Im started to learn TDD just now. And i have some troubles with testing my controllers. So, i will try to explain.
I have a controller:
public AccountController(IStoreService storeService)
{
_storeService = storeService;
}
public virtual ActionResult RenderBalance()
{
var model = _storeService.GetStorePageBalanceModel();
return PartialView("MyControl", model);
}
Here i want to test my RenderBalance action:
public class when_balance_renders
{
private static Mock<IStoreService> storeService = new Mock<IStoreService>();
private static AccountController controller;
private static ActionResult result;
private Establish context = () =>
{
controller = new AccountController(storeService.Object);
result = controller.RenderBalance();
};
private It should_be_not_null_result = () => { result.ShouldNotBeNull(); };
}
But this code doesn't work. I have this error on debug mode:
Could not load file or assembly or one of its dependencies. An attempt was made to load a program with an incorrect format.
How may i fix it? And can you give me some recommendations about testing controllers.
Thanks, Nogin Anton.
If you are just starting out with TDD, try a simpler approach like classical TDD as pointed out here http://martinfowler.com/articles/mocksArentStubs.html
Also if you have this error Could not load file or assembly or one of its dependencies. An attempt was made to load a program with an incorrect format.
There is something very basic setup wrong. Remove lines of code until you can at least compile, then move forward from there.
Related
I have set up a signalR website .net core. My function in my hub is:
public async Task Notify(int id) {
await Clients.All.InvokeAsync("Notified", id);
}
I have also tested this with the following js:
let connection = new signalR.HubConnection(myURL);
connection.on('Notified', data => {
console.log(4, data);
});
connection.start();
The js code seems to work fine and I see the log when I try connection.Invoke('Notify').
Now I have a console app that can needs to make the invoke. I am trying this in two ways and don't mind either solution:
1. A mvc controller within the signalR website that can take the id and invoke 'Notified'.
2. Use the client library Microsoft.AspNetCore.SignalR.Client in the console app.
The way 1 I have only done in classic asp.net like this:
GlobalHost.ConnectionManager.GetHubContext(hubName)
But couldn't find a way to do this in .net core.
Way 2 I have used the library and tried this so far:
var con = new HubConnectionBuilder();
con.WithUrl(myURL);
var connection = con.Build();
connection.InvokeAsync("Notify",args[0]).Wait();
This is the closest I have come to create a connection in the same way as the js code. However this code throws a null pointer when calling connection.InvokeAsync. The connection object is not null. It seems to be an internal object that is null. According to the stack trace the exception is thrown when a MoveNext() function is internally called.
Well looks like both are not currently possible. As of now I just used a forced way which is hopefully temporary.
I have created and used the following base class for hubs:
public abstract class MyHub : Hub
{
private static Dictionary<string, IHubClients> _clients = new Dictionary<string, IHubClients>();
public override Task OnConnectedAsync()
{
var c = base.OnConnectedAsync();
_clients.Remove(Name);
_clients.Add(Name, Clients);
return c;
}
public static IHubClients GetClients(string Name) {
return _clients.GetValueOrDefault(Name);
}
}
GlobalHost is gone. You need to inject IHubContext<THub> like in this sample.
This can be a bug in SignalR alpha1. Can you file an issue on https://github.com/aspnet/signalr and include a simplified repro?
I'm in the process of rewriting one little WPF-App I wrote to make use of ReactiveUI, to get a feeling about the library.
I really like it so far!
Now I've stumbled upon the Throttle method and want to use it when applying a filter to a collection.
