i have an interface that defines some methods and i have N classes that implement it. How can i register all the classes found in all the loaded assemblies with autofac?
You will have to "know about" the assemblies containing the classes, you could perhaps load them yourself with Assembly.Load(..).
From there, it is easy to register the classes:
var assemblies = new[]{....};
builder.RegisterAssemblyTypes(assemblies)
.Where(t => typeof(IMyInterface).IsAssignableFrom(t))
.As<IMyInterface>();
Update: to get to the registered instances, you can use Autofac builtin support for collections:
public class MyService
{
private readonly IEnumerable<IMyInterface> _services;
public MyService(IEnumerable<IMyInterface> services)
{
_services = services;
}
public void DoStuffWithServices()
{
foreach(var svc in _services)
{
...
}
}
}
Related
I was using CastleWindsor in my ASP.NETCore2.2 WebAPI project and was working fine. I'm migrating to ASP.NETCore3.1 now and it doesn't look like CastleWindor has offical support for that so I decided to move to Autofac with minimal changes but having some issues resolving the dependencies.
In my project, I've maintained very loose coupling between different layers in the application namely, business layer, data layer, and translation layer. All of those layers are in their own assemblies. And then in my main project, I've a folder say "dependencies" which will hold all the DLLs of differnet layers. Additionally, I've a separate project that lists all the interfaces that are implemented by the different layers and which needs to be resolved by the IoC container.
The project having all the interfaces looks like this:
namespace Shared.Interfaces
{
public interface IBusinessLayer<T>
{
....
}
public interface IDataLayer<T>
{
....
}
public interface ITranslationLayer<T>
{
....
}
}
The implementing projects looks like this:
namespace POC.Person.BusinessLayer
{
public class BusinessLayer<T> : IBusinessLayer<T> where T : Models.Person
{
...
}
}
namespace POC.Person.DataLayer
{
public class DataLayer<T> : IDataLayer<T> where T : Models.Person
{
...
}
}
namespace POC.Person.TranslationLayer
{
public class TranslationLayer<T> : ITranslationLayer<T> where T : Models.Person
{
...
}
}
Using Autofac in my migrated .netcore3.1 project, Startup.cs looks like this:
public void ConfigureServices(IServiceCollection services)
{
services.AddControllers();
//and other codes
}
public void ConfigureContainer(ContainerBuilder builder)
{
builder.RegisterModule(new DependencyResolver());
}
DependencyResolver is a class that inherits from Autofac.Module, which is again in a separate assembly in different project which looks like this:
namespace IOC.Autofac
{
public class DependencyResolver: Autofac.Module
{
protected override void Load(ContainerBuilder builder)
{
base.Load(builder);
// get our path to dependencies folder in the main project
var path = Directory.GetCurrentDirectory() + "\\dependencies\\";
//get all the assemblies inside that folder
List<Assembly> assemblies = new List<Assembly>();
foreach (string assemblyPath in Directory.GetFiles(path, "*.dll", SearchOption.AllDirectories))
{
var assembly = System.Runtime.Loader.AssemblyLoadContext.Default.LoadFromAssemblyPath(assemblyPath);
assemblies.Add(assembly);
}
// Register and resolve the types with the container
builder
.RegisterAssemblyTypes(assemblies.ToArray())
.AsClosedTypesOf(typeof(IBusinessLayer<>))
.AsClosedTypesOf(typeof(IDataLayer<>))
.AsClosedTypesOf(typeof(ITranslationLayer<>))
.AsImplementedInterfaces()
.InstancePerRequest();
}
}
}
I'm getting this error and I've not been able to fix it:
":"Unable to resolve service for type 'Shared.Interfaces.IBusinessLayer`1[Models.Person]' while attempting to activate 'POC.Person.Controllers.PersonController'.","
Inside my controller I've injection which looks like this:
namespace POC.Person.Controllers
{
public class PersonController : ControllerBase
{
private readonly IBusinessLayer<Models.Person> _bl;
public PersonController(IBusinessLayer<Models.Person> bl)
{
_bl = bl;
}
//other codes
}
}
Program.cs looks like this:
namespace POC.Person
{
public class Program
{
public static void Main(string[] args)
{
var host = BuildWebHost(args);
host.Build().Run();
}
public static IHostBuilder BuildWebHost(string[] args)
{
return Host.CreateDefaultBuilder(args)
.UseServiceProviderFactory(new AutofacServiceProviderFactory())
.ConfigureWebHostDefaults(webBuilder =>
{
webBuilder.UseKestrel()
.UseStartup<Startup>()
.UseIIS()
.UseIISIntegration();
;
}).ConfigureAppConfiguration((context, config) =>
{
var builtConfig = config.Build();
});
}
}
}
It looks like with autofac involving generics, registering and resolving the type is not that straight forward?
