I am exposing my repository operations through web api. Repository has been implemented with Entity framework and Unit Of Work Pattern. I have many instances of the same database. Each one represent the data of a different Client. Now the issue is how can I set the connection string dynamically through each webapi call? Should I get connection string parameter with each call ? Or I should host web Api per client ?
Based on the information provided, I would use the same controller and look up the connection string rather than rather than hosting separate Web API instances for each client. There would be more complexity in hosting multiple instances and given the only difference indicated is the connection string, I do not think the complexity would be justified.
The first thing we will need to do is determine which client is calling in order to get the appropriate connection string. This could be done with tokens, headers, request data, or routing. Routing is simplest and most generally accessible to clients, so I will demonstrate using it; however, carefully consider your requirements in deciding how you will make the determination.
[Route( "{clientId}" )]
public Foo Get( string clientId ) { /* ... */ }
Next we need to get the right DbContext for the client. We want to keep using DI but that is complicated in that we do not know until after the Controller is created what connection string is needed to construct the object. Therefore, we need to inject some form of factory rather than the object itself. In this case we will represent this as a Func<string, IUnitOfWork> with the understanding it takes the 'clientId' as a string and returns an appropriately instantiated IUnitOfWork. We could alternatively use a named interface for this.
[RoutePrefix("foo")]
public class FooController : ApiController
{
private Func<string, IUnitOfWork> unitOfWorkFactory;
public FooController( Func<string, IUnitOfWork> unitOfWorkFactory )
{
this.unitOfWorkFactory = unitOfWorkFactory;
}
[Route( "{clientId}" )]
public Foo Get( string clientId )
{
var unitOfWork = unitOfWorkFactory(clientId);
// ...
}
}
All that remains is configuring our dependency injection container to provide us that Func<string, IUnitOfWork>. This could vary significantly between implementation. The following is one possible way to do it in Autofac.
protected override void Load( ContainerBuilder builder )
{
// It is expected `MyDbContext` has a constructor that takes the connection string as a parameter
// This registration may need to be tweaked depending on what other constructors you have.
builder.Register<MyDbContext>().ForType<DbContext>().InstancePerRequest();
// It is expected `UnitOfWork`'s constructor takes a `DbContext` as a parameter
builder.RegisterType<UnitOfWork>().ForType<IUnitOfWork>().InstancePerRequest();
builder.Register<Func<string, Bar>>(
c =>
{
var dbContextFactory = c.Resolve<Func<string, DbContext>>();
var unitOfWorkFactory = c.Resolve<Func<DbContext, IUnitOfWork>>();
return clientId =>
{
// You may have injected another type to help with this
var connectionString = GetConnectionStringForClient(clientId);
return unitOfWorkFactory(dbContextFactory(connectionString));
};
});
}
Autofac is used since comments indicates Autofac is currently being used, though similar results would be possible with other containers.
With that the controller should be able to be instantiated and the appropriate connection string will be used for each request.
Example registration based on linked project:
builder.Register<Func<string, IEmployeeService>>(
c =>
{
var dbContextFactory = c.Resolve<Func<string, IMainContext>>();
var unitOfWorkFactory = c.Resolve<Func<IMainContext, IUnitOfWork>>();
var repositoryFactory = c.Resolve<Func<IMainContext, IEmployeeRepository>>();
var serviceFactory = c.Resolve<Func<IUnitOfWork, IEmployeeService>>();
return clientId =>
{
// You may have injected another type to help with this
var connectionString = GetConnectionStringForClient(clientId);
IMainContext dbContext = dbContextFactory(connectionString);
IUnitOfWork unitOfWork = unitOfWorkFactory(dbContext);
IEmployeeRepository employeeRepository = repositoryFactory(dbContext);
unitOfWork.employeeRepositoty = employeeRepository;
return serviceFactory(unitOfWork);
};
});
If you find the registration grows too cumbersome because of needing to do a little wiring manually, you probably need to look at updating (or creating a new) container after you have determined the client so that you can rely more on the container.
