Testing with your own test doubles
Note
EF6 Onwards Only - The features, APIs, etc. discussed in this page were introduced in Entity Framework 6. If you are using an earlier version, some or all of the information does not apply.
When writing tests for your application it is often desirable to avoid hitting the database. Entity Framework allows you to achieve this by creating a context – with behavior defined by your tests – that makes use of in-memory data.
Options for creating test doubles
There are two different approaches that can be used to create an in-memory version of your context.
- Create your own test doubles – This approach involves writing your own in-memory implementation of your context and DbSets. This gives you a lot of control over how the classes behave but can involve writing and owning a reasonable amount of code.
- Use a mocking framework to create test doubles – Using a mocking framework (such as Moq) you can have the in-memory implementations of you context and sets created dynamically at runtime for you.
This article will deal with creating your own test double. For information on using a mocking framework see Testing with a Mocking Framework.
Testing with pre-EF6 versions
The code shown in this article is compatible with EF6. For testing with EF5 and earlier version see Testing with a Fake Context.
Limitations of EF in-memory test doubles
In-memory test doubles can be a good way to provide unit test level coverage of bits of your application that use EF. However, when doing this you are using LINQ to Objects to execute queries against in-memory data. This can result in different behavior than using EF’s LINQ provider (LINQ to Entities) to translate queries into SQL that is run against your database.
One example of such a difference is loading related data. If you create a series of Blogs that each have related Posts, then when using in-memory data the related Posts will always be loaded for each Blog. However, when running against a database the data will only be loaded if you use the Include method.
For this reason, it is recommended to always include some level of end-to-end testing (in addition to your unit tests) to ensure your application works correctly against a database.
Following along with this article
This article gives complete code listings that you can copy into Visual Studio to follow along if you wish. It's easiest to create a Unit Test Project and you will need to target .NET Framework 4.5 to complete the sections that use async.
Creating a context interface
We're going to look at testing a service that makes use of an EF model. In order to be able to replace our EF context with an in-memory version for testing, we'll define an interface that our EF context (and it's in-memory double) will implement.
The service we are going to test will query and modify data using the DbSet properties of our context and also call SaveChanges to push changes to the database. So we're including these members on the interface.
using System.Data.Entity;
namespace TestingDemo
{
public interface IBloggingContext
{
DbSet<Blog> Blogs { get; }
DbSet<Post> Posts { get; }
int SaveChanges();
}
}
The EF model
The service we're going to test makes use of an EF model made up of the BloggingContext and the Blog and Post classes. This code may have been generated by the EF Designer or be a Code First model.
using System.Collections.Generic;
using System.Data.Entity;
namespace TestingDemo
{
public class BloggingContext : DbContext, IBloggingContext
{
public DbSet<Blog> Blogs { get; set; }
public DbSet<Post> Posts { get; set; }
}
public class Blog
{
public int BlogId { get; set; }
public string Name { get; set; }
public string Url { get; set; }
public virtual List<Post> Posts { get; set; }
}
public class Post
{
public int PostId { get; set; }
public string Title { get; set; }
public string Content { get; set; }
public int BlogId { get; set; }
public virtual Blog Blog { get; set; }
}
}
Implementing the context interface with the EF Designer
Note that our context implements the IBloggingContext interface.
If you are using Code First then you can edit your context directly to implement the interface. If you are using the EF Designer then you’ll need to edit the T4 template that generates your context. Open up the <model_name>.Context.tt file that is nested under you edmx file, find the following fragment of code and add in the interface as shown.
<#=Accessibility.ForType(container)#> partial class <#=code.Escape(container)#> : DbContext, IBloggingContext
Service to be tested
To demonstrate testing with in-memory test doubles we are going to be writing a couple of tests for a BlogService. The service is capable of creating new blogs (AddBlog) and returning all Blogs ordered by name (GetAllBlogs). In addition to GetAllBlogs, we’ve also provided a method that will asynchronously get all blogs ordered by name (GetAllBlogsAsync).
using System.Collections.Generic;
using System.Data.Entity;
using System.Linq;
using System.Threading.Tasks;
namespace TestingDemo
{
public class BlogService
{
private IBloggingContext _context;
public BlogService(IBloggingContext context)
{
_context = context;
}
public Blog AddBlog(string name, string url)
{
var blog = new Blog { Name = name, Url = url };
_context.Blogs.Add(blog);
_context.SaveChanges();
return blog;
}
public List<Blog> GetAllBlogs()
{
var query = from b in _context.Blogs
orderby b.Name
select b;
return query.ToList();
}
public async Task<List<Blog>> GetAllBlogsAsync()
{
var query = from b in _context.Blogs
orderby b.Name
select b;
return await query.ToListAsync();
}
}
}
Creating the in-memory test doubles
Now that we have the real EF model and the service that can use it, it's time to create the in-memory test double that we can use for testing. We've created a TestContext test double for our context. In test doubles we get to choose the behavior we want in order to support the tests we are going to run. In this example we're just capturing the number of times SaveChanges is called, but you can include whatever logic is needed to verify the scenario you are testing.
