Click here to Skip to main content
65,938 articles
CodeProject is changing. Read more.
Articles
(untagged)

Beginning ASP.NET using VB.NET

0.00/5 (No votes)
3 Sep 2001 1  
This chapter will provide you with a step-by-step introduction to ASP.NET using VB.NET.
Sample Image
Title Beginning ASP.NET using VB.NET
Authors Rob Birdwell, Ollie Cornes, Chris Goode, John Kauffman, Ajoy Kristnamoorthy, Juan T Llibre, Christopher L Miller, Neil Raybould, David Sussman, Chris Ullman
Publisher Wrox
Published Aug 2001
ISBN 1861005040
Price US 39.99
Pages 1000

Getting Started With ASP.NET

ASP.NET is a new and powerful technology for writing dynamic web pages. It's a convergence of two major Microsoft technologie's, Active Server Pages (ASP) and .NET. ASP is a relative old-timer on the web computing circuit and has provided a sturdy, fast, and effective way of creative dynamic web pages for more than five years now. .NET is the new kid on the block and is a whole suite of technologies designed by Microsoft with the aim of revolutionizing the way in which all programming development is conducted in the future and the way companies carry out business. Therefore, as a conjunction of the two, ASP.NET is a way of creating dynamic web pages while making use of the innovations present in .NET.

The first great thing to note about ASP.NET is that you don't need to know anything about ASP to get started. All you need is a little bit of HTML knowledge for building your own web pages, and you're away! ASP.NET is a more powerful technology than its older namesake, not only can it produce dynamic web pages, but it can tailor them to the browser you are using. Better still, it comes complete with a wide range of predefined controls ready for you to use in your own projects, saving you time, and making you more productive.

So, what can you do with ASP.NET? It'd probably be quicker to list what you can't! One of the most eye-catching new innovations is the way you can create your applications, so you ca write them in VB.NET, Jscript, C# (a new Java-like language from Microsoft), or even a combination of them all � you'll choose the best language for the job, or whichever language compliments your skills best.

Within your applications ASP.NET allows you to customise pages for a particular user, keep track of user's details as they move through a website and store information about them in a database of self-describing XML file. You can alter the layout of pages at the click of a button, add and remove files from your machines (if you have the appropriate permissions) and even draw on the logic of other applications without having to download them first.

 

In this first chapter we'll be mainly concerned with ASP.NET's installation process. We'll start with a quick introduction to the world of web servers, dynamic web pages, and a little bit about what ASP.NET is, but what we really aim to achieve is to get you running a fully functional web server, with a fully functional ASP.NET installation. By the end of the chapter you'll have created a short ASP.NET test page, to check that both the web server and ASP.NET are working as intended. Don't worry we'll have a look at some of the most common pitfalls encountered, just in case things don't go as planned!

The topics to be discussed are:

  • Static Web Pages
  • Dynamic Web Pages
  • An overview of the different technologies for creating dynamic web pages, including ASP.NET
  • Installing Internet Information Services (IIS)
  • Installing the .NET Framework
  • Testing and Troubleshooting your installation

What Is A Static Web Page?

If you surf around the Internet today, you'll see that there are lots of static web pages out there. What do we mean by a static web page? Essentially, it's a page whose content consists of some HTML code that was typed directly into a text editor and saved as an .htm or .html file. Thus, the author of the page has already completely determined the exact content of the page, in HTML, at some time before any user visits the page.

Static web pages are often quite easy to spot; sometimes you can pick them out by just looking at the content of the page. The content (text, images, hyperlinks, etc.) and appearance of a static web page is always the same � regardless of who visits the page, or when they visit, or how they arrive at the page, or any other factors.

For example, suppose we create a page called Welcome.htm for our website, by writing some simple HTML like this:

<hmtl>
<head><title>A Welcome Message</title></head>
<body>
  <h1>Welcome</h1>
  Welcome to our humble web site. Please feel free to view our
  <a HREF="contents.htm">list of contents</a>.
  <br><br>
  If you have any difficulties, you can
  <a href="mailto:webmaster@wrox.com">send e mail to the webmaster</a>.
</body>
</html>

Whenever any client comes to our site to view this page, it will look like this. The content of the page was determined before the request was made � at the time the webmaster saved the .htm file to disk:

How Are Static Web Pages Served?

Ok, so let's think for a moment about how a static, pure-HTML page finds its way onto a client browser:

  1. A web author writes a page composed of pure HTML, and saves it within an .htm file on the server

  2. Sometime later, a user types a page request into their browser, and the request is passed from the browser to the web server

  3. The web server locates the .htm page and converts it to an HTML stream 
  4. The web server sends the HTML stream back across the network to the browser
  5. The browser processes the HTML and displays the page

Static, pure-HTML files like Welcome.htm make perfectly serviceable web pages. We can even spruce up the presentation and usability of such pages by adding more HTML to create frames and tables. However, there's only so much we can achieve by writing pure HTML, precisely because their content is completely determined before the page is ever requested.

The Limitations of Static Web Pages

For example, suppose we want to enhance our Welcome page � so that it displays the current time or a special message that is personalized for each user. These are simple ambitions, but they are impossible to achieve using HTML alone. If you're not convinced, try writing a piece of HTML for a web page that displays the current time, like this:

As you type in the HTML, you'll soon realize the problem � you know that the user will request the page sometime, but you don't know what the time will be when they do so! Hard-coding the time into your HTML will result in a page that will lways claim that the time is the same (and will almost always display the wrong time).

In other words, trying to write pure HTML for a web page that displays the time � but you can't be sure of the exact time that the web page should display until the time the page is requested. It can't be done using HTML alone.

Also HTML offers no features for personalizing your web pages, each web page that is served is the same for every user. There's also no security with HTML, the code is there for everybody to view, and there's nothing to stop you from copying somebody else's HTML code and using it in your own web page. Static pages might be very fast, as quick as copying a small file over a network, but they are quite limited without any dynamic features.

Since we can't create our page by saving our hard-coded HTML into a file before the page is requested, what we need is a way to generate the HTML after the page is requested. There are two ways of doing this; we'll look at them both now. Before we go any further, we need to make sure everybody is up to speed on the terminology we've introduced here.

What is a Web Server?

A web server is a piece of software that manages web pages and makes them available to 'client' browsers � via a local network or over the Internet. In the case of the Internet, the web server and browser are usually on two different machines, possibly many miles apart. However, in a more local situation, we might set up a machine that runs the web server software, and then use a browser on the same machine to look at its web pages. It makes no difference whether we access a remote web server (that is, a web server on a different machine to our browser application) or a local one (web server and browser on the same machine), since the web server's function � to make web pages available to all � remains unchanged. It might well be that you are the only person with access to our web server on your own machine, as would be case if you were running a web server from our home machine. Nevertheless, the principles remain the same.

While there are many web servers available (the commonest ones being Apache, IIS and Iplanet's Enterprise server) we're only going to talk about one in this book Microsoft's IIS 5. This is because it is the only web server that will run ASP.NET. The web server comes as part of the installation for both Windows 2000 and Windows XP. IIS version 5.0 comes with Windows 2000, and IIS version 5.1 with Windows XP; however, there is very little to distinguish the two, and we shall treat them in this chapter as the same product. We'll look at how we go about installing IIS shortly; however first let's take a look at its role in helping to create dynamic web pages.

How are Dynamic Web Pages Served?

To fully understand the nature of dynamic web pages, we first need to look at the limitations of what we can and can't do with a static web page.

Two Ways of providing Dynamic Web Page Content

Even though we're only going to be creating dynamic web pages in this book using one of these methods, you need to be aware of the two different ways of doing it, as the underlying principles for both feature heavily throughout the book.

Client-Side Dynamic Web Pages

In the client-side model, modules (or plug-ins) attached to the browser do all the work of creating dynamic pages. The HTML code is typically sent to the browser along with a separate file containing a set of instructions, which is referenced from within the HTML page. However, it is also quite common to find these instructions intermingled with the HTML codes. The browser then uses them to generate pure HTML for the page when the user requests the page � in other words, the page is generated dynamically on request. This produces a HTML page, which is sent back to the browser.

So, in this model, our set of five steps now becomes six:

  1. A web author writes a set of instructions for creating HTML, and saves it within an .htm file. The author also writes a set of instructions in a different language. This might be contained within the .htm file, or within a separate file.

  2. Sometime later, a user types a page request into their browser, and the request is passed from the browser to the web server.