You can change the connectionstring per DbContext instance
Example:
public class AwesomeContext : DbContext
{
public AwesomeContext (string connectionString)
: base(connectionString)
{
}
public DbSet<AwesomePeople> AwesomePeoples { get; set; }
}
And then use your DbContext like this:
using(AwesomeContext context = new AwesomeContext("newConnectionString"))
{
return context.AwesomePeoples.ToList();
}
Depending on how many ConnectionStrings there are you can make a DB table for the client / constring mapping or save it in the solution (array for example).
If you can't/don't want to change the constructor you can do it later as well
Add this to your DbContext override:
public void SetConnectionString(string connectionString)
{
this.Database.Connection.ConnectionString = connectionString;
}
And call the method before you do any DB operations:
using(AwesomeContext context = new AwesomeContext())
{
context.SetConnectionString(ConfigurationManager.ConnectionStrings["newConnectionString"].ConnectionString)
return context.AwesomePeoples.ToList();
}
Related
Situation
We have a controller where users can submit any number of E-Mail addresses to invite other (potential) members as friends. If an address is not found in the database, we send an E-Mail message to that user. Since the user does not has to wait for this process to complete in order to continue working this is done asynchronously.
Sending E-Mails can take a long time if servers respond slowly, are down or overloaded. The E-Mail sender should update the database according to the status received from the E-Mail server, so for example setting the friend request into "Error" state, when a permanent failure occurs, for example if the address does not exists. For this purpose, the E-Mail component implements the function SendImmediateAsync(From,To,Subject,Content,Callback,UserArg). After the message has been delivered (or it failed), the Callback is called with certain arguments about the Delivery state.
When it eventually calls the delegate, the DbContext object has already been disposed (since the controller has been too) and I cannot manually create a new one using new ApplicationDbContext() because there is no constructor that accepts a connection string.
Question
How can I write to the database long after the controller has been disposed? I have not yet figured out how to manually create a DbContext object for myself. An object of type ApplicationDbContext is passed to the constructor of the Controller and I hoped I could instantiate one for myself, but the Constructor has no arguments I can supply (for example connection string). I want to avoid to manually create an SQL Connection and assemble INSERT statements manually and would prefer to work with the entity model we have already set up.
Code
The code shows the affected segment only without any error checking for readability.
[Authorize]
public class MembersController : Controller
{
private ApplicationDbContext _context;
public MembersController(ApplicationDbContext context)
{
_context = context;
}
[HttpPost]
[ValidateAntiForgeryToken]
public IActionResult Friends()
{
MailHandler.SendImmediateAsync(FROM,TO,SUBJECT,CONTENT,
delegate (Guid G, object any)
{
//THIS IS NOT WORKING BECAUSE _context IS DISPOSED
var ctx = _context;
Guid Result = (Guid)any; //user supplied argument
if (G != Guid.Empty)
{
ctx.MailConfirmation.Add(new MailConfirmation()
{
EntryId = Result,
For = EntryFor.FriendRequest,
Id = G
});
if (G == MailHandler.ErrorGuid)
{
var frq = _context.FriendRequest.SingleOrDefault(m => m.Id == Result);
frq.Status = FriendStatus.Error;
ctx.Update(frq);
}
ctx.SaveChanges();
}
}, req.Id);
//rendering view
}
}
First of all, when you are using EF Core with ASP.NET Core's dependency injection, each DbContext instance is scoped per-request, unless you have specified otherwise in ".AddDbContext". This means you should not attempt to re-use an instance of DbContext after that HTTP request has completed. See https://docs.asp.net/en/latest/fundamentals/dependency-injection.html#service-lifetimes-and-registration-options
DbContextOptions, on the other hand, are singletons and can be re-used across requests.
If you need to close the HTTP request and perform an action afterwards, you'll need to create a new DbContext scope an manage it's lifetime.