This is my ViewModel:
namespace ReactiveUIThrottle
{
public class MainViewModel : ReactiveObject
{
private string _filter;
public string Filter { get => _filter; set => this.RaiseAndSetIfChanged(ref _filter, value); }
private readonly ReactiveList<Person> _persons = new ReactiveList<Person>();
private readonly ObservableAsPropertyHelper<IReactiveDerivedList<Person>> _filteredPersons;
public IReactiveDerivedList<Person> Persons => _filteredPersons.Value;
public MainViewModel()
{
Filter = string.Empty;
_persons.AddRange(new[]
{
new Person("Peter"),
new Person("Jane"),
new Person("Jon"),
new Person("Marc"),
new Person("Heinz")
});
var filterPersonsCommand = ReactiveCommand.CreateFromTask<string, IReactiveDerivedList<Person>>(FilterPersons);
this.WhenAnyValue(x => x.Filter)
// to see the problem
.Throttle(TimeSpan.FromMilliseconds(2000), RxApp.MainThreadScheduler)
.InvokeCommand(filterPersonsCommand);
_filteredPersons = filterPersonsCommand.ToProperty(this, vm => vm.Persons, _persons.CreateDerivedCollection(p => p));
}
private async Task<IReactiveDerivedList<Person>> FilterPersons(string filter)
{
await Task.Delay(500); // Lets say this takes some time
return _persons.CreateDerivedCollection(p => p, p => p.Name.Contains(filter));
}
}
}
The filtering itself works like a charm, also the throttling, when using the GUI.
However, I'd like to unittest the behavior of the filtering and this is my first attempt:
[Test]
public void FilterPersonsByName()
{
var sut = new MainViewModel();
sut.Persons.Should().HaveCount(5);
sut.Filter = "J";
sut.Persons.Should().HaveCount(2);
}
This test fails because the collection still has 5 people.
When I get rid of the await Task.Delay(500) in FilterPersons then the test will pass, but takes 2 seconds (from the throttle).
1) Is there a way to have the throttle be instant within the test to speed up the unittest?
2) How would I test the async behavior in my filter?
I'm using ReactiveUI 7.x
Short answers:
Yes, by making sure you're using CurrentThreadScheduler.Instance when running under test
Instead of using CurrentThreadScheduler, use a TestScheduler and manually advance it
The longer answer is that you need to ensure your unit tests can control the scheduler being used by your System Under Test (SUT). By default, you'll generally want to use CurrentThreadScheduler.Instance to make things happen "instantly" without any need to advance the scheduler manually. But when you want to write tests that do validate timing, you use a TestScheduler instead.
If, as you seem to be, you're using RxApp.*Scheduler, take a look at the With extension method, which can be used like this:
(new TestScheduler()).With(sched => {
// write test logic here, and RxApp.*Scheduler will resolve to the chosen TestScheduler
});
I tend to avoid using the RxApp ambient context altogether for the same reason I avoid all ambient contexts: they're shared state and can cause trouble as a consequence. Instead, I inject an IScheduler (or two) into my SUT as a dependency.
I want to built a program which is get problems and I found that my first batches of tests involving custom components would tend to follow code:
import mx.core.Application;
import mx.events.FlexEvent;
import flexunit.framework.TestCase;
public class CustomComponentTest extends TestCase {
private var component:CustomComponent;
public function testSomeAspect() : void {
component = new CustomComponent();
component.addEventListener(FlexEvent.CREATION_COMPLETE,
addAsync(verifySomeAspect, 5000));
component.height = 0;
component.width = 0;
Application.application.addChild(component);
}
public function verifySomeAspect(event:FlexEvent) : void {}
override public function tearDown() : void {
try {
if (component) {
Application.application.removeChild(component);
component = null;
}
} catch (e:Error) {
}
}
First, you need to make sure the component has been fully initialized before you can reliably verify anything about it, and in Flex this happens asynchronously after it has been added to the display list. So you need to setup a callback (using FlexUnit's addAsync function) to be notified when that's happened.
