public string ResolveUrl(string url)
{
string urlToResolve = url;
int nPound = url.LastIndexOf("#");
int nSlash = url.LastIndexOf("/");
if ((nPound > -1) && (nSlash > -1) && ((nSlash + 1) == nPound))
{
// We have been given a somewhat strange URL. It has a foreward slash (/) immediately followed
// by a pound sign (#) like this xxx/#yyy. This sort of oddly shaped URL is what you get when
// you use named anchors instead of pages in the url attribute of a sitemapnode in an ASP.NET
// sitemap like this:
//
// <siteMapNode url="#Introduction" title="Introduction" description="Introduction" />
//
// The intend of the sitemap author is clearly to create a link to a named anchor in the page
// that looks like these:
//
// <a id="Introduction"></a> (XHTML 1.1 Strict compliant)
// <a name="Introduction"></a> (more old fashioned but quite common in many pages)
//
// However, the sitemap interpretter in ASP.NET doesn't understand url values that start with
// a pound. It prepends the current site's path in front of it before making it into a link
// (typically for a TreeView or Menu). We'll undo that problem, however, by converting this
// sort of oddly shaped URL back into what was intended: a simple reference to a named anchor
// that is expected to be within the current page.
urlToResolve = url.Substring(nPound);
}
else
{
urlToResolve = AdaptedControl.ResolveClientUrl(urlToResolve);
}
// And, just to be safe, we'll make sure there aren't any troublesome characters in whatever URL
// we have decided to use at this point.
string NewUrl = AdaptedControl.Page.Server.HtmlEncode(urlToResolve);
return(NewUrl);
}