  3. The web server locates the .htm page, and may also have to locate a second file that contains the instructions.

  4. The web server sends both the newly created HTML stream and instructions back across the network to the browser.

  5. A module within the browser processes the instructions and returns it as HTML within the .htm page � only one page is returned, even if two were requested.
  6. The HTML is then processed by the browser which displays the page

Client-side technologies have fallen out of favor in recent times, as they take a long time to download, especially if we have to download a second file with a separate set of instructions. In some cases, we might have to download several files of separate instructions. A second drawback is that each browser interprets these instructions in different ways, so we have no way of guaranteeing that if Internet Explorer understands them, whether Netscape Navigator or Opera will. An other major drawbacks are that it is a problem to write client-side code that uses server-side resources  such as databases, because it is interpreted at client-side. Also all code for client-side scripting is available to everybody, which can be undesirable.

Server-Side Dynamic Web Pages

With the server-side model, the HTML source is sent to the web server with an intermingled set of instructions. Again this set of instructions will be used to generate HTML for the page at the time the user requests the page. Once again, the page is generated dynamically upon request. Our set of five steps once more becomes six, however, with the subtle twist regarding where the processing of instructions is done:

  1. A web author writes a set of instructions for creating HTML, and saves these instructions within a file

  2. Sometime later, a user types a page request into their browser, and the request is passed from the browser to the web server

  3. The web server locates the file of instructions

  4. The web server follows the instructions in order to create a stream of HTML

  5. The web server sends the newly created HTML stream back across the network to the browser

  6. The browser processes the HTML and displays the page

The twist is that all the processing is done on the server, before the page is sent back to the browser. One of the key advantages this has over the client-side model is that only the HTML code describing the finished page is actually sent to the browser. This means that our page's logic is hidden away on the server, and that we can safely assume that most browsers should be able to at least have a go at displaying it. ASP.NET as you might have gathered, follows the server-side model.

In fact either process of serving a dynamic web page is only slightly different from the process of serving a static web page � there's just one extra step involved (Step 5 on the client or Step 4 on the server). But in both cases this difference is crucial � the HTML that defines the web page is not generated until after the web page has been requested. For example, we can use either technique to write a set of instructions for creating a page that displays the current time:

<html>
<head><title>The Punctual Web Server</title></head>
<body>
  <h1>Welcome</h1>
  In Webserverland, the time is exactly
  <INSTRUCTION: write HTML to display the current time>
</body>
</html>

In this case, we can compose most of the page using pure HTML. It's just that we can't hard-code the current time. Instead, we can write a special code (which would replace the highlighted line here) that instructs the web server to generate that bit of HTML during Step 5 on the client, or Step 4 on the server, at the time the page is requested. We'll return to this example later in the chapter, and we'll see how to write the highlighted instruction using ASP.NET.

Now we're going to look at the various different technologies, including ASP.NET and see how the logic is supported in each.

An Overview of the Technologies

We just seen that there are also two distinct models for providing dynamic content. ASP.NET falls into the server-side model. However, we're going to look at what we consider to be the most important technologies in both models, as we will reference some of the client-side models in later chapters, particularly if we mention old style ASP. Not all of the technologies work in the same way as ASP.NET, but they all allow the user to achieve the same end-result � that of dynamic web applications. If ASP.NET is not an ideal solution to your problems, then you might want to consider these following technologies, taking into account the following questions:

  • Are they supported on the platform you use?
  • Are they difficult to learn?
  • Are they easy to maintain?
  • Do they have a long-term future?
  • Do they have extra capabilities, such as being able to parse XML?
  • Are a lot of people already using them � are there a lot of tools available?
  • Are the support, skills, and knowledge required to use them readily available?

We're now going to give a quick overview of what each one does, and in doing so, try to give you an idea of where ASP.NET (and the ASP technology that preceded it) fits in to the big picture.

Client-Side Technologies For Providing Dynamic Content

Each of these technologies relies on a module (or plug-in) built into the browser to process the instructions we talked about earlier. The client-side technologies are a mishmash of scripting languages, controls, and fully fledged programming languages.

JavaScript

JavaScript is the original browser scripting language, and is not to be confused with Java. Java is a complete application programming language in its own right. Netscape had originally developed a scripting language, known as LiveScript, to add interactivity to their web server and browser range. It was introduced in the release of the Netscape 2 browser, when Netscape joined forces with Sun and in the process, they changed its name to JavaScript. JavaScript borrows some of its syntax and basic structures from Java (which in turn borrowed ideas from C), but has a different purpose � and evolved from different origins (LiveScript was developed separately to Java).

For example, while JavaScript can control browser behavior and content, it isn't capable of controlling features such as file handling. In fact, JavaScript is actively prevented from doing this for security reasons. Think about it: you wouldn't want a web page capable of deleting files on your hard drive, now would you? Meanwhile, Java can't control the browser as a whole, but it can do graphics and perform network and threading functions.

Javascript is much easier to learn than Java. It is designed to create small, efficient, applications that can do many things, from performing repetitive tasks, to handling events generated by the user (such as mouse clicks, keyboard responses, and so on).

Microsoft introduced their own version of JavaScript, known as JScript, in Internet Explorer 3.0 and have supported it ever since right up to, and including IE6. It has only minor differences from the Netscape version of the language, although in older versions of both browsers, the differences were originally quite a lot wider.

VBScript

In Internet Explorer 3.0, Microsoft also introduced their own scripting language, VBScript, which was based on their Visual Basic programming language. VBScript was intended to be a direct competitor to JavaScript. In terms of functionality, there isn't much difference between the two; it's more a matter of personal preference � VBScript has a similarly reduced functionality. Visual Basic developers sometimes prefer VBScript because VBScript is, for the most part, a subset of Microsoft's Visual Basic language. However, it enjoys one advantage that makes it more attractive to novice programmers, in that, unlike JavaScript, it isn't case-sensitive and is therefore less fussy about the particulars of the code. Although this "advantage", makes it a lot slower and less efficient.

The biggest drawback is that there isn't a single non-Microsoft browser that supports VBScript for client-side scripting. For a short while there were some proprietary plug-ins for Netscape that provided VBScript support, but these never took off. You'll find that JavaScript is much more widely used and supported. If you want to do client-side scripting of web pages on the Internet then JavaScript is the only language of choice. Indeed Microsoft themselves have replaced VBScript in their .NET framework, with VB.NET. VBScript should only be considered when working on Intranet pages where it is known that all clients are IE on Windows.

With both JavaScript and VBScript there is a module, known as a script engine, built into the browser that dynamically processes the instructions, or script, as it is known in this case.

ActiveX Controls

An ActiveX control is a self-contained program (or component), written in a language such as C++ or Visual Basic. When added to a web page, it provides a specific piece of client-side functionality, such as a bar chart, timer, client authentication, or database access. ActiveX controls are added to HTML pages via the <object> tag, which is now part of the HTML standard. ActiveX controls can be executed by the browser when they are embedded in a web page.

There is a catch. ActiveX controls were developed by Microsoft, and despite being compatible with the HTML standard, they are not supported on any Netscape browser prior to version 6 (which, at time of writing, was still in beta) without an ActiveX plug-in. Without this, they will only function on Internet Explorer. Also, unlike VBScript, ActiveX is able to manipulate items on the user's machine such as the files or Windows registry. For this reason it is very often considered a security risk and is not even allowed through firewalls. Consequently, ActiveX controls still can't really be considered either a common or a cross-platform way of making your pages dynamic and are falling out of use.

Java applets

Java is a cross-platform language for developing applications. When Java first hit the Web in the mid-1990s, it created a tremendous stir. The idea is to use Java code in the form of applets, which are essentially Java components that can be easily inserted into web pages with the aid of the <applet> tag.

Java enjoys better functionality than scripting languages, offering better capabilities in areas such as graphic functions and file handling. Java is able to provide these powerful features without compromising security because the applets run in what is known as a sandbox � which prevents a malicious program downloaded from the web from doing damage to your system. Java also boasts strong database support through JDBC.

Microsoft and Netscape browsers both have built-in Java support via something known as the Java Virtual Machine (JVM), and there are several standard <object> and non-standard <applet> tags that are used to add Java applets to a web page. These tags tell the browser to download a Java file from a server and execute it with the Java Virtual Machine built into the browser. Of course, this extra step in the web page building phase means that Java applets can take a little while to download, and can take even longer to process once on the browser. So, while smaller Java applets (that provide features such as drop-down menus and animations) are very popular on the Web, larger ones are still not as widespread as scripted pages.