Second of all, you can overload DbContext's base constructor and pass in DbContextOptions directly. See https://docs.efproject.net/en/latest/miscellaneous/configuring-dbcontext.html
Together, this is what a solution might look like.
public class MembersController : Controller
{
private DbContextOptions<ApplicationDbContext> _options;
public MembersController(DbContextOptions<ApplicationDbContext> options)
{
_options = options;
}
[HttpPost]
[ValidateAntiForgeryToken]
public IActionResult Friends()
{
MailHandler.SendImmediateAsync(FROM,TO,SUBJECT,CONTENT, CreateDelegate(_options) req.Id);
}
private static Action<Guid, object> CreateDelegate(DbContextOptions<ApplicationDbContext> options)
{
return (G, any) =>
{
using (var context = new ApplicationDbContext(options))
{
//do work
context.SaveChanges();
}
};
}
}
public class ApplicationDbContext : DbContext
{
public ApplicationDbContext(DbContextOptions<ApplicationDbContext> options) : base (options) { }
// the rest of your stuff
}
This assumes, of course, that your "MailHandler" class is properly using concurrency to run the delegate so it doesn't block the thread processing the HTTP request.
Why not just pass the dbContext as the userArgs to your SendImmediateAsync? Then the dbContext will not get disposed and can be passed back when you do the callback. I'm pretty sure that should work.
I have an n-tier application, whereas the core web service is built with Web API. many of the web service's methods are set as HTTPGET and accept a DTO object as parameter. my client app, built with MVC 5 is using HttpClient to call this API.
so it seems that by using client.PostAsJsonAsync() I can pass an object, whereas client.GetAsync() doesn't allow me to do that. this forces me to explicitly specify the properties of DTO in the URL, which works, but seem a bit redundant.
Can somebody explain why this is not possible through a GET call and suggest a better practice?
Why does passing data in the URI seem redundant? The HTTP spec says that GET methods are not to use content sent in the body. This is primarily to facilitate caches being able to cache responses based only on the URI, method and headers. Requiring caches to parse the body of a message to identify a resource would be very inefficient.
Here is an basic extension method that will do the grunt work for you,
public static class UriExtensions
{
public static Uri AddToQuery<T>(this Uri requestUri,T dto)
{
Type t = typeof (T);
var properties = t.GetProperties();
var dictionary = properties.ToDictionary(info => info.Name,
info => info.GetValue(dto, null).ToString());
var formContent = new FormUrlEncodedContent(dictionary);
var uriBuilder = new UriBuilder(requestUri) {Query = formContent.ReadAsStringAsync().Result};
return uriBuilder.Uri;
}
}
and assuming you have a DTO like this,
public class Foo
{
public string Bar { get; set; }
public int Baz { get; set; }
}
you can use it like this.
[Fact]
public void Foo()
{
var foo = new Foo()
{
Bar = "hello world",
Baz = 10
};
var uri = new Uri("http://example.org/blah");
var uri2 = uri.AddToQuery(foo);
Assert.Equal("http://example.org/blah?Bar=hello+world&Baz=10", uri2.AbsoluteUri);
}
I use DictionaryAdapter to retrieve settings from appSettings section of my asp.net website.
The IoC configuration is done once, at the startup time and all kinds of different interfaces with getters are being registered with using single Configuration.AppSettings object:
var dictionaryAdapterFactory = new DictionaryAdapterFactory();
container.Register(
Types
.FromAssemblyNamed(assemblyName)
.Where(t => t.Name.EndsWith("AppSettings"))
.Configure(
component => component.UsingFactoryMethod(
(kernel, model, creationContext) =>
dictionaryAdapterFactory.GetAdapter(creationContext.RequestedType, ConfigurationManager.AppSettings))));
The appSettings section hosted in Web.config file works fine, but it has its drawback when I want to update some settings during runtime. As it is web.config file, the whole app is restarted. I would like to be able to modify configuration at runtime without restarting website as a side effect. Therefore, I moved into separate file:
<appSettings configSource="AppSettings.config">
Now, changes are being reflected when retrieving them via ConfigurationManager.AppSettings["key"], but they are not reflected when accessing via
dynamic interfaces from DictionaryAdapter.
Is there any way to have tell DA to watch for the changes in source and not cache the values?
Although I didn't find the exact answer, I found a workaround. Instead of 'binding' DA directly to ConfigurationManager, i bind to a simple proxy that wraps CM:
public class AppSettingsProxy : NameValueCollection
{
public override string Get(string name)
{
return ConfigurationManager.AppSettings[name];
}
public override string GetKey(int index)
{
return ConfigurationManager.AppSettings[index];
}
}
Then jus tchange binding to my proxy instance:
container.Register(
Types
.FromAssemblyNamed(assemblyName)
.Where(t => t.Name.EndsWith("AppSettings"))
.Configure(
component => component.UsingFactoryMethod(
(kernel, model, creationContext) =>
dictionaryAdapterFactory.GetAdapter(creationContext.RequestedType, appSettingsProxy))));
The above works for me. While I can modify my website's settings at runtime without an restart, value changes now are reflected via dynamically generated proxes over my settings interfaces.