We've also created a TestDbSet that provides an in-memory implementation of DbSet. We've provided a complete implemention for all the methods on DbSet (except for Find), but you only need to implement the members that your test scenario will use.
TestDbSet makes use of some other infrastructure classes that we've included to ensure that async queries can be processed.
using System;
using System.Collections.Generic;
using System.Collections.ObjectModel;
using System.Data.Entity;
using System.Data.Entity.Infrastructure;
using System.Linq;
using System.Linq.Expressions;
using System.Threading;
using System.Threading.Tasks;
namespace TestingDemo
{
public class TestContext : IBloggingContext
{
public TestContext()
{
this.Blogs = new TestDbSet<Blog>();
this.Posts = new TestDbSet<Post>();
}
public DbSet<Blog> Blogs { get; set; }
public DbSet<Post> Posts { get; set; }
public int SaveChangesCount { get; private set; }
public int SaveChanges()
{
this.SaveChangesCount++;
return 1;
}
}
public class TestDbSet<TEntity> : DbSet<TEntity>, IQueryable, IEnumerable<TEntity>, IDbAsyncEnumerable<TEntity>
where TEntity : class
{
ObservableCollection<TEntity> _data;
IQueryable _query;
public TestDbSet()
{
_data = new ObservableCollection<TEntity>();
_query = _data.AsQueryable();
}
public override TEntity Add(TEntity item)
{
_data.Add(item);
return item;
}
public override TEntity Remove(TEntity item)
{
_data.Remove(item);
return item;
}
public override TEntity Attach(TEntity item)
{
_data.Add(item);
return item;
}
public override TEntity Create()
{
return Activator.CreateInstance<TEntity>();
}
public override TDerivedEntity Create<TDerivedEntity>()
{
return Activator.CreateInstance<TDerivedEntity>();
}
public override ObservableCollection<TEntity> Local
{
get { return _data; }
}
Type IQueryable.ElementType
{
get { return _query.ElementType; }
}
Expression IQueryable.Expression
{
get { return _query.Expression; }
}
IQueryProvider IQueryable.Provider
{
get { return new TestDbAsyncQueryProvider<TEntity>(_query.Provider); }
}
System.Collections.IEnumerator System.Collections.IEnumerable.GetEnumerator()
{
return _data.GetEnumerator();
}
IEnumerator<TEntity> IEnumerable<TEntity>.GetEnumerator()
{
return _data.GetEnumerator();
}
IDbAsyncEnumerator<TEntity> IDbAsyncEnumerable<TEntity>.GetAsyncEnumerator()
{
return new TestDbAsyncEnumerator<TEntity>(_data.GetEnumerator());
}
}
internal class TestDbAsyncQueryProvider<TEntity> : IDbAsyncQueryProvider
{
private readonly IQueryProvider _inner;
internal TestDbAsyncQueryProvider(IQueryProvider inner)
{
_inner = inner;
}
public IQueryable CreateQuery(Expression expression)
{
return new TestDbAsyncEnumerable<TEntity>(expression);
}
public IQueryable<TElement> CreateQuery<TElement>(Expression expression)
{
return new TestDbAsyncEnumerable<TElement>(expression);
}
public object Execute(Expression expression)
{
return _inner.Execute(expression);
}
public TResult Execute<TResult>(Expression expression)
{
return _inner.Execute<TResult>(expression);
}
public Task<object> ExecuteAsync(Expression expression, CancellationToken cancellationToken)
{
return Task.FromResult(Execute(expression));
}
public Task<TResult> ExecuteAsync<TResult>(Expression expression, CancellationToken cancellationToken)
{
return Task.FromResult(Execute<TResult>(expression));
}
}
internal class TestDbAsyncEnumerable<T> : EnumerableQuery<T>, IDbAsyncEnumerable<T>, IQueryable<T>
{
public TestDbAsyncEnumerable(IEnumerable<T> enumerable)
: base(enumerable)
{ }
public TestDbAsyncEnumerable(Expression expression)
: base(expression)
{ }
public IDbAsyncEnumerator<T> GetAsyncEnumerator()
{
return new TestDbAsyncEnumerator<T>(this.AsEnumerable().GetEnumerator());
}
IDbAsyncEnumerator IDbAsyncEnumerable.GetAsyncEnumerator()
{
return GetAsyncEnumerator();
}
IQueryProvider IQueryable.Provider
{
get { return new TestDbAsyncQueryProvider<T>(this); }
}
}
internal class TestDbAsyncEnumerator<T> : IDbAsyncEnumerator<T>
{
private readonly IEnumerator<T> _inner;
public TestDbAsyncEnumerator(IEnumerator<T> inner)
{
_inner = inner;
}
public void Dispose()
{
_inner.Dispose();
}
public Task<bool> MoveNextAsync(CancellationToken cancellationToken)
{
return Task.FromResult(_inner.MoveNext());
}
public T Current
{
get { return _inner.Current; }
}
object IDbAsyncEnumerator.Current
{
get { return Current; }
}
}
}
Implementing Find
The Find method is difficult to implement in a generic fashion. If you need to test code that makes use of the Find method it is easiest to create a test DbSet for each of the entity types that need to support find. You can then write logic to find that particular type of entity, as shown below.