Autofac does not currently support registering open generics whilst assembly scanning. It's a long-running known issue. You can do assembly scanning, you can register open generics, you can't do both at the same time. There are some ideas in that linked issue on ways some folks have solved it.
Out of the box, the scanning logic would, thus, be reduced to:
builder
.RegisterAssemblyTypes(assemblies.ToArray())
.AsImplementedInterfaces()
.InstancePerRequest();
You need to register generics separately, like:
builder
.RegisterGeneric(typeof(TranslationLayer<>))
.As(typeof(ITranslationLayer<>));
I'm thinking of using MEF to solve a plugin management requirement. In the blurb it says "no hard dependencies" but as far as I can see, there is a hard dependency on the import/export interface.
My concern is this. My extendable app is written by me. Plugins are written by third parties. So lets say we all start off with V1. My app defines a IPlugin interface that the plugin 'parts' need to implement. We deploy the app and users install a bunch of third party plugins. All well and good.
Now I upgrade my app and I want to add a new method to the plugin interface. The way I see it I have 2 choices:
Edit the interface - probably bad, and this would break existing plugins because they would no longer correctly implement the interface.
Create a new 'V2' interface, that inherits from the original
public interface IPluginV2 : IPlugin {}
Now I have a problem. My users all have a bunch of 3rd party plugins implementing IPlugin, but I now require them to implement IPluginV2. I presume these 3rd party plugins will no longer work, until the developers implement the new interface.
Does MEF have a way to handle this situation? I'm really looking for a way that lets me evolve my app while having old plugins continue to work without having to be rebuilt. Whats the best way of handling that?
For versioning, you will probably want an interface for each version and the adapter pattern to go between them. It is how System.AddIn handles versioning, and it works for MEF, too.
Let's say we have the following types for the V1 of your application:
public interface IPlugin
{
string Name { get; }
string Publisher { get; }
string Version { get; }
void Init();
}
This is the only contract for our V1 plugin-aware app. It is contained in assembly Contracts.v1.
Then we have a V1 plugin:
[Export(typeof(IPlugin))]
public class SomePlugin : IPlugin
{
public string Name { get { return "Some Plugin"; } }
public string Publisher { get { return "Publisher A"; } }
public string Version { get { return "1.0.0.0"; } }
public void Init() { }
public override string ToString()
{
return string.Format("{0} v.{1} from {2}", Name, Version, Publisher);
}
}
Which is exported as IPlugin. It is contained in assembly Plugin.v1 and is published on the "plugins" folder under the application base path of the host.
Finally the V1 host:
class Host : IDisposable
{
CompositionContainer _container;
[ImportMany(typeof(IPlugin))]
public IEnumerable<IPlugin> Plugins { get; private set; }
public Host()
{
var catalog = new DirectoryCatalog("plugins");
_container = new CompositionContainer(catalog);
_container.ComposeParts(this);
}
public void Dispose() { _container.Dispose(); }
}
which imports all IPlugin parts found in folder "plugins".
Then we decide to publish V2 and because we want to provide versioning we will need versionless contracts:
public interface IPluginV2
{
string Name { get; }
string Publisher { get; }
string Version { get; }
string Description { get; }
void Init(IHost host);
}
with a new property and a modified method signature. Plus we add an interface for the host:
public interface IHost
{
//Here we can add something useful for a plugin.
}
Both of these are contained in assembly Contracts.v2.
To allow versioning we add a plugin adapter from V1 to V2:
class V1toV2PluginAdapter : IPluginV2
{
IPlugin _plugin;
public string Name { get { return _plugin.Name; } }
public string Publisher { get { return _plugin.Publisher; } }
public string Version { get { return _plugin.Version; } }
public string Description { get { return "No description"; } }
public V1toV2PluginAdapter(IPlugin plugin)
{
if (plugin == null) throw new ArgumentNullException("plugin");
_plugin = plugin;
}
public void Init(IHost host) { plugin.Init(); }
public override string ToString() { return _plugin.ToString(); }
}
This simply adapts from IPlugin to IPluginV2. It returns a fixed description and in the Init it does nothing with the host argument but it calls the parameterless Init from the V1 contract.