Although the popularity of Java today isn't quite what some people expected, it makes an ideal teaching tool for people wishing to break out into more complex languages; and its versatility makes it well suited for programming web applications.

Curl

A very recent innovation comes from a company partly set up by Tim Berners-Lee (the innovator behind the Web and the HTML language). Curl is another programming language like Java, but unlike Java, where a second file (or more) has to be downloaded with the HTML file, it completely replaces the HTML source and the Java files. It relies on a Curl plug-in having been installed on your browser first, and currently only works on very recent browsers. The advantage are that the download time is faster than Java, and also you don't have to worry about integrating different languages into the page, as Curl is capable of providing the same features as both Java and JavaScript.

Curl is still in the very early stages of development, although the first version has been released, and more details can be obtained at http://www.curl.com.

Server-Side Technologies For Providing Dynamic Content

Each of these technologies relies on a modular attachment added onto the web server rather than the browser. Consequently, only HTML, and any client-side script, is sent back to the browser by the web server. In other words, none of the server-side code is sent back. Server-side technologies have a more consistent look and feel than client-side ones, and it doesn't take that much extra learning to move between some of the server-side technologies (excepting CGI).

CGI

The Common Gateway Interface (CGI) is a mechanism for creating scripts on the server, which can then be used to create dynamic web applications. CGI is a module that is added to the web server. It has been around for quite a bit longer than even ASP, and right now, a large proportion of dynamically created web pages are created using CGI and a scripting language. However, it's incorrect to assume that CGI does the same job as ASP.NET or ASP. Rather, CGI allows the user to invoke another program (such as a Perl script) on the web server to create the dynamic web page, and the role of CGI is to pass the user supplied data to the this program for processing. However, it does provide the same end result � a dynamic web application.

You should be aware that CGI has some severe shortcomings:.

  • It is not easy for a beginner to learn how to program such modules.

  • CGI requires a lot of server resources, especially in a multiuser situation.

    It adds an extra step to our server�side model of creating dynamic content: namely, it's necessary to run a CGI program to create the dynamic page, before the page is processed on the server.

    What's more, the format in which CGI receives and transmits data means that the data is not easily manipulated by many programming languages, so you need one with good facilities for manipulating text and communicating with other software. The most able programming languages that can work on any operating system for doing this are C, C++ and Perl. While they can adequately do the job for us, they are some of the more complex languages to learn. Visual Basic doesn't offer adequate text handling facilities, and is therefore rarely used with CGI.

    Despite this, CGI is still very popular with many big web sites, particularly those running on UNIX operating systems. It also runs on many different platforms, which will ensure its continued popularity.

    ASP

    Active Server Pages (ASP) is now dubbed "Classic ASP" and if you see this term in the book, we will be using it to describe any ASP that isn't ASP.NET. ASP commonly relied on either of the JavaScript or VBScript scripting languages (although it was also possible to use any scripting language installed on Windows, such as PerlScript) to create dynamic web pages. ASP is a module (the asp.dll file) that you attach to your web server, and it then processes the JavaScript/VBScript on the web server, and turns it into HTML, before sending it into the server, rather than doing it on the browser.

    ASP lets us use practically any of the functionality provided by Windows, such as database access, e-mailing, graphics, networking, and system functions, and all from within a typical ASP page. However, ASP's shortcomings are that it is very, very slow performance wise. It is also restricted to using only scripting languages. It can't do all the things that a fully-fledged programming language can. Secondly, the scripting languages, being like "junior" versions of full programming languages, took a lot of shortcuts to make the language smaller. Some of these shortcuts make their programs longer and more complicated than is otherwise necessary. As we're going to see, ASP.NET rectifies a lot of this by making code more structured, easier to understand, and shorter.

    JSP

    JavaServer Pages (JSP) is a technology that allows you to combine markup (HTML or XML) with Java code to dynamically generate web pages. The JSP specification is implemented by several web servers, as opposed to ASP which is only supported under IIS, and plug-ins are available that allow you to use JSP with IIS 4.0/5.x. One of the main advantages of JSP is the portability of code between different servers. JSP is also very powerful, faster than ASP, and instantly familiar to Java programmers. It allows the Java program to leverage the aspects of the Java2 platform such as JavaBeans and the Java 2 libraries. JavaServer Pages isn't directly related ASP, but it does boast the ability to embed Java code into your web pages using server-side tags. More details can be found at the official site at http://www.javasoft.com/products/jsp/index.html and at the JSP FAQ at http://www.esperanto.org.nz/jsp/jspfaq.html.

    ColdFusion

    ColdFusion (http://www.macromedia.com/software/coldfusion/) also enables servers to access data as the server builds an HTML page. ColdFusion is a module installed onto your web server. Like ASP, ColdFusion pages are readable by any browser. ColdFusion also utilizes a proprietary set of tags, which are processed by the ColdFusion Server software. This server software can run on multiple platforms, including IIS, Netscape Enterprise Server and Unix/Apache. The major difference is that while ASP.NET solutions are built primarily with programming languages and objects, ColdFusion utilizes HTML-like tags, which encapsulate functionality. A drawback is that the ColdFusion software doesn't come for free and indeed you could find yourself paying well in excess of a thousand dollars for the privilege of running Cold Fusion on your web server.

    PHP

    PHP (originally Personal Home Pages, but more recently PHP HyperText Preprocessor) is another scripting language for creating dynamic web pages. When a visitor opens the page, the server processes the PHP commands and then sends the results to the visitor's browser, just as with ASP.NET or ColdFusion. Unlike ASP.NET or ColdFusion, however, PHP is open-source and cross-platform. PHP runs on Windows NT and many Unix versions, and it can be built as an Apache module and as a binary that can run as a CGI. When built as an Apache module, PHP is especially speedy. A downside is that you have to download PHP separately and go through a series of quite complex steps to install it and get it working on your machine. Also PHP's session management was non-existent until PHP 4, and still is even now,inferior to ASP's even now.

    PHP's language syntax is similar to C and Perl. This might prove a barrier to people with no prior programming experience, but if you have a background in either language then you might want to take a look. PHP also has some rudimentary object-oriented features, providing a helpful way to organize and encapsulate your code. You can find more information about PHP at http://www.php.net.

    ASP.NET

    So why are you telling me about all these other technologies if we're only going to be learning about ASP.NET you might be wondering? Hopefully you'll see a similarity between the technologies, and this will aid your understanding of ASP.NET.

    ASP.NET also relies on a module attached to the web server. However, the ASP.NET module (which is a physical file called aspnet_isapi.dll) doesn't do all of the work itself; it passes some on to the .NET Framework to do the processing for it. Rather than going into ASP.NET in this subsection here, it's time to start talking about it as a separate entity in its own right, as this is the focus of the book.

    What is ASP.NET?

    We're going to be asking this question a lot throughout the book, and each time we ask it, we're going to give you a slightly more in-depth answer. If we were we to give you a full answer now, you'd be overwhelmed by as-yet meaningless jargon. So, you'll probably be aware of some unanswered questions each time we describe it.

    Our original definition, right at the very start of the chapter, was "ASP.NET is a new and powerful technology for creating dynamic web pages", and this still holds true. However, as you now know, it isn't the only way to deliver dynamic web pages, so let's refine our definition a little to read:

    A new and powerful server-side technology for creating dynamic web pages.

    Secondly, ASP.NET isn't the only thing that we're interested in. In fact, it's one of a set of technologies that comprise the Microsoft .NET Framework. For now, you can think of this as a giant toolkit for creating all sorts of applications, and in particular, for creating applications on the Web. When we come to install ASP.NET, we will also be installing the .NET Framework at the same time, and we'll be using bits and pieces of the .NET Framework throughout the book.

    How does ASP.NET differ from ASP?

    Steady on! We're just getting to this part. ASP, as we've said is restricted to using scripting languages, mainly JavaScript or VBScript (although it can be any scripting language supported by the Windows system). We add ASP code to our pages in the same way as we do client-side script, and this leads to problems such as messy coding and restricted functionality. ASP.NET has no such problems.

