN=RIGHT>" & _
FormatDateTime(objField.Value, 2)
Case adVarNumeric
strResult = strResult & "<TD ALIGN=RIGHT>" & _
FormatNumber(objField.Value)
Case adInteger
strResult = strResult & "<TD ALIGN=RIGHT>" & _
FormatNumber(objField.Value, 0)
Case adBigInt
strResult = strResult & "<TD ALIGN=RIGHT>" & _
FormatNumber(objField.Value, 0)
Case adTinyInt
strResult = strResult & "<TD ALIGN=RIGHT>" & _
FormatNumber(objField.Value, 0)
Case adSmallInt
strResult = strResult & "<TD ALIGN=RIGHT>" & _
FormatNumber(objField.Value, 0)
'... You can add special formatting for more datatypes
' here if you like...
Case Else
strResult = strResult & "<TD>" & objField.Value
End Select
Example 6-23. The Generic Helper Function CreateList Creates a List of Records, Allowing the
User to Select One or More (continued)
208 Chapter 6: Database Reuse
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
CreateList provides a tabular list of all of the records in a particular
database table. The “Delete an Existing Record” administration page
can be difficult to navigate (not to mention slow to load) if there are
hundreds or thousands of records in the database table. The most
obvious solution to this problem is to page the returned records.
This possibility is discussed later in this chapter, in the “Limitations
and Possible Enhancements” section.
The “Edit an Existing Record” administration page
The “Edit an Existing Record” administration page is a bit like a child to the “Insert
a New Record” and “Delete an Existing Record” administration pages. The “Edit an
Existing Record” administration page has similar functionality to the “Delete an
Existing Record” administration page in that before a particular record can be
edited, it must be selected from a list of all possible records. The “Edit an Existing
Record” administration page is also similar to the “Insert a New Record” administration
page in that once a record is selected for editing, the form used to allow
End If
strResult = strResult & "</TD>" & vbCrLf
End If
End If
Next
strResult = strResult & "</TR>" & vbCrLf
objRS.MoveNext
Loop
strResult = strResult & "</TABLE>" & vbCrLf
objRS.Close
Set objRS = Nothing
strResult = strResult & "<P><INPUT TYPE=SUBMIT VALUE="""
If bolDelete then
strResult = strResult & "Delete Selected Records"">"
Else
strResult = strResult & "Edit Selected Record"">"
End If
strResult = strResult & vbCrLf & "</FORM>" & vbCrLf & vbCrLf
CreateList = strResult
End Function
'*****************************************************
End Class
%>
Example 6-23. The Generic Helper Function CreateList Creates a List of Records, Allowing the
User to Select One or More (continued)
Creating Reusable Administration Pages 209
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
the user to change the values of a particular record is identical to the form used
for creating a new record.
The “Edit an Existing Record” administration page form’s similarity to the “Delete
an Existing Record” administration page form can be seen in a screenshot. When a
user needs to edit a record, she will first be shown a listing of available records to
choose from. Figure 6-11 shows this form, from which the user can select a particular
record to edit from the Catalog table.
The screenshot in Figure 6-11 looks very similar to the screenshot in Figure 6-10.
The only visual difference between the user selecting a record for editing, or
selecting one or more records for deleting, is that when selecting a record for editing,
each record has a radio button next to it. When selecting one or more records
to delete, each listed record has a checkbox next to it. These form elements were
chosen due to their nature; a radio button limits the user to selecting one option
out of several, whereas checkboxes permit several options to be chosen from a list.
As with both the “Insert a New Record” and “Delete an Existing Record” administration
pages, from the end developer’s standpoint, creating the “Edit an Existing
Record” administration page requires just a few lines of code. Example 6-24 contains
the code in SelectRecord.asp that generated the screenshot in Figure 6-11.
Figure 6-11. When editing an existing record, users must first choose a particular record to
edit
Example 6-24. Creating the “Edit an Existing Record” Administration Page Form Is Trivial for
the End Developer
<% @LANGUAGE = "VBSCRIPT" %>
<% Option Explicit %>
210 Chapter 6: Database Reuse
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
In previous code examples, the SQLOLEDB provider has been used
as the database provider for the Connection object. In Example 6-24,
however, the Jet 4.0 provider is used. This is just to illustrate how to
use different database providers for the administration page generation
scripts.
Note that SelectRecord.asp, shown in Example 6-24, displays the first of two forms
for the “Edit an Existing Record” administration page. A server-side include is used
to import the contents of the file /CODEREUSE/SchemaInfo.Class.asp, which contains
the AdminPageGenerator class definition. The IdentityColumnHint and ForeignKeyHint
methods are used to provide hints about the structure of the Product
table.
The code for UpdateRecordList, which is shown in Example 6-25, is strikingly similar
to the code for DeleteRecordForm—both methods call the CreateList helper
function, which was shown in Example 6-23. In fact, the only difference between
UpdateRecordList and DeleteRecordForm is that UpdateRecordList calls CreateList
with the bolDelete parameter set to False.

<%
Dim strTableName
strTableName = "Product"
Dim objAdmin
Set objAdmin = new AdminPageGenerator
objAdmin.IdentityColumnHint = "ProductID"
objAdmin.ForeignKeyHint "ProductTypeid", "Product Type ~Name~"
%>
<HTML><BODY>
<h1>Select a <%=UCase(strTableName)%> to Edit</h1><BR>
<%
Dim strConnString
strConnString = "Provider=Microsoft.Jet.OLEDB.4.0;" & _
"Data Source=C:\InetPub\Products.mdb"
'Display the form
Response.Write objAdmin.UpdateRecordList(strTableName, strConnString, "", "")
Set objAdmin = Nothing 'Clean up!
%>
</BODY></HTML>
Example 6-24. Creating the “Edit an Existing Record” Administration Page Form Is Trivial for
the End Developer (continued)
Creating Reusable Administration Pages 211
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
The remaining method, UpdateRecordForm, creates a form similar to that created
by CreateRecordForm. Unlike CreateRecordForm, UpdateRecordForm requires a
fifth parameter, strWhereClause. StrWhereClause must be a valid SQL WHERE
clause (less the WHERE keyword) returning exactly one record from the database
table. Figure 6-12 contains a screenshot of a particular record being edited for the
Catalog table. Note the similarities between the output generated by the
UpdateRecordForm method in Figure 6-8 and the CreateRecordForm method in
Figure 6-5.
Example 6-26 contains the code for EditSelectedRecord.asp, which displays the
“Edit an Existing Record” form shown in Figure 6-12. Note the second parameter
Example 6-25. The UpdateRecordList Method Creates a Form from Which the User Can Select a
Particular Record to Edit


<%
Class AdminPageGenerator
' ... Properties, event handlers, and other methods omitted for
' brevity. The properties and event handlers are shown in
' Example 6-12 ...
'********************** METHODS **********************
Public Function UpdateRecordList(strTableName, strConnString, _
strUser, strPass)
'Present a form that lists the database columns with a
'radio button next to each column. The user can then select
'what column he'd like to edit.
'Assign the private member variables the values
strConnectionString = strConnString
strUserName = strUser
strPassword = strPass
'Populate the objColumns and objForeignKeys objects
'Create a connection to the database
Dim objConn
Set objConn = Server.CreateObject("ADODB.Connection")
objConn.Open strConnectionString, strUserName, strPassword
objColumns.PopulateColumns strTableName, objConn, ""
objForeignKeys.PopulateForeignKeys strTableName, objConn
UpdateRecordList = CreateList(False, objConn, strTableName)
objConn.Close
Set objConn = Nothing
End Function
'*****************************************************
End Class
%>
212 Chapter 6: Database Reuse
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
in the UpdateRecordForm method call. This second parameter is the str-
WhereClause parameter, which must uniquely select a single record from the
table specified by strTableName.
Figure 6-12. When editing an existing record, the form fields have the current column values
already entered
Example 6-26. Creating the “Edit an Existing Record” Form Is Similar to Creating Other
Administration Page Forms
<% @LANGUAGE = "VBSCRIPT" %>
<% Option Explicit %>

<%
Dim strTableName
strTableName = "Product"
Dim objAdmin
Set objAdmin = new AdminPageGenerator
%>
<HTML><BODY>
<h1>Edit a <%=UCase(strTableName)%></h1><BR>
<%
Creating Reusable Administration Pages 213
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
In its current form, EditSelectedRecord.asp isn’t very useful at all, since there are
several hardcoded values. For one, the table name is hardcoded, displaying only a
form for the Product table administration page. Secondly, the strWhereClause
in the UpdateRecordForm method is also hardcoded (ProductID = 1). Therefore,
EditSelectedRecord.asp will only display an update form for a particular row
in a particular table. Later in this chapter, we’ll examine how to make a more
dynamic ASP page to generate the “Edit an Existing Record” administration page.
Not surprisingly, the code for UpdateRecordForm (which can be found in
Example 6-27) is similar to CreateRecordForm. Both methods make use of the
CreateForm helper function, differing in the fact that UpdateRecordForm passes in
a value of False to bolNewRecord and CreateRecordForm passes in a value of
True. (The code for CreateForm was presented in Example 6-21.)
Dim strConnString
strConnString = "Provider=Microsoft.Jet.OLEDB.4.0;" & _
"Data Source=C:\InetPub\Products.mdb"
'Display the form
Response.Write objAdmin.UpdateRecordForm(strTableName, "ProductID=1", _
strConnString , "", "")
Set objAdmin = Nothing 'Clean Up!
%>
</BODY></HTML>
Example 6-27. The UpdateRecordForm Calls the Helper Function CreateForm to Generate a
Form for Editing a Particular Record


<%
Class AdminPageGenerator
' ... Properties, event handlers, and other methods omitted for
' brevity. The properties and event handlers are shown in
' Example 6-12 ...
'********************** METHODS **********************
Public Function UpdateRecordForm(strTableName, strWhereClause, strConnString, _
strUser, strPass)
'Assign the private member variables the values
strConnectionString = strConnString
strUserName = strUser
strPassword = strPass
'Create a connection to the database
Dim objConn
Set objConn = Server.CreateObject("ADODB.Connection")
objConn.Open strConnectionString, strUserName, strPassword
'Populate the objColumns and objForeignKeys objects
Example 6-26. Creating the “Edit an Existing Record” Form Is Similar to Creating Other
Administration Page Forms (continued)
214 Chapter 6: Database Reuse
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Inserting, Updating, and Deleting Database Records
So far, we’ve examined how to gather database table information and display
administration page forms. We have paid little attention to how a record will be
added, updated, or deleted from the database. In this section, we will examine
how to gather the input from the administration page forms and apply the correct
changes to the correct database table.
A single ASP page will be responsible for making all database changes specified
by the various administration page forms. As discussed previously, the file
/CODEREUSE/SchemaInfo.Class.asp, which contains the class definition for
AdminPageGenerator, also has a constant named strAction declared at the
beginning of the file. This constant should be set to the name of the single ASP
page that performs any database alterations. As mentioned previously, for these
examples strAction is set to /CODEREUSE/AdminPageAction.asp using the following
command:
Const strAction ="/CODEREUSE/AdminPageAction.asp"
AdminPageAction.asp utilizes—you guessed it—a class to assist in making any
database changes. This class, ModifyDatabase, contains four methods to assist
with committing the changes made by a user in one of the administration forms.
These methods are InsertRecord, DeleteRecords, UpdateRecord, and UpdateSelect.
Keep in mind that ModifyDatabase is needed only when a user submits an
administration form created by AdminPageGenerator.
Before we look at the methods, however, let’s take a moment to examine the
properties and event handlers of the ModifyDatabase class.
As with the other classes discussed in this chapter, the source code
for the ModifyDatabase class should be placed in its own file. I
chose to place ModifyDatabase in the file /CODEREUSE/
DatabaseModification.Class.asp.
objColumns.PopulateColumns strTableName, objConn, strWhereClause
objForeignKeys.PopulateForeignKeys strTableName, objConn
objConn.Close
Set objConn = Nothing 'Clean up!
UpdateRecordForm = CreateForm(False, strTableName)
End Function
'*****************************************************
End Class
%>
Example 6-27. The UpdateRecordForm Calls the Helper Function CreateForm to Generate a
Form for Editing a Particular Record (continued)
Creating Reusable Administration Pages 215
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
The ModifyDatabase class’s properties and event handlers
The ModifyDatabase class has only a couple of properties. Since the class’s
duties focus on inserting, deleting, or updating database records, it comes as no
surprise that three of the four properties provide database connection information.
These three database connection properties—ConnectionString, UserName,
and Password—are read/write properties and are used when connecting to the
database.
The fourth property, Action, is read-only, and specifies the database action to be
taken: inserting, updating, deleting, or selecting a record to update. Since there
are four administration forms, and all administration forms, when submitted, lead
to /CODEREUSE/AdminPageAction.asp (which employs the ModifyDatabase
class), there needs to be four types of possible Actions. Each of the four administration
forms creates a HIDDEN form field named Action. It is through this
HIDDEN form field that ModifyDatabase determines what type of action needs to
be carried out.
The Initialize event handler of the ModifyDatabase class sets the Action property
and ensures that the class is used only from /CODEREUSE/AdminPageAction.asp
(or whatever URL is specified by the strAction constant in /CODEREUSE/
SchemaInfo.Class.asp).
If any errors occur along the way, such as an invalid Action property being passed
in from the administration page form, an error is raised. Recall that error handling
was discussed in detail in Chapter 3, Exception Handling. The properties and
event-handling code for the ModifyDatabase class are shown in Example 6-28.
Example 6-28. The Properties and Initialize Event Handler of the ModifyDatabase Class
Determine What Database Action Will Be Performed

<%
Class ModifyDatabase
'************** MEMBER VARIABLES *********************
Private strConnectionString
Private strUserName
Private strPassword
Private strDBAction
'*****************************************************
'******************* EVENT HANDLERS ******************
Private Sub Class_Initialize()
If UCase(Request.ServerVariables("SCRIPT_NAME")) <> _
Ucase(strAction) then
Err.Raise vbObjectError + 1234, "ModifyDatabase Class", _
"ModifyDatabase can only be instantiated " & _
216 Chapter 6: Database Reuse
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
"in /CODEREUSE/AdminPageAction.asp"
End If
'Read in our action: UPDATE, INSERT, DELETE
strDBAction = UCase(Request("Action"))
If strDBAction <> "INSERT" And strDBAction <> "UPDATE" And _
strDBAction <> "DELETE" And strDBAction <> "UPDATESELECT" then
'Invalid Action Parameter
Err.Raise vbObjectError + 1236, "ModifyDatabase Class", _
"Invalid Action - must be set to INSERT, UPDATE, or DELETE"
End If
End Sub
'*****************************************************
'****************** GET PROPERTIES *******************
Public Property Get Action() 'A read-only property
Action = strDBAction
End Property
Public Property Get ConnectionString()
ConnectionString = strConnectionString
End Property
Public Property Get UserName()
UserName = strUserName
End Property
Public Property Get Password()
Password = strPassword
End Property
'*****************************************************
'****************** LET PROPERTIES *******************
Public Property Let ConnectionString(strConn)
strConnectionString = strConn
End Property
Public Property Let UserName(strUser)
strUserName = strUser
End Property
Public Property Let Password(strPass)
strPassword = strPass
End Property
'*****************************************************
End Class
%>
Example 6-28. The Properties and Initialize Event Handler of the ModifyDatabase Class
Determine What Database Action Will Be Performed (continued)
Creating Reusable Administration Pages 217
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
As shown in Example 6-28, the ModifyDatabase class must include
/CODEREUSE/SchemaInfo.Class.asp. This is due to the fact that the
UpdateSelect method of ModifyDatabase uses an instance of the
AdminPageGenerator class to create the second form of the “Edit
an Existing Record” administration page. The file /CODEREUSE/
SchemaInfo.Class.asp must be included since it contains the class
definition for AdminPageGenerator.
ModifyDatabase’s methods
The ModifyDatabase class contains four methods, one method for each of the
four possible Action types. These methods have the following definitions:
Public Sub InsertRecord() 'Called from Insert a New Record Admin. Page Form
Public Sub UpdateRecord() 'Called when the user has finished editing a particular
'record in the Edit an Existing Record Admin. Page Form
Public Sub UpdateSelect() 'Called when the user selects a record to edit in
'the Edit an Existing Record Admin. Page Form
Public Sub DeleteRecords() 'Called from Delete an Existing Record Admin. Page
/CODEREUSE/AdminPageAction.asp, the form-processing script that is called when
a user submits an administration page form, calls one of these four methods,
depending on the Action property. If the InsertRecord, UpdateRecord, or
DeleteRecords method is called, /CODEREUSE/AdminPageAction.asp then redirects
the user to the page specified by the end developer in the administration
page generation script. (Recall that the AdminPageGenerator class has a
Redirect property; this property determines where the user is sent after a record
is inserted, updated, or deleted from the database in /CODEREUSE/
AdminPageAction.asp.)
Since ModifyDatabase does the actual work involved in handling a submitted
administration page form, the code for /CODEREUSE/AdminPageAction.asp is fairly
simple. Example 6-29 contains the complete code for AdminPageAction.asp.
Example 6-29. AdminPageAction.asp Is Visited When an Administration Page Form Is
Submitted
<% @LANGUAGE="VBSCRIPT" %>
<% Option Explicit %>

<%
Dim objDB
Set objDB = New ModifyDatabase
objDB.ConnectionString = Request("ConnectionString")
objDB.UserName = Request("UserName")
218 Chapter 6: Database Reuse
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Note that if a database record is inserted, updated, or deleted, the
user is whisked away via a Response.Redirect rather than a Server.
Transfer. I chose to use this method because I didn’t want the user’s
browser to still show that they were at /CODEREUSE/Admin-
PageAction.asp.
Since the ModifyDatabase class is only used from /CODEREUSE/Admin-
PageAction.asp, which is only visited when an administration page form is submitted,
the ModifyDatabase methods have access to the form field values passed
from the administration page form. All administration page forms pass six HIDDEN
form fields to /CODEREUSE/AdminPageAction.asp. These form fields, along with a
description, are listed in Table 6-7.
objDB.Password = Request("Password")
'Decide what ModifyDatabase method to call (based upon the Action property)
Select Case objDB.Action
Case "INSERT":
objDB.InsertRecord 'Insert a new database record
Case "UPDATE":
objDB.UpdateRecord 'Update an existing database record
Case "DELETE":
objDB.DeleteRecords 'Delete an existing database record
Case "UPDATESELECT":
objDB.UpdateSelect 'Display the UpdateRecord form
End Select
'If we actually inserted, updated, or deleted a record, redirect the user!
If objDB.Action <> "UPDATESELECT" then
Response.Redirect Request("Redirect")
End If
Set objDB = Nothing 'Clean up...
%>
Table 6-7. Each Administration Page Form Passes Six HIDDEN Form Fields to
AdminPageAction.asp
HIDDEN Form
Field Name Description
Action The Action property determines what database action is to take
place when the form is submitted. Each administration page form
has a unique value for Action. The legal values for Action are
INSERT, UPDATESELECT, UPDATE, and DELETE, and are used for
the “Insert a New Record,” the selecting a record to edit, the “Edit
an Existing Record,” and the “Delete an Existing Record” administration
pages, respectively.
Example 6-29. AdminPageAction.asp Is Visited When an Administration Page Form Is
Submitted (continued)
Creating Reusable Administration Pages 219
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
These four administration page forms also have their own unique sets of form
fields that are passed along to AdminPageAction.asp once the administration page
form is submitted. Table 6-8 lists the distinctive form fields passed from each of
the four administration page forms.
TableName Specifies the name of the database table the administration page
is for.
Redirect Specifies the URL to redirect the user to once an actual insert,
update, or delete is committed to the database in
AdminPageAction.asp.
ConnectionString Specifies the connection string for the database connection.
UserName Specifies the username for the database connection.
Password Specifies the password for the database connection.
Table 6-8. Each Administration Page Form Also Passes Its Own, Unique Set of Form Fields to
AdminPageAction.asp
Administration
Page Form Form Field Name Description
Insert a New
Record
col_ColumnName Each form field that represents a column in
the database table has the column as the
form field name, prefixed with col_.
Edit an Existing
Record—Selecting
a Particular
Record to Edit
FKName For each foreign key hint supplied by the
end developer, there exists a HIDDEN form
field named FKName. The form field’s value
is set to the column name that the foreign
key hint applies to.
FKValue Like FKName, for each foreign key hint supplied
by the end developer, there exists a
HIDDEN form field named FKValue. The
form field’s value is set to the value of the
foreign key hint.
ID Each listed record contains a radio button
named ID. The value of each radio button is
a SQL WHERE clause (less the WHERE keyword
itself) that uniquely identifies the row.
Edit an Existing
Record—
Update the
Information for
a
Particular,
Selected Record
PrimaryKeyColumns When updating a particular record in the
database, it is important to be able to select
the correct record for editing. This form
field contains a valid SQL WHERE clause
(less the WHERE keyword itself) that
uniquely identifies the row being edited.
Table 6-7. Each Administration Page Form Passes Six HIDDEN Form Fields to
AdminPageAction.asp (continued)
HIDDEN Form
Field Name Description
220 Chapter 6: Database Reuse
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Inserting a record into the database
A single call to the ModifyDatabase class’s InsertRecord method adds a new
record with the values chosen by the user into the database table specified by the
TableName form field. InsertRecord, like the other three methods of
ModifyDatabase, takes no parameters. InsertRecord, shown in Example 6-30,
starts off by ensuring that the all-important TableName form field value was
passed in correctly. If TableName was not passed in, then InsertRecord cannot
possibly continue, since it has no idea which database table needs a new record
added; therefore, in such a case, InsertRecord raises an error.
All four of the ModifyDatabase methods handle errors rather
ungracefully. If a needed form field is not passed in, the behavior of
these methods is chaotic at best, perhaps leading to database inconsistencies.
Therefore, if any of these methods becomes the slightest
bit confused, it throws its arms up into the air, raising an error.
Next, a connection to the database is established using the end developer-set database
connection properties. A Recordset object is explicitly created, and a snapshot
of the database table specified by TableName is grabbed. The AddNew
method is then used to add a new record to the table.
A For Each ... Next loop then steps through the columns in the database table. If
a value was entered for a particular database column (that is, Request("col_
ColumnName") returns a value), then the column’s value is set to the value specified
by Request("col_ColumnName"). Once all the records have been iterated,
the Update method is used to commit the changes to the database.
col_ColumnName Each form field that represents a column in
the database table has the column as the
form field name, prefixed with col_.
Delete an Existing
Record
ID Each listed record contains a radio button
named ID. The value of each radio button is
a SQL WHERE clause (less the WHERE keyword
itself) that uniquely identifies the row.
Example 6-30. InsertRecord Adds a New Record to the Table Specified by the TableName Form
Field Value

<%
Class ModifyDatabase
' ... The properties and event handlers have been omitted for brevity
Table 6-8. Each Administration Page Form Also Passes Its Own, Unique Set of Form Fields to
AdminPageAction.asp (continued)
Administration
Page Form Form Field Name Description
Creating Reusable Administration Pages 221
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
' here, but are listed in Example 6-28 ... The three other
' methods have been left out as well.
'********************** METHODS **********************
Public Sub InsertRecord()
'This method adds a new record to the database
'Read in the tablename and determine what request collection we're using
Dim strTableName
strTableName = Request("TableName")
'If tablename doesn't exist, that's really bad. Throw an error
If Len(strTableName) = 0 then
Err.Raise vbObjectError + 1235, "ModifyDatabase Class", _
"TableName was not correctly passed in."
End If
'Open up a connection to the database
Dim objConn
Set objConn = Server.CreateObject("ADODB.Connection")
objConn.Open strConnectionString, strUserName, strPassword
'Retrieve a picture of the database table we're inserting into
Dim objRS
Set objRS = Server.CreateObject("ADODB.Recordset")
objRS.MaxRecords = 1
objRS.Open "[" & strTableName & "]", objConn, _
adOpenForwardOnly, adLockOptimistic, adCmdTable
objRS.AddNew 'Add a new record
Dim objField, strFormName
For Each objField in objRS.Fields
'If there exists a column in the database table that matches
'the name of a column passed in from the form with a col_ prefix,
'set the database column's value to the form element's value
If Len(Request("col_" & objField.Name)) > 0 then
objField.Value = Request("col_" & objField.Name)
End If
Next
objRS.Update 'commit the changes
'Clean up...
objRS.Close
Set objRS = Nothing
objConn.Close
Set objConn = Nothing
End Sub
'*****************************************************
End Class
%>
Example 6-30. InsertRecord Adds a New Record to the Table Specified by the TableName Form
Field Value (continued)
222 Chapter 6: Database Reuse
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Selecting a particular record to edit
To edit an existing record, the user must first select which record, exactly, he’d
like to edit. The UpdateRecordList method of the AdminPageGenerator class created
the appropriate administration page form, allowing the user to pick a particular
record to edit. Once the record has been chosen and the form submitted, the
user is sent to AdminPageAction.asp, just like he is with any other administration
page form.
Once the user has selected a record to edit, he needs to be presented with a form
similar to the one in the “Insert a New Record” administration page. However, the
“Edit an Existing Record” form the user needs to be presented with should have
the selected record’s column values already entered into the various form fields.
Fortunately, a class and method already exist to handle this for us! The
AdminPageGenerator’s UpdateRecordForm method contains such functionality.
Therefore, in the “Edit an Existing Record” administration page, when the user
selects a record to edit, we need to display yet another administration form. For
this reason, the file that contains the AdminPageGenerator class needs to be
imported into the file that contains the ModifyDatabase class via a server-side
include.
The UpdateSelect method of the ModifyDatabase class is given the responsibility
for creating an appropriate administration page form. To do this, it must know
what foreign key hints are needed for the database table specified by TableName.
This information is available, of course, since we passed a number of FKName and
FKValue HIDDEN form fields from the first “Edit an Existing Record” administration
page form (see Example 6-24).
The UpdateSelect method generates the appropriate “Edit an Existing Record”
administration page form by creating an AdminPageGenerator class instance. The
needed foreign key hints are set, as well as the Redirect property. Finally, the
administration page form is generated with a call to the UpdateRecordForm
method. Example 6-31 contains the source code for UpdateSelect.
Example 6-31. UpdateSelect Is the Only ModifyDatabase Method that Doesn’t Modify the
Database; Rather, It Generates the Correct “Edit an Existing Record” Administration Page
Form

<%
Class ModifyDatabase
' ... The properties and event handlers have been omitted for brevity
' here, but are listed in Example 6-28 ... The three other
' methods have been left out as well.
'********************** METHODS **********************
Public Sub UpdateSelect()
Creating Reusable Administration Pages 223
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
'This function needs to create the Update form
Dim objAdmin
Set objAdmin = new AdminPageGenerator 'Our instance of AdminPageGenerator
'Read in the tablename and determine what request collection we're using
Dim strTableName, objRequestCol, strIDs
strTableName = Request("TableName")
'If tablename doesn't exist, that's really bad. Throw an error
If Len(strTableName) = 0 then
Err.Raise vbObjectError + 1235, "ModifyDatabase Class", _
"TableName was not correctly passed in."
End If
'Make sure the ID form field exists
strIDs = Request("ID")
If Len(strIDs) = 0 then
'The user did not select a item to edit
Response.Write "You did not select an item to edit."
Response.End
End If
'We need to handle foreign key hints
Dim aHintNames, aHintValues
aHintNames = split(Request("FKName"), ",")
aHintValues = split(Request("FKValue"), ",")
Dim iLoop
For iLoop = LBound(aHintNames) to UBound(aHintNames)
objAdmin.ForeignKeyHint Trim(aHintNames(iLoop)), _
Trim(aHintValues(iLoop))
Next
'Set the redirect URL if needed
If Len(Request("Redirect")) > 0 then
objAdmin.Redirect = Left(Request("Redirect"), _
InStr(1, Request("Redirect"), "?") - 1)
End If
'Display the "Edit an Existing Record" administration page form
Response.Write "<HTML><BODY><H1>Edit an Existing " & _
UCase(strTableName) & "</H1><BR>" & vbCrLf
Response.Write objAdmin.UpdateRecordForm(strTableName, strIDs, _
strConnectionString, _
strUserName, strPassword)
Response.Write "</BODY></HTML>"
End Sub
'*****************************************************
End Class
%>
Example 6-31. UpdateSelect Is the Only ModifyDatabase Method that Doesn’t Modify the
Database; Rather, It Generates the Correct “Edit an Existing Record” Administration Page Form
(continued)
224 Chapter 6: Database Reuse
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Editing an existing database record
Editing the columns of an existing database record is handled by the
UpdateRecord method of the ModifyDatabase class. The complete code for the
UpdateRecord method can be found in Example 6-32. The code for UpdateRecord
is very similar to that of the InsertRecord method shown earlier in Example 6-30.
The only major difference is that UpdateRecord doesn’t use the ADO Recordset
object’s AddNew method to create a new record in the table; rather, it uses a
SELECT statement to grab a specific record from the table.
To obtain the correct, specific record from the database table, UpdateRecord relies
on the PrimaryKeyColumns HIDDEN form variable passed in through the “Edit an
Existing Record” administration page form. (See Table 6-8 for a description of the
PrimaryKeyColumns form field, and refer to Example 6-21 to see the code for the
CreateForm method of the AdminPageGenerator class, which creates the “Edit an
Existing Record” administration page form.)
Specifying the Foreign Key Hints
In the UpdateSelect method, the foreign key hints used by the end developer
for the particular table need to be set. The UpdateRecordList method of the
AdminPageGenerator class, which displays the administration page form that
allows the user to select a particular record to edit, provides two HIDDEN form
fields for each foreign key hint (see Table 6-8 and Example 6-23).
When multiple form fields have the same name, say FormName, the Request
object provides a comma-delimited list of the form field values when the form
field FormName is requested. In UpdateSelect, the split function is used to convert
this comma-delimited list into an array. The array is then iterated, and at
each iteration, the ForeignKeyHint method of the AdminPageGenerator class
is called.
This approach ensures that each foreign key hint specified by the end developer
in the administration page form from which the user selected a record to
edit is also specified for the administration page form generated by
UpdateRecordForm method.
To learn more about how to use split to iterate through a series of form fields
with the same name, be sure to read the following articles:
• “Parsing with join and split,” found at http://www.4guysfromrolla.com/
webtech/050999-1.shtml
• “Passing Arrays from One ASP Page to Another,” found at http://www.
4guysfromrolla.com/webtech/101999-1.shtml
Creating Reusable Administration Pages 225
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Note that three different situations can cause the UpdateRecord method to raise an
error:
• TableName was not correctly passed in. If TableName is not supplied,
UpdateRecord has no idea what database table to update.
• PrimaryKeyColumns was not correctly passed in. Since this form field
uniquely identifies a record in the table specified by TableName, if this value
is not supplied, UpdateRecord cannot continue, since it does not know which
record to edit from TableName.
• The query to return a single database record returns no records at all. If this
happens, a record to update does not exist, which is bad. This rare situation
can happen if, between the time a record was selected for editing and the
time the user submitted his changes to the record, another user deleted the
record in question or altered its primary key value(s). (This possibility is discussed
in more detail in “Limitations and Possible Enhancements.”)
Example 6-32. UpdateRecord Applies Changes to an Already Existing Record
in the Database Table