And finally the V2 host:
class HostV2WithVersioning : IHost, IDisposable
{
CompositionContainer _container;
[ImportMany(typeof(IPluginV2))]
IEnumerable<IPluginV2> _pluginsV2;
[ImportMany(typeof(IPlugin))]
IEnumerable<IPlugin> _pluginsV1;
public IEnumerable<IPluginV2> Plugins
{
get
{
return _pluginsV1.Select(p1 => new V1toV2PluginAdapter(p1)).Concat(_pluginsV2);
}
}
public HostV2WithVersioning()
{
var catalog = new DirectoryCatalog("plugins");
_container = new CompositionContainer(catalog);
_container.ComposeParts(this);
}
public void Dispose() { _container.Dispose(); }
}
which imports both IPlugin and IPluginV2 parts, adapts each IPlugin into IPluginV2 and exposes a concatenated sequence of all discovered plugins. After the adaptation is completed, all plugins can be treated as V2 plugins.
You can also use the adapter pattern on the interface of the host to allow V2 plugins to work with V1 hosts.
Another approach would be the autofac IoC that can integrate with MEF and can support versioning using adapters.
Just a couple of suggestions (which I have not tested) that may help brainstorming a solution for you:
If using MEF, use different AggregateCatalog for each of the versions. That way you could maintain both V1 and V2 plugins available
If not using MEF, a dynamic loaded DLL from the third party should return the current interface version it has implemented and you can choose which calls you could make depending on the version number
Did that help?
Cheers, dimamura
I'm building a small Nancy web project.
In a method of one of my classes (not a nancy module), I would like to basically do:
var myThing = TinyIoC.TinyIoCContainer.Current.Resolve<IMyThing>();
However, there is only one registration in .Current (non public members, _RegisteredTypes) which is:
TinyIoC.TinyIoCContainer.TypeRegistration
Naturally, in my above code, I'm getting:
Unable to resolve type: My.Namespace.IMyThing
So, I guess I'm not getting the same container registered in my bootstrapper?
Is there a way to get at it?
EDIT
To flesh out a bit more of what I'm trying to do:
Basically, my url structure looks something like:
/{myType}/{myMethod}
So, the idea being, going to: /customer/ShowAllWithTheNameAlex would load the Customer service, and execute the showAllWithTheNameAlex method
How I do this is:
public interface IService
{
void DoSomething();
IEnumerable<string> GetSomeThings();
}
I then have an abstract base class, with a method GetService that returns the service.
It's here that i'm trying to use the TinyIoC.TinyIoCContainer.Current.Resolve();
In this case, it would be TinyIoC.TinyIoCContainer.Current.Resolve("typeName");
public abstract class Service : IService
{
abstract void DoSomething();
abstract IEnumerable<string> GetSomeThings();
public static IService GetService(string type)
{
//currently, i'm doing this with reflection....
}
}
Here's my implementation of the service.
public class CustomerService : Service
{
public void DoSomething()
{
//do stuff
}
public IEnumerable<string> GetSomeThings()
{
//return stuff
}
public IEnumerable<Customer> ShowAllWithTheNameAlex()
{
//return
}
}
Finally, I have my Nancy Module, that looks like:
public class MyModule : NancyModule
{
public MyModule()
{
Get["/{typeName}/{methodName}"] = p => ExecuteMethod(p.typeName, p.methodName);
}
private dynamic ExecuteMethod(string typeName, string methodName)
{
var service = Service.GetService(typeName);
var result = service.GetType().GetMethod(methodName).Invoke(service, null);
//do stuff
return result; //or whatever
}
}
#alexjamesbrown - The short answer is, you don't. Nancy was specifically designed so that you did not deal with the container directly. You mention that the class, that you want to take a dependency on IMyThing, is not a NancyModule. Well this is not an issue, as long as one of your modules has a reference to it, then those dependencies can also have their own dependencies that will be satisfied at runtime.
public interface IGreetingMessageService
{
string GetMessage();
}
public class GreetingMessageService: IGreetingMessageService
{
public string GetMessage()
{
return "Hi!";
}
}
public interface IGreeter
{
string Greet();
}
public class Greeter
{
private readonly IGreetingMessageService service;
public Greeter(IGreetingMessageService service)
{
this.service = service;
}
public string Greet()
{
return this.service.GetMessage();
}
}
public class GreetingsModule : NancyModule
{
public GreetingModule(IGreeter greeter)
{
Get["/"] = x => greeter.Greet();
}
}
The above will work just fine and Greeter will have it's dependency on IGreetingMessageService satisfied at runtime
I have had a very similar issue, needing to "share" the container. The reason this is an issue is that my program runs as a service using Nancy self hosting to provide a REST API. My modules have dependencies which are injected by Nancy itself, but the other parts of the app which are not referenced from modules also need dependencies injected.