    First off ASP.NET allows you to use a far greater selection of full programming languages and also allows you to utilize to the full the rich potential of the .NET Framework.  It helps you create faster, more reliable dynamic web pages with any of the programming languages supported by the .NET Framework. Typical languages supported natively are VB.NET, C# and a new version of JavaScript called JScript.NET. On top of this it is expected that third party developers will create versions of Perl, Python, and many others to work in ASP.NET. And no, before you ask, we don't expect you to be know any of these programming languages. We're going to choose one language, VB.NET, and teach you ASP.NET with it. We've chosen VB.NET as it's arguably the simplest for beginners, and it can do pretty much anything that the other languages we mentioned can as well. Lastly, and most importantly we've chosen VB.NET as it comes free with ASP.NET � so when you install ASP.NET you get VB.NET as well.

    At this stage, you might be thinking, "Hang on, I've got to figure out VB.NET, then I've got to get a handle on ASP.NET � that sounds like an awful lot to learn." Don't worry; you won't be learning two languages. ASP.NET, as we said right from the beginning, is not a language � it is a technology. This technology is accessible, via a programming language. What we're going to be doing is teaching you ASP.NET features as we teach you VB.NET. So in other words, you will be creating your web pages using VB.NET and using ASP.NET to drive it. However, before you rush out and get a VB.NET book instead, we will be approaching the language from the angle of creating dynamic web pages only.

    In summation, ASP.NET is a server-side technology that lets us use fully-fledged programming languages to create your web pages.

    I'm still confused about ASP, ASP.NET, and VB.NET

    It's really important to get these terms separate and distinct in your mind, so before we move on to actually installing and running ASP.NET, we're going to go back and redefine them just to make sure:

    • ASP � a server-side technology for creating dynamic web pages that only lets you use scripting languages
    • ASP.NET � a server-side technology for creating dynamic web pages that lets you use any fully-fledged programming language supported by .NET
    • VB.NET � our chosen programming language for writing code in ASP.NET

    Now it's time to get it all installed.

    The Installation Process

    Installation is going to be done in three steps. We're going to install the web server first, next, we're going to install the prerequisites required for ASP.NET to work, and then, lastly, we're going to install ASP.NET Premium Edition or .NET Framework SDK (which also contains ASP.NET).

    SDK stands for Software Development Kit, and the only real difference with SDK's is the huge amounts of extra documentation and examples they supply.

    Anybody who is familiar with ASP might be used to it being installed automatically with the web server, and thereby doing it all in one step. This is true � classic ASP is still installed with the web server, however ASP.NET currently is only available as a separate download. This means you will have to download ASP.NET from Microsoft's web site or from CD (if you have one). However, before you can install ASP.NET, it is necessary to have a working web server.

    If you have installed IIS 5.x already, or have installed either the Windows 2000 Server or Advanced Server operating system, then the good news is that you can skip this section, and go straight onto the section about installing the .NET Framework. However for the rest of us, you will have to pay careful attention to the next section.

    Installing the IIS 5.x Web Server

    We'll look at the installation process for IIS on Windows 2000 Professional and Windows XP Professional together, as they don't differ significantly. The main difference is that Windows 2000 installs IIS 5.0, while Windows XP installs IIS 5.1. The options for installing are exactly the same, the only thing that might differ is the look of the dialog boxes. However, the options you need to select are still the same.

    Before you install it though, it's worth noting that we might not have to do much in this initial stage, as it's possible you're already running IIS 5.x. We'll describe a process for checking whether this is the case as part of the installation process. You should also note that to install anything (not just ASP.NET, but literally anything) on Windows 2000/XP you need to be logged in as a user with administrative rights. If you're uncertain of how to do this, we suggest you consult your Windows documentation. Right let's get started!

    Try It Out � Locating and/or installing IIS 5.x on my Web Server machine

    1. Go to the control panel (Start | Settings | Control Panel) and select the Add/Remove Programs icon. The following dialog will appear, displaying a list of your currently installed programs:

    2. Select the Add/Remove Windows Components icon on the left side of the dialog, to get to the screen that allows you to install new windows components:

    3. Locate the Internet Information Services (IIS) entry in the dialog, and note the checkbox that appears to its left. Unless you installed Windows 2000 via a custom install and specifically requested IIS, it's most likely that the checkbox will be unchecked (as shown above).

    4. If the checkbox is cleared, then check the checkbox and click on Next to load Internet Information Services 5.x. You might be prompted to place your Windows 2000/XP installation disk into our CD-ROM drive. It will take a few minutes to complete. Then go to Step 5.

      OR

      If the checkbox is checked then you won't need to install the IIS 5.x component � it's already present on your machine. Go to Step 6 instead.

    5. Click on the Details button � this will take you to the dialog shown below. There are a few options here, for the installation of various optional bits of functionality. For example, if the World Wide Web Server option is checked then our IIS installation will be able to serve and manage web pages and applications. If you're planning to use FrontPage 2000 or Visual InterDev to write your web page code, then you'll need to ensure that the FrontPage 2000 Server Extensions checkbox is checked. The Internet Information Services Snap-In is also very desirable, as you'll see later in the chapter, so ensure that this is checked too:

      For the purpose of this installation, make sure all the checkboxes in this dialog are checked; then click on OK to return to the previous dialog.

    6. There's one other component that we'll need to install, for use later in this book � it's the Script Debugger. If you scroll to the foot of the Windows Components Wizard dialog that we showed above, you'll find a checkbox for Script Debugger. If it isn't already checked, check it now and click on Next to complete the installation. Otherwise, if both IIS 5.x and the script debugger are already present, you can click on Cancel to abort the process:

    How It Works

    Web Services starts up automatically as soon as your installation is complete, and thereafter, whenever you boot up Windows � so you don't need to run any further startup programs, or click on any short-cuts as you would to start up Word or Excel.

    IIS installs most of its bits and pieces on your hard drive, under the \WinNT\system32\inetsrv directory; however, more interesting to us at the moment is the \InetPub directory that is also created at this time. This directory contains subdirectories that will provide the home for the web page files that we create.

    If you expand the InetPub directory, you'll find that it contains several subdirectories:

  • \iissamples\homepage contains some example classic ASP pages

  • \iissamples\sdk contains a set of subdirectories that hold classic ASP pages which demonstrate the various classic ASP objects and components.

  • \scripts is an empty directory, where ASP.NET programs can be stored.

  • \webpub is also empty. This is a 'special' virtual directory, used for publishing files via the Publish wizard. Note that this directory only exists if you are using Windows 2000 Professional Edition.

    \wwwroot is the top of the tree for your web site (or web sites). This should be your default web directory. It also contains a number of subdirectories, which contain various bits and pieces of IIS. This directory is generally used to contain subdirectories which hold the pages that make up our web site � although, in fact, there's no reason why you can't store your pages elsewhere. We'll be discussing the relationship between physical and virtual directories later in this chapter.

  • \ftproot, \mailroot and \nntproot should form the top of the tree for any sites that use FTP, mail, or news services, if installed.

  • In some versions of Windows , you will find an \AdminScripts folder which contains various VBScript files for performing some common "housekeeping" tasks on the web server, allowing you to stop and start services.

    Working with IIS

    Having installed IIS web server software onto our machine, we'll need some means of administering its contents and settings. In this section, we'll meet the user interface that is provided by IIS 5.x.

    In fact, some versions of IIS 5.x provide two user interfaces the MMC, and the PWS interface. We're only going to look at one, as the other version is now obsolete. The version we will use is the Microsoft Management Console (MMC) that is a generic way of managing all sorts of services. Let's take a quick look at it now.

    The Microsoft Management Console (MMC)

    The beauty of the MMC is that it provides a central interface for administrating all sorts of services that are installed on your machine. We can use it to administrate IIS � but in fact, when we use it to administrate other services the interface looks roughly the same. The MMC is provided as part of the Windows 2000 operating system � in fact, the MMC also comes with older Windows server operating systems.

    The MMC itself is just a shell � on its own, it doesn't do much at all. If we want to use it to administer a service, we have to add a snap-in for that service. The good news is that IIS 5.x has its own snap-in. Whenever you need to administer IIS, you can simply call up the Internet Services Manager MMC console by selecting Start |Control Panel |Administrative Tools |Internet Services Manager.

    Having opened the IIS snap-in within the MMC, you can perform all of your web management tasks from this window. The properties of the web site are accessible via the Default Web Site node. We'll be using the MMC more a little later in the chapter.

    Testing the Installation

    The next thing to do is test the web server to see if it is working correctly, and serving pages as it should be. We've already noted that the web services should start as soon as IIS has been installed, and will restart every time you start your machine. In this section, we'll try that out.