Lately i've been just manually calling the methods that the runtime would call for you in the necessary places, so now my tests tend to look more like this:
import flexunit.framework.TestCase;
public class CustomComponentTest extends TestCase {
public function testSomeAspect() : void {
var component:CustomComponent = new CustomComponent();
component.initialize();
component.validateProperties();
}
This is much easier to follow, but it kinda feels like I'm cheating a little either way. The first case is slamming it into the current Application (which would be the unit test runner shell app), and the latter isn't a "real" environment.I was wondering how other people would handle this sort of situation?
I can agree that the second version is shorter, but I'm not sure that I think it's easier to follow. The test does a lot of things that you wouldn't normally do, whereas the first example is more true to how you would use the component outside the test environment.
Also, in the second form you have to make sure that you do exactly what the framework would do, miss one step and your test isn't relevant, and each test must repeat this code. Seems to me it's better to test it in a situation that is as close to the real thing as possible.
You could have a look at dpUint's sequences, they made component testing a little more declarative:
public function testLogin():void {
var passThroughData:Object = new Object();
passThroughData.username = "myuser1";
passThroughData.password = "somepsswd";
var sequence:SequenceRunner = new SequenceRunner(this);
sequence.addStep(new SequenceSetter(form.usernameTI,
{text:passThroughData.username}));
sequence.addStep(new SequenceWaiter(form.usernameTI,
FlexEvent.VALUE_COMMIT, 100));
sequence.addStep(new SequenceSetter(form.passwordTI,
{text:passThroughData.password}));
sequence.addStep(new SequenceWaiter(form.passwordTI, FlexEvent.VALUE_COMMIT, 100));
sequence.addStep(new SequenceEventDispatcher(form.loginBtn,
new MouseEvent("click", true, false)));
sequence.addStep(new SequenceWaiter(form, "loginRequested", 100));
sequence.addAssertHandler(handleLoginEvent, passThroughData);
sequence.run();}
I have a solution with libraries (DLLs) which are used in 2 identical projects (one for WP7, another for WP8). In one of the libraries I have the code which determines the version of the application.
private static Version mVersion;
public static Version Version {
get {
if (mVersion == default(Version)) {
var lcAssembly = Assembly.GetExecutingAssembly();
var parts = lcAssembly.FullName.Split(',');
var lcVersionStr = parts[1].Split('=')[1];
mVersion = new Version(lcVersionStr);
}
return mVersion;
}
}
The problem is that this code returns the version number of the library itself because of this Assembly.GetExecutingAssembly() code. How to get a MAIN Assembly version and not DLL's?
That's a great question on code-sharing between WP7 and WP8.
The simplest way for you to do that would be to read the AppManfiest.xml file at run-time, get the EntryType and use that to get at the entry point Assembly instance. Here's how a sample AppManfiest.xml looks like once MSBuild did its magic on it:
<Deployment xmlns="http://schemas.microsoft.com/client/2007/deployment" xmlns:x="http://schemas.microsoft.com/winfx/2006/xaml" EntryPointAssembly="myAssembly" EntryPointType="myNamespace.App" RuntimeVersion="4.7.50308.0">
<Deployment.Parts>
<AssemblyPart x:Name="myAssembly" Source="myAssembly.dll" />
</Deployment.Parts>
</Deployment>
And here's how you would read the file, get the attributes, then get the entry point type and finally the entry point assembly:
private void MainPage_Loaded(object sender, RoutedEventArgs e)
{
var appManfiest = XElement.Load("AppManifest.xaml");
var entryAssemblyName = appManfiest.Attribute("EntryPointAssembly").Value;
var entryTypeName = appManfiest.Attribute("EntryPointType").Value;
Type entryType = Type.GetType(entryTypeName + "," + entryAssemblyName);
Assembly entryAssembly = entryType.Assembly;
}
That's a simple solution and it works. However, that isn't the cleanest architectural solution. The way I'd implement this solution is to have an interface declared in the shared library, both WP7 and WP8 implement that interface and register their implementation with an IoC container.
