# Friday, 02 October 2009

One of the differences between ASP.NET MVC  and traditional ASP.NET web forms is the URL typed by the user.

In traditional ASP.NET, a URL points to a physical page on disc. Typically one can infer the location of that page by examining the URL. For example, the URL

/Folder1/MyPage.aspx

likely points to a file named MyPage.aspx located in a folder named Folder1.

By contrast, a URL in an ASP.NET MVC application points to an action in a controller. For example, the URL

/Customer/Edit/2

Tells the MVC framework to pass a parameter of "2" to an action method named Edit in a Customer controller class.

ASP.NET MVC (which I’ll call MVC going forward) uses routing to map a URL to a specific action. Routing is typically configured in the  Application_OnStart method of the global.asax file.  An MVC application contains a System.Web.Routing.RouteTable, which contains a collection of routes on which it can act. When you create a new MVC project, a default route is added to this collection by adding the following code in global.asax.

routes.MapRoute(
    "Default",                                              // Route name
    "{controller}/{action}/{id}",                           // URL with parameters
    new { controller = "Home", action = "Index", id = "" }  // Parameter defaults
    );

This code adds a route named "Default" to the route table. This route tells MVC to look for a URL structured similar to the following

Controller/action/id

When it encounters such a URL, it looks for a controller class with a name based on the first part of the URL. By convention, MVC looks for controllers in the "Controllers" folder and is named with the suffix "Controller". So if the first part of the URL is "Customer", MVC will look in the controllers folder for a class derived from System.Web.Mvc.Controller named "CustomerController".

The second part of the URL identifies an action. The action is a method within the controller that returns a  System.Web.Mvc.ActionResult. The ActionResult identifies the view used to render a response (more on that in a later article). So if the second part of the URL is "Edit", MVC will look for a method in the controller class named "Edit" that returns an ActionResult.

The third part of the above URL defines a parameter for the action, in this case a parameter named "id".

The following URL

/Customer/Edit/2

causes MVC to search for a class named Controllers\CustomerController, to call the Edit method in this class and to pass the value 2 to the id parameter of the Edit method.

What if the URL does not contain all three parts? The Default route above specifies a default value for each of the parts. If the third part of the URL is omitted, no parameter is passed to the Action. For a URL with no id parameter, we will need to create an action method that does not expect a parameter. If the second part of the URL is omitted, the default action is "Index", so MVC will look for a method named "Index" in the controller class. If the first part of the URL is omitted, it defaults to "Home" so MVC will look for a Controller class named "HomeController" in the Controllers folder.

You can use this same syntax to add more routes to your application. Simply add more calls to routes.MapRoute() in Global.asax. MVC will match URLs to these routes in the order listed.

Notice that the RouteTable and RouteCollection are part of the System.Web.Routing namespace and not part of any MVC namespace. This should tell you that the routing engine is not restricted to MVC applications. It can be used in any ASP.NET application if you want cleaner URLs.

If you find a URL like /Customer/Edit/2 more asthetically pleasing than /Customers/EditCustomer.aspx?CustID=2, you should investigate the routing engine.

ASP.NET | MVC