<%
Class ModifyDatabase
' ... The properties and event handlers have been omitted for brevity
' here, but are listed in Example 6-28 ... The three other
' methods have been left out as well.
'********************** METHODS **********************
Public Sub UpdateRecord()
'This method updates a record in the database
'Read in the tablename and determine what request collection we're using
Dim strTableName
strTableName = Request("TableName")
'If tablename doesn't exist, that's really bad. Throw an error
If Len(strTableName) = 0 then
Err.Raise vbObjectError + 1235, "ModifyDatabase Class", _
"TableName was not correctly passed in."
End If
'Read in the PrimaryKeyColumns Value
Dim strPKColumns
strPKColumns = Request("PrimaryKeyColumns")
'If strPKColumns doesn't exist, that's really bad. Throw an error
If Len(strPKColumns) = 0 then
Err.Raise vbObjectError + 1237, "ModifyDatabase Class", _
"PrimaryKeyColumns needed for an UPDATE."
End If
'Open up a connection to the database
226 Chapter 6: Database Reuse
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Deleting an existing database record
The DeleteRecords method of the ModifyDatabase class can be used to delete
one or more records from a database table. Recall that the “Delete an Existing
Record” administration page form listed each record in a particular database table,
providing a checkbox next to each record. By using checkboxes, as opposed to
radio buttons, the user can check more than one record, thereby deleting several
records in one fell swoop.
Dim objConn
Set objConn = Server.CreateObject("ADODB.Connection")
objConn.Open strConnectionString, strUserName, strPassword
'Retrieve a picture of the database table we're inserting into
Dim objRS, strSQL
strSQL = "SELECT * FROM [" & strTableName & "] WHERE " & _
strPKColumns
Set objRS = Server.CreateObject("ADODB.Recordset")
objRS.MaxRecords = 1
objRS.Open strSQL, objConn, adOpenForwardOnly, adLockOptimistic
'Make sure we are on a record
If objRS.EOF then
'this is bad...
Err.Raise vbObjectError + 1380, "ModifyDatabase Class", _
"Record to update not found: Failed on - " & strSQL
End If
Dim objField, strFormName
For Each objField in objRS.Fields
'If there exists a column in the database table that matches
'the name of a column passed in from the form with a col_ prefix,
'set the database column's value to the form element's value
If Len(Request("col_" & objField.Name)) > 0 then
objField.Value = Request("col_" & objField.Name)
End If
Next
objRS.Update 'commit the changes
'Clean up...
objRS.Close
Set objRS = Nothing
objConn.Close
Set objConn = Nothing
End Sub
'*****************************************************
End Class
%>
Example 6-32. UpdateRecord Applies Changes to an Already Existing Record
in the Database Table (continued)
Creating Reusable Administration Pages 227
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
As described in Table 6-8, each checkbox in the “Delete an Existing Record”
administration page form is given the same name, ID, which contains a statement
to uniquely identify the row. As with the foreign key hints in the UpdateSelect
method, the value of the ID form fields will be available as a comma-delimited list.
Example 6-33, which contains the code for DeleteRecords, demonstrates how this
comma-delimited list can be used to delete one or more records with a single SQL
statement.
Example 6-33. The DeleteRecords Method Can Delete One or More Records from a Database
Table with a Single SQL Statement

<%
Class ModifyDatabase
' ... The properties and event handlers have been omitted for brevity
' here, but are listed in Example 6-28 ... The three other
' methods have been left out as well.
'********************** METHODS **********************
Public Sub DeleteRecords()
'This method deletes one or more records from the database
'Read in the tablename and determine what request collection we're using
Dim strTableName, objRequestCol, strIDs
strTableName = Request("TableName")
'If tablename doesn't exist, that's really bad. Throw an error
If Len(strTableName) = 0 then
Err.Raise vbObjectError + 1235, "ModifyDatabase Class", _
"TableName was not correctly passed in."
End If
'Make sure the ID form field exists
strIDs = Request("ID")
If Len(strIDs) = 0 then
'The user did not select a item to delete
Response.Write "You did not select an item to delete."
Response.End
End If
'Open up a connection to the database
Dim objConn
Set objConn = Server.CreateObject("ADODB.Connection")
objConn.Open strConnectionString, strUserName, strPassword
Dim strSQL
strSQL = "DELETE FROM [" & strTableName & "] WHERE "
strSQL = strSQL & Replace(strIDs, ",", " OR ")
objConn.Execute strSQL 'Execute the SQL statement
'Clean up...
objConn.Close
228 Chapter 6: Database Reuse
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Tying the Administration Pages Together
Now that we have all of the classes needed to gather table and column information,
display administration page forms, and store changes to the database, we’re
nearly done with the generic administration page creation application. Despite all
of the hard work put into the various administration page classes thus far, and
despite all of the energy put toward making these classes as reusable as possible,
the generic administration page generation application still is not nearly as reusable
as it could be.
For example, in our database example with the Product, Catalog, ProductType,
and CatalogProduct tables, to create the needed administration pages, the end
developer would be required to create a total of 12 ASP pages. These pages would
Set objConn = Nothing
End Sub
'*****************************************************
%>
Deleting Multiple Records
The DeleteRecords method needs to be able to delete more than one record
from a database table. While a single SQL statement for each record deletion
could be used, such an approach would result in poor performance when multiple
records should be deleted.
There is more than one way to delete multiple records in one SQL statement.
Example 6-33 illustrates one such way, by simply creating a lengthy SQL statement
in the form of:
DELETE FROM TableName
WHERE ID = Value1 OR
ID = Value2 OR . . .
ID = ValueN
Another way to accomplish this same feat using a more compact and more elegant
SQL statement is to use SQL’s set notation. N records can be deleted from
a table with a SQL statement such as:
DELETE FROM TableName
WHERE ID IN (Value1, Value2, . . . ValueN)
SQL set notation is very powerful and very useful, but most importantly, it’s
very cool. To learn more about SQL set notation, check out the following article,
“Using SQL Set Notation to do Batch Deletes,” at http://www.
4guysfromrolla.com/webtech/092899-1.shtml.
Example 6-33. The DeleteRecords Method Can Delete One or More Records from a Database
Table with a Single SQL Statement (continued)
Creating Reusable Administration Pages 229
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
consist of an Insert, Edit, and Delete page for each database table. Each of these
pages would use the AdminPageGenerator class to create the correct administration
page form.
It would be nice to have a single, generic ASP page that served as an administration
page menu, listing each database table with options for inserting, editing, or
deleting records. Furthermore, it would be nice if, as new database tables were
added to the database, the end developer could visit this single page to add the
Insert, Edit, and Delete administration pages for the new tables.
When the user selects an administration page for a particular database table, the
generic menu page should redirect the user to a single, generic ASP page responsible
for displaying the correct administration page form. By having a single, generic
menu page send the user to a generic administration page, our administration
page application will have become truly reusable. Since each administration page
form, when submitted, sends the user to a single ASP page responsible for updating
the database with the requested changes, the entire generic administration
page generation application has been boiled down to three ASP pages! Figure 6-13
illustrates the movement experienced throughout the three ASP pages as the user
visits various administration page forms.
Before we examine the code for the generic menu page, you can see a screenshot
of the menu in Figure 6-14.
To assist in creating this administration page menu, we’ll need a couple of new
classes. The first class, MenuItem, stores the information needed to create a particular
table’s administration pages. Information such as the table name and the various
“hints” are included in this class. The second class, AdminPageMenu, is used to
generate the actual administration page menu, and contains information pertinent
to all the various tables’ administration pages, such as connection information to
the database.
MenuItem’s methods, properties, and event handlers
Each database table has information unique to itself. Since the
AdminPageGenerator class cannot determine these table-specific intricacies on its
own, it relies on “hints” from the end developer. The MenuItem class serves as a
container to store various hints and other table-specific information. The properties
for the MenuItem include: TableName, Redirect, IdentityColumnHint, Form-
Hints, FKHints, FormNameHints, and FormDescriptionHints. The last four “hint”
properties are Dictionary object instances, storing zero to many end developersupplied
hints.
The TableName property simply contains the name of the database table. Identity-
ColumnHint contains an end developer-supplied hint as to which database column
is an AutoNumber column. Recall that if a database table contains no records,
230 Chapter 6: Database Reuse
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
it cannot determine which column, if any, is an AutoNumber column. Therefore,
the end developer should supply an IdentityColumnHint if the table contains an
AutoNumber column. The Redirect property determines where the user is sent
after /CODEREUSE/AdminPageAction.asp inserts, updates, or deletes a record from
the database. The diagram in Figure 6-13 assumes that the Redirect property is set
to AdminMenu.asp. However, if the end developer would rather have a database
table’s administration page redirect to some sort of confirmation page, all he needs
to do is set the Redirect property to the URL of the confirmation page.
The MenuItem class also contains four methods (AddFormHint, AddFKHint,
AddFormNameHint, and AddFormDescriptionHint), which are responsible for adding
hints to the four various Dictionary objects. These four methods are
Figure 6-13. An user’s visit to any of the administration pages occurs within three ASP pages
AdminMenu.asp
When the user wishes to visit an
administration page, she should be
taken to this file. AdminMenu.asp lists
the various database tables, and under
each table name a list box is presented.
The user can then choose if she would
like to insert, edit, or delete a record for
that particular database table.
No matter what administration page
form option the user selects and no
matter what database table she chooses
to work with, she is sent to
/CODEREUSE/DisplayAdminPage.asp.
/CODEREUSE/DisplayAdminPage.asp
This page needs to display the proper
administration page form for the proper
database table. Only one of three
administration page forms can be
displayed here: Insert a New Record,
Select a Record to Edit, and Delete an
Existing record. Once any of these three
administration page forms are submitted,
the user is taken to/CODEREUSE/
AdminPageAction.asp, where the actual
changes are committed to the database.
Once the user selects an administration
page form for a particular database,
she is taken to/CODEREUSE/
DisplayAdminPage.asp.
/CODEREUSE/AdminpageAction.asp
This page is only reachable through /CODEREUSE/
DisplayAdminPage.asp. If the user had just completed
an Insert a New Record or Delete an Existing Record
administration page form, a record is inserted or
deleted from the database, and the user is sent back
to /CODEREUSE/AdminMenu.asp. If, however, the user
has just selected a record to edit, the Edit an Existing
Record form is displayed. When this form is submitted,
/CODEREUSE/AdminPageAction.asp is called again, the
record is updated in the database, and the user is
returned to /CODEREUSE/AdminMenu.asp.
When an administration page form is
submitted, AdminPageAction.asp is
visited.
If the user selected a record to edit, then
she is shown the Edit an Existing Record
administration page form.
If a record was inserted, updated, or deleted
from the database, the user is redirected to
/CODEREUSE/Display/AdminPage.asp.
Creating Reusable Administration Pages 231
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
functionally identical to the hint-adding methods in the AdminPageGenerator
class (listed in Example 6-18).
Example 6-34 lists the MenuItem class, along with all of its methods, properties, and
event handlers. The MenuItem class should be placed in a file named /CODEREUSE/
AdminPageMenu.asp.
Figure 6-14. A single ASP page is generated to serve as a menu to all the available
administration pages
232 Chapter 6: Database Reuse
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Example 6-34. A MenuItem Class Instance Contains Information Pertinent for Creating a
Particular Database Table’s Administration Page Forms
<%
Class MenuItem
'************** MEMBER VARIABLES *********************
Private strTableName
Private strRedirect
Private strIdentityColumn
'Hint Dictionary objects
Private objFormHints
Private objFKHints
Private objFormNameHints
Private objFormDescriptionHints
'*****************************************************
'******************* EVENT HANDLERS ******************
Private Sub Class_Initialize()
Set objFormHints = Server.CreateObject("Scripting.Dictionary")
Set objFKHints = Server.CreateObject("Scripting.Dictionary")
Set objFormNameHints = Server.CreateObject("Scripting.Dictionary")
Set objFormDescriptionHints = Server.CreateObject("Scripting.Dictionary")
End Sub
Private Sub Class_Terminate()
Set objFormHints = Nothing
Set objFKHints = Nothing
Set objFormNameHints = Nothing
Set objFormDescriptionHints = Nothing
End Sub
'*****************************************************
'****************** LET PROPERTIES *******************
Public Property Let TableName(str)
strTableName = str
End Property
Public Property Let IdentityColumnHint(strIdentity)
strIdentityColumn = strIdentity
End Property
Public Property Let Redirect(str)
strRedirect = str
End Property
'*****************************************************
'****************** GET PROPERTIES *******************
Public Property Get TableName()
TableName = strTableName
End Property
Public Property Get IdentityColumnHint()
IdentityColumnHint = strIdentityColumn
End Property
Creating Reusable Administration Pages 233
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Public Property Get Redirect()
Redirect = strRedirect
End Property
Public Property Get FormHints()
Set FormHints = objFormHints
End Property
Public Property Get FKHints()
Set FKHints = objFKHints
End Property
Public Property Get FormNameHints()
Set FormNameHints = objFormNameHints
End Property
Public Property Get FormDescriptionHints()
Set FormDescriptionHints = objFormDescriptionHints
End Property
'*****************************************************
'********************** METHODS **********************
Public Sub AddFormHint(strName, strValue)
If objFormHints.Exists(strName) then
objFormHints(strName) = strValue
Else
objFormHints.Add strName, strValue
End If
End Sub
Public Sub AddFKHint(strName, strValue)
If objFKHints.Exists(strName) then
objFKHints(strName) = strValue
Else
objFKHints.Add strName, strValue
End If
End Sub
Public Sub AddFormNameHint(strName, strValue)
If objFormNameHints.Exists(strName) then
objFormNameHints(strName) = strValue
Else
objFormNameHints.Add strName, strValue
End If
End Sub
Public Sub AddFormDescriptionHint(strName, strValue)
If objFormDescriptionHints.Exists(strName) then
objFormDescriptionHints(strName) = strValue
Else
objFormDescriptionHints.Add strName, strValue
End If
End Sub
Example 6-34. A MenuItem Class Instance Contains Information Pertinent for Creating a
Particular Database Table’s Administration Page Forms (continued)
234 Chapter 6: Database Reuse
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Testing the MenuItem class
The MenuItem class is used to add a new administration page to the menu (see
Figure 6-14 for a screenshot). To accommodate such functionality, the MenuItem
class has a number of read/write properties. Example 6-35 contains a simple test
driver for the MenuItem class. In the example, a couple of the properties are set,
creating a MenuItem instance for the Product table.
The output of the test driver in Example 6-35, when viewed through
a browser, is simply “Creating a menu item for Product.”
AdminPageMenu’s properties and event handlers
The AdminPageMenu class is responsible for generating the menu interface shown
in Figure 6-14. Since every single administration page form is shown through a sin-
'*****************************************************
End Class
%>
Example 6-35. The MenuItem Class Describes a Menu Item for the Administration
Page Menu
<%@ LANGUAGE = "VBSCRIPT" %>
<% Option Explicit %>

<%
'Specify the redirect property (where the user should be taken to after
'/CODEREUSE/AdminPageAction.asp updates the database)
Const strRedirect = "/AdminMenu.asp"
'Create an instance of the MenuItem class
Dim objMenuItem
Set objMenuItem = New MenuItem
'Set some of the MenuItem proeprties
objMenuItem.TableName = "Product"
objMenuItem.Redirect = strRedirect
objMenuItem.IdentityColumnHint = "ProductID"
objMenuItem.AddFKHint "ProductTypeid", "Product Type ~Name~"
'Output some of these values
Response.Write "Creating a menu item for <B>" & objMenuItem.TableName
Response.Write "</B><BR>"
Set objMenuItem = Nothing 'Clean up
%>
Example 6-34. A MenuItem Class Instance Contains Information Pertinent for Creating a
Particular Database Table’s Administration Page Forms (continued)
Creating Reusable Administration Pages 235
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
gle ASP page (/CODEREUSE/DisplayAdminPage.asp), the information needed to
display a particular administration page form must be passed from the menu page
(AdminMenu.asp) to a page that can correctly display the proper administration
page form. As shown in Figure 6-13, the page that is responsible for displaying the
administration page form has been named /CODEREUSE/DisplayAdminPage.asp.
However, you can specify that the users be sent to a different page once they
choose an administration page from the menu. A constant, strAction, defines
what ASP page the user should be sent to from the administration page menu
(AdminMenu.asp).
Recall that database table-specific information already exists in each MenuItem
instance. However, we have not yet accounted for certain pieces of information
needed by every database table’s administration page, namely the database connection
information. The AdminPageMenu class contains the information required
by all administration pages.
AdminPageMenu contains three properties used for maintaining database connection
information: ConnectionString, UserName, and Password. The
AdminPageMenu class also contains a private member variable, objMenuDict,
which is a Dictionary object instance containing MenuItem instances. For each
database table that needs to be listed on the administration page menu, a
MenuItem instance representing this database table’s information must exist in the
objMenuDict Dictionary object. In the next section, “AdminPageMenu’s methods,”
we’ll look at how MenuItem instances are inserted into the objMenuDict
member variable.
The AdminPageMenu class source code should be placed in the same file as the
MenuItem class, /CODEREUSE/AdminPageMenu.asp. Example 6-36 contains the
properties and event handlers for the AdminPageMenu class.
Example 6-36. The AdminPageMenu Class Contains Database
Connection Information
<%
'strAction specifies where the Adminitration Menu should take the user once
'they select a particular administration menu option.
Const strAction = "/CODEREUSE/DisplayAdminPage.asp"
Class AdminPageMenu
'************** MEMBER VARIABLES *********************
Private objMenuDict
Private strConnString
Private strUserName
Private strPassword
'*****************************************************
'******************* EVENT HANDLERS ******************
Private Sub Class_Initialize()
'Initialize objMenuDict
236 Chapter 6: Database Reuse
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
The strAction constant appearing at the beginning of
Example 6-36 indicates where the user should be sent once he or
she selects a particular administration page to visit from the administration
page menu.
AdminPageMenu’s methods
Since the AdminPageMenu class is responsible for generating the actual administration
page menu, a method is needed to accomplish this. This method, Generate-
Menu, iterates through the objMenuDict collection. For each MenuItem instance,
GenerateMenu creates a form containing a list box with three options: Insert a
New Record, Edit an Existing Record, and Delete an Existing Record. The form
also contains a number of HIDDEN form variables that /CODEREUSE/
DisplayAdminPage.asp needs to present the user with the correct administration
page form.
Set objMenuDict = Server.CreateObject("Scripting.Dictionary")
End Sub
Private Sub Class_Terminate()
Set objMenuDict = Nothing 'Clean up...
End Sub
'*****************************************************
'****************** GET PROPERTIES *******************
Public Property Get Count()
Count = objMenuDict.Count
End Property
'*****************************************************
'****************** LET PROPERTIES *******************
Public Property Let ConnectionString(str)
strConnString = str
End Property
Public Property Let UserName(str)
strUserName = str
End Property
Public Property Let Password(str)
strPassword = str
End Property
'*****************************************************
' ... The methods of AdminPageMenu are presented in Example 6-37 ...
End Class
%>
Example 6-36. The AdminPageMenu Class Contains Database
Connection Information (continued)
Creating Reusable Administration Pages 237
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
AdminPageMenu also contains a method for adding a MenuItem instance to the
objMenuDict member variable. The method, AddMenuItem, expects a MenuItem
instance as a single parameter, and adds this instance to objMenuDict.
The two methods of AdminPageMenu can be seen in Example 6-37. Note that
when the FORM tag is created, the ACTION property is set to a constant named
strAction. strAction specifies the URL the user is sent to once he selects the
administration page form he’d like to visit. strAction, a constant that needs to be
defined before the AdminPageMenu class definition, should be set to /CODEREUSE/
DisplayAdminPage.asp.
Example 6-37. The GenerateMenu Method of the AdminPageMenu Class Creates the HTML to
Display the Administration Page Menu
<%
'strAction specifies where the Administration Menu should take the user once
'they select a particular administration menu option.
Const strAction = "/CODEREUSE/DisplayAdminPage.asp"
Class AdminPageMenu
' ... The properties and event handlers for AdminPageMenu are
' presented in Example 6-36 ...
'********************** METHODS **********************
Public Sub AddMenuItem(objMenuItem)
'Add a new MenuItem instance to the objMenuDict Dictionary object
objMenuDict.Add objMenuItem.TableName, objMenuItem
End Sub
Public Function GenerateMenu()
'This function returns an HTML string containing the Admin. Page Menu
Dim strResult, strName, objMenuItem, strTmp, objFormHints, _
objFKHints, objFormNameHints, objFormDescriptionHints
strResult = "<CENTER>" & vbCrLf
For Each strName in objMenuDict
Set objMenuItem = objMenuDict(strName)
strResult = strResult & "<FORM METHOD=POST ACTION=""" & _
strAction & """>" & vbCrLf
'********* Create HIDDEN variables **********
strResult = strResult & "<INPUT TYPE=HIDDEN NAME=ConnectionString " & _
"VALUE=""" & strConnString & """>" & vbCrLf
strResult = strResult & "<INPUT TYPE=HIDDEN NAME=UserName " & _
"VALUE=""" & strUserName & """>" & vbCrLf
strResult = strResult & "<INPUT TYPE=HIDDEN NAME=Password " & _
"VALUE=""" & strPassword & """>" & vbCrLf
strResult = strResult & "<INPUT TYPE=HIDDEN NAME=TableName " & _
"VALUE=""" & objMenuItem.TableName & """>" & vbCrLf
238 Chapter 6: Database Reuse
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
strResult = strResult & "<INPUT TYPE=HIDDEN NAME=Redirect " & _
"VALUE=""" & objMenuItem.Redirect & """>" & vbCrLf
strResult = strResult & "<INPUT TYPE=HIDDEN NAME=IdentityColumnHint " & _
"VALUE=""" & objMenuItem.IdentityColumnHint & """>" & vbCrLf
Set objFormHints = objMenuItem.FormHints
Set objFKHints = objMenuItem.FKHints
Set objFormNameHints = objMenuItem.FormNameHints
Set objFormDescriptionHints = objMenuItem.FormDescriptionHints
For Each strTmp in objFormHints
strResult = strResult & "<INPUT TYPE=HIDDEN NAME=FormHintsName " & _
"VALUE=""" & strTmp & """>" & vbCrLf
strResult = strResult & "<INPUT TYPE=HIDDEN NAME=FormHintsValue " & _
"VALUE=""" & objFormHints(strTmp) & """>" & vbCrLf
Next
For Each strTmp in objFKHints
strResult = strResult & "<INPUT TYPE=HIDDEN NAME=FKHintsName " & _
"VALUE=""" & strTmp & """>" & vbCrLf
strResult = strResult & "<INPUT TYPE=HIDDEN NAME=FKHintsValue " & _
"VALUE=""" & objFKHints(strTmp) & """>" & vbCrLf
Next
For Each strTmp in objFormNameHints
strResult = strResult & "<INPUT TYPE=HIDDEN NAME=FormNameHintsName" & _
" VALUE=""" & strTmp & """>" & vbCrLf
strResult = strResult & "<INPUT TYPE=HIDDEN NAME=" & _
"FormNameHintsValue VALUE=""" & _
objFormNameHints(strTmp) & """>" & vbCrLf
Next
For Each strTmp in objFormDescriptionHints
strResult = strResult & "<INPUT TYPE=HIDDEN NAME=" & _
"FormDescriptionHintsName " & _
"VALUE=""" & strTmp & """>" & vbCrLf
strResult = strResult & "<INPUT TYPE=HIDDEN NAME=" & _
"FormDescriptionHintsValue VALUE=""" & _
objFormDescriptionHints(strTmp) & """>" & vbCrLf
Next
Set objFormHints = Nothing
Set objFKHints = Nothing
Set objFormNameHints = Nothing
Set objFormDescriptionHints = Nothing
'********* End Create HIDDEN variables **********
strResult = strResult & "<B><CODE>" & objMenuItem.TableName & _
"</CODE>" & " Table Administration</B><BR>" & vbCrLf
strResult = strResult & "<SELECT SIZE=1 NAME=Action>" & vbCrLf & _
"<OPTION VALUE=INSERT>Insert a New Record</OPTION>" & vbCrLf & _
"<OPTION VALUE=UPDATE>Edit an Existing Record</OPTION>" & vbCrLf & _
"<OPTION VALUE=DELETE>Delete an Existing Record</OPTION>" & vbCrLf & _
"</SELECT>" & vbCrLf
Example 6-37. The GenerateMenu Method of the AdminPageMenu Class Creates the HTML to
Display the Administration Page Menu (continued)
Creating Reusable Administration Pages 239
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Generating a menu with the AdminPageMenu and MenuItem classes
Now that we’ve looked at the properties and methods of the MenuItem and
AdminPageMenu classes, its time we examine how to apply these classes to generate
the administration page menu. In this example, we’ve used AdminMenu.asp as
the ASP page that generates the administration page menu.
The source code for AdminMenu.asp can be seen in Example 6-38. Note that a
server-side include is used to import the contents of /CODEREUSE/
AdminPageMenu.asp, which contains the class definition for AdminPageMenu and
MenuItem. Also notice that the Redirect property for each MenuItem instance is
set to /AdminMenu.asp.
strResult = strResult & "<P><INPUT TYPE=SUBMIT VALUE="" Go! "">" & _
vbCrLf
strResult = strResult & "</FORM><P><HR><P>" & vbCrLf & vbCrLf
Set objMenuItem = Nothing
Next
strResult = strResult & "</CENTER>" & vbCrLf
GenerateMenu = strResult
End Function
'*****************************************************
End Class
%>
Example 6-38. A Single Instance of the AdminPageMenu Class Is All It Takes to Generate the
Administration Page Menu
<%@ LANGUAGE = "VBSCRIPT" %>
<% Option Explicit %>

<HTML>
<BODY>
<CENTER>
<FONT SIZE=+2>
Administration Page Menu
</FONT>
</CENTER>
<P><HR><P>
<%
Const strRedirect = "/AdminMenu.asp"
Dim objAdminMenu, objMenuItem
Set objAdminMenu = New AdminPageMenu
objAdminMenu.ConnectionString = "Provider=SQLOLEDB;Data Source=mitchell;" & _
"Initial Catalog=ProductInfo;User ID=sa;Password="
Example 6-37. The GenerateMenu Method of the AdminPageMenu Class Creates the HTML to
Display the Administration Page Menu (continued)
240 Chapter 6: Database Reuse
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
In Example 6-38, a MenuItem instance is created for each database table that
needs a set of administration pages. There may be times when you don’t want to
allow users to alter particular database tables. To prevent them from doing so, simply
do not add a MenuItem instance for that database table. (Note that there is no
MenuItem instance being added for the ProductType table in Example 6-38.) Furthermore,
if your data model expands to encompass new tables in the future, creating
administration pages for those tables is as simple as adding a MenuItem
instance in AdminMenu.asp.
A screenshot of the code presented in Example 6-38 can be seen in Figure 6-15.
Note that the administration page menu in Figure 6-15 is nearly identical to that in
Figure 6-14. The only difference is the administration page menu in Figure 6-15 is
lacking an administration page option for the ProductType table. There may be
'Add a MenuItem instance to objAdminMenu for each database table
Set objMenuItem = New MenuItem
objMenuItem.TableName = "Product"
objMenuItem.Redirect = strRedirect
objMenuItem.IdentityColumnHint = "ProductID"
objMenuItem.AddFKHint "ProductTypeid", "Product Type ~Name~"
objAdminMenu.AddMenuItem objMenuItem
Set objMenuItem = Nothing
Set objMenuItem = New MenuItem
objMenuItem.TableName = "Catalog"
objMenuItem.Redirect = strRedirect
objMenuItem.IdentityColumnHint = "CatalogID"
objAdminMenu.AddMenuItem objMenuItem
Set objMenuItem = Nothing
Set objMenuItem = New MenuItem
objMenuItem.TableName = "CatalogProduct"
objMenuItem.Redirect = strRedirect
objMenuItem.AddFKHint "ProductID", "Product ~Quantity~ at $~Price~"
objMenuItem.AddFKHint "CatalogID", "Catalog ~Name~"
objAdminMenu.AddMenuItem objMenuItem
Set objMenuItem = Nothing
'Display the administration page menu
Response.Write objAdminMenu.GenerateMenu()
Set objAdminMenu = Nothing 'Clean up
%>
</BODY></HTML>
Example 6-38. A Single Instance of the AdminPageMenu Class Is All It Takes to Generate the
Administration Page Menu (continued)
Creating Reusable Administration Pages 241
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
situations in which you wish to present administration pages only to a subset of
database tables. Since the end developer can limit what administration page
options appear in the administration page menu, such control is possible.
It would also be nice to be able to limit what type of administration
pages each database table could have. For example, the end developer
may not want anyone deleting records from the Catalog table.
Adding such functionality would be relatively simple and would
serve as a great exercise for the interested reader.
Figure 6-15. A generic Administration Page Menu allows the user to select from three
administration pages for each database table
242 Chapter 6: Database Reuse
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
/CODEREUSE/DisplayAdminPage.asp: the single administration page
form generation script
Whenever the user selects an administration page to visit from the administration
page menu, he is taken to /CODEREUSE/DisplayAdminPage.asp. This ASP page is
responsible for presenting the user with the correct administration page form. All
the information needed to display the administration page form is available, having
been passed in from AdminMenu.asp.
To display the proper administration page, the proper AdminPageGenerator
method must be called after the various “hints” and properties have been
entered. Example 6-39 shows the code for /CODEREUSE/DisplayAdminPage.asp,
which creates the appropriate administration page form using the
AdminPageGenerator class.
Example 6-39. /CODEREUSE/DisplayAdminPage.asp Is Responsible for Creating the Proper
Administration Page Form
<% @LANGUAGE="VBSCRIPT" %>
<% Option Explicit %>

