I am trying to figure out the reason why my unit test would fail when run together with other unit tests in the solution but pass when run alone. Can anyone show me what am I missing?
The SUT is a class called CompositeClient that is essentially a wrapper class around two other clients. It's main idea is to give priority to one of those clients to be called.
public class CompositeClient : IReceiverChannel
{
private static readonly List<IReceiverChannel> ReceiverChannels = new List<IReceiverChannel>();
public CompositeClient(IReceiverChannel priority, IReceiverChannel normal)
{
ReceiverChannels.Add(priority);
ReceiverChannels.Add(normal);
}
public async Task<IEnumerable<Request>> ReceiveBatchAsync(int batchSize)
{
var req = new List<Request>();
foreach (var channel in ReceiverChannels)
{
req.AddRange(await channel.ReceiveBatchAsync(batchSize - req.Count).ConfigureAwait(false));
if (req.Count >= batchSize)
{
break;
}
}
return req;
}
}
Running the unit test below with all the other unit tests in the solution yield me a failed result. But if I run this test alone, it will pass.
[TestMethod]
public async Task ReceivedRequestShouldComeFromPriorityClientFirst()
{
var normalPriorityClient = GetNormalClientMock();
var highPriorityClient = GetPriorityClientMock();
var compositeClient = new CompositeClient(highPriorityClient, normalPriorityClient);
var requests = await compositeClient.ReceiveBatchAsync(1);
requests.Should().HaveCount(1);
requests.First().Origin.Should().BeSameAs("priority");
normalPriorityClient.CallCount.Should().Be(1); // It will fail here with actual CallCount = 0.
highPriorityClient.CallCount.Should().Be(0);
}
private static ReceiverChannelMock GetNormalClientMock()
{
return new ReceiverChannelMock("normal");
}
private static ReceiverChannelMock GetPriorityClientMock()
{
return new ReceiverChannelMock("priority");
}
private class ReceiverChannelMock : IReceiverChannel
{
private readonly string name;
public ReceiverChannelMock(string name)
{
this.name = name;
}
public int CallCount { get; private set; }
public Task<IEnumerable<Request>> ReceiveBatchAsync(int batchSize)
{
this.CallCount++;
return Task.FromResult<IEnumerable<Request>>(
new List<Request>
{
new Request
{
Origin = this.name
}
});
}
}
Tools used:
Visual Studio 2013
.NET Framework 4.5.2
Resharper 9.2
FluentAssertion
As David pointed out, I overlooked the static field that I declared in the CompositeClient class. Removing the static keyword solved the issue.
Related
I'm trying to customize some of nUnits behaviour, however I'm constantly hitting a brick wall
because of nUnits heavy use of code reflection. Test methods (and also setup methods etc) are passed all the way down, deep into the framework, and are converted into a delegate at the latest step possible.
The classes I'm interested in are called TestCommands and only there the framework becomes functional.
For reference here is a snippet I found in nUnits source of the TestMethodCommand class which propably is the bread and butter test execution delegate.
public class TestMethodCommand : TestCommand
{
private readonly TestMethod testMethod;
private readonly object[] arguments;
public TestMethodCommand(TestMethod testMethod) : base(testMethod)
{
this.testMethod = testMethod;
this.arguments = testMethod.Arguments;
}
public override TestResult Execute(TestExecutionContext context)
{
object result = InvokeTestMethod(context); // missing a branch deciding about sync vs. async
// missing some code that checks object against "expected result"
return context.CurrentResult;
}
private object InvokeTestMethod(TestExecutionContext context)
{
return testMethod.Method.Invoke(context.TestObject, arguments);
}
}
I'm puzzled why nUnit couldn't wrap the test method into an Func<object> way way sooner and just pass the context along. As it stands for now if I don't have a MethodInfo nUnit can't run it.
In case you wonder, here is an example of a thing I want to do but I ran into the same problem in other instances as well.
[Scenario(When: "Device Registration reads out PCB Type",
Then: "Device Type might change")]
public void Identifier_Changes_Are_Recognized()
{
var changedType = reference.ChangeType(DeviceType.Terminal);
var changedID = reference.ChangeID(123456);
Assert.Multiple(() =>
{
AssertIsSameDevice(reference, changedType);
AssertIsDifferentDevice(reference, changedID);
});
}
This scenario attribute is supposed to print a small description like so.
public void RunBeforeTest()
{
var text = new MultiLineText
("Scenario:",
"\tGiven:\t" + When,
"\tThen:\t" + Then,
"-------------\n"
);
Console.WriteLine(text);
}
I reaaallly want to tell nUnit "Look, here is an action, please run it" but for the time beeing this seems very hard to achieve. Did anyone else here run in these kinds of problems?