DictionaryAdapter does not itself by default cache the values. Here's a passing test to prove that.
public interface IFoo
{
string Foo { get; set; }
}
[Test]
public void Adapter_does_not_cache_values_once_read()
{
var dict = new NameValueCollection { { "Foo", "Bar" } };
var adapter = (IFoo)factory.GetAdapter(typeof(IFoo), dict);
var value = adapter.Foo;
dict["Foo"] = "Baz";
var value2 = adapter.Foo;
Assert.AreNotEqual(value, value2);
Assert.AreEqual("Baz", value2);
}
Are you sure you're not caching the value yourself in your code? Can you reproduce the behaviour in a test?
I have a Web Service for my ASP.NET MVC 3 app which creates a new Sales Order in AX. In AX I have created an OnlineOrder class which has 1 method for now. It is to generate a Sales Reference. Below is the code in my web service:
public AxaptaObject order;
public void CreateOrder()
{
AxaptaStart();
order = Axapta.CreateAxaptaObject("OnlineOrder");
AxaptaStop();
}
public string GetSalesRef(string username, string delivery, string reference)
{
AxaptaStart();
string number = order.Call("orderCreate", username, delivery, reference).ToString();
AxaptaStop();
return number;
}
Then in my Controller I call these methods:
client.CreateOrder();
string number = client.GetSalesRef(user.Username, order.deliverymethod, order.custorder).ToString();
This doesn't work and there is no exception info to show, its just a blank message. I believe the order AxaptaObject is not of type OnlineOrder so it can't call those methods. How would I instantiate the object to use the orderCreate method?
EDIT:
If I do:
public string CreateOrder(string username, string delivery, string reference)
{
AxaptaStart();
order = Axapta.CreateAxaptaObject("OnlineOrder");
string number = order.Call("orderCreate", username, delivery, reference).ToString();
AxaptaStop();
return number;
}
This works, but this isn't a valid solution as I'd like to add more methods to my OnlineOrder object in the future and I dont want to call them all in 1 method on my web service
This will never work as the order object is sort of closed when you call the AxaptaStop method.
I will suggest creating a class implementing IDisposable, then call AxaptaStop in the Dispose method. The AxaptaStart call could go to the constructor. This will allow you to scope the AX context like:
using (var axWrap = new AxWrap())
{
var order = axWrap.onlineOrder();
var number = axWrap.orderCreate(order, username, delivery, reference);
}
The Dispose is automatically called by the using statement.
I'm using EF 4.1 to build a domain model. I have a Task class with a Validate(string userCode) method and in it I want to ensure the user code maps to a valid user in the database, so:
public static bool Validate(string userCode)
{
IDbSet<User> users = db.Set<User>();
var results = from u in users
where u.UserCode.Equals(userCode)
select u;
return results.FirstOrDefault() != null;
}
I can use Moq to mock IDbSet no problem. But ran into trouble with the Where call:
User user = new User { UserCode = "abc" };
IList<User> list = new List<User> { user };
var users = new Mock<IDbSet<User>>();
users.Setup(x => x.Where(It.IsAny<Expression<Func<User, bool>>>())).Returns(list.AsQueryable);
Initialization method JLTi.iRIS3.Tests.TaskTest.SetUp threw exception.
System.NotSupportedException: System.NotSupportedException: Expression
references a method that does not belong to the mocked object:
x => x.Where<User>(It.IsAny<Expression`1>()).
Other than creating a level of indirection (eg, using a ServiceLocator to get an object that runs the LINQ and then mock that method) I can't think of how else to test this, but I want to make sure there is no way before I introduce another layer. And I can see this kind of LINQ queries will be needed quite often so the service objects can quickly spiral out of control.
Could some kind soul help? Thanks!