<%
'This ASP page is called when the user submits the form in AdminMenu.asp.
'This page needs to display the proper administration page...
Dim strTableName, strFormAction, strConnString, strUserName, strPassword
strTableName = Request("TableName")
strFormAction = Request("Action")
strConnString = Request("ConnectionString")
strUserName = Request("UserName")
strPassword = Request("Password")
Dim objAdmin
Set objAdmin = New AdminPageGenerator
'******** BEGIN: ADD ALL HINTS ENTERED BY END DEVELOPER ****************
If Len(Request("IdentityColumnHint")) > 0 then
objAdmin.IdentityColumnHint = Request("IdentityColumnHint")
End If
If Len(Request("Redirect")) > 0 then
objAdmin.Redirect = Request("Redirect")
End If
Dim aTmpNames, aTmpValues, strTmpName, strTmpValue, iLoop
'Add Form Hints
strTmpName = Request("FormHintsName")
strTmpValue = Request("FormHintsValue")
aTmpNames = split(strTmpName, ",")
aTmpValues = split(strTmpValue, ",")
Creating Reusable Administration Pages 243
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
For iLoop = LBound(aTmpNames) to UBound(aTmpNames)
objAdmin.FormHint Trim(aTmpNames(iLoop)), Trim(aTmpValues(iLoop))
Next
'Add FK Hints
strTmpName = Request("FKHintsName")
strTmpValue = Request("FKHintsValue")
aTmpNames = split(strTmpName, ",")
aTmpValues = split(strTmpValue, ",")
For iLoop = LBound(aTmpNames) to UBound(aTmpNames)
objAdmin.ForeignKeyHint Trim(aTmpNames(iLoop)), Trim(aTmpValues(iLoop))
Next
'Add Form Name Hints
strTmpName = Request("FormNameHintsName")
strTmpValue = Request("FormNameHintsValue")
aTmpNames = split(strTmpName, ",")
aTmpValues = split(strTmpValue, ",")
For iLoop = LBound(aTmpNames) to UBound(aTmpNames)
objAdmin.FormNameHint Trim(aTmpNames(iLoop)), Trim(aTmpValues(iLoop))
Next
'Add Form Description Hints
strTmpName = Request("FormDescriptionHintsName")
strTmpValue = Request("FormDescriptionHintsValue")
aTmpNames = split(strTmpName, ",")
aTmpValues = split(strTmpValue, ",")
For iLoop = LBound(aTmpNames) to UBound(aTmpNames)
objAdmin.FormDescriptionHint Trim(aTmpNames(iLoop)), Trim(aTmpValues(iLoop))
Next
'********** END: ADD ALL HINTS ENTERED BY END DEVELOPER ****************
'Output the FORM
Response.Write "<HTML><BODY><h1>" & vbCrLf
'Decide what administration page form to display
Select Case strFormAction
Case "INSERT":
Response.Write "Insert a New " & UCase(strTableName) & "</H1>" & vbCrLf
Response.Write objAdmin.CreateRecordForm(strTableName, strConnString, _
strUserName, strPassword)
Case "UPDATE":
Response.Write "Select an Existing " & UCase(strTableName) & _
" to Edit</H1>" & vbCrLf
Response.Write objAdmin.UpdateRecordList(strTableName, strConnString, _
strUserName, strPassword)
Case "DELETE":
Example 6-39. /CODEREUSE/DisplayAdminPage.asp Is Responsible for Creating the Proper
Administration Page Form (continued)
244 Chapter 6: Database Reuse
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
With AdminMenu.asp displaying the administration page menu, /CODEREUSE/
DisplayAdminPage.asp serving as the sole administration page form generation
script, and /CODEREUSE/AdminPageAction.asp making all requested database
modifications, the generic administration page generation application now has the
flow illustrated in Figure 6-13.
Limitations and Possible Enhancements
Every application has its shortcomings and room for improvements; when using an
application, it is important that you are aware of its limitations. The generic administration
page generation application is not without its fair share of handicaps.
For starters, when creating robust, reusable code, it is important to ensure that the
code will work under a variety of different circumstances. Unfortunately, the
administration page generation is limited to working with only two database providers:
the SQLOLEDB provider, and the Jet 4.0 (or greater) provider. The application
is limited to these two providers due to its use of the adSchemaForeignKeys
schema, which currently is only supported by the SQLOLEDB and Jet 4.0 database
providers. Since the majority of ASP sites use either Microsoft SQL Server or
Microsoft Access, this isn’t too great of a problem, but it would have been nice to
have the administration page generation application work for all database
providers.
Another limitation of the system is the difficulty presented in obtaining information
on AutoNumber columns. While AutoNumber information can be found from
a SQL Server database through examination of the syscolumns system table, the
only generic way to obtain the AutoNumber column for a particular database table
is to use the dynamic Properties collection of the Recordset object. This presents a
problem when attempting to determine the AutoNumber column in a table with
no records. To compensate for this, the IdentityColumnHint property is available
in the AdminPageGenerator class, but this solution is far from graceful.
Response.Write "Delete an Existing " & UCase(strTableName) & "</H1>" & vbCrLf
Response.Write objAdmin.DeleteRecordForm(strTableName, strConnString, _
strUserName, strPassword)
Case Else:
'Uh-oh, shouldn't get here...
Response.Write "Invalid Action parameter specified..."
Response.End
End Select
Response.Write "</BODY></HTML>"
Set objAdmin = Nothing 'Clean up...
%>
Example 6-39. /CODEREUSE/DisplayAdminPage.asp Is Responsible for Creating the Proper
Administration Page Form (continued)
Creating Reusable Administration Pages 245
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Another limitation of the administration page reveals itself when a database table
has a foreign key column that is nullable. A simple enhancement could be made
to check whether a foreign key column allowed NULLs, and if so, the list box of
available foreign key values could also contain an N/A entry.
Yet another shortcoming of the application arises when there are multiple users
making concurrent or near-concurrent changes to the same database tables
through the administration pages. For example, imagine that the following situation
occurred:
• User A chooses to Edit an Existing Record from the Catalog table. Our user
chooses to edit the “Fall, 2000” catalog record, and is whisked away to the
administration page form where he can alter various column values.
• While User A is making some changes to the “Fall, 2000” catalog record, User
B decides to delete the “Fall, 2000” catalog record from the Catalog table.
Uh-oh!
• User A finishes making his changes and submits the form. An error is
returned (since the UpdateRecord method of the ModifyDatabase class
couldn’t grab the record being edited from the Catalog table), confusing and
irritating User A.
This shortcoming is common among Internet applications, or any distributed application
for that matter, and is a difficult one to overcome. Unfortunately, this is a
limitation the users of the administration page generation application will have to
live with. Since it is unfeasible and impractical to implement some sort of locking
system on these database tables, concurrent users working with the same database
table may inadvertently cause errors to occur for one another.
A note on performance
The administration page generation application was written to be as robust and
reusable as possible. Furthermore, an important goal of the application was ease
of use. Often there is a tradeoff between reusability and performance. Likewise, a
tradeoff also exists between code readability and performance. Classes enhance
readability and maintainability, but are often not as efficient as their procedural
counterparts.
How important is performance, though? For a web application that will be used by
many concurrent users, performance is key. Computers have trained us to be
impatient, and there’s nothing worse than having to wait for your computer! With
the administration page generation application, however, performance is not paramount,
since just a few “superusers” will use it occasionally. Therefore, it’s not terribly
important to have a blazingly fast administration page. With an application
focusing so heavily on reusability, readability and maintainability are far more
important qualities than performance.
246 Chapter 6: Database Reuse
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Note that we now have a very powerful, generic administration page generation
application. To add a new set of administration pages for a new database table, a
new MenuItem instance needs to be added to /CODEREUSE/DisplayAdminPage.asp.
That’s all that needs to be done! In the beginning of this chapter, I groaned at how
much work was involved in creating administration pages the classical way. With
this application, administration pages to insert, update, and delete records from a
database table require only five lines of code!
Hopefully this chapter has illustrated the benefits of code reuse. While building
reusable applications takes a tremendous amount of upfront effort (I spent over
two weeks creating the various classes presented in this chapter), the rewards are
immeasurable. The end results were definitely worth the initial hard work!
Further Reading
It is an understatement to say that this chapter discussed numerous topics. As
such, certain topics were skimmed over a bit. This futher reading should provide
more detailed information about a number of topics covered in this chapter.
• The technical documentation for database schemas is available on Microsoft’s
site at http://msdn.microsoft.com/library/psdk/dasdk/mdam2ppd.htm.
• To learn how to use the METADATA tag to import DDL constants, be sure to
read “Using METADATA to Import DDL Constants,” available at http://www.
4guysfromrolla.com/webtech/110199-1.shtml.
• If you are looking for some more detailed information on how to use various
data providers with ADO, be sure to check out “Using Providers with ADO” at
http://msdn.microsoft.com/library/psdk/dasdk/mdap99m7.htm.
• To learn how to create a generic database insertion script similar to the one
used in the InsertRecord method of the ModifyDatabase class, be sure to
read “A Very Generic Database Insertion ASP Page,” at http://www.
4guysfromrolla.com/webtech/122299-1.shtml.
• In this chapter we discussed a number of advanced database techniques. For
database-related questions and answers, be sure to check out the Database
ListServs over at http://www.asplists.com/asplists/database.asp.
247
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Chapter 7 7
Using Components
One of the most powerful features of ASP is its tight integration with COM. Due to
this relationship, Active Server Pages can easily use COM objects, thereby extending
the feature set of ASP. Microsoft provides several ready-to-use COM objects
with IIS, which will be the focus of this chapter; however, there is an entire market
of third-party COM objects created by companies independent from Microsoft,
which will be discussed in length in the next chapter. Furthermore, you, the developer,
can create your own custom COM objects for use in your ASP projects,
which we’ll touch upon briefly in this chapter.
Being able to use COM objects in ASP pages—be it Microsoft-created, third-partycreated,
or custom-built—is what gives Active Server Pages their great flexibility.
Without the ability to use COM objects, ASP would be painfully limited in its
scope. After all, ASP only provides a handful of intrinsic objects for the developer
to use. These built-in ASP objects don’t grant the developer much more than the
ability to read from the HTTP request stream, write to the HTTP response stream,
and gather information about the web server. To extend ASP, components must be
used.
To fully understand the importance of being able to use COM components in an
ASP page, imagine for a moment that suddenly, ASP’s ability to use COM objects is
taken away. Since ADO is nothing more than a collection of Microsoft-provided
COM components that aid in database interactions, if an ASP page could not
instantiate and utilize a COM component, developers would not be able to create
data-driven web sites using Active Server Pages.
Using a high-level programming language like Java, Visual C++, or Visual Basic,
developers can create custom COM components for use in ASP pages. Using custom
COM components is a virtual necessity for large web sites with complex
248 Chapter 7: Using Components
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
business rules. In this chapter we’ll examine what benefits custom COM components
bring to the table and when they should be used. We’ll also discuss Windows
Script Components, which are scriptable COM objects, and compare them to
classical, binary COM components.
In this chapter we’ll also look at some of the lesser-known Microsoft COM components.
Almost every developer is familiar with ADO, but not as many have had
working experience with Ad Rotator, Browser Capabilities, and Content Linker.
COM—A Quick Overview
This book has focused on code reuse and design for ASP applications. Throughout
the past seven chapters, we’ve created various scripts and classes that could
easily be reused in other ASP scripts and web projects with minimal effort. While
following rigorous ASP application design techniques makes future ASP projects a
breeze, it does not lend itself well to future projects using other development
tools.
For example, if you wanted to create a banner rotation system on your web site,
you could write a class that would randomly pick one advertising banner from a
set of potential banners based upon some developer-defined weighting of the various
banners. Creating such a system wouldn’t be too difficult, and could be done
in a highly reusable way using classes. Now, imagine that you were interested in
creating a random image-viewing program using Visual C++. You ought to be able
to reuse the random banner rotation script to display a random image, since both
programs are incredibly alike. However, since the banner rotation system was written
as a VBScript class, you cannot directly use it in your Visual C++ random
image viewing program.
A COM component to assist in banner rotation already exists. This
component, named Ad Rotator, is discussed in detail in this chapter.
It has long been the dream of developers to be able to write and compile a program
just once, and have it able to be used by any program on any platform on
any computer with maximum ease. COM, which stands for Component Object
Model, is a Microsoft-backed standard for creating “use anywhere, anytime, anyhow”
components. With COM, a developer only needs to write the code and compile
the object once. With this compiled object in hand, any programming language
or tool that can utilize COM can then create an instance of the COM object.
Lesser-Known Microsoft COM Components 249
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
This chapter does not contain a thorough explanation of COM. If
you are interested in learning the nitty-gritty details of COM, I highly
suggest you read Inside COM, by Dale Rogerson, published by
Microsoft Press.
COM components can be created in a wide array of programming languages,
including Java, Visual C++, and Visual Basic. A fully functional COM component
can be instantiated and used from a wide array of programming tools and languages.
If you are proficient in any of these languages, you can create custom
COM objects for use in your ASP pages. Later in this chapter, we’ll discuss the
benefits of custom COM objects and when they should be used. This chapter does
not, however, delve into a detailed discussion on how to build custom COM components.
For a great resource on building custom COM components for use in ASP
pages, be sure to read Shelly Powers’s Developing ASP Components (O’Reilly &
Associates).
In summary, COM components allow for the development of components in a
plethora of languages that can be compiled once and used virtually anywhere.
COM components can also be developed using any ActiveX Scripting
Engine, such as VBScript, JScript, PerlScript, or Python. We will
discuss these COM components, referred to as Windows Script Components,
in the section “Building COM Components with Script” later
in this chapter.
Lesser-Known Microsoft COM
Components
Microsoft has created a wide range of COM components for developers to freely
use in their applications. Many of these COM components are intended for use in
a number of application settings. For example, ActiveX Data Objects (ADO), which
were used extensively in Chapter 6, Database Reuse, are nothing more than a collection
of COM objects to aid with database access. Due to the nature of COM,
ADO is not limited to use in Active Server Pages alone. In fact, ADO is commonly
used in stand-alone Visual Basic and Visual C++ applications that need to access a
database.
Microsoft provides a bevy of useful COM components for Active Server Pages
other than ADO. In this section, we will quickly examine the basic functionality of
250 Chapter 7: Using Components
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
these lesser-known COM components, and then look at how to improve their
functionality with the same design techniques we’ve used throughout the first
seven chapters. Table 7-1 shows a listing of the COM components included with
IIS for ASP. This chapter will examine the first two, Ad Rotator and Content Linker,
in detail.
For a list of the various methods and properties for these Installable
Active Server Pages Components, check out http://msdn.microsoft.
com/library/psdk/iisref/comp275c.htm.
Creating a Random Banner Rotation System
with Ad Rotator
Nearly every web site these days supports itself financially through the sale of
advertising banners. For example, the web site I run, 4GuysFromRolla.com, offers
free ASP information to tens of thousands of developers each day. To help pay for
the web hosting for 4GuysFromRolla.com, I sell advertising space to companies
Table 7-1. Installable ASP Components
Component Name Description
Ad Rotator Provides a random banner rotation system.
Content Linker Useful for creating a table of contents page for a collection of
related URLs. Also provides navigational functionality to step
through related URLs sequentially.
Content Rotator Randomly displays text or HTML on a web page.
Browser Capabilities Useful for determining your visitor’s browser’s capabilities. For
example, this component can be used to determine if your visitor
can support client-side VBScript, ActiveX controls, and other
information.
Counters Provides for web page hit counting, letting your visitors see how
many people have visited a given page.
Permission Checker Can be used to determine if a user has access to a certain file or
directory. Especially useful when Basic or NT Challenge/
Response password authentication has been enabled.
MyInfo Keeps track of web server information in one object. The MyInfo
component tracks information like the name, address, and phone
number of the webmaster, company, or organization running the
web site.
Tools A nifty little collection of helpful tools for accomplishing basic
server-side tasks.
Lesser-Known Microsoft COM Components 251
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
that specialize in ASP web hosting or companies that build custom COM
components.
In Chapter 8, Enhancing Your Web Site with Third-Party Components,
we’ll examine some of these custom COM components.
Multiple advertisers mean, of course, that multiple banners are available for display
on the user’s browser. To display them, a banner rotation system is essential.
A banner rotation system is one that randomly displays a single advertising banner
from a list of advertising banners.
Microsoft provides a free custom COM component to assist in creating a banner
rotation system. This component, named Ad Rotator, is installed by default with
Personal Web Server, IIS 4.0, and IIS 5.0.
Creating a rotator schedule file
The Ad Rotator generates the HTML to display an advertising banner randomly
selected from a rotator schedule file. The rotator schedule file contains information
on each of the banners in the banner rotation system. When clicked, an
advertising banner should redirect the user to the appropriate URL. For example, if
you were running a banner for 4GuysFromRolla.com on your site, when the user
clicked the banner, they would be automatically transferred to http://www.
4GuysFromRolla.com.
The Ad Rotator does this in an indirect way. Rather than just displaying a banner
that is hyperlinked to the appropriate URL, like so:
<A HREF="http://www.4GuysFromRolla.com">
<IMG SRC="/images/4GuysBanner.gif">
</A>
the Ad Rotator uses a redirection file, which is an optional ASP page that can be
used to record how many times a specific banner was clicked. Figure 7-1 illustrates
the control flow when using the redirection file.
All this information—the redirection file URL, the banner information, and other
tidbits of information—is stored in the rotator schedule file. The rotator schedule
file is a simple text file that has the form shown in Example 7-1.
252 Chapter 7: Using Components
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
The first four lines in the rotator schedule file apply to all the banners in the rotation.
The REDIRECT line specifies the URL of the redirection file. Recall that the
redirection file script is responsible for sending the user to a URL when a specific
banner is clicked. Later in this chapter, in the section “The redirection file,” we’ll
examine the details of a redirection file.
The URL of the redirection file should not be relative to the directory of the ASP
page using the Ad Rotator. Non-relative URLs can be specified in one of two ways:
either through a full URL listing (http://www.yourserver.com/scripts/somescript.asp)
or by specifying a fully qualified virtual path (/scripts/somescript.asp). A relative
URL, like somescript.asp, limits the directories in which pages utilizing the Ad Rotator
component can be found.
Figure 7-1. The redirection file is an intermediate step occurring between a user clicking a
banner and that user being redirected to the banner’s corresponding URL
Example 7-1. The Rotator Schedule File Contains Information on Each Banner in the Rotation
REDIRECT RedirectionFileURL
WIDTH BannerWidthInPixels
HEIGHT BannerHeightInPixels
BORDER BorderValueForImages
*
Path to banner image 1
Banner redirection URL 1
ALT tag text 1
Weight 1
Path to banner image 2
Banner redirection URL 2
ALT tag text 2
Weight 2
Path to banner image N
Banner redirection URL N
ALT tag text N
Weight N
SomeASPPage uses the
Ad Rotator component to
display a random banner
advertisement.
4Guys Banner
SomeASPPage.asp
The redirection file logs
that a user clicked on a
particular banner and then
redirects the user to the
clicked banner's URL.
RedirectionFile.asp
The user clicks on the
banner and is redirected
to the redirection file.
http://www.4GuysFromRolla.com
Lesser-Known Microsoft COM Components 253
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
The REDIRECT line in the rotator schedule file is optional; however,
it is strongly recommended that you supply it. If you do not supply a
REDIRECT line, when the user clicks a banner, the ASP page on
which the banner appeared will be reloaded with a new query string
indicating the redirect URL specified for the particular banner (which
we’ll discuss shortly).
The following three lines—WIDTH, HEIGHT, and BORDER—are all optional. The
WIDTH line specifies the width, in pixels, of all the banners in the rotator schedule
file. If WIDTH is omitted, a value of 440 is used. The HEIGHT line indicates the
height, in pixels, of all the banners in the rotator schedule file. If HEIGHT is omitted,
a value of 60 is used. The BORDER line specifies the BORDER value in the IMG
tag, which indicates whether a border should be placed around an image. If the
BORDER property is not specified, it defaults to a value of 1.
The asterisk following the first four lines separates the global properties from the
banner-specific properties and is not optional. Therefore, as an absolute minimum,
the first two lines of the rotator schedule file should be:
REDIRECT BannerRedirectURL
*
After the asterisk, four lines are needed for each banner in the rotation. The four
lines provide the following information for each specific banner:
• The path to the image. As with the global REDIRECT line, this should be a nonrelative
URL (http://www.yourserver.com/images/someBanner.gif or /images/
someBanner.gif ). If you specify the image as a relative URL (someBanner.gif ),
you will experience broken images when using the same rotator schedule file
in ASP pages in a different directory.
• The URL the user should be redirected to when they click the banner. The
URL is oftentimes a fully qualified web address, like http://www.someserver.
com. When a banner is clicked, this URL will be passed to the redirection file
through a query string variable.
• The ALT tag text. An image’s ALT tag is displayed while the graphic is loading
or if the visitor has images disabled or is surfing from a text-only browser
(such as lynx). Also, in both Internet Explorer and Netscape for Windows, if
the user moves her mouse over the image, the ALT tag appears as a tool tip.
• The weight of the banner. This weight specifies how often the banner is displayed,
relative to the other banners in the rotation system. If the banner rotation
system contained two banners, one with a weight of 2 and the other with
a weight of 1, the first banner would be displayed, on average, twice as often
as the second. Commonly, developers will ensure that the various weights add
254 Chapter 7: Using Components
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
up to 100 so that by looking at a specific banner’s weight, one can quickly
determine the percentage of times the banner will be displayed.
Example 7-2 contains a working rotator schedule file with three banners.
Our rotator schedule file starts by specifying our redirection file, /CODEREUSE/
AdRedirect.asp. When any of the banners displayed by Ad Rotator are clicked, the
user will be taken to this script. The next three lines contain global information on
each of our banners for this rotator schedule file. These first four parameters that
specify global properties for all advertisement banners are separated from the banner-
specific information by a line containing just an asterisk. This line is vitally
important, and if omitted, will generate errors when the Ad Rotator attempts to display
a banner. Following the asterisk, each of the groups of four lines describes a
banner in the rotation.
The rotator schedule file shown in Example 7-2 is used shortly in a couple of
examples. For these examples, assume the rotator schedule file has been saved as
/CODEREUSE/AdRot.asp.
Often developers will create rotator schedule files with a .txt extension.
This can be dangerous if you wish to keep the rotator schedule
file from prying eyes. If the rotator schedule file is saved with a .
txt extension and placed in a web-accessible directory, anyone who
knows or guesses the filename can easily view the source of the
rotator schedule file. Therefore, it is recommended that you give the
rotator schedule file an .asp extension or place it in a directory that
has HTTP read access disabled.
Example 7-2. AdRot.asp, a Rotator Schedule File Containing Three Banners
REDIRECT /CODEREUSE/AdRedirect.asp
WIDTH 468
HEIGHT 60
BORDER 1
*
/images/4GuysFromRolla.gif
http://www.4GuysFromRolla.com
When you think ASP, think 4GuysFromRolla.com!
40
/images/ASPMessageboard.gif
http://www.ASPMessageboard.com
Got ASP Questions? Find Answers!
40
/images/OReilly.gif
http://www.oreilly.com
Books by O'Reilly
20
Lesser-Known Microsoft COM Components 255
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Displaying a banner
To display a banner, use the GetAdvertisement method of the Ad Rotator. The
GetAdvertisement method has the following definition:
GetAdvertisement(RotatorScheduleFile)
The GetAdvertisement method returns HTML that displays a hyperlinked image.
Therefore, to display an advertisement, simply use the following code:
<%
Dim objAdRot
Set objAdRot = Server.CreateObject("MSWC.AdRotator")
Response.Write objAdRot.GetAdvertisement(RotatorScheduleFile)
%>
It’s that simple. Example 7-3 contains an ASP page that displays a random banner
from the rotator schedule file shown in Example 7-2. The output of the script in
Example 7-3, when viewed through a browser, can be seen in Figure 7-2.
The GetAdvertisement method simply returns the HTML to display a particular
banner with a hyperlink to the ASP page specified by the REDIRECT line in the
Example 7-3. Use the GetAdvertisement Method Where You Wish to Have an Ad Inserted
<%@ LANGUAGE = "VBSCRIPT" %>
<% Option Explicit %>
<%
Dim objAdRot
Set objAdRot = Server.CreateObject("MSWC.AdRotator")
%>
<HTML>
<BODY>

<CENTER>
<%=objAdRot.GetAdvertisement("/CODEREUSE/AdRot.asp")%>
</CENTER>
<H1>Welcome to our Website!</H1>
Blah blah blah blah
<P>

<CENTER>
<%=objAdRot.GetAdvertisement("/CODEREUSE/AdRot.asp")%>
</CENTER>
</BODY>
</HTML>
<%
Set objAdRot = Nothing 'Clean up!
%>
256 Chapter 7: Using Components
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
rotator schedule file. The HTML generated by GetAdvertisement for the top banner
in Figure 7-2 (with line breaks added for readability) is:
<A HREF="/CODEREUSE/AdRedirect.asp?
url=http://www.ASPMessageboard.com&
image=/images/ASPMessageboard.gif" >
<IMG SRC="/images/ASPMessageboard.gif"
ALT="Got ASP Questions? Find Answers!"
WIDTH=468 HEIGHT=60 BORDER=1>
</A>
When clicked, the banner directs the user to the redirection file, passing the banner’s
particular image path and redirection URL through the query string.
The redirection file
The redirection file, which is specified by the REDIRECT line in the rotator schedule
file, is visited whenever a user clicks on a banner advertisement generated by
the Ad Rotator component. This file should contain a redirection script that is
responsible for handling any click-through statistical logging and for forwarding
the user to the correct URL.
Figure 7-2. The rotator schedule file is consulted to display a random banner at the top and
bottom of the document
Lesser-Known Microsoft COM Components 257
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Note that your code in the redirection script is responsible for forwarding the user
to the banner ad’s URL; the Ad Rotator component doesn’t handle that process
automatically. The minimal code needed for the redirection file is simply:
<% Response.Redirect Request.QueryString("URL") %>
which will send the user to the banner’s specific URL.
The redirection file’s only requirement is to send the user to the clicked banner’s
URL. Before sending the user off the site, though, the redirection script can take a
moment to record the fact that the particular banner was clicked.
Tracking banner statistics
Often advertisers (who pay their hard-earned money to purchase banner advertisements)
are interested in viewing various statistics for their banners. The three
major statistics advertisers are interested in are impressions, click-throughs, and
click-through percentage. A banner impression occurs when a banner is displayed
to the user, whereas a click-through only occurs when a user clicks on the displayed
banner. The click-through percentage is the number of click-throughs
divided by the number of impressions. These three statistics give the advertiser a
pretty good view of their banner advertising’s effectiveness.
To keep track of these statistics, we’ll need to use a database table. This database
table, AdInfo, contains the needed rows for advertising statistics tracking, and is
described in Table 7-2.
Table 7-2. The AdInfo Table Is Designed to Track Advertising Banner Statistics
Column Description
URL Specifies the URL to which the banner points (i.e., http://www.
someserver.com/ ). Forms a composite primary key with the
ImageURL column.
ImageURL Specifies the URL for the banner graphic (i.e., /images/banners/
someBanner.gif ). Forms a composite primary key with the URL
column.
Impressions An integer field with a default of zero. Keeps track of the total number
of impressions for a particular banner.
ClickThroughs An integer field with a default of zero. Keeps track of the total number
of click-throughs for a particular banner.
Description Provides a textual description of the banner to be used in a report
screen. This report screen should show the description, impressions,
click-throughs, and click-through percentages for each banner.
258 Chapter 7: Using Components
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
There is no need to have a ClickThroughPercentage column in
the AdInfo table, since the click-through percentages can be determined
by dividing the value of ClickThroughs by the value of
Impressions.
With the Ad Rotator, calculating the total number of click-throughs is rather simple.
Each time a banner is clicked, the redirection file is visited. Since the URL of
the banner image and the URL to redirect to are passed to the redirection file, a
simple UPDATE statement could be issued to increment the number of
ClickThroughs.
Example 7-4 presents a redirection file that increments the number of clickthroughs
for the clicked banner. Note that the script assumes a row has already
been entered into the AdInfo table that contains the URL and image URL. Therefore,
for this approach to accurately track the click-through stats, before adding a
new banner to the rotator schedule file, be sure to add a corresponding row to the
AdInfo database table.
Example 7-4. The Redirection Script Can Track Click-Through Statistics
<% @LANGUAGE="VBSCRIPT" %>
<% Option Explicit %>
<%
'Read in the URL/Image
Dim strURL, strImageURL
strURL = Request.QueryString("url")
strImageURL = Request.QueryString("image")
'Now, open a connection to the database. We need to
'increment the click-through count by 1 for this banner
Dim objConn
Set objConn = Server.CreateObject("ADODB.Connection")
objConn.Open "DSN=AdvertisingInfo"
Dim strSQL
strSQL = "UPDATE AdInfo SET ClickThroughs = ClickThroughs + 1 " & _
"WHERE URL = '" & strURL & "' AND ImageURL = '" & strImageURL & "'"
objConn.Execute strSQL 'Execute the SQL statement
objConn.Close 'Clean up...
Set objConn = Nothing
'Redirect the user
Response.Redirect strURL
%>
Lesser-Known Microsoft COM Components 259
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
The banner rotation script makes tracking click-through statistics simple. However,
tracking impression statistics with Ad Rotator is far from easy; it can be done,
but requires some clever (and messy) hacks. You may wonder how important
impression tracking is. One may think that advertisers are only interested in how
many people actually clicked on their banner, but in my experience, I’ve found
that many advertisers regard impressions as just as important. Furthermore, since
advertisers usually prefer to buy blocks of impressions (that is, an advertiser may
be willing to pay x dollars for y banner impressions), being able to track the number
of impressions is important as well.
Since the GetAdvertisement method of the Ad Rotator component simply spits out
HTML, there are no straightforward hooks or event handlers to use to increment
the number of impressions when a particular banner is shown. Therefore, if tracking
banner impressions is critical, many developers will give up on Ad Rotator and
look for other solutions.
There are a number of articles sprinkled throughout the Web on
how to create a custom banner rotation system using ASP. Here are
a few of these articles:
• “Creating a Banner Rotation System” at http://www.
4guysfromrolla.com/webtech/061399-4.shtml.
• “Dynamic Banner Administration” at http://www.15seconds.com/
issue/pv980220.htm.
• “Rotating Banner Ads Using a Database” at http://www.
4guysfromrolla.com/webtech/091299-1.shtml.
However, using a bit of a hack, we can implement banner impression tracking
with the Ad Rotator. For this hack to work properly, it is essential that for each
banner you wish to add to the rotator schedule file, you first add a record to the
AdInfo table.
OK, now comes the hack! Rather than simply calling the GetAdvertisement method
and sending it directly to the HTML output stream, we can instead store it to a
variable. Having intercepted it, we can quickly scan it, looking for the URL
(wedged between url= and the following ampersand (&)) and the ImageURL
(wedged between the image= and the closing quotes). Since these two pieces of
information uniquely identify each record in the AdInfo table, once we have the
values for URL and ImageURL, we can increment the correct row’s Impressions
column! A huge, ugly, disgraceful hack, I know, but with the Ad Rotator, there is
no cleaner way to track impressions.
260 Chapter 7: Using Components
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
To accomplish this, we’ll create a function named ShowAd, which returns the
HTML to generate the clickable advertising banner. ShowAd has the following
definition:
Function ShowAd(AdRotatorInstance, strScheduleFile)
The AdRotatorInstance must be a valid Ad Rotator object instance. The
strScheduleFile parameter (the same parameter required by the GetAdvertisement
method, incidentally) expects a virtual path to the rotator schedule file.
ShowAd will call the GetAdvertisement method of the AdRotatorInstance component,
saving the HTML for analysis. The URL and ImageURL values are then
picked out of the saved HTML, and a database call is executed to increment the
number of impressions for the specific banner.
The full source code for ShowAd can be seen in Example 7-5.
In Example 7-3, we used the GetAdvertisement method to display a banner. This
method, of course, lacks impression tracking. To display a banner and increment
Example 7-5. Impression Tracking Is Possible with the Ad Rotator Component
Function ShowAd(objAdRotator, strScheduleFile)
'Save the HTML
Dim strHTML
strHTML = objAdRotator.GetAdvertisement(strScheduleFile)
'Now, get the URL (between url= and &) and ImageURL (between image and ")
Dim strURL, strImageURL, iImageStart
strURL = Mid(strHTML, InStr(1, strHTML, "url=") + 4, _
InStr(1, strHTML, "&") - InStr(1, strHTML, "url=") - 4)
iImageStart = InStr(1, strHTML, "image=") + 6
strImageURL = Mid(strHTML, iImageStart, InStr(iImageStart, strHTML, """") _
- iImageStart)
'Update the database info
Dim objConn
Set objConn = Server.CreateObject("ADODB.Connection")
objConn.Open "DSN=AdvertisingInfo", "sa"
Dim strSQL
strSQL = "UPDATE AdInfo SET Impressions = Impressions + 1 " & _
"WHERE ImageURL = '" & strImageURL & "' AND URL = '" & _
strURL & "'"
objConn.Execute strSQL
objConn.Close
Set objConn = Nothing
ShowAd = strHTML
End Function
Lesser-Known Microsoft COM Components 261
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
its Impressions, use the ShowAd function where you previously used the GetAdvertisement
method, like so:
<%
'Create an instance of the Ad Rotator
Dim objAdRot
Set objAdRot = Server.CreateObject("MSWC.AdRotator")
%>

