Posts Tagged ‘TeamFoundationServer’
Like so many things in the programing world, establishing a connection to a TF server can be done in a couple of different ways. I have previously specified everything that you need to start developing an application that will interface with TF server, so I will skip the introduction to TFS development and get my hands dirty right away.
From you code, C# code that is, you can establish a connection to TF server in two basic ways. Both of two create an instance of a TeamFoundationServer class. One is trough a service class TeamFoundationServerFactory and the other is creating an instance of a TeamFoundationServer class manually through a public constructor.
The main difference between the two is that the TeamFoundationServerFactory static class will cache the connected servers, so that every time you request a server, a request is not being sent to the TF server, thus not clogging the server with requests.
TeamFoundationServer server = TeamFoundationServerFactory.GetServer("http://" + serverName + ":" + serverPort);
Now you have your instance of a TeamFoundationServer. Each time you call the GetServer(String) method it will check if the server has been already accessed, and if it was the method will return a cached instance otherwise it will create a new one.You other option is to create an instance of a TeamFoundationServer on your own.
TeamFoundationServer server = new TeamFoundationServer("http://" + serverName + ":" + serverPort);
If you it this way a request will be sent to the TF server to check if it exists, if you do that every time you need something from the server, you will generate a lot of request to the server.Basically if you use the TeamFoundationServerFactory class the framework will handle the caching of the server for you, if you create an instance on your own you will have to store that instance and keep track of it your self.
The TeamFoundationServer constructor way has one benefit over the GetServer(String) method. The benefit is seen when you get around to authenticating the users. When you want to provide user authentication through the GetServer method, you will use the overload that has looks like GetServer(String, ICredentialsProvider)
using the UICredentialsProvider class like so:
TeamFoundationServer server = TeamFoundationServerFactory.GetServer("http://" + serverName + ":" + serverPort, new UICredentialsProvider()); server.EnsureAuthenticated();
What this will do, it will create a pop up requesting the user credentials every time the user needs to be re-authenticated. The EnsureAuthenticated() method will first check if the user has already been authenticated, and if not it will create a login pop up for user credentials then authenticating the user.In your software you might want to save the user credentials to use them any time the user needs to authenticate. That’s when TeamFoundationServer Constructor (String, ICredentials, ICredentialsProvider) constructor comes into play. With this constructor you can specify in addition to a UICredentialsProvider object that is used as a fall back if the authentication fails, separate NetworkCredential object to be used for authentication.
NetworkCredential networkCredentials = new NetworkCredential( userName, password, domain); TeamFoundationServer server = new TeamFoundationServer( "http://" + serverName + ":" + serverPort, networkCredentials, new UICredentialsProvider()); server.EnsureAuthenticated();
This way the code will try to authenticate first with the selected NetworkCredentials and if that fails it will use the UICredentialsProvider to create the pop up and ask the user for credentials. This way you can save and reuse the users credentials when you want to, and if the authentication fails the framework will ask the user for credentials on its own.
I was having trouble connection to my company’s TF server, since our workstations are in one domain and the TF server is in a separate domain, with an ISA and a Web filtering server in between. I was getting the HTTP error code 407 “proxy authentication failed”, until I found a way to deal with it I lost around 4 hours. The solution is quite simple you just add this to to your config file.
<system.net> <defaultProxy enabled="true" useDefaultCredentials="true"> <bypasslist> <add address="serverName" /> </bypasslist> </defaultProxy> </system.net>
When you put this in the application will by pass the proxy for the specified network addresses.