Are there possibly ways to achieve what I'm trying to do? Maybe create my own TestCommand, but as I mentioned, these objects only get created very deep into the framework.
OP here (Writing from my home account)
I looked more into this and actually found a working solution:
public class ArbitraryCodeExecutionWrapper : DelegatingTestCommand
{
public ArbitraryCodeExecutionWrapper(TestCommand innerCommand) : base(innerCommand)
{
}
public Action<TestExecutionContext> BeforeTest { get; init; } = _ => { };
public Action<Test, TestResult> AfterTest { get; init; } = (_,_) => { };
public override TestResult Execute(TestExecutionContext context)
{
BeforeTest(context);
var result = innerCommand.Execute(context);
AfterTest(context.CurrentTest, result);
return result;
}
}
public class NUnitTestCommandWrapperAttribute : Attribute, IWrapTestMethod
{
protected virtual void BeforeTest(TestExecutionContext context)
{
}
protected virtual void AfterTest(Test test, TestResult result)
{
}
public TestCommand Wrap(TestCommand command)
=> new ArbitraryCodeExecutionWrapper(command)
{
BeforeTest = BeforeTest,
AfterTest = AfterTest
};
}
public class ScenarioAttribute : NUnitTestCommandWrapperAttribute
{
public string When { get; init; } = "";
public string Then { get; init; } = "";
protected override void BeforeTest(TestExecutionContext context)
{
var text = new MultiLineText
("Scenario:",
"\tGiven:\t" + When,
"\tThen:\t" + Then
);
Console.WriteLine(text);
}
protected override void AfterTest(Test test, TestResult result)
{
Console.WriteLine("After Test");
}
}
[TestFixture]
public class TestCodeExecution
{
[Test]
[Scenario(When = "nUnit Comes here",
Then = "Print Hello World")]
public void Try_Out_Code_Execution()
{
Console.WriteLine("Hello World");
}
}
public class MultiLineText
{
private List<string> items = new();
public static implicit operator string(MultiLineText text) => text.ToString();
public MultiLineText(params string[] lines)
{
items = lines.ToList();
}
public override string ToString() => string.Join("\n", items);
}
Please see the update at the bottom!
I am setting up my framework and currently have 4 Tests. Individually they all run like a charm. However when I try to run all 4 in parallel (I have set up the Parallelizable attribute up correctly and am calling tests from different classes not within the same method) I am consistently getting several errors that seem to jump around each test. These are the messages that I am getting each run:
Again These objects are found when the tests are run individually. I am not sure what code I need to show in order to help. Please advise.
UPDATE** #Chris my suspicions are the same. I think my tests are confusing the same driver when looking for objects. If that is the case can someone please advise how to better handle this, my browser class is what is calling the driver.
public static class Browser
{
private static IWebDriver driver;
private static string baseURL = "someURL";
public static ISearchContext Driver { get { return driver; } }
internal static bool WaitUntilElementIsDisplayed(By element, int timeout)
{
for (int i = 0; i < timeout; i++)
{
if (ElementIsDisplayed(element))
{
return true;
}
Thread.Sleep(1000);
}
return false;
}
internal static IWebElement FindElement(By by)
{
return driver.FindElement(by);
}
public static bool ElementIsDisplayed(By element)
{
var present = false;
driver.Manage().Timeouts().ImplicitlyWait(System.TimeSpan.FromSeconds(0));
try
{
present = driver.FindElement(element).Displayed;
}
catch (NoSuchElementException)
{ }
driver.Manage().Timeouts().ImplicitlyWait(TimeSpan.FromSeconds(10));
return present;
}
public static void Initialize()
{
var options = new InternetExplorerOptions();
options.IntroduceInstabilityByIgnoringProtectedModeSettings = true;
options.EnsureCleanSession = true;
options.IgnoreZoomLevel = true;
driver =
new InternetExplorerDriver(
#"C:Myfilepath",
options, TimeSpan.FromMinutes(10));
Goto("");
}
public static void CleanUp()
{
driver.Close();
driver.Quit();
}
public static void Goto(string URL, bool userBaseURL = true)
{
if (userBaseURL)
driver.Navigate().GoToUrl(string.Format("{0}/{1}", baseURL, URL));
else
driver.Navigate().GoToUrl(URL);
}
}
Newest Update: per the recommendation I have removed the static references but could someone help me with the syntax on creating an instance of the driver within my current code
public class Pages
{
private T GetPage<T>() where T : new()
{
var page = new T();
PageFactory.InitElements(Browser.Driver, page);
return page;
}
public LoginPage Login
{
get { return GetPage<LoginPage>(); }
}
public RegisterPage Register
{ get { return GetPage<RegisterPage>(); } }
public SearchPage Search
{ get { return GetPage<SearchPage>(); } }
}
I am not sure how to create an instance of Browser.Driver Please help!