<CENTER>
<%=ShowAd(objAdRot, "/CODEREUSE/AdRot.asp")%>
</CENTER>
Note that the ShowAd method expects an instance of the Ad Rotator
as its first parameter. Be sure to have created such an instance of the
Ad Rotator object before calling the ShowAd method.
Providing impression and click-through tracking via a class
Clearly, implementing impression tracking adds considerable complexity in using
the Ad Rotator component. As we’ve seen countless times before in this book,
when faced with a task with complicated implementation details, it is often wise to
encapsulate the task in a class. Therefore, let us create a class to ease impression
and click-through tracking with the Ad Rotator component.
This class, AdRotation, will have an Ad Rotator instance as a member variable,
which will be used to generate the HTML for the advertisements. Along with a single
private member variable, the AdRotation class contains the following
properties:
ScheduleFile
Specifies the rotator schedule file the Ad Rotator component should consult
when displaying a banner.
ConnectionString
Specifies the connection string to the database that contains the AdInfo table.
UserName
Specifies the username to use when connecting to the database, if one is
needed.
Password
Specifies the password to use when connecting to the database, if one is
needed.
Example 7-6 contains the properties and event handlers for the AdRotation class.
This class definition should be placed in the file /CODEREUSE/AdRotation.Class.asp.
262 Chapter 7: Using Components
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Example 7-6. The AdRotation Class Contains Three Database Connection Properties and a
Property for the Rotator Schedule File
<%
Class AdRotation
'************** MEMBER VARIABLES *********************
Private objAdRotator
Private strScheduleFile
'Database connection info
Dim strConnectionString
Dim strUserName
Dim strPassword
'*****************************************************
'******************* EVENT HANDLERS ******************
Private Sub Class_Initialize()
'Create an instance of the Ad Rotator component
Set objAdRotator = Server.CreateObject("MSWC.AdRotator")
End Sub
Private Sub Class_Terminate()
Set objAdRotator = Nothing 'clean up!
End Sub
'*****************************************************
'****************** GET PROPERTIES *******************
Public Property Get ScheduleFile()
ScheduleFile = strScheduleFile
End Property
Public Property Get ConnectionString()
ConnectionString = strConnectionString
End Property
Public Property Get UserName()
UserName = strUserName
End Property
Public Property Get Password()
Password = strPassword
End Property
'*****************************************************
'****************** LET PROPERTIES *******************
Public Property Let ScheduleFile(str)
strScheduleFile = str
End Property
Public Property Let ConnectionString(str)
strConnectionString = str
End Property
Public Property Let UserName(str)
strUserName = str
End Property
Lesser-Known Microsoft COM Components 263
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
The AdRotation class has only two methods: ShowAd and SendOffSite. The
ShowAd method randomly picks a banner based upon the rotator schedule file
and updates the number of Impressions in the AdInfo table, returning the
HTML to generate the banner. This method should be called each time you wish
to display an advertising banner. SendOffSite, on the other hand, is only called
once from the banner rotation script. SendOffSite increments the ClickThroughs
column value and uses a Response.Redirect to send the user on to the URL the
clicked banner points to.
Example 7-7 contains the source code for the two AdRotation methods. Note that
the ShowAd method is nearly identical to the ShowAd function presented in
Example 7-5, and the SendOffSite method is nearly identical to the redirection file
shown in Example 7-3.
Public Property Let Password(str)
strPassword = str
End Property
'*****************************************************
'... The two methods for the AdRotation class have been omitted for
brevity; they are both presented in Example 7-7 ...
End Class
%>
Example 7-7. The Two AdRotation Methods Assist with Impression and Click-Through
Statistic Tracking
<%
Class AdRotation
'... The properties and event handlers for the AdRotation class
were presented in Example 7-6 ...
'********************** METHODS **********************
Public Function ShowAd()
'Save the HTML
Dim strHTML
strHTML = objAdRotator.GetAdvertisement(strScheduleFile)
'Now, get the URL (between url= and &) and ImageURL (between image and ")
Dim strURL, strImageURL, iImageStart
strURL = Mid(strHTML, InStr(1, strHTML, "url=") + 4, _
InStr(1, strHTML, "&") - InStr(1, strHTML, "url=") - 4)
iImageStart = InStr(1, strHTML, "image=") + 6
strImageURL = Mid(strHTML, iImageStart, InStr(iImageStart, strHTML, """") _
- iImageStart)
'Update the database info
Dim objConn
Example 7-6. The AdRotation Class Contains Three Database Connection Properties and a
Property for the Rotator Schedule File (continued)
264 Chapter 7: Using Components
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Before you use the AdRotation class in a production environment,
be sure to add exception-handling code in both ShowAd and Send-
OffSite. Both methods assume the information needed to extract URL
and ImageURL is present and in the correct format.
Set objConn = Server.CreateObject("ADODB.Connection")
objConn.Open strConnectionString, strUserName, strPassword
Dim strSQL
strSQL = "UPDATE AdInfo SET Impressions = Impressions + 1 " & _
"WHERE ImageURL = '" & strImageURL & "' AND URL = '" & _
strURL & "'"
objConn.Execute strSQL 'Update the impressions
objConn.Close
Set objConn = Nothing 'close/clean up
ShowAd = strHTML
End Function
Public Sub SendOffSite()
'Read in the URL/Image
Dim strURL, strImageURL
strURL = Request.QueryString("url")
strImageURL = Request.QueryString("image")
'Now, open a connection to the database. We need to
'increment the click-through count by 1 for this banner
Dim objConn
Set objConn = Server.CreateObject("ADODB.Connection")
objConn.Open strConnectionString, strUserName, strPassword
Dim strSQL
strSQL = "UPDATE AdInfo SET ClickThroughs = ClickThroughs + 1 " & _
"WHERE URL = '" & strURL & "' AND ImageURL = '" & strImageURL & "'"
objConn.Execute strSQL 'Execute the SQL statement
objConn.Close 'Clean up...
Set objConn = Nothing
Response.Redirect strURL 'Redirect the user
End Sub
'*****************************************************
End Class
%>
Example 7-7. The Two AdRotation Methods Assist with Impression and Click-Through
Statistic Tracking (continued)
Lesser-Known Microsoft COM Components 265
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Using the AdRotation class
As we’ve examined throughout this book, classes hide implementation details and
mask complexity through encapsulation. As we’ll see in a moment, by wrapping
the impression and click-through tracking features into a class, the Ad Rotator is
now even easier to use than before! In Example 7-3, we looked at how to display
an advertisement using the Ad Rotator. Example 7-8 shows how to use the
AdRotation class not only to display an add, but also to increment the ad’s
impression count.
In Example 7-8, the AdRotation class has been placed inside a file
named /CODEREUSE/AdRotation.Class.asp, and is imported into the
code using a server-side include.
Example 7-8. Use the ShowAd Method to Display a Banner Advertisement with the
AdRotation Class
<%@ LANGUAGE = "VBSCRIPT" %>
<% Option Explicit %>

<%
'Create an instance of the AdRotation class
Dim objAdRot
Set objAdRot = New AdRotation
'Set the database connection properties and the schedule file
objAdRot.ConnectionString = "DSN=AdvertisingInfo"
objAdRot.ScheduleFile = "/CODEREUSE/AdRot.asp"
'Wherever an ad is to be displayed, simply call the ShowAd() method and
'send its output to the client!
%>
<HTML>
<BODY>

<CENTER>
<%=objAdRot.ShowAd()%>
</CENTER>
<H1>Welcome to our Website!</H1>
Blah blah blah blah
<P>

<CENTER>
<%=objAdRot.ShowAd()%>
</CENTER>
</BODY>
</HTML>
266 Chapter 7: Using Components
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
When a banner is clicked, the user will be taken to the redirection file, which
should have been specified in the first line of the rotator schedule file (which contains
the REDIRECT keyword). From the redirection file, we will need to send the
user to the URL the clicked banner points to, but first we need to update the banner’s
ClickThroughs value. Both of these tasks are accomplished through the
SendOffSite method.
Using the AdRotation class, all we need to do in the redirection file is:
• Create an instance of the AdRotation class.
• Set its properties.
• Call the SendOffSite method.
That’s all! The SendOffSite method will correctly increment the ClickThroughs
value and redirect the user to the proper URL! Example 7-9 shows the source code
for the redirection file.
When using the AdRotation class, you must have the AdInfo table properly set
up and have an entry in the table for every banner in the rotator schedule file.
(The database table AdInfo was presented in Table 7-2.) With the current implementation,
this results in having to enter the same information twice: once in the
AdInfo table and once in the rotator schedule file. An industrious developer
would create a script that would accomplish both tasks in one fell swoop.
<%
Set objAdRot = Nothing 'Clean up!
%>
Example 7-9. With the AdRotation Class, the Redirection File Needs Only to Call the
SendOffSite Method
<% @LANGUAGE="VBSCRIPT" %>
<% Option Explicit %>

<%
'Create an instance of the AdRotation class
Dim objAdRot
Set objAdRot = New AdRotation
'Set the properties...
objAdRot.ScheduleFile = "/CODEREUSE/AdRot.asp"
objAdRot.ConnectionString = "DSN=AdvertisingInfo"
objAdRot.SendOffSite 'Increment ClickThroughs and redirect the user!
Set objAdRot = Nothing 'Clean up...
%>
Example 7-8. Use the ShowAd Method to Display a Banner Advertisement with the
AdRotation Class (continued)
Lesser-Known Microsoft COM Components 267
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Possible enhancements to the AdRotation class
One obvious enhancement to the AdRotation class would be to add a reporting
method. This method, when called, would return either an HTML string or a
Recordset object that contained the various banner information in the AdInfo
table. The pertinent information would be each banner’s description, impressions,
click-throughs, and click-through percentages.
Another needed enhancement is some sort of administration tool. Currently, banner
information must be stored in two places: the rotator schedule file and the
database. It is essential that the banner’s image URL and redirect URL in the rotator
schedule file match the AdInfo table’s ImageURL and URL columns, respectively.
Rather than having to make these two identical entries by hand, an
administration page that made the needed changes in both the rotator schedule
file and the database would be ideal, saving time and errors resulting from typos.
Did we really accomplish anything?
At the start of this chapter, I listed some of the benefits of COM objects over
VBScript classes. Namely, COM objects are highly portable, able to be accessed by
multiple development tools and programming languages. VBScript classes, on the
other hand, are only really useful in programming tools that can utilize VBScript,
such as ASP, client-side browser code, and the Windows Script Host.
However, as soon as I introduced a useful COM component (the Ad Rotator), I
almost immediately provided a more functional version using a class. Isn’t this
chapter about components and not classes? While this VBScript AdRotation class
is not portable, it is still worth the time to develop the class. When developing
components, I find it intelligent to keep the components fairly simple, allowing a
particular component to have only a small subset of core features.
Once this focused component has been created, developers working with various
development tools and programming languages can extend the core functionality
of the portable component using non-portable methods. That is exactly what we
have done with the AdRotation class.
Managing Web Site Content with the Content Linker
One of the greatest challenges of running a web site is adding new content. As a
site grows, adding a new web page to the site often involves adding links to several
other pages. Microsoft provides a free component with IIS 4.0 and IIS 5.0
called the Content Linker, which makes managing content-rich, frequently updated
web sites much easier.
Specifically, the Content Linker provides dynamic hyperlinks between a series of
ordered web pages. To use the Content Linker, a Content Linker list file is needed,
which provides an ordered list of related URLs and descriptions. When the Content
268 Chapter 7: Using Components
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Linker is used in one of the pages listed in the Content Linker list file, hyperlinks
can be generated to point to the next and previous URLs in the list. To add a new
URL to the list of related URLs, simply add a reference to the new web page in the
Content Linker list file. We’ll look at the specifics of the Content Linker list file
shortly.
Content Linker is especially useful in web sites that contain several related web
pages, especially if those web pages can be related sequentially. If you think of a
web site’s layout in terms of a tree structure (see Figure 7-3), after using the Content
Linker, it quickly becomes apparent that it was designed for sites with a flat
organization of web pages. In sites with a deep organizational structure, the Content
Linker can still be used, but will not be as useful or helpful. Figure 7-3 shows
both flat and deep tree structures representing two web site layouts.
A great example of the Content Linker in use is LearnASP.com, a site dedicated to
providing various ASP tutorials. LearnASP.com breaks down each tutorial into several
related lessons, giving the site a flat web page structure. The Content Linker
component is used to generate dynamic hyperlinks between each lesson and tutorial.
Take a moment to check out the LearnASP.com tutorial layout (http://www.
LearnASP.com/learn).
Figure 7-3. A flat web site design lends itself to use of the Content Linker
A flat web site design
A deep web site design
Lesser-Known Microsoft COM Components 269
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
The Content Linker list file
To use the Content Linker, a Content Linker list file is needed. The Content Linker
list file contains a list of ordered URLs, each appearing on a separate line. Each
URL can contain three tab-delimited values:
URL Description Comment
The URL field is the web page URL. Unfortunately, URL must be a relative URL.
URLs in the form of http://someAddress are invalid and will be ignored by the Content
Linker component. Furthermore, non-relative URLs in the form /Address are
also invalid. I find this to be extremely annoying and a painful shortcoming of the
Content Linker component.
The Description parameter provides a description for the specific URL. For
example, if you ran a web site with several related products for sale, the Content
Linker would be a wise choice to use to provide a simple means of navigation
among the products. Since the URL might be cryptic (something like Product.
asp?PID=5623932, perhaps), a textual description of the page would be in order.
This description should appear in the Description parameter in the Content
Linker list file.
The final parameter, Comment, is not used at all by the Content Linker component.
Rather, it serves as a place to add comments for the developer creating the
Content Linker list file.
In the Content Linker list file, only the URL parameter is required.
Description and Comment are optional.
Example 7-10 contains a valid Content Linker list file for a web site that provides a
step-by-step tutorial for learning JScript’s control structures. For this entire example,
all our files will be placed in the /JscriptTutorial directory. The Content Linker
list file shown in Example 7-10 should be named JScriptContLink.asp.
Example 7-10. The Content Linker List File Provides a List of Related Web Pages
default.asp Tutorial Menu Menu page for the tutorial
if.asp if ... else Statement Tutorial
for.asp for Statement Tutorial
while.asp while Loops Tutorial
switch.asp switch Statement Tutorial
270 Chapter 7: Using Components
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
As with the Ad Rotator’s rotator schedule file, the Content Linker list
file is often given a .txt extension. Such an extension, however,
enables users to download your Content Linker list file over the
Web. By giving the file an .asp extension, IIS will attempt to process
the file as an ASP page before sending the output to the client. Since
the Content Linker list file does not contain valid ASP code, an error
will be generated, and the user will not see the file’s source code. Of
course, if this is not a concern to you, feel free to give the Content
Linker list file a .txt extension.
In our example, default.asp will serve as a menu, containing a link to each of the
tutorials. All of the other files in the Content Linker list file contain a tutorial on a
particular JScript control structure.
Providing a next and previous hyperlink in each tutorial
In each tutorial it would be nice to provide a next and previous hyperlink, so that
the user can quickly and easily traverse through the various JScript control structure
tutorials. Of course, this is what the Content Linker is intended for, and now
that we have our Content Linker list file, adding such dynamic hyperlinks is a
breeze.
To use the Content Linker in an ASP page, we must first create an instance of it.
To do this, we can use the following code:
Dim objContentLinkerInstance
Set objContentLinkerInstance = Server.CreateObject("MSWC.NextLink")
The Content Linker component contains eight methods that can be used to generate
dynamic hyperlinks among web pages in the Content Linker list file. These
methods are:
GetListCount(ContLinkList)
Returns the number of web pages represented by the Content Linker list file
ContLinkList (essentially returns the number of lines in the text file
ContLinkList).
GetListIndex(ContLinkList)
Returns the index number of the current page within the Content Linker list
file. The various lines in the Content Linker list file are given index numbers
sequentially, with the first line having an index of 1. If a page whose URL is
not found in ContLinkList executes GetListIndex, a value of 0 will be
returned.
Lesser-Known Microsoft COM Components 271
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
GetNextURL(ContLinkList)
Returns the next URL in the Content Linker list file. If the page that executes
the GetNextURL method is the last file listed in the Content Linker list file, the
GetNextURL method returns the URL of the first item in the Content Linker list
file.
GetNextDescription(ContLinkList)
Returns the description of the next page in the Content Linker list file. If a
description was not specified in the Content Linker list file for the next URL, a
blank string is returned.
GetPreviousURL(ContLinkList)
Returns the previous URL in the Content Linker list file.
GetPreviousDescription(ContLinkList)
Returns the description of the previous page in the Content Linker list file.
GetNthURL(ContLinkList, Index)
Returns the URL of the page with the specified Index value.
GetNthDescription(ContLinkList, Index)
Returns the description of the page with the specified Index value.
To display the next and previous links in our JScript control structure tutorial
pages, we’ll use the GetPreviousURL, GetPreviousDescription, GetNextURL, and
GetNextDescription methods. Example 7-11 contains the source code for if.asp,
which contains a (very) short tutorial on using the if ... else JScript control
structure.
Example 7-11. The if ... else Tutorial Contains a Dynamic Hyperlink to the Next and Previous
Tutorials
<% @LANGUAGE="VBSCRIPT" %>
<% Option Explicit %>
<%
'Create an instance of the Content Linker
Dim objContLink
Set objContLink = Server.CreateObject("MSWC.NextLink")
%>
<HTML>
<BODY>
<CENTER><H1>if ... else Control Structure Tutorial</H1></CENTER>
<HR NOSHADE>
<PRE>
if (<i>condition</i>)
<i>statement</i>
[else <i>statement</i>]
</PRE>
<P>