There is an article on MSDN highlighting how to mock using moq:
The gist of it is to represent linq to entities operations with linq to objects.
var mockSet = new Mock<DbSet<Blog>>();
mockSet.As<IQueryable<Blog>>().Setup(m => m.Provider).Returns(data.Provider);
mockSet.As<IQueryable<Blog>>().Setup(m => m.Expression).Returns(data.Expression);
mockSet.As<IQueryable<Blog>>().Setup(m => m.ElementType).Returns(data.ElementType);
mockSet.As<IQueryable<Blog>>().Setup(m => m.GetEnumerator()).Returns(data.GetEnumerator());
As Ladislav points out there are disadvantages to this as Linq To Objects is simply different to Linq to Entities so it may result in false positives. But it now being an MSDN article it does point that it is at least possible and perhaps recommended in some cases?
One thing that may of changed since the original answers to this post is that the Entity Framework team have opened up areas of Entity Framework in EF 6.0 to make it easier to mock it's inners.
Although I have not tried this, because IDBSet implements IEnumerable you might have to mock the enumerator method so the linq statements will pick up your list of users. You don't actually want to mock linq but by the looks of your code you want to test whether you are finding the right user based on the UserCode which I think is a valid unit test.
var user = new User { UserCode = "abc" };
var list = new List<User> { user };
var users = new Mock<IDbSet<User>>();
users.Setup(x => x.GetEnumerator()).Returns(list.GetEnumerator());
You might get a conflict with the non-generic version of the GetEnumerator but it this might help you on the right track.
Then you have to then place the mocked object on the data context which depends on other code that we don't see.
As I know Moq is able to set up only virtual methods of mocked object itself but you are trying to set up extensions (static) method - no way! These methods are absolutely outside of your mock scope.
Moreover that code is hard to test and requires too much initialization to be able to test it. Use this instead:
internal virtual IQueryable<User> GetUserSet()
{
return db.Set<User>();
}
public bool Validate(string userCode)
{
IQueryable<User> users = GetUserSet();
var results = from u in users
where u.UserCode.Equals(userCode)
select u;
return results.FirstOrDefault() != null;
}
You will just need to set up GetUserSet to return your list. Such testing has some major issues:
You are not testing the real implementation - in case of EF mocking sets is stupid approach because once you do it you change linq-to-entities to linq-to-objects. Those two are totally different and linq-to-entities is only small subset of linq-to-objects = your unit tests can pass with linq-to-objects but your code will fail at runtime.
Once you use this approach you cannot use Include because include is dependent on DbQuery / DbSet. Again you need integration test to use it.
This doesn't test that your lazy loading works
The better approach is removing your linq queries from Validate method - just call them as another virtual method of the object. Unit test your Validate method with mocked query methods and use integration tests to test queries themselves.
I found it easier just to write the stub:
internal class FakeDbSet<T> : IDbSet<T>where T : class
{
readonly HashSet<T> _data;
readonly IQueryable _query;
public FakeDbSet()
{
_data = new HashSet<T>();
_query = _data.AsQueryable();
}
public virtual T Find(params object[] keyValues)
{
throw new NotImplementedException("Derive from FakeDbSet<T> and override Find");
}
public T Add(T item)
{
_data.Add(item);
return item;
}
public T Remove(T item)
{
_data.Remove(item);
return item;
}
public T Attach(T item)
{
_data.Add(item);
return item;
}
public void Detach(T item)
{
_data.Remove(item);
}
Type IQueryable.ElementType
{
get { return _query.ElementType; }
}
Expression IQueryable.Expression
{
get { return _query.Expression; }
}
IQueryProvider IQueryable.Provider
{
get { return _query.Provider; }
}
IEnumerator IEnumerable.GetEnumerator()
{
return _data.GetEnumerator();
}
IEnumerator<T> IEnumerable<T>.GetEnumerator()
{
return _data.GetEnumerator();
}
public TDerivedEntity Create<TDerivedEntity>() where TDerivedEntity : class, T
{
return Activator.CreateInstance<TDerivedEntity>();
}
public T Create()
{
return Activator.CreateInstance<T>();
}
public ObservableCollection<T> Local
{
get
{
return new ObservableCollection<T>(_data);
}
}