    In order to test the web server, we'll start up a browser and try to view some web pages that we know are already placed on the web server. In order to do that, we'll need to type a URL (Uniform Resource Locator)into the browser's Address box, as we often do when browsing on the Internet. The URL is an http://... web page address which indicates which web server to connect to, and the page we want to view.

    What URL do we use in order to browse to our web server? If your web server and web browser are connected by a local area network, or if you're using a single machine for both web server and browser, then it should be enough to specify the name of the web server machine in the URL.

    Identifying your Web Server's Name

    By default, IIS will take the name of your web server from the name of the computer. You can change this in the machine's network settings. If you haven't set one, then Windows will generate one automatically � note that this automatic name won't be terribly friendly; probably something  along the lines of "P77RTQ7881". To find the name of your own web server machine, select Start | Settings | Network and Dial-up Connections, and from the Advanced menu select Network Identification. The Network Identification tab will display your machine name under the description Full computer name:

    My machine has the name chrisu, and (as you can see here, and in the earlier screenshot of the MMC dialog) my web server has adopted the same name. Browsing to pages on this machine across a local area network (or, indeed, from the same machine), I can use a URL that begins http://chrisu/� 

    There are a couple of alternatives if you're using the same machine as both web server and browser. Try http://127.0.0.1/� � here, 127.0.0.1 is the loopback address that causes requests to be sent to a web server on the local machine. Alternatively, try http://localhost/� � 'localhost' is an alias for the 127.0.0.1 address � you may need to check the LAN settings (in your browser's options) to ensure that local browsing is not through a proxy server.

    Throughout the book, in any examples that require you to specify a web server name, the server name will be shown as localhost, implicitly assuming that your web server and browser are being run on the same machine. If they reside on different machines, then you simply need to substitute the computer name of the appropriate web server machine.

    Browsing to a Page on your Web Server

    Now you know the name of your web server, and that web services are running; you can view some classic ASP pages hosted on your web server by browsing to them with your web browser. Let's test out this theory by viewing our default home page:

    Try It Out � Testing the Web Service

    • To verify that web services are working, start up your browser and type http://my_server_name/localstart.asp into the address box. (My server is named chrisu, so I typed in http://chrisu/localstart.asp.) Now press Enter; and (if all is well) you should get to see a page like this one:

    Note that the default page we see here uses the .asp extension, denoting a Classic ASP page. Support for ASP3 is provided as part of the standard IIS5.x web server program.

    What do you do if this doesn't work?

    If you don't get this page, then take a look at the following steps as we try to resolve the problem. If it's not working correctly, then most likely you'll be greeted with this screen:

    If you get this page then it can mean a lot of things, however one of the most likely problems is that your Web services are not switched on. To switch on Web services, you'll first need to start the IIS admin snap-in that we described earlier in the chapter (select Start | Run, type MMC and hit OK; then select Open from the MMC's Console menu and locate the iis.msc file from the dialog. Alternatively, just use the shortcut that you created there).

    Now, click on the + of the root node in the left pane of the snap-in, to reveal the Default sites. Then right-click on Default Web Site, and select Start:

     

    If it's still not working then here are a few more suggestions, which are based on particular aspects of your PC's setup. If you're running on a network and using a proxy server (a piece of software that manages connections from inside a firewall to the outside world � don't worry if you don't have one, they're mainly used by big businesses), there's a possibility that this can prevent your browser from accessing your web server. Most browsers will give you an opportunity to bypass the proxy server:

    • If you're using Internet Explorer, you need to go to View | Internet Options (IE4) or Tools | Internet Options (IE5/IE6) and select the Connections tab. In IE5/IE6 press the LAN Settings button and select Bypass the proxy server for local addresses. In IE4, this section forms part of the Connections dialog.

    • If you're using Netscape Navigator (either version 4.x or 6.x) and you are having problems then you need to turn off all proxies and make sure you are accessing the Internet directly. To do this, select Edit | Preferences; in the resulting dialog select Advanced | Proxies from the Category box on the left. Then on the right, select the Direct Connection to Internet option, and hit OK. Although you won't be browsing online to the Internet, it'll allow Netscape Navigator to recognize all variations of accessing local ASP.NET pages � such as http://127.0.0.1, http://localhost, etc.

    You may hit a problem if your machine name is similar to that of some web site out there on the Internet � for example, if your machine name is jimmyd but there also happens to be a public web site out there called http://www.jimmyd.com. When you type http://jimmyd into your browser's address box, expecting to view a page on your local web server, you unexpectedly get transported to http://www.jimmyd.com instead. If this is happening to you, then you need to make sure that you're not using a proxy server in your browser settings � again, this can be disabled using the Internet Options | Connection dialog or the Edit | Preferences dialog.

    Lastly, if your web server is running on your home machine with a modem, and you get an error message informing you that your web page is offline, this could in fact be a misperception on the part of the web server. This can be corrected by changing the way that your browser looks for pages. To do this, select View | Internet Options (IE4) or Tools | Internet Options(IE5/IE6), choose the Connections tab and select Never dial a connection.

    Of course, you might encounter problems that aren't answered above. In this case, the chances are that it's related to your own particular system setup. We can't possibly cover all the different possible configurations here; but if you can't track down the problem, you may find some help at one of the web sites and newsgroups listed later in this chapter.

    Managing Directories on your Web Server

    Before we install ASP.NET, we need to make one last pit stop in IIS. This is because when you come to run your ASP.NET pages, you need to understand where to place your pages, and how to make sure you have the permission to access them. As this is governed by IIS, now seems as good a time as any.

    These days, many browsers are sufficiently advanced that you can use them to locate and examine files and pages that exist on your computer's hard disk. So, for example, you can start up your browser, type in the physical location of a web page (or other file) such as C:\My Documents\mywebpage.html, and the browser will display it. However, this isn't real web publishing at all:

    First, web pages are transported using a protocol called HTTP � the HyperText Transfer Protocol. Note that the http:// at the beginning of a URL indicates that the request is being sent by HTTP. Requesting C:\My Documents\mywebpage.html in your browser doesn't use HTTP, and this means that the file is not delivered and handled in the way a web page should be. No server processing is done in this case.  We'll discuss this in greater detail when we tackle HTTP in Chapter 2.

  • Second, consider the addressing situation. The string C:\My Documents\mywebpage.html tells us that the page exists in the \My Documents directory of the C:drive of the hard disk of the machine on which the browser is running. In a network situation, with two or more computers, this simply doesn't give enough information about the web server.

    However, when a user browses (via HTTP) to a web page on some web server, the web server will need to work out where the file for that page is located on the server's hard disk. In fact, there's an important relationship between the information given in the URL, and the physical location (within the web server's file system) of the file that contains the source for the page.

    Virtual Directories

    So how does the relationship between the information given in the URL, and physical location work? In fact, it can work by creating a second directory structure on the web server machine, which reflects the structure of your web site. It sounds like it could be complicated, but it doesn't have to be. In fact, in this book it's going to be very simple.

    The first directory structure is what we see when we open Windows Explorer on the web server � these directories are known as physical directories. For example, the folder C:\My Documents is a physical directory.

    The second directory structure is the one that reflects the structure of the web site. This consists of a hierarchy of virtual directories. We use the web server to create virtual directories, and to set the relationship between the virtual directories and the real (physical) directories.

    When you try to visualize a virtual directory, it's probably best not to think of it as a directory at all. Instead, just think of it as a nickname or alias for a physical directory that exists on the web server machine. The idea is that, when a user browses to a web page that is contained in a physical directory on the server, they don't use the name of the physical directory to get there, instead, they use the physical directory's nickname.