<TABLE WIDTH=100% BORDER=0>
272 Chapter 7: Using Components
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
A screenshot of if.asp, when viewed through a browser, can be seen in Figure 7-4.
Note the two hyperlinks at the bottom of the page, providing easy access to the
next and previous URLs in the Content Linker list file. As you can see, the Description
field from the Content Linker list file is used as the hypertext, and if we were
to examine the page’s HTML source, we would find that the URL field is used as
the value of the HREF attribute for forward and backward navigation.
To provide a previous and next hyperlink on every JScript Control Structure tutorial,
simply cut and paste the HTML TABLE near the bottom of Example 7-11. (Of
course, rather than cutting and pasting the code into each tutorial page, it is
<TR><TD WIDTH=50% ALIGN=LEFT>
<A HREF="<%=objContLink.GetPreviousURL("JScriptContLink.asp")%>">
<< <%=objContLink.GetPreviousDescription("JScriptContLink.asp")%>
</A>
</TD><TD WIDTH=50% ALIGN=RIGHT>
<A HREF="<%=objContLink.GetNextURL("JScriptContLink.asp")%>">
<%=objContLink.GetNextDescription("JScriptContLink.asp")%> >>
</A>
</TD></TR>
</TABLE>
</BODY>
</HTML>
<%
Set objContLink = Nothing 'Clean up!
%>
Figure 7-4. Each tutorial provides dynamic next and previous hyperlinks
Example 7-11. The if ... else Tutorial Contains a Dynamic Hyperlink to the Next and Previous
Tutorials (continued)
Lesser-Known Microsoft COM Components 273
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
recommended that you place the table in a separate file and use a server-side
include in all tutorial pages.)
In the next section, we will look at displaying a dynamically created
table of contents as the tutorial’s start page (/JscriptTutorial/
default.asp)!
With the Content Linker list file in place, to create a new JScript Control Structure
tutorial, simply perform the following three tasks:
1. Create the new tutorial, saving it in the /JscriptTutorial directory.
2. Add the code that generated the dynamic next and previous links in our new
tutorial.
3. Add the new tutorial’s URL and description to JScriptContLink.asp, the Content
Linker list file for the JScript Control Structure tutorials.
It’s that simple! There’s no need to go poking through existing tutorials and
updating links to include the new tutorial. Rearranging the order of the tutorials is
also a breeze with the Content Linker. If we were not using the Content Linker
and wished to completely reorder the tutorials, we’d have to edit the hardcoded
next/previous links in all the tutorials. With only four tutorials, this isn’t much of a
headache, but imagine if we had hundreds!
However, since we are using the Content Linker, we can reorder the tutorials by
simply rearranging the order of the tutorials in the Content Linker list file. Since
the next and previous links are generated upon demand, after saving the changes
to the Content Linker list file, the next and previous links on the various tutorials
will automatically reflect the changes!
Creating an index listing of the JScript Control Structure tutorials
When users first visit our JScript Control Structure tutorial section, it might be nice
to show them a quick overview of the available tutorials, rather than immediately
starting on the first tutorial. Such a listing of available tutorials would be a pain to
manage without Content Linker, especially if new tutorials were added frequently.
With the Content Linker component, however, maintaining a listing of available
tutorials is quite simple. For our example, default.asp will contain a listing of available
tutorials. Figure 7-5 presents a screenshot of what we want default.asp to
look like when viewed through a browser.
To generate such an index listing, we’ll loop from the first tutorial’s index value to
the maximum index value of the various pages in the Content Linker list file.
Recall that the GetListCount method returns the total number of pages in the
274 Chapter 7: Using Components
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Content Linker list file. Also, recall that two methods exist that allow us to pick out
a specific URL and description: GetNthURL and GetNthDescription.
To create a listing of available tutorials, we’ll loop through the available indexes,
displaying a hyperlink for each index using the GetNthURL and GetNthDescription
methods. Example 7-12 contains the source code of default.asp, displaying a
listing of the available tutorials.
Figure 7-5. The first page in the JScript Control Structure tutorial series should present a
listing of the available tutorials
Example 7-12. Listing the Available Tutorials Is a Simple Task with the Content Linker
<% @LANGUAGE="VBSCRIPT" %>
<% Option Explicit %>
<%
'Instantiate the Content Linker component
Dim objContLink
Set objContLink = Server.CreateObject("MSWC.NextLink")
%>
<HTML>
<BODY>
<CENTER><H1>JScript Control Structure Tutorial</H1></CENTER>
<HR NOSHADE>
Welcome to the JScript Control Structure Tutorial! In this
tutorial you will find syntactical information on all of JScript's
control structures!
<P>
<%
Enhancing Microsoft’s COM Components 275
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Enhancing Microsoft’s COM Components
As we saw in the Ad Rotator example, VBScript classes can be used to enhance
existing COM components. Rarely will you find an existing COM component that
meets every single one of your requirements for a given project. Rather than scrapping
the notion of using the existing COM component, though, why not use it as a
foundation and use VBScript classes to include the needed extra functionality?
This is exactly what was done in this chapter with the Ad Rotator. A random banner
rotation application with impression and click-through tracking was needed.
Ad Rotator met some of those requirements, but not all. Rather than dismissing Ad
Rotator altogether, we embraced the random banner display and click-through
tracking capabilities of Ad Rotator and wrote our own VBScript wrapper class to
add the impression tracking. Existing COM components are great building blocks,
and in my opinion, should be used as often as possible.
Using existing COM components is a lot like buying an existing house and applying
your own add-ons, as opposed to building your own made-to-order home. For
example, if you wanted a home with a deck and the only homes on the market
were those without decks, you would have two options:
• Buy an existing home and add on a deck.
• Build a new home from scratch with, of course, a deck.
Buying an existing home and adding a new deck requires less time, effort, and
money than building a new home from scratch does. Similarly, using existing COM
components and adding the extra needed functionality is a lot less time-consuming,
and less prone to error, than building your own COM component from
scratch.
'Since default.asp is listed as the first entry in the Content Linker file,
'start the For loop at 2, so as not to include default.asp in the listing of
'available tutorials
Dim iLoop
For iLoop = 2 to objContLink.GetListCount("JScriptContLink.asp")
%>
<LI><A HREF="<%=objContLink.GetNthURL("JScriptContLink.asp", iLoop)%>">
<%=objContLink.GetNthDescription("JScriptContLink.asp", iLoop)%>
</A><BR>
<%
Next
Set objContLink = Nothing 'Clean up...
%>
</BODY>
</HTML>
Example 7-12. Listing the Available Tutorials Is a Simple Task with the Content Linker
276 Chapter 7: Using Components
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Building Components
Since COM components can be created in a number of high-level programming
languages, such as Visual C++, Java, and Visual Basic, if you are proficient in any
of these languages, you can create your own custom COM components (or COM
objects, as they are sometimes called). For an object to be considered a COM
object (and therefore to be able to be instantiated from the wide range of development
tools and programming languages), it must follow certain guidelines. These
low-level guidelines are far past the scope of this book, and will not be discussed
in detail.
Thankfully, high-level programming languages ensure that most of these low-level
implementation details are taken care of for you. For example, when creating a
COM component in Visual Basic, these low-level details can be completely
ignored, and the developer can focus on creating classes that will be translated
into full-fledged COM components by Visual Basic.
This book also refrains from stepping through the process of creating a COM
object with any programming language. There are already several great books on
the topic. If you are looking for an entire book dedicated to building COM objects
in various languages, let me recommend Shelly Powers’s Developing ASP Components
(O’Reilly), which shows how to create components in Java, Visual C++, and
Visual Basic in great detail.
What this book does focus on is a particular type of custom COM component, one
used to ensure business logic integrity.
Wrapping Business Logic Within a COM Object
Custom COM components usually fall into one of two categories:
• Components that provide some type of added functionality to the web site
• Components that define the business logic a site must abide by
The two Microsoft COM components we examined earlier in this chapter were
both examples of COM components that provided added functionality. The Ad
Rotator component provided a generic advertising banner rotation system, while
the Content Linker provided a simpler means for adding and updating content to a
large web site. In Chapter 8, we’ll look at a number of commercially available
COM components that also fall within this category.
The second category of custom COM components is not nearly as glamorous as
the first, but in large web sites, is incredibly important. Pause for a moment and
think of any large e-commerce web site. When purchasing an item from that site, a
number of complicated transactions must occur: taxes must be computed; credit
card balances must be checked; orders must be sent to the warehouse via some
Building Components 277
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
mechanism. Now, imagine that all of these complicated transactions were handled
by logic in each ASP page from which a user could purchase an item.
While ASP may be able to handle such complex business rules, it would not be
wise to place complicated business rules within an ASP page, for a number of reasons.
Imagine, for a moment, that all the business rules were hardcoded into the
applicable ASP pages. What would happen if a tax law changed, requiring the
taxes to be calculated differently? Since the tax calculations were hardcoded into a
number of ASP pages, a developer must hunt through the site, making the needed
changes. The more changes the developer is required to make, the more likely he
is to miss a needed change or commit a typo, resulting in a tax-calculation error.
As we’ve seen throughout this book, complex tasks become incredibly simpler
when the task is wrapped inside of a class. With a class, the end developer can
treat the class like a black box and not worry about how the black box performs
its tasks. With classes, the responsibility for a particular task is removed from the
developer and placed squarely on the shoulders of the class. Custom COM components
that handle business logic provide a similar black box interface.
In an e-commerce web site, business logic might include tax information, credit
card handling, shipping procedures, and mechanisms to notify complementary
computer systems (such as sending an order to the warehouse, which might use a
different computer system). If changes occur in the business logic, only the custom
COM component’s code needs to be modified. All the ASP pages using the
COM component are unaware of the business logic changes that occurred.
Custom COM components responsible for business logic also provide a level of
abstraction between an ASP page and the database. In a large, data-driven web
site that does not use COM objects for business logic, when some piece of information
needs to be committed or retrieved from the database, the ASP page is
responsible for executing the SQL statement. In this book, we’ve looked at several
examples of using the ADO objects to insert, update, and delete database records.
However, in a large site with several developers and a complex data model, having
the ASP pages handle database transactions is asking for trouble. Since similar
database actions are taken on various ASP pages, there are bound to be typos or
mistakes, resulting in database or ASP errors. Granted, stored procedures could be
used to help alleviate this problem, but what happens if the data model is
changed? Those ASP pages that reference tables that no longer exist or have been
altered need to be fixed to reflect the changes in the data model.
On a large web site, where there may be hundreds of ASP pages, such a data
model change can be a logistical nightmare. When custom COM components are
involved, the ASP pages make calls to the COM component; it is then the responsibility
of the component to make the appropriate database access calls. That way,
278 Chapter 7: Using Components
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
if there is a data model change, only the custom COM object needs to be modified;
the ASP pages that reference the COM component can go unchanged!
Custom COM components can therefore encapsulate the database and business
logic, providing a black box for ASP pages to use. Custom COM components
make up the middle layer of a three-tiered web site. A three-tiered web site is one
that consists of three parts, each distinct from one another and responsible for a
unique set of tasks. The three tiers, which are illustrated in Figure 7-6, include:
• A Data Services tier, which includes data storage tools, such as databases.
• A Business Services tier, which includes components that contain business
logic information.
• A User Services tier, which contains the front end of the application that is presented
to the user.
For a more detailed explanation of the motivation behind and the
advantages of a three-tiered client-server application, be sure to read
“Building Three-Tiered Client-Server Business Solutions,” a Microsoft
white paper available at http://www.microsoft.com/TechNet/Analpln/
3tierwp.asp.
Imagine the steps that occur when a user visits an order-tracking page in a threetiered
e-commerce web site. When a user enters his or her ordering specifics, the
Figure 7-6. Large, robust web sites should consist of three distinct tiers
ASP PAGES
ON THE
SITE
ASP PAGES
ON THE
SITE
ASP PAGES
ON THE
SITE
ProductOrder
Component
USER SERVICES
OrderTracking
Component
TaxInfo
Component
InventoryInfo
Component
TaxInfo
Database
Inventory
Database
BUSINESS SERVICES
DATA SERVICES
Building Components 279
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
ASP page creates an instance of the OrderTracking component (see Figure 7-6).
This component may interact with other custom COM components (such as the
InventoryInfo component). The order-tracking ASP page, though, is not concerned
with how the OrderTracking component provides the order information
details.
As illustrated in Figure 7-6, the OrderTracking component performs some calculations
and then requests information from the InventoryInfo component. This
component grabs the requested information from the Inventory database, which is
in the Data Services tier.
Using custom COM components also makes for easier-to-read, simplified ASP
code. Imagine the several lines of cryptic ASP code that would be needed to handle
a sale from an e-commerce web site. Using custom COM components that
assume responsibility for the business logic, the ASP code would be greatly simplified
to something like:
'Instantiate the object
Dim objBuySomething
Set objBuySomething = Server.CreateObject("Product.Purchase")
'Set the properties
objBuySomething.ProductID = 458734582
objBuySomething.Quantity = 1
objBuySomething.CreditCardNumber = 1234432112344321
objBuySomething.AddressLine1 = ...
'...
'Now that all the properties have been set, make the purchase
objBuySomething.Purchase
Using COM Objects in a Multienvironment Scenario
One of the major advantages of COM components is they can be used by a vast
array of high-level languages and programming tools. This enables developers to
create a single COM component that can be used in a wide array of situations.
This fact is especially useful when creating custom COM objects to encapsulate
business logic.
In the previous section, “Wrapping Business Logic Within a COM Object,” we
looked at how to wrap the business logic involved in an e-commerce web site into
a set of COM objects. These COM objects housed business logic like sales tax
information, and were used as a layer of abstraction between the ASP pages and
the data store.
Might there be other applications besides an ASP web site that need access to
these business rules? For example, imagine that besides having an e-commerce
site, the company also takes orders over the phone. Employees using a simple
Visual Basic program would likely fill these phone orders. This Visual Basic
280 Chapter 7: Using Components
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
program would need to adhere to the same business logic rules that customers
shopping on the web site follow. Therefore, the custom COM components developed
for the e-commerce site could also be used in the phone-ordering program.
Building COM Components with Script
Despite the performance and reusability advantages of custom COM components
written in a high-level programming language in your ASP pages, such COM components
suffer a couple of shortcomings that make them difficult and annoying to
use.
For starters, ASP pages are not written in the existing programming languages that
can create compiled COM objects. While this may seem irrelevant, imagine that
you already have a complex routine written in VBScript in an ASP page that performs
the business logic your site needs to adhere to. To put this in, say, a Visual
Basic COM object, you must translate the VBScript to Visual Basic syntax. While
such a translation may be light work, what if you wrote your ASP code in JScript
or PerlScript, or if you needed to create your COM component using Java or Visual
C++? Such a translation would be exceedingly difficult.
Another shortcoming of COM components occurs when you need to make a
change to an existing component that’s currently in use. Making a change requires,
at minimum, recompilation and reregistration of the component, and at worst, a
restart of the web services. Anyone who’s developed COM components with a
high-level language is far too familiar with this annoyance.
To address these shortcomings, Microsoft offers Windows Script Components
(WSC), which, as the name implies, are COM components created with script as
opposed to a high-level programming language. Since Windows Script Components
can be created using any ActiveX scripting engine, you can simply cut and
paste your ASP code and create a COM component, regardless of what scripting
language you used to create your ASP page! Also, since these are scripts, no
lengthy recompilation process is needed when changes need to be made to an
existing component.
Once you create a COM object and make it publicly available, any
future changes made to the object must not break existing code. If
you need to make changes to a COM object that will cause existing
code to break, it is essential that you create a new component rather
than alter the existing one.
To be able to create COM components with script, you must download and install
the Windows Script Component from Microsoft’s Script site, http://www.microsoft.
com/msdownload/vbscript/scripting.asp. Windows Script Components are simple
Building Components 281
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
text files and must be created using a predefined XML format. Rather than forcing
developers to fully learn the rather terse XML format, Microsoft provides a freely
available Windows Script Component Wizard. In this section we’ll look at how to
create Windows Script Components using this wizard.
For starters, you’ll need to get yourself a copy of the Windows Script Component
Wizard from the Microsoft Scripting Site, http://msdn.microsoft.com/scripting. At
the time of this writing, the wizard was directly available at http://msdn.microsoft.
com/scripting/scriptlets/wz10en.exe.
For this example, let’s create a COM component that would be similar to a compiled
custom COM component. Assume we have a database table named
Employee that has the following columns:
FirstName
The employee’s first name
LastName
The employee’s last name
Salary
The employee’s salary
BirthDate
The employee’s date of birth
Our COM component will contain a property for each column in the Employee
table and a single method, AddEmployee, that will add a new row to the
Employee table. The COM component will apply some business logic, ensuring
that the employee is over 18 years of age and that the Salary is greater than or
equal to $25,000. By placing this database insertion into a component, we are
encapsulating the database complexity from the ASP developer. The developer
does not need to concern himself with the structure of the Employee table or
worry about adhering to any business rules.
Now let’s create our Windows Script Component! Once the Windows Script Component
Wizard has been downloaded and installed, start the wizard by going to
Start . Programs . Microsoft Windows Script . Windows Script Component Wizard.
The first step of this six-step wizard prompts for the name and ProgID of your
component, as well as the version and filename for the script file.
The ProgID is what you will use in an ASP page to instantiate the
Windows Script Component:
Dim objInstance
Set objInstance = _
Server.CreateObject("ProgID")
282 Chapter 7: Using Components
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Let’s name our component Employee and give it a ProgID of Employee.Insert.
A screenshot of the first page of the Windows Script Component Wizard can be
seen in Figure 7-7.
In Step 2 of the wizard, we are asked to specify the scripting language to use, as
well as what special implementations to support. To have access to the built-in
ASP objects, be sure to select the “Support Active Server Pages” option. Note that
the Error checking and Debugging options have been disabled. These should be
enabled only when in the development stages of the component, and not when
the component is ready to be publicly used. If selected, the Error checking option
will display error messages interactively; if the debugging option is selected, the
Microsoft Script Debugger can be launched when an error occurs.
Figure 7-8 shows what Step 2 of the wizard should look like.
Step 3 prompts for the properties of the component and allows us to define
whether the property is read/write, read-only, or write-only, as well as to set the
property’s default value. For this example, we need a read/write property for each
database column in the Employee table. Figure 7-9 provides a screenshot of Step 3
of the Windows Script Component Wizard.
Step 4 prompts for the component’s methods. Since we will only have one
method, AddEmployee, there is only one entry here. Furthermore, this method
expects no parameters, since there is a property for each database column;
Figure 7-7. Enter the component’s registration and file information in Step 1 of the wizard
Building Components 283
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Figure 7-8. In Step 2 of the wizard, specify the scripting language that you wish to use and
whether to implement ASP support
Figure 7-9. Specify the component’s properties in Step 3 of the wizard
284 Chapter 7: Using Components
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
therefore, the parameters entry is left blank for the AddEmployee method. A
screenshot of Step 4 of the wizard can be seen in Figure 7-10.
Step 5 of the wizard prompts for any events for the component. For this example,
we will not add any events, so proceed on to the next step. Step 6, which is the
final step in the wizard, provides summary information. Once you’ve checked to
make sure everything is in order, go ahead and click Finish. This will create a shell
for the Windows Script Component. It is now up to us to fill in the code for the
AddEmployee method. The wizard has already written the rest of the component’s
code for us!
Dissecting the Windows Script Component file
The generated Windows Script Component file contains several sections. Take a
moment to look at the component’s source. At the start of the document there is
the XML tag, since Windows Script Components are XML-formatted documents:
<?xml version="1.0"?>
The remainder of the file’s contents are placed within a <component> ...
</component> block.
Figure 7-10. Specify the component’s methods in Step 4 of the wizard
Building Components 285
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
The first tag in the <component> tag is the <?component?> tag. This tag indicates
whether or not the component supports error checking and debugging. The
<?component?> tag has the following form:
<?component error="boolean" debug="boolean" ?>
If error checking and debugging were not selected in Step 2 of the
wizard, as was the case in our example, then this tag will not exist.
For completeness, feel free to add it in:
<?component error="false" debug="false" ?>
Next, a <registration> tag defines the description of the class, its ProgID,
ClassID, and version number. Note that except for the ClassID, these values were
entered into Step 1 of the wizard:
<registration
description="Employee"
progid="Employee.Insert"
version="1.00"
classid="{b31aab60-1fb8-11d4-8013-0000216d54d6}"
>
</registration>
The next block of code defines what properties and methods are public. Public
properties and methods in a Windows Script Component are synonymous to public
properties and methods in a VBScript class. Also note that each of the properties
is marked as both read and write with the <get/> and <put/> tags:
<public>
<property name="FirstName">
<get/>
<put/>
</property>
<property name="LastName">
<get/>
<put/>
</property>
<property name="Salary">
<get/>
<put/>
</property>
<property name="BirthDate">
<get/>
<put/>
</property>
<method name="AddEmployee">
</method>
</public>
286 Chapter 7: Using Components
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Next, the <implements> tag is used to indicate that this component has access to
the built-in ASP objects:
<implements type="ASP" id="ASP"/>
The remainder of the file is a <script> block that contains the actual code for the
component (in this case, VBScript code). Note that the <script> block is immediately
followed by a funky-looking <![CDATA[ tag, and right before the closing
script tag, its ending ]]> tag. This is a special XML reference indicating that the
text between the <![CDATA[ and ]]> tags is source code and not to be interpreted
by the XML parser. If the <![CDATA[ tag were left out, the parser would
raise an error if any reserved XML names or characters were found in the component’s
source code.
The CDATA tag is similar to the opening and closing HTML comment
tags () that should be used when writing client-side
JavaScript. Both indicate to the parser that the content between these
tags is not to be interpreted by the parser.
At this time, the AddEmployee method is just a skeleton function:
function AddEmployee()
AddEmployee = "Temporary Value"
end function
It is up to us to add the necessary code. We do need to make one other slight
modification to the WSC file before writing our code. Since the AddEmployee
method will add a record to a database, we will need an instance of the ADO
Connection object. This should be declared using the <object> tag after the
<implements> tag and before the <script> tag:
<object id="objConn" progid="ADODB.Connection" />
The ID parameter in the <object> tag defines how the object will be referred to
in our component’s code.
Now that everything is ready to go, let’s write the code for the AddEmployee
method. This method needs to insert a record into the Employee table based upon
the four properties. If one of these properties is not supplied, or if the supplied
value of the property violates one of the business logic rules, an error will be
raised. Otherwise, if the data is valid, a record will be inserted into the database.
Example 7-13 contains the source code for the AddEmployee method.
Building Components 287
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Note that the AddEmployee method begins by checking to ensure that all data is
passed in, that the Employee being added is at least 18 years of age, and that the
salary is at least $25,000. If any of these checks fail, an error is raised using the
Raise method of the Err object. Since the error-checking option was not selected in
Step 2 of the wizard, when an error is raised it will be passed to the calling application
(the ASP script using this component). Assuming these checks all pass, a
row is added to the Employee table.
Example 7-13. The AddEmployee Method, upon Validating the Inputs, Inserts a Row into the
Employee Table
function AddEmployee()
'Ensure that all of the properties have been entered
If Len(CStr(FirstName)) = 0 or Len(CStr(LastName)) = 0 _
or Len(CStr(Salary)) = 0 or Len(CStr(BirthDate)) = 0 then
'Raise an error
Err.Raise vbObjectError + 1000, "Employee.Insert", "Property not supplied."
End If
'Ensure that the business logic rules have not been violated
'Is employee 18 or over as of today?
If DateDiff("yyyy", BirthDate, Date()) < 18 then
'Raise an error
Err.Raise vbObjectError + 1001, "Employee.Insert", "Employee under 18!"
End If
'Is the salary under 25k?
If Salary < 25000 then
'Raise an error
Err.Raise vbObjectError + 1002, "Employee.Insert", _
"Employee Salary cannot be less than $25,000!"
End If
'OK, there is no problem with the insertion data, so insert a
'record into the Employee table!
Dim strSQL
strSQL = "INSERT INTO Employee (FirstName, LastName, Salary, BirthDate) " & _
"VALUES('" & Replace(FirstName, "'", "''") & "','" & _
Replace(LastName, "'", "''") & "'," & Salary & ",'" & _
BirthDate & "')"
'Open the connection to the database
Const strConnection = "DSN=EmployeeDatabase"
objConn.ConnectionString = strConnection
objConn.Open
objConn.Execute strSQL 'Add the record
objConn.Close 'Close the database connection
end function
288 Chapter 7: Using Components
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
The validation in the AddEmployee method is far from complete.
Measures should be taken not only to ensure that values were
entered for each of the parameters, but that they match their
expected datatypes. For example, an error should be raised if the
Salary property is assigned a value of “More, please.”
Before using this component in an ASP page, we must register it. To register a
Windows Script Component, right-click on the Windows Script Component file
and select the Register option. Figure 7-11 shows how to register the component.
Once the component is registered, you can create an instance of the object in an
Active Server Page (or in a Visual Basic, Visual C++, or Java program). Also, if you
ever need to alter the component, all you need to do is simply edit the appropriate
.wsc file! There is no complicated unregistering/reregistering as there can be
with compiled COM objects, and you don’t need to stop and restart the web services
when making changes to an existing Windows Script Component.
Example 7-14 provides a quick example of how to use the Windows Script Component
we just created within an ASP page. Note that instantiating and using a
Windows Script Component is identical to instantiating and using a compiled COM
component. The ASP developer does not need to concern herself with whether
Figure 7-11. To register a Windows Script Component, right-click the .wsc file and
choose Register
Building Components 289
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
the COM component was created as a binary or is in script form. Note that an
instance of the Windows Script Component we just created is instantiated using
the ProgID that we specified in Step 1 of the wizard (Employee.Insert).
Windows Script Components versus classical, binary COM components
Since Windows Script Components are interpreted scripts as opposed to compiled
binary files, they will obviously suffer from poor performance when compared to
binary files. Therefore, if you expect your site to have many concurrent users, it is
best to stick with binary COM objects.
Also, since classical COM components are created using high-level languages,
chances are you will be developing them using a mature development environment.
With Windows Script Components, however, you will likely be using a
development tool like Notepad to edit the component! While this is not a huge
issue when dealing with a single, small component, imagine that you needed to
develop a slew of large, robust, interactive components. Managing these large,
interrelated files using Notepad would be difficult, at best, especially if a team of
developers were working on the project.
A side benefit of compiled COM components is the source is hidden
from prying eyes. Developers creating COM components for sale
would want to make sure that the final COM component product is
in binary form so that those purchasing the components cannot
share the component’s source code with other potential customers.
Example 7-14. Using a Windows Script Component Is Identical to Using a Classical, Binary
COM Component
<% @LANGUAGE="VBSCRIPT" %>
<% Option Explicit %>
<%
'Create an instance of the Employee.Insert Windows Script Component
Dim objEmployeeInsert
Set objEmployeeInsert = Server.CreateObject("Employee.Insert")
'Set the Employee.Insert properties
objEmployeeInsert.FirstName = "Scott"
objEmployeeInsert.LastName = "Mitchell"
objEmployeeInsert.Salary = 78000
objEmployeeInsert.BirthDate = "8/1/78"
objEmployeeInsert.AddEmployee 'Add the employee to the database
Set objEmployeeInsert = Nothing 'Clean up!
%>
290 Chapter 7: Using Components
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
The advantages Windows Script Components have over compiled COM components
include the ease in making changes to existing components and the ability
to dump ASP code into a component with little or no translation. Windows Script
Components should be used as a stepping-stone to compiled COM objects. For
example, in the development stages, use Windows Script Components. During
development, you’ll enjoy being able to quickly make changes to existing, in-use
components. As the project grows larger and moves toward shipping, be sure to
move the components to a compiled, binary form, where they’ll benefit from
improved performance.
Personally, I find Windows Script Components to be very neat and cool. In smaller
projects, I have a hard time convincing myself that the advantages of using COM
objects outweigh the burdens involved in needing to make changes to the COM
object as the project progresses. Since altering an existing COM object requires
recompilation and reregistration, it’s much easier to not use components. With
Windows Script Components, however, that is no longer true! Also, I find it really
cool that you can grab a snippet of useful ASP code and quickly stick it into a
Windows Script Component, creating a COM component from working ASP code
with little or no translation.
Further Reading
If you are in charge of a large ASP/IIS web site, or are working on one, chances
are you are familiar with building and using custom COM components. If you do
not have much experience with custom COM objects, these articles should help!
• I know I mentioned it earlier, but if you need information on Microsoft’s
Browser Capabilities component, you should definitely check out ASPTracker:
http://www.asptracker.com.
• For some real-world code examples of using the Content Linker, check out the
“Content Link Tutorial” at http://www.learnasp.com/learn/cl.asp.
• For a plethora of articles on building components for ASP applications, check
out the Component Building section on 15Seconds.com, http://www.
15seconds.com/focus/Component%20Building.htm.
• For a good article on the pros of placing business logic in custom COM components,
be sure to read “Using Business Objects in your Web Application”,
http://www.4guysfromrolla.com/webtech/110198-1.shtml.
• For a ListServer discussion on component building, check out http://www.
asplists.com/asplists/components.asp.
• For a good example of how to create a Windows Script Component be sure to
check out “A Simple Windows Scripting Component” at http://www.asptoday.
com/articles/19990729.htm.
291
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Chapter 8 8
Enhancing Your Web Site
with Third-Party Components
One of ASP’s greatest features is its tight integration with COM. As we examined in
Chapter 7, Using Components, Active Server Pages can instantiate complex COM
objects written in high-level languages like Visual C++, Visual Basic, or Java.
Microsoft provides several such components for use in Active Server Pages, such
as the ActiveX Data Objects (ADO), the Ad Rotator, and the Content Linker.
As the popularity of Active Server Pages has grown, many developers have started
their own companies that focus solely on developing COM components for use in
Active Server Pages. These components can be used to greatly enhance the functionality
provided by your web site. In this chapter, we’ll look at several of these
components, focusing on what they do, when they should be used, and how to
use them.
ASP was designed to intrinsically support only a small core of functionality, with
the premise that if a developer needed further functionality, he could create his
own COM component. Some people find this to be a shortcoming of ASP. For
example, to open, read, and write to files with ASP, you must use the
FileSystemObject; other server-side web scripting languages, like Perl, have filehandling
capabilities already built in. While needing to instantiate a component to
simply read a text file may seem like a superfluous performance hit, by omitting
file-handling capabilities, ASP is more streamlined.
Personally, I find ASP’s implementation to be ideal. ASP is very streamlined, consisting
of only a few needed built-in objects. All other functionality should be (and
can be) imported from a COM component. Tools to develop COM components are
widely accessible (Visual Basic, for instance, is the most popular programming language
in the world, with an installed base in the millions), and both ASP and COM
have spawned a lively third-party market that focuses on add-on tools for Active
292 Chapter 8: Enhancing Your Web Site with Third-Party Components
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Server Pages. The components presented in this chapter help to greatly extend the
capabilities of ASP pages.
Executing DOS and Windows
Applications on the Web Server
with ASPExec
There may be times when you’d like an ASP page to be able to execute an application
on the web server. For example, you may wish to provide the webmaster
with the ability to execute certain maintenance programs residing on the web
server through a web page. In “Executing Applications on a Server Through an
ASP Web Page” (available at http://www.4guysfromrolla.com/webtech/072199-2.
shtml), author Neema Moraveji discusses how he needed to be able to remotely
start a setup program on the web server from a web page. In “Creating a Component
Using Visual C++ to Manipulate Virtual Directories” (available at http://www.
15seconds.com/issue/990107.htm), author Shai Vaingast needed to be able to execute
a command-line program through an ASP page.
Rarely will you want to let your users execute an application on the
web server. Imagine the performance ramifications of hundreds of
users running an application simultaneously on your web server!
In both of these articles, the authors turned to the ASPExec component, which
allows developers to execute DOS and Windows applications on the web server
through an ASP page. ASPExec is a free component available from Stephen
Genusa’s company, ServerObjects (download it now from http://www.serverobjects.
com/products.htm#free). Once you download the component, you must use
regsvr32 to install it before you can use it in your ASP pages. To do so, copy
ASPExec.DLL to the Windows system directory (Windows\System or Winnt\
system32). Next, run regsvr32 to register the DLL as follows:
regsvr32 aspexec.dll
Instructions on installing the ASPExec component are included with
the download at http://www.serverobjects.com/products.htm#free.
ASPExec’s Properties and Methods
The ASPExec object contains four properties and three methods. The ASPExec
properties can be seen in Table 8-1 and the methods in Table 8-2.
Executing DOS and Windows Applications on the Web Server with ASPExec 293
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Executing a Command-Line Program with ASPExec
ASPExec can execute both command-line programs (using ExecuteDOSApp) and
Windows programs (using ExecuteWinApp or ExecuteWinAppAndWait). To execute
a command-line program using ASPExec, you must first create an instance of
the ASPExec component. To do so, use the following code:
Dim objASPExecInstance
Set objASPExecInstance = Server.CreateObject("ASPExec.Execute")
Next, set the Application and Parameters properties to indicate the program you’d
like to execute and the command-line arguments you wish to pass to the application.
Once these two properties have been set, simply call the ExecuteDOSApp
method to execute the DOS application.
One such DOS application you can execute through an ASP page is ping. ping
sends an echo request to a server. When the server receives this request, it returns
a confirmation. Oftentimes ping is used to ensure that a web site is up and functional.
ping has many command-line argument options, but the only required one
is the IP, hostname, or domain name of the server to contact.
Table 8-1. ASPExec Contains Four Properties
Property Description
Application Specifies the path (optional) and the executable’s filename.
Parameters Specifies the command-line parameters. The property’s value
is a single string containing all command-line parameters
separated by spaces.
TimeOut Specifies the amount of time to wait in milliseconds if either
the ExecuteDOSApp or ExecuteWinAppAndWait method is
used. Its default value is 30 milliseconds.
ShowWindow Specifies whether or not a window is displayed on the web
server for the executing application (only applicable when
either the ExecuteWinAppAndWait or ExecuteWinApp
method is used.)
Table 8-2. ASPExec Contains Three Methods
Method Description
ExecuteDOSApp Executes a DOS application specified by the Application
property, returning the output of the application.
ExecuteWinAppAndWait Executes the Windows application specified by the Application
property and waits for the timeout specified by the
TimeOut property.
ExecuteWinApp Executes the Windows application specified by the Application
property.
294 Chapter 8: Enhancing Your Web Site with Third-Party Components
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Try it out! Drop to the command prompt (go to Start . Run and
enter command if you are using Windows 95/98, cmd if you are using
Windows NT or Windows 2000) and type in ping www.
4GuysFromRolla.com. This will send an echo request to the web
server 4GuysFromRolla.com runs on.
For a full list of ping’s command-line arguments, enter ping /? at
the command prompt.
Example 8-1 contains an ASP page that will execute ping through an ASP page.
The page first creates a form in which the developer can enter the name of the
computer he wishes to ping. When the developer enters a domain name (or IP or
hostname) to ping, the form reloads the page and the ASPExec component is
instantiated. Its Application property is set to ping and its Parameters property is
set to the hostname entered by the user in the form. The code in Example 8-1,
which should be saved as /AspPing.asp, is nearly identical to one of the example
ASP pages that is included with the ASPExec download.
Example 8-1. You Can Execute ping from an ASP Page Using ASPExec
<% @LANGUAGE="VBSCRIPT" %>
<% Option Explicit %>
<html>
<body>
<H3>ASPExec Ping Test</H3>
<% if Request.QueryString("host") = "" then %>

