Getting the absolute path in ASP.NET (part 2)

In a previous blog, I received many comments on various ways I could get the application root.  I am thankful for such comments.  What most of the commenters did not realize is that I need the resolution of the url in a custom http module, where Control.ResolveUrl is not practical to use.  And since I am using vanity urls (phantom paths that have no mapping to physical files or pages), I needed the resolution without the use of any System.IO objects.

However, I did make revisions to my method, and here it is:

1: public static string ResolveUrl(string url) 
2: { 
3:  if (url==null) throw new ArgumentNullException("url", "url can not be null"); 
4:  if (url.Length == 0) throw new ArgumentException("The url can not be an empty string", "url"); 
5:  
6:  // there is no ~ in the first character position, just return the url 
7:  if (url[0] != '~') return url; 
8:  
9:  string applicationPath = HttpContext.Current.Request.ApplicationPath; 
10:  
11:  // there is just the ~ in the URL, return the applicatonPath 
12:  if (url.Length == 1) return applicationPath; 
13:  
14:  // assume url looks like ~somePage 
15:  int indexOfUrl=1; 
16:  
17:  // determine the middle character 
18:  string midPath = (applicationPath.Length >1 )? "/" : string.Empty; 
19:  
20:  // if url looks like ~/ or ~\ change the indexOfUrl to 2 
21:  if (url[1] == '/' || url[1] == '\\') indexOfUrl=2; 
22:  
23:  return applicationPath + midPath + url.Substring(indexOfUrl); 
24: }// ResolveUrl

Any comments?