Path.Combine for URLs?

Question

Path.Combine is handy, but is there a similar function in the .NET framework for URLs?

I'm looking for syntax like this:

Url.Combine("http://MyUrl.com/", "/Images/Image.jpg")

which would return:

"http://MyUrl.com/Images/Image.jpg"

Answer 1

Path.Combine does not work for me because there can be characters like "|" in QueryString arguments and therefore the URL, which will result in an ArgumentException.

I first tried the new Uri(Uri baseUri, string relativeUri) approach, which failed for me because of URIs like http://www.mediawiki.org/wiki/Special:SpecialPages:

new Uri(new Uri("http://www.mediawiki.org/wiki/"), "Special:SpecialPages")

will result in Special:SpecialPages, because of the colon after Special that denotes a scheme.

So I finally had to take mdsharpe/Brian MacKays route and developed it a bit further to work with multiple URI parts:

public static string CombineUri(params string[] uriParts)
{
    string uri = string.Empty;
    if (uriParts != null && uriParts.Count() > 0)
    {
        char[] trims = new char[] { '\\', '/' };
        uri = (uriParts[0] ?? string.Empty).TrimEnd(trims);
        for (int i = 1; i < uriParts.Count(); i++)
        {
            uri = string.Format("{0}/{1}", uri.TrimEnd(trims), (uriParts[i] ?? string.Empty).TrimStart(trims));
        }
    }
    return uri;
}

Usage: CombineUri("http://www.mediawiki.org/", "wiki", "Special:SpecialPages")

Answer 2

There is a Todd Menier's comment above that Flurl includes a Url.Combine.

More details:

Url.Combine is basically a Path.Combine for URLs, ensuring one and only one separator character between parts:

var url = Url.Combine(
    "http://foo.com/",
    "/too/", "/many/", "/slashes/",
    "too", "few?",
    "x=1", "y=2"
// result: "http://www.foo.com/too/many/slashes/too/few?x=1&y=2" 

Answer 3

Combining multiple parts of a URL could be a little bit tricky. You can use the two-parameter constructor Uri(baseUri, relativeUri), or you can use the Uri.TryCreate() utility function.

In either case, you might end up returning an incorrect result because these methods keep on truncating the relative parts off of the first parameter baseUri, i.e. from something like http://google.com/some/thing to http://google.com.

To be able to combine multiple parts into a final URL, you can copy the two functions below:

    public static string Combine(params string[] parts)
    {
        if (parts == null || parts.Length == 0) return string.Empty;

        var urlBuilder = new StringBuilder();
        foreach (var part in parts)
        {
            var tempUrl = tryCreateRelativeOrAbsolute(part);
            urlBuilder.Append(tempUrl);
        }
        return VirtualPathUtility.RemoveTrailingSlash(urlBuilder.ToString());
    }

    private static string tryCreateRelativeOrAbsolute(string s)
    {
        System.Uri uri;
        System.Uri.TryCreate(s, UriKind.RelativeOrAbsolute, out uri);
        string tempUrl = VirtualPathUtility.AppendTrailingSlash(uri.ToString());
        return tempUrl;
    }

Full code with unit tests to demonstrate usage can be found at https://uricombine.codeplex.com/SourceControl/latest#UriCombine/Uri.cs

I have unit tests to cover the three most common cases:

Answer 4

Ryan Cook's answer is close to what I'm after and may be more appropriate for other developers. However, it adds http:// to the beginning of the string and in general it does a bit more formatting than I'm after.

Also, for my use cases, resolving relative paths is not important.

mdsharp's answer also contains the seed of a good idea, although that actual implementation needed a few more details to be complete. This is an attempt to flesh it out (and I'm using this in production):

C#

public string UrlCombine(string url1, string url2)
{
    if (url1.Length == 0) {
        return url2;
    }

    if (url2.Length == 0) {
        return url1;
    }

    url1 = url1.TrimEnd('/', '\\');
    url2 = url2.TrimStart('/', '\\');

    return string.Format("{0}/{1}", url1, url2);
}

VB.NET

Public Function UrlCombine(ByVal url1 As String, ByVal url2 As String) As String
    If url1.Length = 0 Then
        Return url2
    End If

    If url2.Length = 0 Then
        Return url1
    End If

    url1 = url1.TrimEnd("/"c, "\"c)
    url2 = url2.TrimStart("/"c, "\"c)

    Return String.Format("{0}/{1}", url1, url2)
End Function

This code passes the following test, which happens to be in VB:

<TestMethod()> Public Sub UrlCombineTest()
    Dim target As StringHelpers = New StringHelpers()

    Assert.IsTrue(target.UrlCombine("test1", "test2") = "test1/test2")
    Assert.IsTrue(target.UrlCombine("test1/", "test2") = "test1/test2")
    Assert.IsTrue(target.UrlCombine("test1", "/test2") = "test1/test2")
    Assert.IsTrue(target.UrlCombine("test1/", "/test2") = "test1/test2")
    Assert.IsTrue(target.UrlCombine("/test1/", "/test2/") = "/test1/test2/")
    Assert.IsTrue(target.UrlCombine("", "/test2/") = "/test2/")
    Assert.IsTrue(target.UrlCombine("/test1/", "") = "/test1/")
End Sub

Answer 5

I used this code to solve the problem:

string[] brokenBaseUrl = Context.Url.TrimEnd('/').Split('/');
string[] brokenRootFolderPath = RootFolderPath.Split('/');

for (int x = 0; x < brokenRootFolderPath.Length; x++)
{
    //if url doesn't already contain member, append it to the end of the string with / in front
    if (!brokenBaseUrl.Contains(brokenRootFolderPath[x]))
    {
        if (x == 0)
        {
            RootLocationUrl = Context.Url.TrimEnd('/');
        }
        else
        {
            RootLocationUrl += String.Format("/{0}", brokenRootFolderPath[x]);
        }
    }
}

Answer 6

Rules while combining URLs with a URI

To avoid strange behaviour there's one rule to follow:

• The path (directory) must end with '/'. If the path ends without '/', the last part is treated like a file-name, and it'll be concatenated when trying to combine with the next URL part.
• There's one exception: the base URL address (without directory info) needs not to end with '/'
• the path part must not start with '/'. If it start with '/', every existing relative information from URL is dropped...adding a string.Empty part path will remove the relative directory from the URL too!

If you follow rules above, you can combine URLs with the code below. Depending on your situation, you can add multiple 'directory' parts to the URL...

        var pathParts = new string[] { destinationBaseUrl, destinationFolderUrl, fileName };

        var destination = pathParts.Aggregate((left, right) =>
        {
            if (string.IsNullOrWhiteSpace(right))
                return left;

            return new Uri(new Uri(left), right).ToString();
        });