<form action="AspPing.asp" method=get>
Enter Host to Ping:
<input type=text size=45 name=host value="localhost"><P>
<input type="Submit">
</form>
<% else
Dim Executor, strResult
'Create an instance of the ASPExec object
Set Executor = Server.CreateObject("ASPExec.Execute")
'Indicate that we want to run the ping program with the hostname as the
'only parameter
Executor.Application = "ping"
Executor.Parameters = Request.QueryString("host")
'Execute ping, storing the results in strResult
strResult = Executor.ExecuteDosApp
Executing DOS and Windows Applications on the Web Server with ASPExec 295
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Figure 8-1 shows /AspPing.asp when first visited. Note that a form is displayed
where the user can enter the hostname of the computer he or she wishes to ping.
Once the user enters a hostname (or an IP or a domain name) and submits the
form, /AspPing.asp will be called again. This time, however, an instance of the
ASPExec component will be created and ping will be executed with the hostname
entered by the user as the only command-line argument. The output is returned by
the ExecuteDOSApp method and stored in strResult. This output is then displayed.
Figure 8-2 shows the output of /AspPing.asp after the user has selected a
hostname to ping.
The ASPExec download includes several examples of using ASPExec.
One example uses ASPExec to output the contents of a DOS command
(such as dir C:\). This idea can easily be extended into a
remote DOS window on a web page, which would serve as a great
administrative tool for remote webmasters. (Of course, such an
application is also a scary security risk. With the proper procedures,
however, you could guarantee that only a select set of users have
access to such a powerful administration tool.)
'Output the contents of strResult
Response.Write "<pre>" & strResult & "</pre>"
end if
%>
</body>
</html>
Figure 8-1. First, the user enters the hostname of the computer to ping
Example 8-1. You Can Execute ping from an ASP Page Using ASPExec (continued)
296 Chapter 8: Enhancing Your Web Site with Third-Party Components
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Obtaining Detailed Information About
Your Users’s Browsers
One of the difficulties in designing web pages for the Internet is your users may
be using a number of different browsers. The fierce competition between Netscape
and Microsoft has led to differing “standards.” Internet Explorer, for example, supports
client-side VBScript and ActiveX controls. Netscape, however, supports Java-
Script as the only client-side scripting language, and ActiveX controls can only be
used with a plug-in. Similarly, both Internet Explorer and Netscape have their own
protocol and extensions for Dynamic HTML and cascading style sheets.
While Internet Explorer and Netscape are by far the two most popular browsers
on the market, there are many other browsing options available. If users may visit
your site through the WebTV browser, AOL’s custom browser, lynx (a text-based
Figure 8-2. The results of ping are output
Obtaining Detailed Information About Your Users’s Browsers 297
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Unix browser), Opera, or any other non-mainstream browser, it is important that
you ensure that your site still looks pleasant and is easy to use. For example, if
you had a set of web pages that utilized DHTML and a user visited your site using
a browser that didn’t support DHTML (older versions of Internet Explorer and
Netscape, lynx, etc.), you would want to redirect the user to a set of pages that
accomplished the same tasks but refrained from using DHTML.
Using Microsoft’s Browser Capabilities Component
As mentioned in Chapter 7, Microsoft provides a free COM component to accomplish
this task: the Browser Capabilities component. This component has a couple
of drawbacks, though. For starters, to use the Browser Capabilities component, a
browser capabilities file is needed and must be updated each time a new browser
comes to market or a new browser version is released. This file, named browscap.
ini, contains information about all of the popular browser types and each
browser’s capabilities.
For the latest version of browscap.ini, visit http://www.asptracker.com.
Another disadvantage of the Browser Capabilities component is it only reports on
a small set of available browser properties. Furthermore, the Browser Capabilities
component only indicates whether or not the user’s browser supports a certain
property. For example, Internet Explorer 5.0 supports client-side JavaScript; however,
users can easily disable JavaScript support. The Browser Capabilities component,
therefore, would indicate that a visitor who was using Internet Explorer 5.0
supported client-side JavaScript, even if the user had disabled JavaScript from their
browser.
A More Versatile Solution: cyScape’s BrowserHawk
To compensate for the disadvantages of the Browser Capabilities component, you
can use BrowserHawk, a third-party component from cyScape, Inc. Like the
Browser Capabilities component, BrowserHawk maintains a list of potential
browser types and versions and their various capabilities. Unlike the Browser
Capabilities component, however, BrowserHawk will automatically check for and
download updates to this list of the various browsers’ capabilities. Furthermore,
BrowserHawk contains a much more thorough list of browser capabilities than the
Browser Capabilities component.
298 Chapter 8: Enhancing Your Web Site with Third-Party Components
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Determining the User’s Browser Information
How does a component like the Browser Capabilities component know what
browser a visitor is using?
Each time a browser requests a web page from a web site, the browser sends
along some information about itself in the request header. This information,
referred to as the User-Agent string, contains the name and version of the web
browser and the visitor’s operating system.
To access this string in your ASP page, examine the HTTP_USER-AGENT variable
in the Request.ServerVariables collection. Creating a simple ASP page with
the following code:
<%=Request.ServerVariables("HTTP_USER-AGENT")%>
I received the following output when visiting the page with an Internet
Explorer 5.0 browser:
Mozilla/4.0 (compatible; MSIE 5.0; Windows 98; DigExt)
The Browser Capabilities component takes this string and searches for an entry
in the browscap.ini file that corresponds to the browser type and version indicated
by the User-Agent HTTP header. Following this header in the browscap.
ini file is a list of capabilities supported by the particular browser. The following
is a partial list of the capabilities under the IE 5.0 header in the browscap.
ini file:
[IE 5.0]
browser=IE
Version=5.0
majorver=5
minorver=0
frames=True
tables=True
cookies=True
backgroundsounds=True
vbscript=True
javascript=True
javaapplets=True
ActiveXControls=True
Win16=False
beta=False
AOL=False
MSN=False
CDF=True
DHTML=True
XML=True
Obtaining Detailed Information About Your Users’s Browsers 299
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
A free 30-day evaluation copy of BrowserHawk can be downloaded
from cyScape’s web site at http://www.cyscape.com/products/bhawk/
start.asp. At the time this book was published, BrowserHawk offered
Standard, Professional, and Enterprise editions costing $119, $274,
and $489 USD, respectively.
Once BrowserHawk is installed, you can start using it in your ASP pages to detect
your visitor’s browser information! BrowserHawk contains an extremely lengthy
set of properties for each possible browser type and version. Table 8-3 contains a
list of some of the more useful properties. A complete list can be found online at
http://www.cyscape.com/showbrow.asp.
Table 8-3. BrowserHawk Indicates the Properties Your Users’s Browsers Support
Property Description
ActiveXControls Boolean; indicates whether or not the user’s browser supports
ActiveX controls.
Cookies Boolean; indicates whether or not the user’s browser supports
cookies.
CookiesEnabled Boolean; indicates whether or not the visitor has cookies
enabled. For browsers that accept cookies, this property checks
to ensure the user has cookies enabled. (For a browser that supports
cookies but has disabled cookies, the Cookies property
would return True, while the CookiedEnabled property would
return False.)
DHTML Boolean; indicates whether or not the user’s browser supports
DHTML.
Frames Boolean; indicates whether or not the visitor’s browser supports
frames.
Height Returns the height of the visitor’s screen resolution. For example,
a visitor who was viewing your site at 800x600 would have
a Height property of 600.
IPAddr Returns the IP address of the visitor.
JavaApplets Boolean; indicates whether or not the user’s browser supports
Java applets.
JavaEnabled Boolean; similar to the CookiesEnabled property, this property
indicates whether or not the visitor has Java applets enabled.
JavaScript Boolean; indicates whether or not the user’s browser supports
client-side JavaScript.
JavaScriptEnabled Boolean; similar to the CookiesEnabled property, this property
indicates whether or not the visitor has client-side JavaScript
enabled.
300 Chapter 8: Enhancing Your Web Site with Third-Party Components
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
The Enabled properties—CookiesEnabled, JavaEnabled, and
JavaScriptEnabled—only work with the Professional or Enterprise
editions of BrowserHawk.
Using the BrowserHawk component
Once you have run through the BrowserHawk installation process, the Browser-
Hawk DLL will be registered on your machine. You are now ready to start using
BrowserHawk in your ASP pages! To create an instance of the BrowserHawk component,
use the following lines of code:
Dim objBrowserHawkInstance
Set objBrowserHawkInstance = Server.CreateObject("cyScape.browserObj")
Before reading the Height, Width, JavaScriptEnabled, or JavaEnabled properties,
you must call the GetExtProperties method. This method sends a blank page to
the client to assess which of these properties, if any, are disabled. Likewise, before
reading the CookiesEnabled property, call the CookieDetector method. The
BrowserHawk documentation details what properties need to have a method
called before being read.
Example 8-2 demonstrates how to use the BrowserHawk component to determine
what properties your visitor supports. Note that the GetExtProperties and Cookie-
Detector methods are called prior to listing the capabilities of the user’s browser.
StyleSheets Boolean; indicates whether or not the user’s browser supports
cascading style sheets (CSS).
VBScript Boolean; indicates whether or not the user’s browser supports
client-side VBScript code.
Version Returns the version of the browser.
Width Returns the width of the visitor’s screen resolution. For example,
a visitor who was viewing your site at 800x600 would have a
Width property of 800.
XML Boolean; indicates whether or not the user’s browser supports
XML.
Example 8-2. BrowserHawk Determines What Properties Are Supported
by a Visitor’s Browser
<% @LANGUAGE="VBSCRIPT" %>
<% Option Explicit %>
<%
Dim objBHawk
Table 8-3. BrowserHawk Indicates the Properties Your Users’s Browsers Support (continued)
Property Description
Obtaining Detailed Information About Your Users’s Browsers 301
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
When the ASP page presented in Example 8-2 is visited through a browser, the
output will differ on depending what browser is used to visit the page! I visited
this page with Microsoft’s Internet Explorer 5.0; a screenshot can be seen in
Figure 8-3.
Set objBHawk = Server.CreateObject("cyScape.browserObj")
'Get the extended properties
objBHawk.GetExtProperties
'Call the cookie detector
objBHawk.CookieDetector
%>
<HTML>
<BODY>
<H1>Fun Facts About Your Browser!</H1>
Your resolution: <%=objBHawk.Width%>x<%=objBHawk.Height%><BR>
Cookies Enabled: <%=objBHawk.CookiesEnabled%><BR>
Support VBScript: <%=objBHawk.VBScript%><BR>
Support XML: <%=objBHawk.XML%><BR>
JavaScript Enabled: <%=objBHawk.JavaScriptEnabled%><BR>
Java Applets Enabled: <%=objBHawk.JavaEnabled%>
</BODY>
</HTML>
<%
Set objBHawk = Nothing 'Clean up!
%>
Figure 8-3. As you can see, I have cookies, JavaScript, and Java applets enabled
Example 8-2. BrowserHawk Determines What Properties Are Supported
by a Visitor’s Browser (continued)
302 Chapter 8: Enhancing Your Web Site with Third-Party Components
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Redirecting users based on their browser’s enabled capabilities
Web developers face a real dilemma when trying to create easy-to-use web sites.
While many new technologies have arisen to help make the web experience more
interactive (Macromedia Flash, for example, or Dynamic HTML), these add-ons are
not standard across all browsers and all versions. Therefore, many web developers
choose to create two sets of web pages. One set utilizes the advanced features—
such as cascading style sheets, Java applets, and XML—supported by the
more recent browsers. The second set of pages provides similar functionality but
uses the older technologies—vanilla HTML and simple client-side JavaScript
commands.
When a visitor first arrives at the site, he is redirected to the appropriate set of
pages, depending upon his browser. For example, the following snippet of code
could be used on the start page (/default.asp) to send the user to a certain set of
web pages based upon his browser’s support of DHTML:
<% @LANGUAGE="VBSCRIPT" %>
<% Option Explicit %>
<%
Dim objBHawk
Set objBHawk = Server.CreateObject("cyScape.browserObj")
'Redirect the user based on the browser's ability to support DHTML
If objBHawk.DHTML then
'Supports DHTML
Response.Redirect "/DHTMLSupport/default.asp"
Else
'Does not support DHTML
Response.Redirect "/NonDHTML/default.asp"
End If
%>
Then, for each ASP page in the /DHTMLSupport directory, you’d want to use a
server-side include to add the following ASP code:
<%
Dim objBHawk
Set objBHawk = Server.CreateObject("cyScape.browserObj")
'Redirect the user to the NonDHTML section if they don't support DHTML
If Not objBHawk.DHTML then
'Does not support DHTML
Response.Redirect Replace(Request.ServerVariables("SCRIPT_NAME"), _
"DHTMLSupport", "NonDHTML")
End If
%>
This snippet of code would redirect the user to the proper page in the Non-
DHTML directory if the browser didn’t support DHTML. That way, if some user
entered the URL http://www.yoursite.com/DHTMLSupport/SomePage.asp into their
Grabbing Information from Other Web Servers 303
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
non-DHTML supporting browser, they would be automatically redirected to http://
www.yoursite.com/NonDHTML/SomePage.asp. Similarly, in all of the ASP pages in
the /NonDHTML directory, you’d want to add a similar block of ASP code to the
one above that would check to see if DHTML was supported, and if so, would
redirect the user to the appropriate page in the /DHTMLSupport directory.
Grabbing Information from Other
Web Servers
One of the most frequent questions I hear from ASP developers is “How can I
grab information from another web server?” Specifically, they’re interested in the
HTML generated by a particular URL, be it a static web page or an ASP page.
There are many reasons a developer may be interested in using an ASP page to
obtain a web page fresh from another server. One of the most popular reasons is
to obtain some real-time data. For example, several of the portals maintain
updated stock quotes and weather information. Many developers would like to be
able to snatch this information and display it on their own web page.
Many sites that provide weather or stock information have the actual
temperature or quotes in a specific column and row in an HTML
table. An excellent article on 4GuysFromRolla.com, “Grabbing Table
Columns From Other Web Pages,” by Thomas Winningham, presents
an application to quickly and easily grab a particular column
and row from an HTML table residing on a different web server. The
article is available at: http://www.4guysfromrolla.com/webtech/
031000-1.shtml.
To grab information from a web page, we need look no further than to Stephen
Genusa’s excellent set of objects. The ASPHTTP object, available for $49.95 at http://
www.serverobjects.com/products.htm#asphttp, allows developers to download HTML
content from another web server. Furthermore, while using ASPHTTP, you can save
the downloaded content to a file if you so choose!
When you download the ASPHTTP DLL from ServerObjects.com, you will need to
move it to your Windows System directory and register it using regsvr32. Once it is
registered, you are ready to start using the component in your ASP pages.
ASPHTTP’s Properties and Methods
ASPHTTP contains a vast array of properties. A full listing of these properties can
be seen at: http://www.serverobjects.com/comp/asphttp3.htm. Table 8-4 contains a
listing of some of the more commonly used properties.
304 Chapter 8: Enhancing Your Web Site with Third-Party Components
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
The most common ASPHTTP method is the GetURL method, which returns the
response of the HTTP request to the URL specified by the URL property.
Retrieving a Web Page with ASPHTTP
Since ASPHTTP is designed to request data through the HTTP protocol from other
servers, let’s look at an example of doing just that. When using ASPHTTP, as with
any of the other components discussed in this chapter, you must first create an
instance of the object. To create an instance of ASPHTTP, simply use the following
two lines of code:
Dim objASPHTTPInstance
Set objASPHTTPInstance = Server.CreateObject("AspHTTP.Conn")
To retrieve the contents of an HTML page, all you need to do is set the URL property
accordingly and call the GetURL method. Example 8-3 contains ASP code that
will grab the HTML content from the LearnASP.com homepage and display the
HTML syntax.
Table 8-4. The ASPHTTP Properties
Properties Description
BinaryData Returns the data obtained by ASPHTTP in binary format. This is useful if
you are using ASPHTTP to grab binary images, such as GIFs, from other
web sites.
Headers Returns the response HTTP headers sent by the web server contacted by
ASPHTTP.
Port Indicated what port to connect to on the remote web server (defaults to
80, the standard HTTP port).
PostData Specifies the data to send through the HTTP POST.
SaveFileTo Used to save the resulting HTML or binary data to a file.
URL Specifies the URL to grab the information from (must start with http://).
Example 8-3. Grabbing the HTML of a Web Page on a Remote Web Server Is Easy with the
ASPHTTP Component
<% @LANGUAGE="VBSCRIPT" %>
<% Option Explicit %>
<%
Dim objAspHTTP, strHTML
Set objAspHTTP = Server.CreateObject("AspHTTP.Conn")
'Grab the HTML from http://www.LearnASP.com
objAspHTTP.URL = "http://www.LearnAsp.com/default.asp"
strHTML = objAspHTTP.GetURL()
%>
<HTML>
<BODY>
Grabbing Information from Other Web Servers 305
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Once the GetURL method is called, the LearnASP.com web site is contacted
through an HTTP Get, and the response is stored in the variable strHTML. Understand
that strHTML simply contains the HTML returned by /default.asp on
LearnASP.com. When the contents of strHTML are output, the HTML source is
shown due to the surrounding XMP tags. A screenshot of the code in Example 8-3,
when viewed through a browser, can be seen in Figure 8-4.
Retrieving Binary Data with ASPHTTP
ASPHTTP is not restricted to grabbing just textual HTML information from remote
servers; binary data, such as GIFs, JPGs, and ZIPs, can also be obtained by
ASPHTTP. Furthermore, with the SaveFileTo property, you can save these files to
the web server.
You can also use the SaveFileTo property to save textual HTML data
from remote servers.
Imagine that a News web site had a GIF that showed any major accidents on the
expressways in your location. This GIF was updated as accidents occurred and
were cleared, and was always stored in one location, perhaps http://www.
TrafficNews.com/Chicago/map.gif. If you wanted to provide a picture of this GIF
on your site, you could do so with a simple IMG tag, like so:
<IMG SRC="http://www.TrafficNews.com/Chicago/map.gif">
However, if your site attracts a lot of visitors during the day, the webmasters at
TrafficNews.com might not appreciate your linking directly to their image and eating
up their bandwidth. To appease these webmasters, you might decide to copy
this GIF to your web server. However, if this traffic map is updated frequently, the
copy on your web server would become quickly outdated.
<H1>Here is the HTML to display LearnASP.com!</H1>
<XMP>
<%=strHTML%>
</XMP>
</BODY>
</HTML>
<%
Set objAspHTTP = Nothing 'Clean up!
%>
Example 8-3. Grabbing the HTML of a Web Page on a Remote Web Server Is Easy with the
ASPHTTP Component (continued)
306 Chapter 8: Enhancing Your Web Site with Third-Party Components
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
The following situation would be ideal, and is quite possible using ASPHTTP:
• A user visits a web page on your server that displays the accident map. Before
displaying the map, the date and time the image was saved to the web server
are compared to the current time. If the accident map on the local web server
is less than one hour old, the user is shown that image.
• If, however, the image is more than an hour old, the current accident map is
downloaded from TrafficNews.com, the user is shown the more recent map,
and the downloaded accident map is saved to the local web server.
To accomplish this, we will use an ASP page, /images/TrafficReport.asp, that outputs
the correct traffic report image (either the local copy or the one on the
TrafficNews.com web site) in binary GIF form. Since /images/TrafficReport.asp will
Figure 8-4. The HTML source for http://www.LearnASP.com/default.asp is displayed
Grabbing Information from Other Web Servers 307
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
output the GIF in binary format, you can display the GIF from any other ASP or
HTML page using the IMG HTML tag, like so:
<IMG SRC="/images/TrafficReport.asp">
In the file /images/TrafficReport.asp, we will need to perform the following tasks:
1. Check to see if a version of the traffic report GIF exists on the local web
server. If there is no local copy, proceed to Step 4.
2. Since the file exists, compare its date last modified property to the current
time. If the file is more than an hour old, proceed to Step 4.
3. If we’ve reached this point, the traffic report GIF on the local web server is
less than one hour old, so show this GIF and end.
4. If we’ve reached this point, either a local copy of the traffic report does not
exist, or it is out of date. In either case, use ASPHTTP to grab the recent version
from TrafficNews.com, saving the GIF to the local web server.
Using ASP Pages to Provide Non-HTML Output
By default, ASP pages send HTML to the browser, which renders the output
like any other HTML page. However, ASP pages can send non-HTML output.
When sending data to a client, it is important to let the client know the MIME
type of the data so it can properly display it. The MIME (Multipurpose Internet
Mail Extensions) type lets the browser know how to display the data. HTML
has a MIME type of text/html. For example, if you send a web browser the
binary contents of a GIF file, you need to send the data with a MIME type of
image/gif.
To let the browser know what type of output you are sending, use the ContentType
property of the Response object. Try the following quick example.
Create an ASP page named /ShowImage.asp and enter the following code:
<%
Response.ContentType = "image/gif"
Response.Redirect "/images/SomeImage.gif"
%>
where /images/SomeImage.gif is a valid GIF file on your web server. Then,
from any ASP page, include the following line of HTML:
<IMG SRC="/ShowImage.asp">
When you view an ASP page with the above IMG tag, you will see the contents
of /images/SomeImage.gif !
There are a vast number of MIME types available. A rather thorough list can be
found at: ftp://ftp.isi.edu/in-notes/iana/assignments/media-types/media-types.
308 Chapter 8: Enhancing Your Web Site with Third-Party Components
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
The source code for /images/TrafficReport.asp can be seen in Example 8-4. Note
the comments in Example 8-4 illustrating the start of the four steps outlined.
Example 8-4. The Traffic Report GIF Is Only Downloaded from TrafficNews.com if the Local
Version Is Outdated or Nonexistent
<% @LANGUAGE="VBSCRIPT" %>
<% Option Explicit %>
<%
'We are going to be outputting a GIF, so set the ContentType
Response.ContentType = "image/gif"
'strVirtualName is the virtual file name of the traffic report GIF
'on the local Web server
Const strVirtualFileName = "/TrafficReport.gif"
'strFileName stores the physical file name of the traffic report GIF
Dim strFileName
strFileName = Server.MapPath(strVirtualFileName)
'Create an instance of the FileSystemObject component
Dim objFSO
Set objFSO = Server.CreateObject("Scripting.FileSystemObject")
'******** STEP 1: Determine if the file exists ***********
If objFSO.FileExists(strFileName) then
'Since the file exists, we need to check when it was created
Dim objFile
Set objFile = objFSO.GetFile(strFileName)
'********* STEP 2: Compare the local copy's DateLastModified property
with the current date/time (Now()) ******************
Dim dtDateModified
dtDateModified = objFile.DateLastModified
Set objFile = Nothing
If DateDiff("h", dtDateModified, Now()) = 0 then
'****** STEP 3: The local copy is up to date – show it to the user! ******
Response.Redirect strVirtualFileName
Response.End
End If
End If
Set objFSO = Nothing
'********** STEP 4 **************
'If we get here, the file either doesn't exist on the web server, or it
'is outdated. Use ASPHTTP to grab the current version and save it to disk.
'*********************************
Dim objAspHTTP, strHTML
Set objAspHTTP = Server.CreateObject("AspHTTP.Conn")
'Grab the traffic report banner and save it to disk
objAspHTTP.URL = "http://www.TrafficNews.com/reports/TrafficReport.gif"
Encrypting Information 309
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
In Step 4, the GIF is output using the BinaryWrite method of the
Response object and the BinaryData property of the ASPHTTP
object. This line of code could be replaced by:
Response.Redirect strVirtualFileName
Encrypting Information
Encryption is the process of taking a legible piece of information, such as a text
file or Word document, and converting it into a scrambled, unreadable version.
The process of converting the scrambled version back into a readable version is
referred to as decryption. Encryption is commonly used in sending sensitive information
over an insecure medium. Every communication medium is potentially
insecure, and the Internet’s communication channels are no exception.
Encryption algorithms commonly use a key when encrypting the information. This
key helps ensure that only the intended recipient of the message can decrypt the
message. When sending a sensitive message, the following steps should be taken:
1. Before sending the sensitive message, encrypt it into an unreadable form using
a key.
2. Send the encrypted message over the insecure communication medium.
3. The intended recipient of the message has a key that will decrypt the message,
converting it from the unreadable, scrambled form back into the readable
version that existed prior to Step 1. Once he receives this message, it is
decrypted and read.
There are several potential weak points in these three steps. The key used in Step
1 to encrypt the message is nothing more than a collection of bits (be it a number,
textual password, or some other key). These keys must be of a finite size;
objAspHTTP.SaveFileTo = strFileName
objAspHTTP.GetURL 'Get the file
'Since we grabbed binary data from TrafficNews.com, we must output it using
'the BinaryWrite method of the Response object and the BinaryData property
'of the ASPHTTP component.
Response.BinaryWrite objAspHTTP.BinaryData
Set objAspHTTP = Nothing 'Clean up
%>
Example 8-4. The Traffic Report GIF Is Only Downloaded from TrafficNews.com if the Local
Version Is Outdated or Nonexistent (continued)
310 Chapter 8: Enhancing Your Web Site with Third-Party Components
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
since they are a finite size, there is obviously a finite number of potential keys. If a
rogue developer intercepts the encrypted message, he can attempt to decrypt the
message through brute force. With such an approach, the rogue developer would
try every possible key in an attempt to decrypt the data. While there is no way to
prevent this (since there is no way to make a key infinite in size), the best thing
to do is to make the key so long that it would take the rogue developer too long
to decrypt the encrypted message.
Another potential security hole in message encryption involves the storage and
exchange of keys. The entire encryption process becomes moot if a rogue developer
can discover the intended recipient’s key. This is of special concern in symmetric
cryptography, which we’ll discuss in the next section.
For more information on cryptography basics, be sure to check out ASPEncrypt.
com’s “Crypto 101” section, available at: http://www.aspencrypt.com/crypto101_
terminology.html. In this section, we will be looking at how to perform securityrelated
tasks using ASPEncrypt, a third-party component from Persits.com. ASPEncrypt,
which includes a free version of ASPEmail, costs $249.00 for a single server
license. A free 30-day trial can be downloaded at: http://www.aspencrypt.com/
download.html.
For more information on ASPEncrypt and all of Persits’s product line,
check out http://www.persits.com.
Encrypting Information with a Key
When encrypting information, a key is used to help ensure that only the intended
recipient of the message will be able to decrypt the message. As we discussed in
the previous section, the strength of the encryption algorithm is based on the size
of the key. There are two types of key-based cryptography schemes that can be
employed: symmetric cryptography and asymmetric cryptography. These types of
cryptography differ on how they use keys to encrypt and decrypt information.
Symmetric cryptography
In symmetric cryptography, a single key is used to both encrypt and decrypt the
data. Figure 8-5 graphically illustrates the principles behind symmetric cryptography.
Both the sender and receiver of the encrypted message have to agree on the
secret key. One glaring weakness of this approach is that, at some point, the
sender has to let the receiver know what the secret key is. Since this information
Encrypting Information 311
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
can’t be encrypted (since the receiver wouldn’t have a key to decrypt it), a security
hole exists if this transfer occurs over an insecure communication medium.
The major advantage of this algorithm is it is a fast and efficient one.
Figure 8-5. Symmetric cryptography performs both encryption and decryption with a
single key
Cracking Symmetric Cryptography
As with any key-based encryption algorithm, the encryption algorithm is only
as strong as its key size. DES, a symmetric cryptographic encryption algorithm
that has been a standard since 1976, uses a 56-bit key. This key length was
more than sufficient when the algorithm was first devised, but has since
become far too short. With modern machines, such short keys can be cracked
via a brute force method within hours or days.
As keys grow larger, the time required to apply a brute-force algorithm
increases exponentially, making a brute-force approach for a single computer
a near impossibility. However, if the brute-force algorithm can be distributed
over hundreds or thousands of computers, cracking large keys might be
possible.
There is currently such a project underway at http://distributed.net. By installing
a small program on your computer, during idle periods your computer will
work on a brute-force algorithm to decrypt an encrypted message.
Encryption
Algorithm
Decryption
Algorithm
Secret Key
Why did the
chicken cross
the road? To
get to the
other side!
Plain Text
Message
lknv;f99vmn
xkjc 9039 z
..ø…*0cqfa/
k_()8 9cx´¶•
jcds==980e
Encrypted
Message
Why did the
chicken cross
the road? To
get to the
other side!
Decrypted
Message
Inputs:
the encrypted message
and the key
Inputs:
the plain text message
and the key
Output:
an encrypted
message
Output:
the plain text
message
312 Chapter 8: Enhancing Your Web Site with Third-Party Components
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Asymmetric cryptography
Asymmetric cryptography uses two mathematically related keys. These two keys
include a private key, which is never revealed, and a public key, which is available
to everyone. These two keys are related mathematically in such a way that a document
encrypted by a public key can only be decrypted by the corresponding private
key. So, as Figure 8-6 illustrates, if Person A wishes to send Person B an
encrypted message, Person A will encrypt the message with Person B’s public key.
Once this is accomplished, the only key that will decrypt the message is Person B’s
private key, which is known only by Person B. This has the advantage that anyone
can send Person B an encrypted message that only Person B can decrypt, since
everyone knows Person B’s public key and only Person B knows his private key.
The most common asymmetric cryptography approach is RSA,
named after its inventors: Ron Rivest, Adi Shamir, and Len Adleman.
RSA keys are based on the products of large prime numbers. RSA’s
strength lies in the fact that computers have an extremely difficult
time factoring large numbers.
Figure 8-6. Asymmetric cryptography utilizes mathematically related public and private keys
for encryption and decryption
Encryption
Algorithm
Decryption
Algorithm
Person B's
Public Key
Why did the
chicken cross
the road? To
get to the
other side!
Plain Text
Message
lknv;f99vmn
xkjc 9039 z
..ø…*0cqfa/
k_()8 9cx´¶•
jcds==980e
Encrypted
Message
Why did the
chicken cross
the road? To
get to the
other side!
Decrypted
Message
Inputs:
the encrypted message
and the private key
Inputs:
the plain text message
and the public key from
Person B
Output:
an encrypted
message
Output:
the plain text
message
Person A sends Person B a message
Person B's
Private Key
Encrypting Information 313
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Authentication
Encryption can be used to garble the information sent between two computers,
but how can a user be sure that when he or she receives a piece of information
such as email, it came from whomever claimed to have sent it? While authentication
may not seem vitally important, it is essential to protect from the “Man in the
Middle” attack.
In the “Man in the Middle” attack, a malicious hacker intercepts communications
between a sender and a receiver and masquerades as the sender. For example, say