using System.Linq;
namespace TestingDemo
{
class TestBlogDbSet : TestDbSet<Blog>
{
public override Blog Find(params object[] keyValues)
{
var id = (int)keyValues.Single();
return this.SingleOrDefault(b => b.BlogId == id);
}
}
}
Writing some tests
That’s all we need to do to start testing. The following test creates a TestContext and then a service based on this context. The service is then used to create a new blog – using the AddBlog method. Finally, the test verifies that the service added a new Blog to the context's Blogs property and called SaveChanges on the context.
This is just an example of the types of things you can test with an in-memory test double and you can adjust the logic of the test doubles and the verification to meet your requirements.
using Microsoft.VisualStudio.TestTools.UnitTesting;
using System.Linq;
namespace TestingDemo
{
[TestClass]
public class NonQueryTests
{
[TestMethod]
public void CreateBlog_saves_a_blog_via_context()
{
var context = new TestContext();
var service = new BlogService(context);
service.AddBlog("ADO.NET Blog", "http://blogs.msdn.com/adonet");
Assert.AreEqual(1, context.Blogs.Count());
Assert.AreEqual("ADO.NET Blog", context.Blogs.Single().Name);
Assert.AreEqual("http://blogs.msdn.com/adonet", context.Blogs.Single().Url);
Assert.AreEqual(1, context.SaveChangesCount);
}
}
}
Here is another example of a test - this time one that performs a query. The test starts by creating a test context with some data in its Blog property - note that the data is not in alphabetical order. We can then create a BlogService based on our test context and ensure that the data we get back from GetAllBlogs is ordered by name.
using Microsoft.VisualStudio.TestTools.UnitTesting;
namespace TestingDemo
{
[TestClass]
public class QueryTests
{
[TestMethod]
public void GetAllBlogs_orders_by_name()
{
var context = new TestContext();
context.Blogs.Add(new Blog { Name = "BBB" });
context.Blogs.Add(new Blog { Name = "ZZZ" });
context.Blogs.Add(new Blog { Name = "AAA" });
var service = new BlogService(context);
var blogs = service.GetAllBlogs();
Assert.AreEqual(3, blogs.Count);
Assert.AreEqual("AAA", blogs[0].Name);
Assert.AreEqual("BBB", blogs[1].Name);
Assert.AreEqual("ZZZ", blogs[2].Name);
}
}
}
Finally, we'll write one more test that uses our async method to ensure that the async infrastructure we included in TestDbSet is working.
using Microsoft.VisualStudio.TestTools.UnitTesting;
using System.Collections.Generic;
using System.Linq;
using System.Threading.Tasks;
namespace TestingDemo
{
[TestClass]
public class AsyncQueryTests
{
[TestMethod]
public async Task GetAllBlogsAsync_orders_by_name()
{
var context = new TestContext();
context.Blogs.Add(new Blog { Name = "BBB" });
context.Blogs.Add(new Blog { Name = "ZZZ" });
context.Blogs.Add(new Blog { Name = "AAA" });
var service = new BlogService(context);
var blogs = await service.GetAllBlogsAsync();
Assert.AreEqual(3, blogs.Count);
Assert.AreEqual("AAA", blogs[0].Name);
Assert.AreEqual("BBB", blogs[1].Name);
Assert.AreEqual("ZZZ", blogs[2].Name);
}
}
}