    To see how this might be useful, consider a web site that publishes news about many different sporting events. In order to organize his web files carefully, the Webmaster has built a physical directory structure on his hard disk, which looks like this:

    Now, suppose you visit this web site to get the latest news on the Javelin event in the Olympics. If the URL for this web page were based on the physical directory structure, then the URL for this page would be something like this:

    http://www.oursportsite.com/sportsnews/athletics/field/javelin/default.asp

    That's Okay for the Webmaster, who understands his directory structure; but it's a fairly unmemorable web address! So, to make it easier for the user, the Webmaster can assign a virtual directory name or alias to this directory � it acts just like a nickname for the directory. Here, let's suppose we've assigned the virtual name javelinnews to the c:\inetpub\...\javelin\ directory. Now, the URL for the latest Javelin news is:

    http://www.oursportsite.com/javelinnews/default.asp

    By creating virtual directory names for all the directories (such as baseballnews, 100mnews, 200mnews, etc.) it's easy for the user to type in the URL and go directly to the page they want:

    http://www.oursportsite.com/baseballnews/default.asp

    http://www.oursportsite.com/100mnews/default.asp

    http://www.oursportsite.com/200mnews/default.asp

    Not only does this save the user from long, unwieldy, URLs � it also serves as a good security measure, because it hides the physical directory structure from all the web site visitors. This is good practice, otherwise hackers might be able to work out and access our files if they know what the directory structure looked like. Moreover, it allows the Webmaster's web site structure to remain independent of the directory structure on his hard drive � so he can move files on his disk between different physical folders, drives, or even servers, without having to change the structure of his web pages.  There is a performance overhead to think about as well, as IIS has to expend effort translating out the physical path. It can be a pretty costly performance wise to have too many virtual directories.

    Let's have a crack at setting up our own virtual directories and permissions (please note that these permissions are set automatically if you use the FrontPage editor to create a new site � so don't use FrontPage to set up this site for you unless you know what you're doing).

    Try It Out � Creating a Virtual Directory and Setting Up Permissions

    Now let's take a quick look at how you can create your own virtual directory. We'll use this directory to store the examples that we'll be creating in this book. We don't want to over complicate this example by creating lots of directories, so we'll demonstrate by creating a single physical directory on the web server's hard disk, and using the IIS admin tool to create a virtual directory and make the relationship between the two:

    1. Start Windows Explorer and create a new physical directory named BegASPNET, in the root directory of your hard drive. For example, C:\BegASPNET\:

    2. Next, start up the IIS admin tool (using the MMC, as we described earlier). Right-click on Default Web Site, and from the menu that appears select New | Virtual Directory. This starts the Virtual Directory Creation Wizard, which handles the creation of virtual directories for you and the setting up of permissions as well. You'll see the splash screen first, which looks like this. Click on Next:

    3. Type 5040 (an abbreviation of the book's ISBN, found on the back cover) in the Alias text box; then click Next:

    4. Click on the Browse� button and select the directory \BegASPNET that you created in Step 1. Then click Next:

    5. Make sure that the Read and Run scripts checkboxes are checked, and that the Execute checkbox is empty. Click on Next, and in the subsequent page, click on Finish:

    6. The BegASPNET virtual directory will appear on the tree in the IIS admin window:

    How It Works

    We just created a physical directory called BegASPNET, this directory will be used throughout the book to store our code examples. The download files from wrox.com are also designed to follow this structure. Within this directory we recommend that you create a sub-directory for each of the chapters in order to keep things tidy (this needn't be a virtual directory � just a physical one.)

    You've also created a virtual directory called 5040, which you created as an alias for the physical BegASPNET directory. If when we create Chapter 1 examples you place the ASP.NET files in the physical C:\BegASPNET\ch01 directory; you can use the browser to access pages stored in this folder. You'll need to use the URL http://my_server_name/5040/ch01/�

    You should also note that the URL uses the alias /5040 � IIS knows that this stands for the directory path C:\BegASPNET. When executing ASP.NET pages, you can reduce the amount of typing you need to do in the URL, by using virtual directory names in your URL in place of the physical directory names.

    We also set the permissions read and run � these must be set or the IIS security features will prevent you from running any ASP.NET pages. The Execute checkbox is left empty as allowing others to run applications on your own machine is a sure way of getting viruses or getting hacked. We'll take a closer look at permissions now, as they are so important. If you don't assign them correctly you may find that you're unable to run any ASP.NET pages at all � or worse still, that anybody at all can access your machine, and alter (even delete) your files via the Web.

    Permissions

    As we've just seen, we can assign permissions to a new directory as we create it, by using the options offered in the Virtual Directory Wizard. Alternatively, we can set permissions at any time, from the IIS admin tool in the MMC. To do this, right-click on the 5040 virtual directory in the IIS admin tool, and select Properties. You'll get the following dialog:

    It's quite a complicated dialog, and it contains a lot of options � not all of which we wish to go into now.

    Access Permissions

    The four check boxes on the left are of interest to us, as they govern the types of access for the given directory and dictate the permissions allowed on the files contained within that directory. Let's have a look at what each of these options means:

    Script source access � This permission enables users to access the source code of an ASP.NET page. It's only possible to allow this permission if the Read or Write permission has already been assigned. But we generally don't want our users to be able to view our ASP.NET source code, so we would usually leave this checkbox unchecked for any directory that contains ASP.NET pages. By default, all directories created during setup have Script Source Access permission disabled. You should leave this as is.

    Read � This permission enables browsers to read or download files stored in a home directory or a virtual directory. If the browser requests a file from a directory that doesn't have the Read permission enabled, then the web server will simply return an error message. Note that when the folder has Read permission turned off, HTML files within the folder cannot be read; but ASP.NET code within the folder can still be run. Generally, directories containing information that you want to publish (such as HTML files, for example) should have the Read permission enabled, as we did in our Try It Out.

    Write � If the write permission on a virtual directory is enabled, then users will be able to create or modify files within the directory, and change the properties of these files. This is not normally turned on, for reasons of security and we don't recommend you alter it.

    Directory Browsing If you want to allow people to view the contents of the directory (that is, to see a list of all the files that are contained in that directory), then you can allow this by checking the Directory Browsing option.

    If someone tries to browse the contents of a directory that has Directory Browsing enabled but Read disabled, then they may receive the following message:

    For security reasons, we'd recommend disabling this option unless your users specifically need it � such as when transferring files using FTP (file transfer protocol), from your web site . If you don't know what FTP is, then we recommend this strongly, as you obviously don't need it!

    Execute Permissions

    There's a dropdown list box near the foot of the Properties dialog, labeled Execute permissions � this specifies what level of program execution is permitted on pages contained in this directory. There are three possible values here � None, Scripts only, or Scripts and Executables:

    Setting Execute permissions to None means that users can only access static files, such as image files and HTML files. Any script-based files of other executables contained in this directory are inaccessible to users. If you tried to run an ASP.NET page, from a folder with the permission set to None, we would get the following � note the Execute Access Permission forbidden message in the page:

  • Setting Execute permissions to Scripts Only means that users can also access any script-based pages, such as ASP.NET pages. So if the user requests an ASP.NET page that's contained in this directory, the web server will allow the ASP.NET code to be executed, and the resulting HTML to be sent to the browser.

  • Setting Execute permissions to Scripts and Executables means that users can execute any type of file type that's contained in the directory. It's generally a good idea to avoid using this setting, in order to prohibit users from executing potentially damaging applications on your web server.

    For any directory containing ASP.NET files that you're publishing, the appropriate setting for the Execute permissions is Scripts Only. Now you've started to familiarize yourself with IIS, you're ready to prepare your machine for the installation of ASP.NET itself.

    Prerequisites for installing ASP.NET

    Before you can install ASP.NET or the .NET Framework you will need to install the Microsoft Data Access Components (MDAC) version 2.7 or later. This is a set of components that will enable you to use ASP.NET to communicate with databases and display the contents of your database on a web page. Without these components installed you won't be able to run any of the database examples in this book. This will affect examples as early as Chapter 2, so please don't skip this stage! Although you might already have an earlier version of MDAC installed (such as 2.5 if you're using Windows 2000), unless you have specifically upgraded, in all likelihood you won't have the most up to date version and will still need to upgrade.

    The Microsoft Data Access Components is a small download (roughly 5 or 6 MB) available for free from Microsoft's site at http://www.microsoft.com/data.

    As of writing, Microsoft hadn't yet placed 2.7 at the above URL, as it was still in beta. If you can only find version 2.6 at this location, we suggest trying http://www.microsoft.com/downloads/release.asp?ReleaseID=30134 Alternatively you can look at http://msdn.microsoft.net for a list of the latest .NET resources.

    The installation for MDAC 2.7 is pretty straightforward, but we'll run through it quickly just to make sure that everything is clear.

    Try It Out � Installing MDAC 2.7

    1. MDAC 2.7 comes as file MDAC_typ_dnld.exe that you will need to run. Once you have run it, it will ask you for a location where you wish to download the files. Type in an appropriate location:

    2. In fact rather confusingly, it seems only to download one file, MDAC_TYPE.EXE to the pre-specified location. If you run this EXE file, then it will begin the installation process.