that Person A was interested in obtaining some sensitive information from Person
B. In a normal, asymmetric cryptography scheme, the following events would
unfold:
1. Person A sends Person B his public key and asks for a sensitive document.
2. Person B knows that A has rights to see that sensitive document, so Person B
encrypts the sensitive document with Person A’s public key, so that only Person
A’s private key can decrypt the information.
3. Person A receives the encrypted message containing the sensitive document
and decrypts it with his private key.
This is how it’s supposed to work, but what if a hacker interjected himself into the
middle of this scenario? Imagine that the following sequence of steps occurred:
1. Person A sends Person B his public key and asks for a sensitive document.
2. The hacker intercepts this request and alters the request. The hacker replaces
Person A’s public key in the request with his own public key.
3. Person B receives the request. Thinking the request contains Person A’s public
key, the information is encrypted with the hacker’s public key. The
encrypted message is then sent to Person A.
4. That hacker again intercepts the message en route to Person A. Since the message
was encrypted with the hacker’s public key, the hacker can decrypt the
message with his private key and read the sensitive documents!
5. So as not to alert Person A to a potential security attack, the hacker can
encrypt the plain text secure document with Person A’s public key and send it
to Person A. At this point, Person A can decrypt the message with his private
key. No one except the hacker is aware that any breach in security has
occurred!
Figure 8-7 illustrates the “Man in the Middle” attack.
The reason the “Man in the Middle” attack works is because Person A and Person
B have no means to verify that the message hasn’t been altered in transit. To
314 Chapter 8: Enhancing Your Web Site with Third-Party Components
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
account for this, digital signatures are used. Digital signatures reverse the public
key/private key roles. If a sender encrypts his message using his own private key,
anyone can decrypt it with the sender’s public key. Since the sender is the only
one who has access to his private key, a message received that can be successfully
decrypted by the sender’s public key must have been sent from the sender.
Understand that digital signatures are not used to encrypt data; rather, they are
used solely for authentication. Encryption and authentication are two different
things; the former garbles the message so it is only readable by the intended recipient(
s), while the latter is used to guarantee the message was sent by the person
claiming to have sent it. There is no reason why, though, a message can’t be both
digitally signed and encrypted.
Digital Certificates
Asymmetric cryptography, which is widely used, requires a public and private key
pair for all parties interested in participating in secure communications. Who is
responsible, though, for generating these public and private keys? Obviously, if by
chance two people were using the same public or private keys, all the advantages
for the public/private key system would go flying out the window.
To overcome these problems, digital certificates are used. Digital certificates bind
an individual to a public key. Since asymmetric cryptography fails if multiple users
Figure 8-7. Without authentication, asymmetric cryptography is vulnerable to a “Man in the
Middle” attack
Hacker
The Hacker intercepts Person A's
request to Person B and replaces
Person A's public key with his own
public key.
Person A
Person B
The Hacker encrypts the message with
Person A's public key and sends it back to
Person A. No one is aware that the sensitive
document was compromised.
The Hacker intercepts the message
from Person B and can decrypt the
message with his own private key.
Person B receives the message from the
Hacker, but thinks it is from Person A.
Person B encrypts the sensitive document
using the Hacker's public key and sends
the document to Person A.
A
H
H
A
Encrypting Information 315
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
have identical public or private keys, trusted third-party companies known as
certificate authorities have arisen. The job of certificate authorities is to issue digital
certificates, thereby assigning unique public and private keys. There are many
certificate authorities available that can easily be found by searching the web for
“Certificate Authorities.” A popular certificate authority is VeriSign (http://www.
verisign.com).
For more information on digital certificates, be sure to check out
http://www.aspencrypt.com/crypto101_certs.html.
Client and server certificates
Digital certificates come for both clients (individual users) and servers (web servers,
for example). A web server that wishes to establish the Secure Socket Layer
(SSL) needs to use a server certificate. Individuals who wish to receive encrypted
documents using an asymmetric cryptography algorithm (such as encrypted email)
must have a client digital certificate set up on their computer. In the next section,
“Sending Digitally Signed, Encrypted Email with ASPEncrypt,” we’ll examine how
to set up a client certificate.
Sending Digitally Signed, Encrypted Email
with ASPEncrypt
There are many potential situations in which it would be nice to be able to send
an encrypted email. An online stock-trading web site, for example, might offer its
users an option to have their stock portfolio’s current worth emailed to them at the
close of the market each day. Sending such information in plain text email poses a
potential security threat, since anyone who can intercept the email can easily view
its contents. The email could be most likely be intercepted in one of two ways: by
some hacker who was eavesdropping on the network communications, or by a
nosy technical engineer at the user’s ISP.
To keep the portfolio information secure, one option would be for the online trading
company to send encrypted email. When encrypting email, it is important to
ensure that the email is encrypted so only the person the email is intended for can
decrypt it. Furthermore, it would be nice to encrypt the message using a standard
technology so the user receiving the email can decrypt the message without needing
to download or use any type of add-on component.
Such a scenario is possible using S/MIME Certificate-based encryption. S/MIME,
which stands for Secure/Multipurpose Internet Mail Extensions, aims at adding
316 Chapter 8: Enhancing Your Web Site with Third-Party Components
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
security measures to email messages using the MIME format. S/MIME provides for
secure email by providing two mechanisms that were discussed throughout the
past few pages:
• Authentication
• Encryption
To reiterate, authentication ensures the recipient of an email message that the
email was indeed sent by the person who claimed to send it. Encryption is used to
secure the email message, ensuring that only the person the email is intended for
can read it. To accomplish both of these tasks, S/MIME uses digital certificates.
S/MIME is currently supported by a number of popular email clients,
including Netscape Communicator, Microsoft Outlook, and Microsoft
Outlook Express. A full list of S/MIME-enabled email clients can be
found at: http://www.rsasecurity.com/standards/smime/products.html.
The RSA Security web site also contains a list of S/MIME-compatible
third-party encryption components, which can be found at: http://
www.rsasecurity.com/standards/smime/interop_center.html.
For a web server to send encrypted email to a visitor, that visitor must have a digital
certificate and must present the web server with his public key. For the web
server to digitally sign email, thereby proving to the recipient of the email that it
truly came from the source it claims to have come from, the web server must have
a digital certificate. Therefore, to just encrypt email, only the client is required to
have a certificate; to just sign email, only the web server must have a certificate; to
both encrypt and sign email, both the web server and client must have a
certificate.
In this section, we will look at how to encrypt and digitally sign email using
ASPEncrypt. For information on how to perform a number of other security-related
tasks with ASPEncrypt, be sure to check out the ASPEncrypt task list at http://www.
aspencrypt.com/tasks.html.
Besides being able to authenticate and encrypt email, ASPEncrypt
can also encrypt files on the web server, be used to implement
secure uploads and downloads of sensitive information, and encrypt
sensitive information like credit-card numbers!
Obtaining the client’s certificate
To encrypt email, you need to know the public key of the intended email recipient.
To obtain this information, you need to have access to the recipient’s client
certificate, which you can obtain in one of two ways:
Encrypting Information 317
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
1. Have the client upload their certificate.
2. Obtain the client’s certificate through the ClientCertificate property of the
Request object.
The first method is a little cumbersome for the end user, requiring them to export
their certificate to a text file and then upload that text file. The second method,
using the ClientCertificate property, is preferred, but only works over an SSL connection
(where the URL contains https://). I recommend you use the second
method if possible. The first method is error-prone, since the client has to export
his or her certificate. Furthermore, asking the client to take the time to export their
certificate seems unprofessional to me.
Since the ClientCertificate property requires the web server to use SSL, you must
first set up SSL. There are several great online articles that include examples on
how to set up SSL on IIS:
• “How to Request and Install an SSL Certificate on IIS 4.0,” found at http://www.
4guysfromrolla.com/webtech/062299-1.shtml.
• “SSL: Both Secure and Friendly: Why and How to use SSL,” found at http://
www.asptoday.com/articles/19991206.htm.
• “Untangling Web Security: Getting the Most from IIS Security,” found at http://
msdn.microsoft.com/workshop/server/iis/Websec.asp.
Obtaining Client and Server Certificates
To obtain a client or server certificate, you will need to visit one of the Certificate
Authorities, such as VeriSign. At the time of this writing, VeriSign offers a
free 60-day trail for their client certificates and a free 14-day trial for their server
certificates. After the trial period, client certificates run at $14.95 per year while
various server certificate options can cost from $395 to $1,295.
To obtain a trial client certificate, visit http://www.verisign.com/client/. For a
quick reference guide on how to set up your email client to be able to receive
encrypted email, be sure to read VeriSign’s encrypted email guide: http://www.
verisign.com/securemail/guide/. To help install a client certificate, ASPEncrypt.
com provides an easy-to-use web page at: http://www.aspencrypt.com/get_cert.
htm.
To obtain a trial server certificate, visit http://www.verisign.com/server/. The
server certificate, which is used for authentication, will also enable you to perform
secure web site communications (using SSL) from your web site.
318 Chapter 8: Enhancing Your Web Site with Third-Party Components
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
When installing SSL on your server, be sure to opt to accept client
certificates. The default Client Certificate Authentication option is not
to accept client certificates.
When a web visitor visits a SSL-enabled page that requests a client certificate, she
will be prompted to select which client certificate she’d like to present to the web
server. This certificate contains the piece of information we are after, which is the
client’s public key. Figure 8-8 contains a screenshot of the client certificate selection
dialog box.
The Request object’s ClientCertificate property is a collection containing the client
certificate information. Since the property is a collection, it can be iterated through
using a For Each ... Next loop, as Example 8-5 illustrates.
The client certificate contains information on the issuer of the certificate, the email
address of the client, the certificate’s serial number, the date the certificate is valid
until, and other related information. Figure 8-9 contains the output of the code in
Figure 8-8. When visiting an SSL-enabled web page that accepts client certificates, the visitor
must choose what certificate to present to the web site
Example 8-5. The Contents of the Client Certificate are Displayed While Iterating Through the
ClientCertificate Collection
<% @LANGUAGE="VBSCRIPT" %>
<% Option Explicit %>
<%
Dim strKey
'Loop through each key in the ClientCertificate collection
For Each strKey in Request.ClientCertificate
Response.Write "<B>" & strKey & "</B> - " & _
Request.ClientCertificate(strKey) & "<P>"
Next
%>
Encrypting Information 319
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Example 8-5 when viewed through a browser. Note that the URL in the address
bar is prefixed by the SSL-required https:// instead of the standard http://.
One bit of information contained in the client certificate, which is
used in Example 8-6 to send encrypted email, is the Certificate,
which is a binary representation of the entire certificate.
Sending encrypted email
To send encrypted email using the client’s public key, the ASPEncrypt component
can be used in conjunction with the ASPEmail component. Example 8-6 contains
an ASP page named /SendEncMail.asp that sends encrypted email using the recipient’s
public key. This page needs to accomplish three tasks:
1. Read in the client certificate.
2. Create the email message.
3. Send the encrypted email.
Figure 8-9. The client certificate contains an assortment of information
320 Chapter 8: Enhancing Your Web Site with Third-Party Components
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
These three steps are outlined with comments in the code in Example 8-6.
Example 8-6. ASPEncrypt and ASPEmail Provide for Encrypted Email Delivery
<% @LANGUAGE="VBSCRIPT" %>
<% Option Explicit %>
<%
Dim objMail, objCM, objContext, objBlob
Dim objCert, strMsg, strEmail
Set objMail = Server.CreateObject("Persits.MailSender")
Set objCM = Server.CreateObject("Persits.CryptoManager")
Set objContext = objCM.OpenContext("mycontainer", True)
' Retrieve the client certificate from the ClientCertificate property
Set objBlob = objCM.CreateBlob
'****** STEP 1: Read in the certificate **********
objBlob.Binary = Request.ClientCertificate("Certificate")
'Ensure that the client certificate was correctly received.
If Len(objBlob.Hex) > 0 Then
Set objCert = objCM.ImportCertFromBlob(objBlob)
'See if the certificate contains an email address
strEmail = objCert.Subject("E")
If strEmail <> "" Then
Set strMsg = objContext.CreateMessage
strMsg.AddRecipientCert objCert
'******** STEP 2: Create the Email Message **********
objMail.Host = "mail.yourserver.com"
objMail.Subject = "This is an Encrypted Message"
objMail.From = "Mitchell@4GuysFromRolla.com"
objMail.FromName = "Scott Mitchell"
objMail.AddAddress strEmail
objMail.Body = "This message is encrypted! Only the visitor who supplied " & _
"the certificate will be able to view this message."
'******* STEP 3: Send the encrypted email ***********
objMail.SendEncrypted strMsg
Response.Write "Message was successfully sent to " & strEmail
Else
'the certificate does not contain an email address; can't send the
'encrypted email!
Response.Write "Certificate does not contain an Email address."
End If
Else
'objBlob.Hex = 0, ergo no client certificate was presented
Response.Write "No certificate received."
End if
%>
Encrypting Information 321
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Note that the email is sent using the SendEncrypted method. This method of the
ASPEmail component encrypts the email message using the public key provided in
the client certificate before sending the email message. This encrypted email message
can only be viewed by the client who provided the client certificate and can
be displayed only in an S/MIME-compliant email program.
Only the body of the email is encrypted. The subject, however, is
not encrypted, so be sure not to place any sensitive information in
the email’s subject line.
Sending digitally signed emails
To digitally sign email, the sender of the email must have a certificate. Recall that
when a sender wishes to digitally sign email, he will use his private key to encrypt
the message. If the receiver of the email message can successfully decrypt the
message using the sender’s public key, then the message must have come from
the person claiming to have sent the message, since no one else has access to the
sender’s private key.
The purpose of digitally signing an email is not to hide the message
of the email from prying eyes, but rather to provide authentication.
This will prove the email was really sent from the person who
claimed to have sent it, and that it wasn’t tampered with in transit.
Sending digitally signed email requires the message to be encrypted with the
server certificate’s private key. This sensitive key is located in the registry, and naturally,
should not be able to be read by anyone but the owner of the certificate
(the web server). To access this registry information, you must either set the registry
permissions to allow the anonymous user to have read access or use the
ASPEncrypt object’s LogonUser method to log on as a user who has rights to read
the registry.
To send digitally signed email, we must be able to get our hands on the server
certificate, which contains the private key. Unlike reading the client’s certificate,
there is no simple mechanism to read the server certificate through an ASP page.
To get to the server certificate, we must know the server certificate’s serial
number. The certificate authority assigns a serial number that uniquely identifies
each certificate.
322 Chapter 8: Enhancing Your Web Site with Third-Party Components
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
The server certificate’s serial number can be found through the Internet Services
Manager. Start by opening the Internet Services Manager (Start . Programs .
Administrative Tools . Internet Services Manager). A list of web sites will be
brought up; right-click on the web site that will send digitally signed email and
click on Properties. Select the Directory Security tab and click on View Certificate.
This will bring up a dialog box containing the server certificate’s information. The
second tab, Details, contains the properties. Click on the Serial Number property,
and the certificate’s serial number is displayed in the text box below. Figure 8-10
shows this dialog box and the server certificate’s numerous properties.
Digitally signed email messages are created using the ASPEncrypt and ASPEmail
component in tandem. As we saw in Example 8-6, sending encrypted email utilizes
both the ASPEncrypt component and a special send method from the
ASPEmail component, SendEncrypted. Creating and sending digitally signed email
follows a similar pattern: the ASPEncrypt component must be used to read in the
server certificate’s private key and encrypt the message, and a special ASPEmail
component method (SendSigned) needs to be called to send the signed email.
Example 8-7 contains the source code to send digitally signed email using
ASPEncrypt and ASPEmail.
Sending Digitally Signed Email—A Security Risk?
Logging on as a user with permissions proves to be a security risk, in my opinion,
since the username and password must be entered into the ASP page like
so:
Set CM = Server.CreateObject("Persits.CryptoManager")
CM.LogonUser "Domain", "Administrator", "password"
While anonymous web users are not supposed to be able to see the source to
your ASP pages, there have been a couple of security holes in IIS over the years
that have revealed the source to ASP pages. For example, in ASP’s early days,
there was the ::$DATA bug, in which anyone could view the source to an ASP
page if they entered ::$DATA at the end of the URL in their browser:
http://www.someserver.com/SomeASPPage.asp::$DATA
While the ::$DATA and other similar security holes are patched quickly by
Microsoft, there is still often a period during which unscrupulous individuals
can view the source code to any of your ASP pages. While the probability that
someone may decide to view the source to the one ASP page on your entire
site that contains a superuser’s username and password is extremely low, it is
still something you should take into account before implementing this
approach.
Encrypting Information 323
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Figure 8-10. The server certificate’s serial number can be found through the Internet
Services Manager
Example 8-7. Digitally Signed Email Is Used for Authentication of the Sender
<% @LANGUAGE = "VBSCRIPT" %>
<% Option Explicit %>
<%
Dim objCM, objMail, objStore, objSignerCert, objContext, objMsg
Set objCM = Server.CreateObject("Persits.CryptoManager")
'Log on as a user who has the proper registry permissions
objCM.LogonUser "domain", "Administrator", "password"
'Grab the server certificate by serial number
Set objStore = objCM.OpenStore( "my", True )
Set objSignerCert = objStore.Certificates("60D3 2D8D D833 43DF 079B 83E9 3CA1 AEE3")
'Create the digitally signed email
Set objContext = objCM.OpenContext("my", True )
Set objMsg = objContext.CreateMessage
objMsg.SetSignerCert objSignerCert ' Specify signer certificate
'Send Signed Message
Set objMail = Server.CreateObject("Persits.MailSender")
objMail.Host = "mail.yourcompany.com"
objMail.Subject = "This is a Digitally Signed message"
objMail.From = "Mitchell@4guysfromrolla.com"
objMail.FromName = "Scott Mitchell"
objMail.AddAddress "yourname@yourcompany.com"
324 Chapter 8: Enhancing Your Web Site with Third-Party Components
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
In Example 8-7, we used the LogonUser method to read the server certificate from
the web server’s registry. Also note that we used the SendSigned method of the
ASPEmail component to send the digitally signed email. For more information on
sending digitally signed email using ASPEncrypt and ASPEmail, check out: http://
www.aspencrypt.com/task_mail.html.
Sending digitally signed, encrypted email
If you need to send an email message that contains sensitive material and you
wish to provide authentication so the user can be assured that the email came
from the person claiming to have sent it, you will need to both digitally sign and
encrypt the email message. Recall that digitally signing an email message does not
hide the email’s contents from prying eyes—it only serves to authenticate that the
email was sent by the person who claimed to have sent it. Furthermore, encryption
alone does not provide any sort of authentication of the sender.
ASPEncrypt and ASPEmail can be used together to send digitally signed, encrypted
emails. The ASPEmail component has a special method, SendSignedAndEncrypted,
that is used for sending email messages that need to be both encrypted and
signed. Details on sending signed and encrypted email are available on the ASPEncrypt
web site at: http://www.aspencrypt.com/task_mail.html. Not surprisingly, this
approach involves the union of the code to encrypt the message (shown in
Example 8-6) and the code to digitally sign an email message (shown in
Example 8-7).
Uploading Files from the Browser
to the Web Server
One of the most frequently asked questions on the ASP Messageboard (http://
www.aspmessageboard.com) is how to upload files from the client to the web
server via an ASP page. Allowing your web visitors to upload binary documents
has a number of practical applications. For example, a community site might allow
its members to upload pictures of themselves; a job-hunting site might allow its
applicants to upload their resumes in Word format; a software distribution site, like
Tucows.com, might allow developers to upload their programs for download.
Uploading binary information to a web server is possible using existing HTML
forms. To upload binary data, alter the HTML form tag so it contains an ENCTYPE
and uses the POST METHOD:
objMail.Body = "This message is digitally signed, proving its authenticity!"
objMail.SendSigned objMsg 'Send the signed message
%>
Example 8-7. Digitally Signed Email Is Used for Authentication of the Sender (continued)
Uploading Files from the Browser to the Web Server 325
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
<FORM ENCTYPE="MULTIPART/FORM-DATA"
METHOD="POST"
ACTION="URL for Form processing script">
You must explicitly set METHOD="POST" in your form tag; if you fail
to do this, your ASP page that attempts to save the uploaded file will
report an error. Also, if your visitors are using out-of-date browsers,
they may not be able to upload files to the web server, since older
browsers did not support the multipart/form-data standard.
Internet Explorer started supporting the standard in Version 3.02,
and Netscape has supported the standard since Version 2.0.
Then, to provide a text box and browse button in the form so the user can select a
file to upload, use the INPUT tag with the TYPE set to FILE:
<INPUT TYPE="FILE" NAME="FormElementName">
Clicking the Browse button will open the Choose file dialog box, from which the
user can select a local file to upload.
The multipart/form-data ENCTYPE was adopted with RFC 1867.
More detailed documentation can be found at: http://www.w3.org/
1999/07/NOTE-device-upload-19990706.
While providing the ability for the client to send a binary file up to the web server
through an HTML form is easy, saving this binary data as a file through an ASP
page is not practical without using a third-party component.
ASP pages can access files on the web server using the
FileSystemObject (FSO). This object, however, was built to handle
text files only. Accessing binary files with FSO can be impractical,
especially for large binary files, since a byte-by-byte conversion must
occur each time the FileSystemObject accesses a binary file. Therefore,
file uploads using just the FileSystemObject are quite impractical;
your best bet is to turn to a solid third-party component. For
information on accessing binary files using the FileSystemObject, be
sure to check out the “Further Reading” section.
There are a number of ASP upload components available on the Net, all of which
can be seen at: http://www.aspalliance.com/components/. For this chapter, I will be
discussing how to use Software Artisans’s SA-FileUp component. SA-FileUp, which
can be downloaded from http://www.softartisans.com/softartisans/saf.html, costs
$129 per server or $1,999 for a site license.
326 Chapter 8: Enhancing Your Web Site with Third-Party Components
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
In this section, we will look at how to upload a file from the client to the web
server and how to upload a file from the client into a database BLOB column. If
you don’t already own a copy of SA-FileUp, you will need to download and install
the evaluation copy before proceeding with the examples.
Creating a Simple “Upload a File” ASP Page
SA-FileUp is a very robust component, offering more than the simple ability to
upload a file from the client to the server. We will look at these more advanced
functionalities in a bit, but first I’d like to just go through an example of creating a
simple “Upload a File” ASP page. For this example, we will create two files:
UploadFile.htm, an HTML page using the multipart/form-data method to allow a
user to upload a local file, and SaveUploadedFile.asp, an ASP page that will save
the uploaded file to the web server.
Example 8-8 contains the HTML for UploadFile.htm. Note the use of the attributes
ENCTYPE="form/multi-part", TYPE="FILE", and METHOD="POST"; all of these
are needed to allow for the client to upload a file.
Figure 8-11 shows a screenshot of UploadFile. Note that clicking the Browse button
will open the Choose file dialog box, from which the user can select a local
file to upload.
To save the file, we will use the SA-FileUp component in SaveUploadedFile.asp. To
instantiate the component, use the following code:
Dim objSAFileUpInstance
Set objSAFileUpInstance = Server.CreateObject("SoftArtisans.FileUp")
Before we look at the source code for SaveUploadedFile.asp, let’s take a moment
to examine SA-FileUp’s properties and methods. The pertinent properties are listed
first in Table 8-5, while the pertinent methods are outlined in Table 8-6.
Example 8-8. UploadFile.htm Creates a Form from Which the User Can Select
a Local File to Upload
<HTML>
<BODY>
<FORM ENCTYPE="multipart/form-data" METHOD="POST" ACTION="SaveUploadedFile.asp">
Enter the file to upload:<BR>
<INPUT TYPE="FILE" NAME="UploadFile">
<P><INPUT TYPE="SUBMIT" VALUE="Upload File!"
</FORM>
</BODY>
</HTML>
Uploading Files from the Browser to the Web Server 327
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Figure 8-11. Using an INPUT tag with its TYPE Set to FILE creates a text box and Browse
button to aid the user in selecting a local file to upload
Table 8-5. SA-FileUp Contains Many Properties. The Following Is a Listing of the More
Commonly Used Properties.
Property Description
ContentDisposition Contains the MIME content disposition of the data submitted by
the client. This should be multipart/form-data when
uploading a file.
IsEmpty A Boolean value specifying whether or not the user selected a
file to upload.
MaxBytes Used to specify the largest possible file size that can be
uploaded; useful for preventing malicious users from eating up
your web server’s hard drive space by uploading incredibly
large files. (The default value, 0, specifies that no limit on file
size should be imposed.)
OverWriteFiles A Boolean value that specifies whether an existing file on the
web server should be overwritten if a file with the same filename
is uploaded. The default value for OverWriteFiles is
True.
Path Specifies the default directory for uploading files. If this is not
specified, the system-defined Temporary Path is used.
TotalBytes Reports the total number of bytes written to the web server
from the uploaded file. If MaxBytes is set to a value other than
0, TotalBytes will never exceed the value of MaxBytes.
UserFileName The name and path of the file on the client’s machine that was
selected for uploading.
328 Chapter 8: Enhancing Your Web Site with Third-Party Components
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
The code for SaveUploadedFile.asp, which can be seen in Example 8-9, is fairly
straightforward. We start by performing a little error checking, ensuring that the
user selected a file to upload and that his or her browser can support binary
uploads. Next, we set the Path property to C:\Temp and save the file using the
Save method. This will save the file using the same filename as the file on the client’s
machine. Therefore, if the client uploads the file C:\SomeDirectory\SomeFile.
jpg, the uploaded file will be saved on the web server at C:\Temp\SomeFile.jpg.
Since the OverWriteFiles property defaults to True, if a file named SomeFile.jpg
already exists in the C:\Temp directory, the existing file will be overwritten. If the
file is successfully uploaded, the user will see a confirmation message.
Table 8-6. SA-FileUp’s Methods Serve One Purpose: to Assist in Uploading a File from the
Client to the Server
Method Description
Delete(FileName) Deletes the file FileName from the web server.
Save Saves the uploaded file to disk in the directory specified
by the Path property and with the original file’s
filename (the value of the UserFileName property).
SaveAs(FileName) Saves an uploaded file to the web server with the filename
specified by FileName. If no path is specified in
the FileName parameter, the value of the Path property
is used.
SaveAsBlob(ADOField) Saves an uploaded file to the database. We’ll look at
how to use this method in “Uploading Binary Objects
to a Database.”
SaveInVirtual(VirtualLoc) Saves the uploaded file to the specified virtual location,
VirtualLoc.
Example 8-9. SaveUploadedFile.asp Writes the Uploaded File to the Web
Server’s Disk
<%@ LANGUAGE="VBSCRIPT" %>
<% Option Explicit %>
<%
'Create an instance of SA-FileUp
Dim objUpload
Set objUpload = Server.CreateObject("SoftArtisans.FileUp")
If objUpload.IsEmpty then
'The user did not specify a filename in the text box to upload
Response.Write "Please specify a file to upload."
Response.End
End If
'Check to make sure the multipart/form-data ENCTYPE is supported
'by the user's browser
If objUpload.ContentDisposition <> "form-data" Then
Response.Write "Your browser does not support binary uploads."
Response.End
End If
Uploading Files from the Browser to the Web Server 329
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
When saving an uploaded file to the web server, the IUSR_
machinename account must have write permissions to the directory
specified by the Path property. If the IUSR_machinename account
does not have the required permissions, an error message will be
displayed explaining the problem.
Uploading Binary Objects to a Database
Databases such as Microsoft Access and Microsoft SQL Server can store a number
of different datatypes. The more commonly used datatypes include simple scalar
datatypes, such as numeric and string variables. These databases can also store
binary objects, or BLOBs (Binary Large OBjects). For example, a table storing
employee information (Employee) might contain a column named Picture that
would be a BLOB containing a picture of the employee.
Allowing the user to upload a binary to the server and then save that binary object
to the database is fairly simple using SA-FileUp. In fact, it only requires a slight
modification to the ASP page that is responsible for saving the uploaded file to the
web server. The HTML page that provides the form for selecting a file to upload
does not need any changes.
For this example, let’s assume we have a table named Property that contains
information about various real estate properties. This table contains the following
columns:
PropertyID
A Primary Key/AutoNumber field that uniquely identifies each row in the
table.
'If we've reached this point, the file was successfully uploaded.
'Now we just need to save it to the web server!
'Set the path to save the file to, saving it as the same name as the
'name of the file uploaded by the client.
objUpload.Path = "C:\Temp"
objUpload.Save
'To save the file to the root directory in the web site, use:
'objUpload.SaveInVirtual "/challenges"
Response.Write "Your file (" & objUpload.UserFileName & _
") has been saved to the web server!"
Set objUpload = Nothing
%>
Example 8-9. SaveUploadedFile.asp Writes the Uploaded File to the Web
Server’s Disk (continued)