For example, let's say you need to "DoSomething" in the shared library that's platform version specific. First you'll create have an IDoSomething interface. Let's also assume you have an IoC standing by.
public interface IDoSomething
{
}
public static class IoC
{
public static void Register<T>(T t)
{
// use some IoC container
}
public static T Get<T>()
{
// use some IoC container
}
}
In your WP7 app you'll implement the shared Interface for WP7 and register it once the WP7 starts up.
public App()
{
MainPage.IoC.Register(new MainPage.DoSomethingWP7());
}
private class DoSomethingWP7 : IDoSomething
{
}
You'll also do the same for WP8 in the WP8 app. And in your shared library you can then ask for the relevant interface regardless of its platform version specific implementation:
IDoSomething sharedInterface = IoC.Get<IDoSomething>();
I have a simpler answer. I think you are close with what you are doing. I just used your code with one modification so I can use it with the Telerik controls. Here's what I did. I located your code in my project's App class (codebehind of App.Xaml). I made one change that I think will take care of your problem:
private static Version mVersion;
public static Version Version {
get {
if (mVersion == default(Version)) {
var lcAssembly = typeof(App);
var parts = lcAssembly.FullName.Split(',');
var lcVersionStr = parts[1].Split('=')[1];
mVersion = new Version(lcVersionStr);
}
return mVersion;
}
}
Now I can get the version number by calling "App.Version".
This worked for me:
var appAssembly = Application.Current.GetType().Assembly;
var appAssemblyVersion = appAssembly.GetName().Version;
I tested with WP7.1 and WP8.0.
I have a extension method. Can any one help me how to test this method with Moq?
public static string GetBaseUrl(this UrlHelper urlHelper)
{
Uri contextUri = new Uri(urlHelper.RequestContext.HttpContext.Request.Url, urlHelper.RequestContext.HttpContext.Request.RawUrl);
UriBuilder realmUri = new UriBuilder(contextUri) { Path = urlHelper.RequestContext.HttpContext.Request.ApplicationPath, Query = null, Fragment = null };
string url = realmUri.Uri.AbsoluteUri;
if (url.EndsWith("/"))
{
url = url.Remove(url.Length - 1, 1);
}
return url;
}
many thanks.
As TrueWill points out, you can't use Moq directly with UrlHelper.RequestContext because it isn't virtual. On the other hand, UrlHelper is a public class that you can instantiate for use with unit testing.
At some point, however, you will encounter the need to assign a HttpContextBase to create the UrlHelper, and Moq can help you to do that.
Here's a test that shows that I can at least write a unit test that invokes your GetBaseUrl without throwing any exceptions:
[TestMethod]
public void Test1()
{
var httpCtxStub = new Mock<HttpContextBase>();
httpCtxStub.SetupGet(x => x.Request).Returns(() =>
{
var reqStub = new Mock<HttpRequestBase>();
reqStub.SetupGet(r => r.RawUrl).Returns("http://foo");
reqStub.SetupGet(r => r.Url).Returns(new Uri("http://foo"));
return reqStub.Object;
});
var requestCtx = new RequestContext(httpCtxStub.Object, new RouteData());
var urlHelper = new UrlHelper(requestCtx, new RouteCollection());
var result = urlHelper.GetBaseUrl();
// Assert something
}
However, this isn't the simplest unit test to write and maintain, so I support TrueWill's comment that you might make life simpler for yourself if you hide UrlHelper behind an interface.
The UrlHelper.RequestContext property is non-virtual. Moq isn't going to be of help in this case, to the best of my knowledge.
You could create a wrapper class for UrlHelper that implements an interface, but that would seem to defeat the purpose of using an extension method.
Typemock would probably do what you want, if you have the budget for a commercial program. (I haven't tried it; I use Moq myself.)
Another option would be to write integration tests against this method; while they would run more slowly than unit tests, I suspect this method is unlikely to version often.
A larger issue is coupling to UrlHelper reducing testability in the rest of your application. Perhaps other posters can suggest answers to that issue.