    3. After agreeing to the terms of the license, there's a good chance that you will be asked to reboot our system, it will tell you this in advance.

    4. Then the installation process will continue without requiring further intervention, although you might have to wait for a system reboot, if one was specified earlier.

    You're now ready to install ASP.NET.

    Installing ASP.NET and the .NET Framework SDK

    We're almost ready to install ASP.NET, but there are two important points to be made beforehand.

    First, there are two different types of installation available from Microsoft's site, the .NET Framework SDK and ASP.NET. The .NET Framework SDK already contains ASP.NET, so you do not need to download both separately. You only need to download one. Both downloads contain ASP.NET, VB.NET, and the .NET Framework.

    The ASP.NET Premium Edition download is a smaller, streamlined download that only contains the bare bones needed for you to run ASP.NET and the .NET framework. None of the extra documentation or samples will be included. The size differential between the two is pretty big (ASP.NET is 18MB while the .NET Framework SDK is a staggering 123MB), so unless you have the .NET Framework SDK on CD (which you can order from the Microsoft site) or broadband high-speed Internet access, you'll probably want to download the ASP.NET version.

    Don't worry, this won't affect your ability to run the examples in this book � everything's been written so that it will run on the ASP.NET Premium Edition. While you won't have direct access to the help files, all support materials are available online at Microsoft's http://www.asp.net site.

    Don't worry about replacing an existing Classic ASP installation, since ASP.NET will be installed alongside ASP and they will both continue to work with no action from us.

    We'll now walk you through a typical installation of both ASP.NET Premium Edition and the .NET Framework SDK. The installation process is the same for Windows 2000 and Windows XP, so once again we're only going to detail the installation process on the former. Although the wizard looks a bit different on XP, it asks for exactly the same things.

    Try It Out � Installing ASP.NET

    1. Click on setup.exe and after confirming that you do want to install ASP.NET Premium, and after a short interval, you are propelled into the setup wizard:

    2. Click on Next and accept the License agreement to continue. The next dialog after the license agreement will ask you where you wish to install ASP.NET:

    3. Unless you have good reason to, we suggest leaving the location as the one specified by the setup wizard, and then click on Next. ASP.NET will now install without further intervention:

    4. You will be notified when installation has finished, and unlike with the MDAC 2.7, you probably won't have to reboot. We can now go to the testing section and check everything is working.

    Installing the .NET Framework SDK

    1. Click on setup.exe and after confirming that you do want to install the NET Framework SDK package, and after an interval of a few minutes, you are propelled into the setup wizard:

    2. Click on Next and accept the License agreement to continue. The next dialog after the license agreement will ask you which different pieces of the SDK you need to install. You should check all of them, although if you're short of hard drive space, you could choose to omit the SDK_Samples or documentation. The Software Development Kit is essential:

    3. After clicking on Next you get to a dialog that specifies the destination folder for the different .NET Framework SDK samples and bits and pieces. You can choose to install these wherever you want. More importantly there is a checkbox at the foot of the dialog, which asks you to register environment variables. This checkbox should be checked, as we will use the environment variables in later chapters:

    Click on Next and the .NET Framework SDK will install without further ado. It shouldn't require a reboot.

    Troubleshooting Hints and Tips

    The installation process is very straightforward, and will work on the majority of machines. However, sometimes the particular configuration of your machine will prevent it from installing. Unfortunately we can't cover all of the possible eventualities, but if it doesn't work on yours, you should check that you have enough hard disk space, as this is the most common cause of problems. Also try to ensure that the installation process isn't curtailed half way, as no installer is completely foolproof at removing all the different bits and pieces of the aborted install and it can cause problems when you try to reinstall, and leave you needing to reformat your hard drive to get it to work correctly. Other than that, check the list of newsgroups and resources later in this chapter.

    ASP.NET Test Example

    Ok, we've now reached the crux of the chapter, checking to see if everything is working correctly. Do you remember the punctual web server code that we talked about earlier in the chapter � in which we wanted to write a web page that displays the current time? We'll return to that example now. As you'll see it's quite a simple bit of code, but it should be more than enough to check that ASP.NET is working Okay.

    Try It Out � Your first ASP.NET web page
    1. Open up a text editor and type in the following code:
      <script language="vb" runat="server">
      Sub Page_Load()
      time.text=Hour(Now) & ":" & Minute(Now) & ":" & Second(Now)
      End Sub
      </script>
      <html>
      <head><title>The Punctual Web Server</title></head>
      <body>
        <h1>Welcome</h1>
        In WebServerLand the time is currently:
      <asp:label id="time" runat="server" />
      </body>
      </html>

      We strongly suggest (and will assume throughout) that you use Notepad to code all the examples in this book, since it will always do precisely what you ask it to and no more. It's therefore a lot easier to track down any problems you're having, and is a great deal easier than troubleshooting problems caused by FrontPage or similar web page editors.

    2. Save this page as punctual.aspx. Make sure that you save it in the physical folder you created earlier C:\BegASPNET\Ch01\.

      When you save the file, you should double-check that your new file has the correct suffix. It should be .aspx, since this is how you tell the web server that the page contains ASP.NET code. Be aware that Notepad (and many other text editors) consider .txt to be the default. So in the Save or Save As dialog, make sure that you change the Save As type to read All Files, or All Files(*.*),or enclose the path and filename in quotes.
    3. Now start up your browser and type in the following: http://localhost/5040/punctual.aspx

    4. Now click on the refresh button of the browser and the displayed time will change. In effect the browser is showing a new and different instance of the same page.

    5. Now on your browser select View Source or similar, (depending on which browser you're using) from the browser menu to see the HTML source that was sent from the web server to the browser. The result is shown below. You can see that there is no ASP.NET code to be seen, and nothing before the first <html> tag  � the ASP.NET code has been processed by the web server and used to generate pure HTML, which is hard-coded into the HTML source that's sent to the browser:

      Here, you can see the HTML that was sent to the browser when I Refreshed the page at 15.39:15.

    6. As we mentioned before, you can expect this to work in any browser � because the ASP.NET is processed on the web server. If you have another browser available, try it out.

    How It Works

    Easy wasn't it? (If you didn't get it to work first time, then don't rush off to email technical support just yet � have a little look at the next section, ASP.NET Troubleshooting, first.) Now let's take a look at the ASP.NET code that makes this application tick.

    Of course, there is only one block of ASP.NET code (ignoring the server control) in the whole program. It's enclosed by the <script> and </script> tags:

    <script language="vb" runat="server">
    Sub Page_Load()
    time.text=Hour(Now) & ":" & Minute(Now) & ":" & Second(Now)
    End Sub
    </script>

    The script delimiters specify which code needs ASP.NET to run, and we'll look at them in detail in the next chapter. If we ignore the <script> tags for the time being, and just think of them as ASP.NET code delimiters, then we're left with just three lines. If we further ignore the Sub Page_Load and End Sub lines, which are standard to many ASP.NET programs, and which we'll be discussing in Chapter 3, we're left with one line. This line:

    time.text=Hour(Now) & ":" & Minute(Now) & ":" & Second(Now)

    tells the web server to go off and run the VB.NET Now() function on the web server. The VB.NET Now() function returns the current time at the web server. It returns the values of the Now() function divided up into hour, minute, and second values. The result of this function is returned as part of the <ASP: label> control, further down the page. We'll be looking at this control in Chapter 3.

    If the web server and browser are on different machines, then the time returned by the web server might not be the same as the time kept by the machine you're using to browse. For example, if this page is hosted on a machine in Los Angeles, then you can expect the page to show the local time in Los Angeles � even if you're browsing to the page from a machine in Cairo.

    This example isn't wildly interactive or dynamic, but it illustrates that we can ask the web server to go off and do something for us, and return the answer within the context of an HTML page. Of course, by using this technique with things like HTML forms and other tools, we'll be able to build a more informative, interactive, interface with the user.

    ASP.NET Troubleshooting

    If you had difficulty with the example above, then perhaps you fell into one of the simple traps that commonly snare new ASP.NET programmers, and that can be easily rectified. In this section we'll look at a few common errors and reasons why your script might not run. If you did have problems, maybe this section will help you to identify them.