Remove all references to "static" in your class and create an instance of the class in each test to fix your issue.
...Now change your Page class to accept the driver in the constructor
public class Pages
{
private readonly ISearchContext _context;
public Pages(ISearchContext context)
{
_context = context;
}
private T GetPage<T>() where T : new()
{
var page = new T();
PageFactory.InitElements(_context, page);
return page;
}
public LoginPage Login
{
get { return GetPage<LoginPage>(); }
}
public RegisterPage Register
{ get { return GetPage<RegisterPage>(); } }
public SearchPage Search
{ get { return GetPage<SearchPage>(); } }
}
... then in your test method
var browser = new Browser();
var page = new Page(browser.Driver);
Sorry. Been away and noticed your updates.
I have a separate class helper that I use to return my web driver. I’m using chrome driver and (headless) unit driver, which on my machines polices requires several params to get it running, so a class in its own right makes senses to me. E.g. WebDriverHelper.java. This has several static methods that returns a new instance of the driver of interest.
E.g.
WebDriver myDriver = WebDriverHelper.ChromeDriver();
My ChromeDriver method returns a new driver.
E.g.
return new ChromeDriver;
If you need more detail, let me know and I’ll copy some of my classes when I get in work tomorrow.
Is there in C# some kind of equivalent of ExpectedSystemExit in Java? I have an exit in my code and would really like to be able to test it. The only thing I found in C# is a not really nice workaround.
Example Code
public void CheckRights()
{
if(!service.UserHasRights())
{
Environment.Exit(1);
}
}
Test Code
[TestMethod]
public void TestCheckRightsWithoutRights()
{
MyService service = ...
service.UserHasRights().Returns(false);
???
}
I am using the VS framework for testing (+ NSubstitute for mocking) but it is not a problem to switch to nunit or whatever for this test.
You should use dependency injection to supply to the class being tested an interface that provides an environmental exit.
For example:
public interface IEnvironment
{
void Exit(int code);
}
Let's also assume that you have an interface for calling UserHasRights():
public interface IRightsService
{
bool UserHasRights();
}
Now suppose your class to be tested looks like this:
public sealed class RightsChecker
{
readonly IRightsService service;
readonly IEnvironment environment;
public RightsChecker(IRightsService service, IEnvironment environment)
{
this.service = service;
this.environment = environment;
}
public void CheckRights()
{
if (!service.UserHasRights())
{
environment.Exit(1);
}
}
}
Now you can use a mocking framework to check that IEnvironment .Exit() is called under the right conditions. For example, using Moq it might look a bit like this:
[TestMethod]
public static void CheckRights_exits_program_when_user_has_no_rights()
{
var rightsService = new Mock<IRightsService>();
rightsService.Setup(foo => foo.UserHasRights()).Returns(false);
var enviromnent = new Mock<IEnvironment>();
var rightsChecker = new RightsChecker(rightsService.Object, enviromnent.Object);
rightsChecker.CheckRights();
enviromnent.Verify(foo => foo.Exit(1));
}
Ambient contexts and cross-cutting concerns
A method such as Environment.Exit() could be considered to be a cross-cutting concern, and you might well want to avoid passing around an interface for it because you can end up with an explosion of additional constructor parameters. (Note: The canonical example of a cross cutting concern is DateTime.Now.)
To address this issue, you can introduce an "Ambient context" - a pattern which allows you to use a static method while still retaining the ability to unit test calls to it. Of course, such things should be used sparingly and only for true cross-cutting concerns.
