Last day I was working to deploy a ClickOnce application from an ASP.NET MVC website. Nothing was working, and looking for a solution, I arrived to a method documented like this on the MSDN. Wow… a method to change ASP.Net MVC routes with expressions based on a syntax that nobody knows!

The problem behind the problem

DISCLAIMER: I’m new to MVC. Probably this post will contain mistakes.

Let’s me explain what was my problem step by step:

  • On ASP.Net MVC, the framework by default handles any request.
  • ClickOnce uses its own handlers.
  • When ClickOnce gets behind MVC, it gets “hidden” by ASP.Net MVC handlers.
  • You have to instruct ASP.Net MVC to ignore some requests in order to let them be handled by ClickOnce.

This is the way I arrived at the IgnoreRoute method. I wanted to instruct MVC to ignore any request for a ClickOnce handled URL.

ClickOnce URLs

ClickOnce uses three extensions:

  • .application: This is the entry point to the application
  • .deploy: Depending on the configuration, ClickOnce deploys applications files using their original extension, or files renamed withe .deploy extension. So, this extension refers to application files
  • .manifest: Manifest with metadata

I wanted MVC to ignore any request with this extensions

Back to the IgnoreRoute syntax

IgnoreRoute is an extension method applied to a RouteCollection that receives two parameters:

  • url (System.String): The URL pattern for the route to ignore.
  • constraints [optional] (System.Object): A set of expressions that specify values for the url parameter.

You could expect the url to be some kind of regular expression, but, as you can see here, it uses its own syntax.

There is almost no reference to what “constrainst” means.

The copy-paste syndrome

If you search for documentation on this topic, google will make you walk in circles around this examples:

So, everybody is using two examples:

  • One to ignore a file or folder
  • The other one to ignore certain extension

Well… it was enough for me, but I was feeling like if nobody really has the knowledge about the real syntax of this method.

Researching on the syntax

Viewing examples, I was sure the “constraints” was a collection of regular expressions. I found one page describing the “url” syntax:

My guess about how this stuff works

First parameter (url)

Well, here comes my conclussion:

  • With the first parameter, you can split the url based on the slash character. You can use literals or your own group identifiers. The last group can have a wildcard (*) character, that expands this group among the rest of the slash characters.

Let’s suppose you have the following URL: path/to/mypage

  • path/to/{page}: Captures “mypage” into a group with the name “page”
  • {app}/{*operation}: Captures “path” into “app”, and “to/mypage” into “operation”
  • {*url}: Captures “path/to/mypage” into “url”

Second parameter (constraints)

The second parameter is an object composed of regular expression string. Each regular expression is mapped to the group with the same name captured in the url.

  • {app = “^ab.”, operation = “.cd$}: In the previous examples, tells that app must start with “ab” and operation must end with cd.

Example: Filter ClickOnce urls using a single regular expression

I think this is too complex, and a single regular expression should be enough to filter any url… so here comes the way to filter any url using a single regular expression:

routes.IgnoreRoute("{*url}", new { url = @"(.*)\.application(\?.*)?$" });
routes.IgnoreRoute("{*url}", new { url = @"(.*)\.manifest$" });
routes.IgnoreRoute("{*url}", new { url = @"(.*)\.deploy$" });

And this way we arrive to the end of the post. See you and Happy Coding!