    Program Not Found, or the Result of the ASP.NET isn't being Displayed, or the Browser tries to Download the File

    You'll have this problem if you try to view the page as a local file on your hard drive, like this:

    C:\BegASPNET\ch01\punctual.aspx

    You'll also get this problem if you click on the file in Windows Explorer. If you have Microsoft FrontPage or Visual Studio.NET installed, then it will start up and attempt to help you to edit the code. Otherwise, your browser may display a warning message, or most likely it will ask you which application you wish to use to open up the ASPX file:

     

    Older browsers may try to download the file:

     

    The Problem

    This is because you're trying to access the page in a way that doesn't cause the ASP.NET page to be requested from the web server. Because you're not requesting the page through the web server, the ASP.NET code doesn't get processed � and that's why you don't get the expected results.

    To call the web page through the web server and have the ASP.NET code processed, you need to reference the web server in the URL. Depending on whether you're browsing to the server across a local network, or across the Internet, the URL should look something like one of these:

    http://localhost/5040/ch01/punctual.aspx

    http://www.distantserver.com/5040/ch01/punctual.aspx

    Page Cannot be Displayed: HTTP Error 403

    If you get a 403 error message, then it's probably because you don't have permission to execute the ASP.NET code contained within the page � notice the Execute Access Forbidden Error in the middle of the page:

    As you'll recall, permissions are controlled by the properties of the virtual directory that contains the ASP.NET page. To change these properties, you'll need to start up the IIS admin snap-in in the MMC, as we described earlier in the chapter. Find the BegASP.NET virtual directory in the left pane, right-click on it and select Properties. This will bring up the BegASP Properties dialog that we met earlier in the chapter:

    Here, you'll need to check that the value shown in the Execute Permissions box is "Scripts only" or "Scripts and Executables" �but definitely NOT "None".

    Page Cannot Be Found: HTTP Error 404

    If you get this error message then it means that the browser has managed to connect to the web server successfully, but that the web server can't locate the page you've asked for. This could be because you've mistyped the URL at the browser prompt. In this case, we'll see a message like this:

    If you get this page, then you might suspect one of the following errors:

    • A simple typing error in the URL, for example. http://localhost/5040/ch01/punctually.aspx
    • A wrong directory name, for example. http://localhost/5040/punctual.aspx instead of http://localhost/5040/ch01/punctual.aspx
    • Including a directory separator (/) after the file name, for example. http://localhost/5040/ch01/punctual.aspx/
    • Using the directory path in the URL, rather than using the alias, for example. http://chrisu//BegASPNET/ch01/punctual.aspx
    • Saving the page as .html or .htm, rather than as an .aspx, for example. http://localhost/5040/ch01/punctual.htm

    Or as above, you've used the name of the physical directory rather than the virtual one for example.

    http://localhost/BegASPNET/ch01/punctual.aspx 

    Of course, it may be that you've typed in the URL correctly, and you're still experiencing this error. In this case, the most likely cause is that you have used Notepad to save your file and that (when you saved the file) it used its default Save As Type setting, which is Text Documents (*.txt). This automatically appends a .txt suffix to the end of your file name. In this case, you will unwittingly have finished up with a file called punctual.aspx.txt.

    To check if that is what happened, go to Windows Explorer, and view the (physical) folder that contains the file. Go to the Tools menu and select Folder Options�. Now, in the View tab, ensure that the Hide file extensions for known file types is unchecked, as shown here:

    Now click OK and return to view your file in Windows Explorer. You may well see something like
    the following:

    As you can see, Notepad has been less than honest in its dealings with you: when you thought that you had saved your file as punctual.aspx, it had inconveniently saved it as punctual.aspx.txt. Not surprisingly, your web server won't be able to find your file if it's been renamed accidentally. To correct the filename, right click on the it in the right pane above, select Rename from the dropdown menu that appears and remove the .txt at the end.

    Web Page Unavailable While Off-line

    Very occasionally, you'll come across the following message box:

    This happens because you've tried to request a page and you haven't currently got an active connection to the Internet. This is a misperception by the browser (unless your web server isn't the same machine as the one you're working on) � it is trying to get onto the Internet to get your page when there is no connection, and it's failing to realize that the page you've requested is present on your local machine. One way of retrieving the page is to hit the Connect button in the dialog; but that's not the most satisfactory of solutions (since you might incur call charges). Alternatively, you need to adjust the settings on your browser. In IE5/IE6, select the File menu and uncheck the Work Offline option.

    This could also be caused if you're working on a network and using a proxy server to access the Internet. In this case, you need to bypass the proxy server or disable it for this page, as we described in the section Browsing to a Page on your Web Server, earlier in the chapter. Alternatively, if you're using a modem and you don't need to connect, you can correct this misperception by changing the way that IE looks for pages. To do this, select the Tools | Connections option and select Never dial a connection.

    I Just Get a Blank Page

    If you see an empty page in your browser, then it probably means that you managed to save your punctual.aspx without entering any code into it, or that you didn't remember to refresh the browser.

    The Page Displays the Message but not the Time

    If the web page displays the message "In Webserverland, the time is exactly " � but doesn't display the time � then you might have mistyped the code. For example, you may have mistyped the name of the control:

    time.text=Hour(Now) & ":" & Minute(Now) & ":" & Second(Now)

    and:

    <asp:label id="hour" runat="server" />

    The name of the control "hour", must match the first word in the line of ASP.NET code, otherwise the control won't be able to identify it.

    I Get an Error Statement Citing Server Error

    If you get a message stating that the page cannot be displayed, and citing a server error such as:

    then it means that there's an error in the ASP.NET code itself. Usually, there's additional information provided with the message. For example, you'd get this error message if you omitted the closing </script> tag on your code. To double-check that this isn't the case, use the sample punctual.aspx from the Wrox site at http://www.wrox.com.

    I Have a Different Problem

    If your problem isn't covered by this description, it's worth testing some of the sample ASP.NET pages that are supplied with the QuickStart tutorials at http://www.asp.net. These should help you to check that IIS has actually installed properly. You can always uninstall and reinstall if necessary, although before you try this, rebooting the server first might help.

    You can get support from http://p2p.wrox.com, which is our web site dedicated to support issues in this book. Alternatively, there are plenty of other web sites that are dedicated to ASP and ASP.NET. In fact you will find very few sites focus on just one of the two technologies. Here are just a few:

    • http://www.asp.net
    • http://www.asptoday.com
    • http://www.asp101.com
    • http://asptracker.com
    • http://www.15seconds.com
    • http://www.4guysfromrolla.com

    There are lots of solutions, discussions, and tips on these pages, plus click-throughs to other related pages. Moreover, you can try the newsgroups available on www.asp.net such as aspngfreeforall.

    You should now have successfully downloaded, set up and installed both IIS and ASP.NET, and got your first ASP.NET application up and running. If you've done all that, you can pat yourself on the back, make a cup of tea, and get ready to learn about some of the principles behind ASP.NET in the next chapter.

    Summary

    We started the chapter with a brief introduction to ASP.NET and to dynamic web pages in general and we looked at some of the reasons why you'd want to use a server-side technology for creating web pages. We looked at some of the history behind dynamic web pages, in the form of an overview of the other technologies. This very brief introduction to ASP.NET will be expanded in the next chapter.

    The bulk of the chapter though, was taken up by a description of the installation process. You must have installed IIS 5.0/5.1, MDAC 2.7 and either ASP.NET or the .NET Framework SDK to be able to progress at all with this book, so please don't be tempted to skip parts that might not have worked. We've listed plenty of resources that will help you get everything up and running, and there's rarely a problem that somebody somewhere hasn't encountered before.

    The next chapter covers the software installed with ASP.NET, the .NET Framework, and will build up in much greater detail what ASP.NET does, what the .NET Framework is, and how the two work together.

    Copyright and Authorship Notice

    This chapter is written by Juan T Llibre, Ollie Cornes, Chris Goode, Chis Ullman et al, and taken from "Beginning ASP.NET using VB.NET" published by Wrox Press Limited in August 2001; ISBN 1861005040; copyright � Wrox Press Limited 2000; all rights reserved.

    No part of these chapters may be reproduced, stored in a retrieval system or transmitted in any form or by any means - electronic, electrostatic, mechanical, photocopying, recording or otherwise - without the prior written permission of the publisher, except in the case of brief quotations embodied in critical articles or reviews.

  • License

    This article has no explicit license attached to it but may contain usage terms in the article text or the download files themselves. If in doubt please contact the author via the discussion board below.

    A list of licenses authors might use can be found here