Introduction - What is AJAX?

AJAX, or Asynchronous JavaScript and XML,is a fancy way to use JavaScript and XML to communicate with a web server without refreshing the web page.
You can visit this web page for more information about AJAX; it has a good list of links.

Why use AJAX?

There are a couple of reasons to use AJAX in lieu of the traditional form submission. The first is that it is very light weight: instead of sending all of the form information to the server and getting all of the rendered HTML back, simply send the data the server needs to process and get back only what the client needs to process. Light weight means fast. The second reason to use AJAX is because (as the logo in the link above makes clear) AJAX is cool.

Using AJAX with ASP.NET

Even though AJAX is a client side technology that uses JavaScript, the client-server interaction is very important.

Adam Vandenberg has put together a nice JavaScript wrapper for AJAX. His wrapper even does caching, although that will not be discussed here. His wrapper is used in the code examples and project files.

Using the code

There are four parts of the code that are important to look at:

1. The HTML

   There are two form elements that will interact with AJAX. The input button "btn1" will invoke the AJAX code, which will make a server call and fill the select element "select1".

2. The JavaScript that calls the AJAX.

   The function getOptions() will do the main work.

   Collapse

   // Create the Request object (the AJAX wrapper)
   
   var request = new Request();
   
   // Change this to fit your environment
   
   var url = "http://localhost/ajax/";
   
   function getOptions()
   
   {
   
   // Call the AJAX
   
   // Notice the second parameter is actually a function to handle the
   
   // response
   
   request.GetNoCache(url + "requests/getOptions.aspx",
   
   function(result)
   
   {
   
   if (result.readyState!=ReadyState.Complete)
   
     return;     
   
   if (result.status==HttpStatus.OK && result.responseText != "")
   
   {
   
     // If the request was successfull and returned data
   
     var vals = result.responseText.split("~");
   
     for (i=0; ivar pair = vals[i].split("|");
   
         var op = new Option(pair[1], pair[0], false, false);
   
         var sel = document.getElementById("select1");
   
         sel.options[sel.length] = op;
   
     }
   
     alert("Remember that the new values in form" +
   
           " element 'select1' are not in viewstate." +
   
           " Code appropriately.");
   
   }
   
   else
   
   {
   
     // Handle the failure condition
   
     alert('Get options failed.');
   
   }
   
   }
   
   )
   
   }

3. The ASPX file.

   The important thing here is that the aspx file only returns the string (XML) data from the code behind.

   <%@ Page language="c#"            Codebehind="getOptions.aspx.cs" 

4.  AutoEventWireup="false"
   
   

5.  Inherits="ajax.requests.getOptions" %>
   
   <%=result%>

6. The code behind.

   protected string result = string.Empty;
   
   private void Page_Load(object sender, System.EventArgs e)
   
   {
   
   for (int i=0; i<10; i++)
   
   {
   
   result += i.ToString() + "|option " + i.ToString() + "~";
   
   }
   
   // to drop the last '~'
   
   result = result.Substring(0, result.Length - 1);
   
   }

Introduction

Most developers often have many windows open at once, and so often will use a virtual desktop program to help organize their windows. However, most virtual desktop software is ugly and lacks many useful features. While I don't put every feature you may want in this program, I do give a powerful base implementation of a Vista thumbnail-enhanced virtual desktop program.

How to Use the Program

Obviously, you need Windows Vista to use this program. It was tested on RC1, and some older versions have differing thumbnail API functions that may fail. When you first start the program, a window flashes open and then closes. A new icon is now in the system tray. It installs hotkeys as follows:

• Windows Key + Z - Open switcher
• Windows Key + Numpad 1...4 (with numlock on) - Switch to desktop x

When you hit Windows Key + Z, everything starts to fade to black and any windows you have open will appear as thumbnails in the first desktop in the upper lefthand corner. You can drag windows around and they will move in real time. If you drag them to a different desktop, they will disappear. To move between desktops, either double-click a window on a desktop, or use the arrow keys to select a desktop and then press enter to switch to it. You can also press escape to hide the switcher.

Every time you switch desktops, the switcher window gives a visual representation of the switch.

If you right-click on the tray icon, you are presented with standard options and a menu for switching between desktops.