330 Chapter 8: Enhancing Your Web Site with Third-Party Components
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Description
A textual description of the property.
Cost
The cost of the property.
Picture
A BLOB column that stores a picture of the property. (In SQL Server, this
would have a datatype of image; in Access, this datatype would be OLE
Object.)
Example 8-10 contains the source code for UploadFileToDB.htm, an HTML page
that displays an “Add a New Property” form with three form fields: a text box for
the Description, a text box for the Cost, and a file upload text box/Browse button
to select the property image to save in the database.
To Use BLOBs or Not to Use BLOBs?
If a database table needs to have some binary information associated with it,
such as an employee’s picture, there are two methods that can be used: store
the picture in the database using a BLOB datatype, or store the image on the
web server and store just the path and filename of the image in the database.
Storing the binary object on disk and a path and filename in the database has
one primary advantage: speed. The advantage is especially pronounced if your
database and web server reside on different machines. Since this BLOB must
be transmitted to the database whenever saving the information, and retrieved
from the database whenever it needs to be downloaded or viewed, this can
serve as a bottleneck if there are many simultaneous requests for the binary
data.
However, by placing binary objects on the web server, you are limiting the
scalability of the database. For example, what if a Visual Basic application
wanted to query the database and display the images? Tough luck, since the
images are off on the web server. Or what if you had two or more database
servers spread around the world and you wanted to replicate the database
information? You’d be up a creek if you had the binary objects stored on the
web server and not in the database.
The decision to store binary data in the database or on the web server, then,
is one that depends upon the situation. While storing binary objects on the web
server will likely lead to performance increases, it adversely affects the scalability
of the database application.
Uploading Files from the Browser to the Web Server 331
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
A screenshot of the form created by the HTML code in Example 8-10 can be seen
in Figure 8-12. Note that the form indicates that only GIFs should be uploaded to
the database. This is done so we have consistent BLOBs in the database. For
example, we do not want the user to upload a ZIP file into the Picture column
of the Property table. In SaveUploadedFileToDB.asp, the form-processing script
that is responsible for adding the new row to the Property table, we’ll examine
how to ensure that the file uploaded by the client was a GIF file.
Now that we have created the HTML page that displays the upload form, we need
to create the ASP page that grabs the uploaded file and inserts it into the database
(along with the Description and Cost information). This ASP page,
SaveUploadedFileToDB.asp, is fairly similar to the simple upload file we looked at
Example 8-10. UploadFileToDB.htm: Adds a New Row to the Property Table
<HTML>
<BODY>
<H1>Add a New Property to the Database</H1>
<FORM ENCTYPE="multipart/form-data" ACTION="SaveUploadedFileToDB.asp" METHOD="POST">
Description: <INPUT TYPE=TEXT NAME=Description><BR>
Cost: $<INPUT TYPE=TEXT NAME=Cost SIZE=8><BR>
Enter the picture to upload: (<I>GIFs only!</I>)<BR>
<INPUT TYPE="FILE" NAME="UploadFile">
<P><INPUT TYPE="SUBMIT" VALUE="Add the Property"
</FORM>
</BODY>
</HTML>
Figure 8-12. Adding a new row to the property table includes entering a description and cost
and choosing a GIF file that is a picture of the property
332 Chapter 8: Enhancing Your Web Site with Third-Party Components
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
in Example 8-9, SaveUploadedFile.asp. However, instead of using the Save method
of SA-FileUp, we’ll use the SaveAsBlob method to save the image to the database.
Example 8-11 contains the source code for SaveUploadedFileToDB.asp.
Example 8-11. The SaveAsBlob Method Saves an Uploaded File to a Database
<%@ LANGUAGE="VBSCRIPT" %>
<% Option Explicit %>
<%
'Create an instance of SA-FileUp
Dim objUpload
Set objUpload = Server.CreateObject("SoftArtisans.FileUp")
If objUpload.IsEmpty then
'The user did not specify a filename in the text box to upload
Response.Write "Please specify a file to upload."
Response.End
End If
'Check to make sure the multipart/form-data ENCTYPE is supported
'by the user's browser
If objUpload.ContentDisposition <> "form-data" Then
Response.Write "Your browser does not support binary uploads."
Response.End
End If
'Make sure the image is a GIF
If UCase(Right(objUpload.UserFileName, 4)) <> ".GIF" then
Response.Write "You must upload a GIF of the property."
Response.End
End If
'Grab a snapshot of the table Property
Dim objRS
Set objRS = Server.CreateObject("ADODB.Recordset")
objRS.Open "Property", "DSN=PropertyDB", adOpenStatic, adLockOptimistic, adCmdTable
objRS.AddNew 'Add a new record
'Set the values for our three columns based upon the form values
objRS("Cost") = objUpload.Form("Cost")
objRS("Description") = objUpload.Form("Description")
objUpload.SaveAsBlob objRS("Picture")
objRS.Update 'Save the changes
objRS.Close 'Close and clean up...
Set objRS = Nothing
Response.Write "Your file (" & objUpload.UserFileName & _
") has been saved to the Web server!"
Set objUpload = Nothing
%>
Uploading Files from the Browser to the Web Server 333
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
To prevent users from uploading non-GIF files for the property’s picture,
a quick check is done on the uploaded file’s filename. If the
last four characters are not equal to .gif, the user is shown a message
indicating that they can only upload GIF files for the property’s
picture. A more thorough technique for validating that a given file is
indeed a GIF file can be seen at: http://www.4guysfromrolla.com/
webtech/050300-1.shtml.
Displaying a property’s pictures
Displaying the BLOB object in the Property table is not exactly straightforward. It
would be nice to be able to do:
objRS.Open "Property", "DSN=PropertyDB"
'Display the picture
Response.Write objRS("Picture")
However, displaying binary objects it is not quite that simple. To display the
Picture column, we’ll use a separate ASP page, one called /images/
ShowPropertyPicture.asp, whose sole purpose is to display the proper GIF from
the Property table. This page is strikingly similar to the /images/TrafficReport.asp
Reading Textual Form Information Using SA-FileUp
When submitting a form that contains both textual information (such as the
Description and Cost fields in UploadFileToDB.htm) and binary information
(such as the file we are uploading), the textual data can be picked out in the
form-processing script using the Form collection of the SA-FileUp object. For
example, when setting the Description and Cost columns of the Recordset
object, we read in the user’s entries for those two form variables using:
objUpload.Form("Cost")
and:
objUpload.Form("Description")
Due to the fact that the SA-FileUp component uses a Request.BinaryRead
call to obtain the posted form information, you cannot use the Request.Form
collection in the form-processing script. (If you attempt to use the Request.Form
collection, you’ll receive an error: “Cannot use Request.Form collection after
calling BinaryRead.”) Rather, use the SA-FileUp component’s Form collection.
334 Chapter 8: Enhancing Your Web Site with Third-Party Components
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
page we looked at in Example 8-4. As with /images/TrafficReport.asp, in /images/
ShowPropertyPicture.asp, we will use the Response.ContentType property to let
the browser know we are sending it a GIF. We will then use the BinaryWrite
method of the Response object to send the binary content of the BLOB column to
the browser.
/images/ShowPropertyPicture.asp will show a single property’s picture. To specify
what property’s picture to display, /images/ShowPropertyPicture.asp should be
passed the value of the property’s PropertyID. Example 8-12 contains the source
code for /images/ShowPropertyPicture.asp.
To view a particular property’s image, use an IMG tag. For example, the following
snippet of code would display a property’s picture:
<IMG SRC="/images/ShowPropertyPicture.asp?PropertyID=ValidPropertyID">
Example 8-13 contains the source to DisplayProperties.asp, which will display the
Description and Cost information about all of the properties available, along with a
picture of the property.
Example 8-12. /images/ShowPropertyPicture.asp Sends the Binary Content of the Picture
Column to the Client, Instructing to the Browser to Interpret the Binary Content as a GIF
<% @LANGUAGE="VBSCRIPT" %>
<% Option Explicit %>
<%
'Get the passed-in property ID
Dim iPropertyID
iPropertyID = Request("PropertyID")
'Retrieve exactly one row from the Property table.
Dim strSQL
strSQL = "SELECT Picture FROM Property WHERE PropertyID = " & iPropertyID
Dim objRS
Set objRS = Server.CreateObject("ADODB.Recordset")
objRS.Open strSQL, "DSN=PropertyDB"
'Only display if we found a row
If Not objRS.EOF then
Response.ContentType = "image/gif" 'Tell the browser we are sending it a GIF
Response.BinaryWrite objRS(0) 'Send the BLOB to the client
End If
'Clean Up...
objRS.Close
Set objRS = Nothing
%>
Why Reinvent the Wheel? 335
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Figure 8-13 contains a screenshot of a browser visiting DisplayProperties.asp.
Other SA-FileUp Features
SA-FileUp is a very robust upload component and offers many features not outlined
in this chapter. Some of its especially interesting capabilities include:
• Transactional uploads
• Uploading a file with a progress indicator
• The option to upload an entire directory and all of its files and subdirectories
from the client
A full list of SA-FileUp’s features can be seen at: http://www.softartisans.com/
softartisans/comcom.html.
Why Reinvent the Wheel?
Throughout this chapter, we’ve looked at various third-party components that can
greatly enhance your IIS/ASP web site. While most of these components come at a
cost, it is almost guaranteed to be much lower than the cost in man-hours needed
to develop a similar component inhouse.
Example 8-13. To Display the Picture for a Particular Property, Use an IMG Tag
<%@ LANGUAGE="VBSCRIPT" %>
<% Option Explicit %>
<%
Dim objRS
'Open a snapshot of the Property table
Set objRS = Server.CreateObject("ADODB.Recordset")
objRS.Open "Property", "DSN=PropertyDB"
'Loop through each record in the Property table
Do While Not objRS.EOF
Response.Write "<TABLE ALIGN=CENTER BORDER=0 WIDTH=""80%"">"
Response.Write "<TR><TD><B>" & objRS("Description") & "</B></TD><TD>"
Response.Write FormatCurrency(objRS("Cost"), 0) & "</TD></TR>"
Response.Write "<TR><TD ALIGN=CENTER COLSPAN=2>"
Response.Write "<IMG SRC=""ShowImage.asp?PropertyID=" & objRS("PropertyID") & """>
"
Response.Write "</TD></TR>"
Response.Write "</TABLE><P><HR><P>"
objRS.MoveNext
Loop
'Clean up...
objRS.Close
Set objRS = Nothing
%>
336 Chapter 8: Enhancing Your Web Site with Third-Party Components
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Since its Introduction, this book has promoted code reuse repeatedly for several
reasons:
• Code reuse decreases the time needed to complete a project.
• Code reuse decreases the time needed to complete future projects.
• Code reuse leads to fewer bugs, since the reused code is more thoroughly
tested than newly written code (and it has already been checked time and
time again for typos and silly mistakes).
Figure 8-13. Each property’s description and cost is listed, along with a picture of the property
Further Reading 337
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
COM components, if they are Microsoft-created, third-party created, or developed
in-house, are shining examples of code reuse and are guaranteed to save you time
in development and debugging. My advice: if you can find a third-party COM
component that can meet your needs, by all means, use it!
The only disadvantage of using third-party or Microsoft COM objects
is you don’t have access to the source. This, in my opinion, has two
disadvantages:
• Sometimes it’s nice to be able to see under the hood of the car.
With third-party COM components, you are in the dark as to
how, exactly, they work. While this black-box implementation
has its advantages in encapsulating the implementation complexity,
there are still those times when you need to know what
makes the component tick. For example, use of the Request
object’s QueryString and Form collections is not available when
using SA-FileUp, since SA-FileUp issues a BinaryRead. Since SAFileUp
is a compiled COM component, we cannot see the code
that issues the BinaryRead. While the SA-FileUp help files thoroughly
explain why you can’t use the QueryString and Form
collections on a web page that uses SA-FileUp, it would be nice
to be able to dissect the code ourselves, rather than having to
rely on the help file.
• It’s impossible to reuse COM component code. Perhaps a small
portion of a COM object could be used in a future project. With
VBScript classes (or COM objects developed in-house), you can
take advantage of the existing COM component’s source code,
creating a new component that reuses the needed section of the
source code form the existing component.
Further Reading
With the explosive growth in the popularity of Active Server Pages, an entire component-
development industry has arisen. This industry is a testament to ASP’s ability
to quickly and easily create powerful, dynamic web sites. The links below are
to lists of available third-party components and numerous third-party vendors. If
you need a specific tool for your web site, say a threaded message board system,
check out these lists and vendors to quickly find a complete third-party COM
object that fits your needs.
• For a list of available third-party ASP COM components, check out the component
list on ASPAlliance.com: http://www.aspalliance.com/components/.
• If you are looking for an ASP e-commerce component, look no further than
iisCart: http://www.iiscart2000.com.
338 Chapter 8: Enhancing Your Web Site with Third-Party Components
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
• Doug Dean Consulting builds a number of COM objects that can be used in
an ASP web site. If you are in the market for a threaded forum, calendar, or
web-based memo system, check out: http://www.dougdean.com.
• ASPdb is a component designed to aid in publishing database information on
the Web. Easily display table information in a number of different formats and
styles: http://www.aspdb.com/.
• Looking to add a chat system to your site? Why write your own when you can
use ASPChat? It’s available from ASPHelp.com at http://www.asphelp.com/
ASPChat/ ?
• Offer your users the option to spellcheck their inputs; check out http://www.
xde.net/spellchecker/.
• While the FileSystemObject was developed to work exclusively with text files,
it can be used to access binary files as well. Check out http://www.pstruh.cz/
help/ScptUtl/library.htm to learn how to upload files using just ASP code and
the FileSystemObject!
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
339
Index
<% %> code block delimiters, 22
& bitwise operator, 53
<< bitwise operator, 53
>> bitwise operator, 53
| bitwise operator, 53
~ bitwise operator, 53
^ (caret), bitwise operator, 53
@ symbol, in ASP directives, 20
@ symbol (PerlScript array), 32
% symbol (PerlScript associative array), 32
$ symbol (PerlScript scalar variable), 25
A
ACTION property (FORM tag), 107, 124
Active Server Pages (see ASP)
ActiveX Data Objects (ADO), 142, 249
enumerated constants, 143–145
schemas, 142, 145–149
ActiveX Scripting Engine, 19, 20
Ad Rotator, GetAdvertisement method, 255
Ad Rotator component, 250–267
Redirection file, 252–253, 257–258
Schedule file, 252–255
AddEmployee method, source code
for, 286
AdInfo database table, 257
administration pages
creating page forms, 185–213
“Delete an Existing Record”, 202
“Edit an Existing Record”, 208
“Insert a New Record”, 194–201, 181
interfaces, 138
maximizing code reuse, 228
reusable, 150
AdminPageAction.asp, 214, 217
AdminPageGenerator class, 186–201
member variables, 186
methods, 190
properties, 186
AdminPageMenu class, 229, 234–241
ADO (see ActiveX Data Objects)
adovbs.inc, 143
AdRotation class, 261
methods, 263
adSchemaPrimaryKeys schema,
opening, 162
adSchemaTables, 148
AND operator, 53
applications, design and development
of, 1–2
ASP (Active Server Pages)
code block delimiters (<% and %>), 22
COM, using, 247–290
downloadable source code, x
information on web, ix–xi
pages (see ASP pages)
Release 3.0, exception handling, 58
scripting language, choosing, 19, 24–38
scripts, defining classes, 88
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
340 Index
ASP pages
code reuse, 105
server-side validation, 116
design of, 3
DisplayAdminPage.asp, 242
for displaying GIFs from a database, 333
execution on the server of DOS and
Windows applications, 292
form creation pages, 106–110
form processing scripts, 106–110
form validation, 111–124
for getting content from other
servers, 303
information displayed in, 176
for processing forms, 108
for provision of non-HTML output to
client browsers, 307
ASP upload components, 325
ASPEmail components, 319
SendEncrypted method, 321
ASPEncrypt component, 315–324
range of uses, 316
ASPError object, 30, 42, 57–66
Category property, 62
error reporting, 69
instance of, 60
properties, 58, 62
ASPExec component, 292
command-line program execution, 293
properties and methods, 292
security concerns, 295
ASPHTTP component, 303–309
installing, 303
instantiating, 304
methods, 304
properties, 303
asymmetric cryptography, 312
authentication, 313–314, 321
of email from server, 321
AutoNumber columns, 152, 187
determining, 162
B
banners
impression tracking with Ad
Rotator, 259
statistics, tracking, 257
binary data
client-to-server upload, 324
retrieval with ASPHTTP, 305
Binary Large Objects (see BLOBs)
binary objects, displaying, 333
bitwise operators, 53
bitwise shift-left operator, 53
bitwise shift-right operator, 53
BLOBs (Binary Large Objects), uploading to
a database, 329
browscap.ini, 297, 298
Browser Capabilities component, 297
vs. BrowserHawk, 297
collecting information with, 298
BrowserHawk, 297–303
vs. Browser Capabilities
component, 297
instantiating, 300
browsers
client-side scripting languages,
support, 296
determining configuration settings, 302
properties, 298
request header information, 298
standards, 296
business logic, 5
implementation through ASP pages, 276
C
case-sensitivity, 26
CDATA tag, 286
CDONTS (Collaborative Data Objects for
NT Server), 99
for error reporting, 67
certificate authorities, 315, 317
CheckForError subroutine, 50
for error reporting, 67
class developers, 88
Class statement (VBScript), 89
classes
AdminPageGenerator, 186–201
AdminPageMenu, 229, 234–241
AdRotation, 261
ColumnInformation, 155
Columns, 155
DataTypeFormElementInformation, 176
for databases, 138–246
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Index 341
for form creation and validation, 125
form field element objects, 125
GenerateForm, 128–135
in VBScript, 15, 88
Release 5.0, 78
MenuItem, 229–234, 239
ModifyDatabase, 214
using in server-side form
validation, 121–124
Clear method, 46
click-throughs, 257
client certificates, obtaining, 317
client-side form validation, 111–112
client-side scripting languages, 296
code reuse
administration pages, 228
advantages of, 104
in ASP pages, 105
databases, 105, 138–246
exception handling with subroutines, 50
forms, 124
for server-side form validation, 116
with third-party components, 337
collections, use of properties and
methods, 109
column information, collecting, 155
ColumnInformation class, 155
testing, 165
Columns class, 155, 157
member variable, 158
COM (Component Object Model), 248
inconsistent exceptions and, 47
using in Active Server Pages, 247–290
COM components
advantages of, 279
categories of, 276
compiled vs. interpreted
components, 289
creating with Windows Script
Components Wizard, 280–290
for determining browser capabilities of
users accessing web pages, 297
disadvantage of, 280
for encapsulating database and business
logic, 278
enhancement with non-portable
methods, 267, 275
as examples of code reuse, 337
included with IIS, ASP, 250
third-party, 291
compile-time errors, 40, 63
ComputeTotalWithSalesTax function, 5–7
consistent exceptions, 47
constants
ADO, 143
vbObjectError, 46
constructors, 88, 89
Content Linker, 250, 267–275
list file for, 267, 269, 273
control flow with Server.Execute, 16
CreateList, helper function, 203–208
cryptography, 309–324
asymmetric, 312
symmetric, 310
custom HTTP error pages, creating, 74
D
databases
access to backup while primary is
unavailable, 71
classes, 138–246
code reuse for access to, 105
connecting to with private member
variables, 187
data model for, 150
displaying BLOB datatypes, 333
enumerations in ADO (ActiveX Data
Objects), 143
insertions, deletions, and updates using
an ASP page, 214
requirements for administration
pages, 140
reusable code, 138–246
web pages for administrating, 139
DataTypeFormElement property, 180, 184
DataTypeFormElementInformation
class, 176, 187
member variables, 180
testing, 183
datatypes
JScript, 25
mapping to form elements, 176
setting with DataTypeFormElement, 184
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
342 Index
declarations, JScript variables, 25
decryption, 309
default scripting language, 21
“Delete an Existing Record”, 139
DeleteRecordForm
methods, 203
parameters, 203
DeleteRecords method
ModifyDatabase class, 217, 226
multiple deletions, 228
DES algorithm, 311
design
Active Server Pages, 1–18
scripts, 2
web pages, 2
destructors, 89
digital certificates, 314
digital signatures, 314
digital signing of email, 321
Dim statement, for defining public
properties, 91
directory structure for storage of ASP
modules, 7
displaying information in ASP pages, 176
displaying non-HTML output with
MIME, 307
displaying tables with foreign key
constraints, 192
DLL’s, for server-side includes, 9
dollar sign ($), 25
DOS applications, execution on the server
via an ASP page, 292
dynamic evaluation and execution
security risks, 101
support in VBScript, 78
E
“Edit an Existing Record” option, 139
“Edit an Existing Record” administration
page, 208
“Edit an Existing Record” form, 222
email
acquiring client’s certificate for
encryption, 316
authenticating with the server certificate,
security concerns, 322
digital signing, 321
encrypting
ASPEmail component for, 319
with client’s public key, 319
encryption, 309
of email, 319
using S/MIME, 315
ENCTYPE, for binary uploads, 324
end developer, 88
Err object
methods, 46
Number property, 46
properties in VBScript, 45
error detection, 51
error handling
enabling with VBScript, 44
(see also exception handling)
error messages, 41
error numbers, hexadecimal formatting, 52
error-detection, 42
errors
ASPError object, catching with, 30
compile, 40
halting, 40
in programming, 39
non-halting, 40
runtime, 40
syntactical, 40
(see also exceptions)
event handlers, 89
Columns class, 158
ForeignKeys class, 171
ModifyDatabase class, 215
exception handling, 39–76
AdRotation class, 264
for ASP pages using functions and
subroutines, 48
enabling and disabling scripting
language error handling, 71
error messages, 66
JScript, 51–57
support team notification, 67
(see also error handling)
exceptions, 41
consistent and inconsistent, 47
error messages for, 41
(see also errors)
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Index 343
Execute method, RegExp object, 81, 85
Execute statement, 102
ExecuteDOSApp, 293
ExecuteWinApp, 293
ExecuteWinAppAndWait, 293
external COM object errors, 64
F
file checking, verification of format on
upload, 333
File extension mapping, 9
file naming
Content Linker list file, 270
rotator schedule files, 254
files, uploading from client to server, 324
FileSystemObject
dynamic includes, 11
limitations for binary uploads, 325
filetypes, for server-side includes, 8
foreign key columns, management in
administration pages, 168
foreign key constraints, 153
ForeignKeyExists method, 171
ForeignKeyInformation class, 170
testing, 174
ForeignKeys class, 171
testing, 174
Form collection, 108
form creation web page, 106
form elements, 107
defining datatypes with classes, 176
standardization, 124
form field NAME value, use of quotes
in, 118
form interfaces, for database access, 138
form processing script, 106
form reuse, 105
FORM tag, 107
form validation, 111–124
client-side, 111
server-side, 113
FormElement class, 125
FormHint method, AdminPageGenerator
class, 192
forms
code reuse, 124
creating, 106
formatting of entered data, 117
validation, server-side, 28
G
GenerateForm class, 128–135
GET method, 107
GetAdvertisement method, Ad Rotator, 255
GetColumnInformation method, 158
GetCurrentPrimaryKeys method, 163
GetForeignKeyInformation method, 171
GetLastError method, for retrieval of
ASPError object instance, 60
GetPrimaryKeys method, 163
GetURL method, ASPHTTP
component, 304
GIF files, obtaining from other servers, 305
Global property, RegExp object, 79
H
halting errors, 40
helper function, CreateList, 203–208
HIDDEN form fields, passed to
AdminPageAction.asp, 218
Home Directory tab, Internet Services
Manager, 21
HTML
forms, 107
non-HTML output, displaying, 307
SCRIPT blocks, 23
HTTP errors
automated handling, 59
codes list on Web, 57
I
IgnoreCase property, RegExp object, 79
IIS (Internet Information Server), 24
creating custom error pages, 74
default scripting language, 21
file-extension mapping, 9
specifying handling of HTTP errors, 59
implementation errors, 39
importing one ASP into another, 8
impressions, banner advertisements, 257
include directive
cyclical includes and, 8
file and virtual keywords, 7
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
344 Index
include directive (continued)
operation, 9–10
rules for using, 7
syntax, 7
include files
file extensions, 15
naming conventions, 12
nesting, 8
security, 14
(see also server-side includes)
inconsistent exceptions, 47
inheritance, 88
Initialize event handlers, 89
AdminPageGenerator class, 188
INPUT tags, 108
“Insert a New Record”, 139, 194–201
InsertRecord method, ModifyDatabase
class, 217–222
instances, 86
instantiation
ASPHTTP component, 304
BrowserHawk component, 300
Content Linker, 270
Windows Script Components, 288
internal ASP errors, 64
Internet Information Server (see IIS)
iterating through collections, 109
J
JavaScript, 25
JPG files, obtaining from other servers, 305
JScript
ASP example, 30
bitwise operators, 53
case-sensitivity, 26
creating ASP pages in, 25–31
datatypes, 25
error detection, 51
error handling vs. ASP Error object, 42
Error object, 52
exception handling, 51–57
vs. VBScript, 57
regular expressions in, 27–29
script block, use of, 23
statement termination, 25
throw statement, 54
try ... catch blocks, 51
variables, 25
variants, 25
vs. VBScript, 19
K
key length, security concerns, 311
keys (cryptography), 309
L
language directive, 20
LANGUAGE parameter, 22
logical errors, 39
LogonUser method, ASPEncrypt object, 321
M
“Man in the Middle” attacks, 313
match method (JScript String object), 27
member functions, 90
member variables, 90
AdminPageGenerator class, 186
Columns class, 158
DataTypeFormElementInformation
class, 177, 180
ForeignKeyInformation class, 170
ForeignKeys class, 171
MenuItem class, 229–234, 239
METADATA tag, Global.asa, 143
METHOD property (FORM tag), 107
methods, 90
AdminPageGenerator class, 190
AdRotation class, 263
ASPExec component, 292
ASPHTTP component, 304
Columns class, 159
Content Linker list file, 270
DeleteRecordForm, 203
ForeignKeys class, 171
GenerateForm class, 128
GetLastError, 60
MenuItem class, 230–234
RegExp object, 81
SA-FileUp component, 326
setting for Windows Script
Components, 282
VBScript, 86
VBScript classes, 90
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Index 345
MIME (Multipurpose Internet Mail
Extensions), 307
ModifyDatabase class, 214
DeleteRecords method, 217, 226
InsertRecord method, 217–222
UpdateRecord method, 217, 224
UpdateSelect method, 222
N
NAME property (FORM tag), 124
NAME property, of FORM tag, 107
NaN (Not a Number), 26
non-halting errors, 40
non-HTML output, displaying, 307
NOT operator, 53
Number property (Err object), 46
O
objColumnDict variable, 158
objConn, 166
objDataTypeDict, 177–180
objDataTypeFormElements member
variable, 187
object instances, assigning to properties
with Property Set statement, 98
object-oriented programming, 86
objects, 86
objFKDict, 171
objFormDescriptionDict, 177, 180
objFormHints, 187
objFormNameDict, 177, 180
On Error Goto statement, 44
On Error Resume Next statement, 44, 48
Option Explicit statement, 6, 25
OPTION tags, 108
OR operator, 53
P
parameters
DeleteRecordForm, 203
for OpenSchema, 147
PopulateColumns method, 166
Pattern property, RegExp object, 79
PerlScript, creating ASP pages with, 31–37
Personal Web Server (see PWS)
ping, 293–295
placeholders, 179
PopulateColumns method, 159–163
parameters, 166
PopulateForeignKeys method, 171
POST method, 107
for binary data uploads, 324
primary keys (database), 152
private key (cryptography), 312
Private keyword, 91
private properties, retrieval with Property
Get statement, 92
properties, 86, 90
AdminPageGenerator class, 186
AdRotation class, 261
ASPError object, 58
ASPExec component, 292
ASPHTTP component, 303
Columns class, 158
ForeignKeys class, 171
FORM tag, 107
FormElement class, 125
JScript Error object, 52
MenuItem class, 229
ModifyDatabase class, 215
SA-FileUp component, 326
setting for Windows Script
Components, 282
of VBScript Err object, 45
of VBScript’s RegExp object, 79
Property Get statement, 158
arglist, 94
Property Let statement, 94
Property Set statement, 97
argument list, 98
public key, 312
Public keyword, 91
PWS (Personal Web Server)
default scripting langage, 21
VBScript as default scripting
language, 24
Python, creating ASP pages with, 37
Q
QueryString collection, 108
QueryType, 145
QuotedValue, private member
function, 163
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
346 Index
R
Raise method, 46
Redirect property, 188
redirection file, 251, 256
redirection of users based on browser
capabilities, 302
referential integrity, 154
RegExp object
methods, 81, 85
properties, 79
syntax, 79–81
regular expressions
form validation and, 115, 28
in JScript, 27–29
in PerlScript, 35
in VBScript, 78
RegularExpression in FormElement
class, 127
replace method (JScript String object), 27
Replace method, RegExp object, 81, 85
Request object, ClientCertificate
property, 317
reusable administration pages, 150
reusable database classes, 138–246
RSA keys, 312
runtime errors, 40
S
SA-FileUp, 325–335
properties and methods, 326
sales tax, computing (example code), 5
SaveFileTo property of ASPHTTP, 305
SchemaEnums, 145
schemas (ADO), 142
opening, 145–149
SCRIPT blocks, 22–24
script design, 3
SCRIPT tag (see SCRIPT blocks)
scripting engines, downloads, 51
scripting language, 19–38
choosing, 24
default for ASP, 21
in errors, 62
JScript, 25–31
PerlScript, 31–37
Python, 37
specifying, 20–24
VBScript, 19–24
scripts, advantages of good design, 2
search method (JScript String object), 27
Secure Socket Layer (SSL), 315
setup, 317
Secure/Multipurpose Internet Mail
Extensions (see S/MIME)
security
ASPExec component, 295
client-side form validation, 112
digitally signed email from the
server, 322
dynamic evaluation and execution, 101
server-side form validation, 120
SELECT tags, 108
SendEncrypted method, 321
SendOffSite method, AdRotation class, 263
server certificate, obtaining, 317
Server.Execute method, 5, 12, 16
server-side form validation, 113
server-side includes, 5–15
CheckForError subroutine, 50
declarations of variables, 6
dynamic, 10–12
filetypes, 8
for importing adovbs.inc into ASP
pages, 143
importing one ASP into another, 8
(see also include files)
Server.Transfer method, 5, 17
ShowAd function, 260
ShowAd method, AdRotation class, 263
SIZE property, FormElement class, 127
S/MIME Certificate-based encryption, 315
source code
for AddEmployee method, Windows
Script Component, 286
format,Windows Script
Components, 284
special characters in regular expression
patterns, VBScript, 79–81
SSI (see server-side includes)
statement termination, JScript, 25
static evaluation and static execution, 99
strAction constant, 214
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
Index 347
strColumnName, 170
String object, regular expression matching
methods, 27
symbol (JScript statement terminator), 25
symmetric cryptography, 310
syntactical errors, 40
sysobjects, 142
T
tags
CDATA, 286
Windows Script Components, 284
Terminate event handler, 89
AdminPageGenerator class, 188
Test method, RegExp object, 81, 85
testing
ColumnInformation class, 165
DataTypeFormElementInformation
class, 183
ForeignKeyInformation class, 174
ForeignKeys class, 174
MenuItem class, 234
TEXTAREA tags, 108
third-party components, disadvantages
of, 337
three-tiered web sites, 278
throw statement, 54
tilde-delimited strings as placeholders, 179
tree structures, representing web site
layouts, 268
try ... catch blocks
nesting, 55–57
using for error detection in JScript, 51
TYPE property, FormElement class, 127
U
UpdateRecord method, ModifyDatabase
class, 217, 224
UpdateRecordForm, 211
UpdateRecordList, 210
UpdateSelect method, ModifyDatabase
class, 214–227
UploadFile.htm, HTML component for
file-uploading ASP page, 326
uploading binary objects to a database, 329
uploading files from client to server, 324
URLs, specifying non-relative, 252
User-Agent string, 298
V
ValidateForm.asp, 116–124
VALUE property of the INPUT tag,
FormElement class, 127
values, assigning to private properties
with Property Let statement, 94
var keyword (JScript), 25
variables
JScript, 25
PerlScript, 32–33
variants, 25
vbObjectError, constant, 46
VBScript, 19–24
Boolean values and, 79
case-sensitivity vs. JScript, 26
classes, 15, 88, 90
enabling error handling, 44
Err object, 45
error handling vs. ASP Error object, 42
exception handling vs. JScript, 57
vs. JScript, 19
methods, 90
object-oriented programming, 85–99
properties, 90
RegExp object properties, 79
Release 5.0, new features, 78
script block, use of, 23
virtual keyword, 7
Visual Basic, downloadable source code for
ASP pages, x
W
web browsers (see browsers)
web page design, 2
accomodation of different browsers, 302
impact of browser standards on, 296
web pages
for database administration, 139
displaying content from other
servers, 303
error-handling, 42
use of forms for, 106
Windows applications, executing on server
via ASP page, 292
This is the Title of the Book, eMatter Edition
Copyright © 2000 O’Reilly & Associates, Inc. All rights reserved.
348 Index
Windows Script Component Wizard, 281
Windows Script Components (WSC), 280–290
performance vs. compiled
COM objects, 289
write-only properties, 99
X
XMP tag, 184
XOR operator, 53
Z
ZIP files, obtaining from other servers, 305