For example, you could introduce an ambient context for Environment like so:
public abstract class EnvironmentControl
{
public static EnvironmentControl Current
{
get
{
return _current;
}
set
{
if (value == null)
throw new ArgumentNullException(nameof(value));
_current = value;
}
}
public abstract void Exit(int value);
public static void ResetToDefault()
{
_current = DefaultEnvironmentControl.Instance;
}
static EnvironmentControl _current = DefaultEnvironmentControl.Instance;
}
public class DefaultEnvironmentControl : EnvironmentControl
{
public override void Exit(int value)
{
Environment.Exit(value);
}
public static DefaultEnvironmentControl Instance => _instance.Value;
static readonly Lazy<DefaultEnvironmentControl> _instance = new Lazy<DefaultEnvironmentControl>(() => new DefaultEnvironmentControl());
}
Normal code just calls EnvironmentControl.Current.Exit(). With this change, the IEnvironment parameter disappears from the RightsChecker class:
public sealed class RightsChecker
{
readonly IRightsService service;
public RightsChecker(IRightsService service)
{
this.service = service;
}
public void CheckRights()
{
if (!service.UserHasRights())
{
EnvironmentControl.Current.Exit(1);
}
}
}
But we still retain the ability to unit-test that it has been called:
public static void CheckRights_exits_program_when_user_has_no_rights()
{
var rightsService = new Mock<IRightsService>();
rightsService.Setup(foo => foo.UserHasRights()).Returns(false);
var enviromnent = new Mock<EnvironmentControl>();
EnvironmentControl.Current = enviromnent.Object;
try
{
var rightsChecker = new RightsChecker(rightsService.Object);
rightsChecker.CheckRights();
enviromnent.Verify(foo => foo.Exit(1));
}
finally
{
EnvironmentControl.ResetToDefault();
}
}
For more information about ambient contexts, see here.
I ended up creating a new method which I can then mock in my tests.
Code
public void CheckRights()
{
if(!service.UserHasRights())
{
Environment.Exit(1);
}
}
internal virtual void Exit()
{
Environment.Exit(1);
}
Unit test
[TestMethod]
public void TestCheckRightsWithoutRights()
{
MyService service = ...
service.When(svc => svc.Exit()).DoNotCallBase();
...
service.CheckRights();
service.Received(1).Exit();
}
If your goal is to avoid extra classes/interfaces just to support tests, how do you feel about Environment.Exit action via Property Injection?
class RightsChecker
{
public Action AccessDeniedAction { get; set; }
public RightsChecker(...)
{
...
AccessDeniedAction = () => Environment.Exit();
}
}
[Test]
public TestCheckRightsWithoutRights()
{
...
bool wasAccessDeniedActionExecuted = false;
rightsChecker.AccessDeniedAction = () => { wasAccessDeniedActionExecuted = true; }
...
Assert.That(wasAccessDeniedActionExecuted , Is.True);
}
Is there a possibility to cache a collection, retrieved using WCF from an OData service.
The situation is the following:
I generated a WCF service client with Visual Studio 2015 using the metadata of the odata service. VS generated a class inheriting from System.Data.Services.Client.DataServiceContext. This class has some properties of type System.Data.Services.Client.DataServiceQuery<T>. The data of some of these properties change seldom. Because of performance reasons I want the WCF client to load these properties just the first time and not every time I use it in the code.
Is there a built in possibility to cache the data of these properties? Or can I tell the service client not to load specific proeprties newly every time.
Assuming the service client class is ODataClient and one of its properties is `Area, for now I get the values in the following way:
var client = new ODataClient("url_to_the_service");
client.IgnoreMissingProperties = true;
var propertyInfo = client.GetType().GetProperty("Area");
var area = propertyInfo.GetValue(client) as IEnumerable<object>;
The reason why I do this in such a complicated way is, that the client should be very generic: The properties to be handled can be configured in a configuration file.
* EDIT *
I already tried to find properties in the System.Data.Services.Client.DataServiceContext class or the System.Data.Services.Client.DataServiceQuery<T> class for the caching. But i wasn't able to find any.
To my knowledge there is no "out of the box" caching concept on the client. There are options for caching the output of a request on the server which is something you might want consider as well. Googling "WCF Caching" would get you a bunch of info on this.
Regarding client side caching...#Evk is correct it is pretty straight forward. Here is an sample using MemoryCache.
using System;
using System.Runtime.Caching;
namespace Services.Util
{
public class CacheWrapper : ICacheWrapper
{
ObjectCache _cache = MemoryCache.Default;
public void ClearCache()
{
MemoryCache.Default.Dispose();
_cache = MemoryCache.Default;
}
public T GetFromCache<T>(string key, Func<T> missedCacheCall)
{
return GetFromCache<T>(key, missedCacheCall, TimeSpan.FromMinutes(5));
}
public T GetFromCache<T>(string key, Func<T> missedCacheCall, TimeSpan timeToLive)
{
var result = _cache.Get(key);
if (result == null)
{
result = missedCacheCall();
if (result != null)
{
_cache.Set(key, result, new CacheItemPolicy { AbsoluteExpiration = DateTimeOffset.Now.Add(timeToLive) });
}
}
return (T)result;
}
public void InvalidateCache(string key)
{
_cache.Remove(key);
}
}
}
This is an example of code that uses the cache...