How it Works

I'll go over the internals of the program by class. A lot of tricks are needed in order to get everything working. I have not only classes that will be useful in other projects, but a library I'll do another article on named PPAB Forms.dll. This library contains many useful graphical functions and most importantly, it offers a class called Form that inherits from System.Windows.Forms.Form that displays a layered window. If you don't know what a layered window is, it's a special kind of window that is composited and can have per-pixel alpha information (PPAB stands for per-pixel alpha blending). That's right, before Vista and DWM, there were layered windows. You can use them for all kinds of graphical things. If you've ever used Object Desktop, Konfabulator, ObjectDock, etc, or seen the old Adobe Reader alpha-blended splash screen, you've seen layered windows in action.

Thumbnail Class

This class mainly wraps the Vista thumbnail API. You give it source and destination windows. Then, you set some properties and call UpdateRendering. This starts the rendering. It is disposable, so you can stop the rendering completely by calling dispose.

Introduction

The Cryptography API: Next Generation(CNG) is a new and agile framework in Windows Vista™, which implements an extensible provider model that allows you to load a provider by specifying the required cryptographic algorithm rather than having to hardcode a specific provider.

The advantage is that an algorithm provider can be replaced or upgraded and you will not have to change your code in any way to use the new provider. Also, if some algorithm is determined to be unsafe in the future, a more secure version of that algorithm can be installed with no effect to your code. To facilitate this, you load a CNG provider by identifying the cryptographic algorithm that you require, not the specific provider. Most of the CNG APIs require a provider or an object created by a provider.

In this article, I try to describe the new security feature Cryptography API: Next Generation(CNG) and compare it with an RSA and AES samples, both managed and unmanaged, using "Crypto API" (CAPI before Vista) and how it can be implemented using CNG in Windows Vista. Managed version of CNG is yet to come, if you feel like you want more, wait for the next release of Visual Studio "Orcas".

Background

About RSA

RSA is the established standard for public key encryption. The name RSA is derived from the names of the inventors of this algorithm, which are: Ron Rivest, Adi Shamir and Leonard Adleman. The principle and security of RSA is based on the fact that with today's knowledge, it is not possible to find the prime factors of a big number (n=pq, where p and q are prime numbers) in suitable time.

Short details

• Public Key: n=pq (p and q are big prime numbers)
• e relative prime to (p-1)(q-1)
• Private Key: d e-1 mod ((p-1)(q-1))
• Encryption: c = me mod n
• Decryption: m = cd mod n

About AES

Advanced Encryption Standard (AES), also known as Rijndael, is a symmetric 128-bit block cipher adopted as an encryption standard by the US government.

Short details

• AES operates on a 4×4 array of bytes.
• For encryption, each round of AES (except the last round) consists of four stages: AddRoundKey, Subbytes, Shift rows and Mix columns.
• At each stage, the bytes are manipulated and processed for the next level.

Crytography API: Next Generation(CNG)

CNG provides a set of APIs that are used for performing basic cryptographic operations, such as creating hashes, encrypting, and decrypting data.

Each algorithm class in CNG is represented by a primitive router. Applications making use of the primitive APIs will link to the router binary (Bcrypt.dll in user mode, or Ksecdd.sys in kernel mode), and make calls to the various CNG primitive functions. All of the algorithm primitives are managed by various router components. These routers keep track of each algorithm implementation that has been installed on the system. The router will route each function call to the appropriate primitive provider module.

The following illustration shows the design and function of the CNG cryptographic primitives.

CNG provides primitives for the following classes of algorithms:

• Random Number Generator: This class is used to represent pluggable random number generation (RNG).
• Hashing: This class represents algorithms used for hashing, such as SHA1 and SHA2.
• Symmetric encryption: This class represents algorithms used for symmetric encryption. Some examples are AES, 3DES, and RC4.
• Asymmetric encryption: This class represents asymmetric (public key) algorithms that support encryption, like RSA.
• Signature: This class represents signature algorithms such as DSA and ECDSA. This class can also be used with RSA.
• Secret Agreement: This class represents secret agreement algorithms such as Diffie-Hellman (DH) and elliptical curve Diffie-Hellman (ECDH).