[c#] How can I get the baseurl of site?

I want to write a little helper method which returns the base URL of the site. This is what I came up with:

public static string GetSiteUrl()
{
    string url = string.Empty;
    HttpRequest request = HttpContext.Current.Request;

    if (request.IsSecureConnection)
        url = "https://";
    else
        url = "http://";

    url += request["HTTP_HOST"] + "/";

    return url;
}

Is there any mistake in this, that you can think of? Can anyone improve upon this?

string baseUrl = Request.Url.GetLeftPart(UriPartial.Authority)

That's it ;)

I'm using following code from Application_Start

String baseUrl = Path.GetDirectoryName(HttpContext.Current.Request.Url.OriginalString);

Based on what Warlock wrote, I found that the virtual path root is needed if you aren't hosted at the root of your web. (This works for MVC Web API controllers)

String baseUrl = Request.RequestUri.GetLeftPart(UriPartial.Authority) 
+ Configuration.VirtualPathRoot;

I go with

HttpContext.Current.Request.ServerVariables["HTTP_HOST"]

Please use the below code                           

string.Format("{0}://{1}", Request.url.Scheme, Request.url.Host);

To me, @warlock's looks like the best answer here so far, but I've always used this in the past;

string baseUrl = Request.Url.GetComponents(
    UriComponents.SchemeAndServer, UriFormat.UriEscaped)   

Or in a WebAPI controller;

string baseUrl = Url.Request.RequestUri.GetComponents(
    UriComponents.SchemeAndServer, UriFormat.Unescaped)

which is handy so you can choose what escaping format you want. I'm not clear why there are two such different implementations, and as far as I can tell, this method and @warlock's return the exact same result in this case, but it looks like GetLeftPart() would also work for non server Uri's like mailto tags for instance.

The popular GetLeftPart solution is not supported in the PCL version of Uri, unfortunately. GetComponents is, however, so if you need portability, this should do the trick:

uri.GetComponents(
    UriComponents.SchemeAndServer | UriComponents.UserInfo, UriFormat.Unescaped);

I believe that the answers above doesn't consider when the site is not in the root of the website.

This is a for WebApi controller:

string baseUrl = (Url.Request.RequestUri.GetComponents(
                    UriComponents.SchemeAndServer, UriFormat.Unescaped).TrimEnd('/') 
                 + HttpContext.Current.Request.ApplicationPath).TrimEnd('/') ;

This works for me.

Request.Url.OriginalString.Replace(Request.Url.PathAndQuery, "") + Request.ApplicationPath;

• Request.Url.OriginalString: return the complete path same as browser showing.
• Request.Url.PathAndQuery: return the (complete path) - (domain name + PORT).
• Request.ApplicationPath: return "/" on hosted server and "application name" on local IIS deploy.

So if you want to access your domain name do consider to include the application name in case of:

1. IIS deployment
2. If your application deployed on the sub-domain.

====================================

For the dev.x.us/web

it return this strong text

you could possibly add in the port for non port 80/SSL?

something like:

if (HttpContext.Current.Request.ServerVariables["SERVER_PORT"] != null && HttpContext.Current.Request.ServerVariables["SERVER_PORT"].ToString() != "80" && HttpContext.Current.Request.ServerVariables["SERVER_PORT"].ToString() != "443")
            {
                port = String.Concat(":", HttpContext.Current.Request.ServerVariables["SERVER_PORT"].ToString());
            }

and use that in the final result?

This is a much more fool proof method.

VirtualPathUtility.ToAbsolute("~/");