private class DataAccessTestStub
{
public const string DateTimeTicksCacheKey = "GetDateTimeTicks";
ICacheWrapper _cache;
public DataAccessTestStub(ICacheWrapper cache)
{
_cache = cache;
}
public string GetDateTimeTicks()
{
return _cache.GetFromCache(DateTimeTicksCacheKey, () =>
{
var result = DateTime.Now.Ticks.ToString();
Thread.Sleep(100); // Create some delay
return result;
});
}
public string GetDateTimeTicks(TimeSpan timeToLive)
{
return _cache.GetFromCache(DateTimeTicksCacheKey, () =>
{
var result = DateTime.Now.Ticks.ToString();
Thread.Sleep(500); // Create some delay
return result;
}, timeToLive);
}
public void ClearDateTimeTicks()
{
_cache.InvalidateCache(DateTimeTicksCacheKey);
}
public void ClearCache()
{
_cache.ClearCache();
}
}
And some tests if you fancy...
[TestClass]
public class CacheWrapperTest
{
private DataAccessTestStub _dataAccessTestClass;
[TestInitialize]
public void Init()
{
_dataAccessTestClass = new DataAccessTestStub(new CacheWrapper());
}
[TestMethod]
public void GetFromCache_ShouldExecuteCacheMissCall()
{
var original = _dataAccessTestClass.GetDateTimeTicks();
Assert.IsNotNull(original);
}
[TestMethod]
public void GetFromCache_ShouldReturnCachedVersion()
{
var copy1 = _dataAccessTestClass.GetDateTimeTicks();
var copy2 = _dataAccessTestClass.GetDateTimeTicks();
Assert.AreEqual(copy1, copy2);
}
[TestMethod]
public void GetFromCache_ShouldRespectTimeToLive()
{
_dataAccessTestClass.ClearDateTimeTicks();
var copy1 = _dataAccessTestClass.GetDateTimeTicks(TimeSpan.FromSeconds(2));
var copy2 = _dataAccessTestClass.GetDateTimeTicks();
Assert.AreEqual(copy1, copy2);
Thread.Sleep(3000);
var copy3 = _dataAccessTestClass.GetDateTimeTicks();
Assert.AreNotEqual(copy1, copy3);
}
[TestMethod]
public void InvalidateCache_ShouldClearCachedVersion()
{
var original = _dataAccessTestClass.GetDateTimeTicks();
_dataAccessTestClass.ClearDateTimeTicks();
var updatedVersion = _dataAccessTestClass.GetDateTimeTicks();
Assert.AreNotEqual(original, updatedVersion);
}
}
I have set up a testproject using NUnit and Selenium Webdriver of which you can find a shortened version below.
class ByHolder
{
public readonly string name, path;
public readonly Func<string, By> call;
public ByHolder(string name, string path, Func<string, By> call)
{
this.name = name;
this.path = path;
this.call = call;
}
}
class Page
{
private readonly List<ByHolder> LocatorList = new List<ByHolder>();
public Page()
{
SetUpList();
}
private void SetUpList()
{
AddLocator("Button0", "//button0", By.XPath);
AddLocator("Button1", "button1", By.Id);
...
}
public By Get(string locatorName)
{
var holder = LocatorList.FirstOrDefault(p => p.name.Equals(locatorName));
return holder?.call(holder.path);
}
public void AddLocator(string name, string path, Func<string, By> call)
{
LocatorList.Add(new ByHolder(name, path,call ));
}
}
class PersonelDriver : IWebDriver
{
IWebDriver driver = new FirefoxDriver();
Page page = new Page();
public void Click(string locatorName)
{
driver.FindElement(page.Get(locatorName)).Click();
}
...
}
[TestFixture]
class PageTest
{
private readonly PersonelDriver d = new PersonelDriver();
[Test]
public void ClickTest0()
{
d.Click("Button0");
}
[Test]
public void ClickTest1()
{
d.Click("Button1");
}
...
}
As you can hopefully see I tried implementing a shortened method with a minimum of variables to make longer testcases easier to read mainly for outsiders but also for me, for example.
d.Click("that");
d.EnterText("thisLocator","text");
d.WaitFor("somethingElse");
d.Click("this");
(After using Selenium for some time I find that things can become chaotic quite fast when repeatedly using the driver.FindElement... in the tests themselves.)
Even tough I'm happy with the shortened versions and readability, there is of course no autocomplete or check since i'm handling strings and not IWebElement objects or By references that have been named or put in a specific getter.
What I used to do was the following, but it just felt wrong:
class Locators
{
public By GetButton()
{
return By.Id("button");
}
...
}
I was wondering if there is a way to implement an autocomplete or some other check for the string values when adding for example d.Click("stringvalue");
Thank you in advance.