Multiple containers are not a sensible option here (or anywhere really), I need to share the container between Nancy and the rest of the app.
I simply did the following
(I'm using Autofac but I suspect that TinyIoC in similar)
public class Bootstrapper : AutofacNancyBootstrapper
{
private static readonly Lazy<ILifetimeScope> container = new Lazy<ILifetimeScope>(RegisterTypes);
public static ILifetimeScope Container => container.Value;
protected override ILifetimeScope GetApplicationContainer()
{
return container.Value;
}
// Create container and register my types
private static ILifetimeScope RegisterTypes()
{
var builder = new ContainerBuilder();
// Register all my own types.....
return builder.Build();
}
}
Then, in my main code, I can use the container myself
public class Program
{
public static void Main(string[] args)
{
// Resolve main service with all its dependencies
var service = Bootstrapper.Container.Resolve<Service>();
service.Run();
}
}
As my NancyHost is within the Service, the container is constructed (once) upon its first use in main, this static is then used when Nancy gets round to creating the Bootstrapper itself.
In an ideal world, I wouldn't really want a globally accessible container, normally it would be local to the main function.
In this particular case "not dealing with the container directly" is highly problematic:
public interface IFoo {}
public class Foo : IFoo { public Foo(string bar) {} }
Assume IFoo already is a constructor dependency of a Nancy module.
Note the Foo constructor's string dependency. I need to communicate to the container to use that constructor for an IFoo singleton, when encountered as a Nancy module dependency. I need to register that on the TinyIoC instance NancyFx uses, and pass in the actual value of bar.
I am trying to configure an application such that types from assemblyA can be used by my console to allow for logging in an AOP style. The JournalInterceptor will just write out method calls, input and maybe output arguments to a log file or datastore of some kind.
I can register one type at a time but I would like to register all types in one go. Once I get going I may add some filtering to the registered types but I am missing something.
I am trying to use Classes.FromAssemblyContaining but am not sure how to get at an IRegistration instance for the call to WindsorContainer::Register
Any clues?
// otherAssembly.cs
namespace assemblyA
{
public class Foo1 { public virtual void What(){} }
public class Foo2 { public virtual void Where(){} }
}
// program.cs
namespace console
{
using assemblyA;
public class JournalInterceptor : IInterceptor {}
public class Program
{
public static void Main()
{
var container = new Castle.Windsor.WindsorContainer()
.Register(
Component.For<JournalInterceptor>().LifeStyle.Transient,
// works but can't be the best way
Component.For<Foo1>().LifeStyle.Transient
.Interceptors<JournalInterceptor>(),
Component.For<Foo2>().LifeStyle.Transient,
.Interceptors<JournalInterceptor>(),
// how do I do it this way
Classes.FromAssemblyContaining<Foo1>()
.Pick()
.LifestyleTransient()
.Interceptors<JournalInterceptor>()
);
Foo1 foo = container.Resolve<Foo1>();
}
}
}
Implement a Pointcut. In Castle Windsor this is done by implementing the IModelInterceptorsSelector interface.
It would go something like this:
public class JournalPointcut : IModelInterceptorsSelector
{
public bool HasInterceptors(ComponentModel model)
{
return true; // intercept everything - probably not a good idea, though
}
public InterceptorReference[] SelectInterceptors(
ComponentModel model, InterceptorReference[] interceptors)
{
return new[]
{
InterceptorReference.ForType<JournalInterceptor>()
}.Concat(interceptors).ToArray();
}
}
Then register the Interceptor and the Pointcut with the container:
this.container.Register(Component.For<JounalInterceptor>());
this.container.Kernel.ProxyFactory.AddInterceptorSelector(new JournalPointcut());
For in-depth explanation, you may want to see this recording.
I'm using StructureMap to resolve references to my repository class. My repository interface implements IDisposable, e.g.
public interface IMyRepository : IDisposable
{
SomeClass GetById(int id);
}
An implementation of the interface using Entity Framework:
public MyRepository : IMyRepository
{
private MyDbContext _dbContext;
public MyDbContext()
{
_dbContext = new MyDbContext();
}
public SomeClass GetById(int id)
{
var query = from x in _dbContext
where x.Id = id
select x;
return x.FirstOrDefault();
}
public void Dispose()
{
_dbContext.Dispose();
}
}
Anyway as mentioned I'm using StructureMap to resolve IMyRepository. So when, where and how should I call my dispose method?
WARNING: please note that my views have changed, and you should consider the following advise outdated. Please see this answer for an updated view: https://stackoverflow.com/a/30287923/264697
While DI frameworks can manage lifetime of objects for you and some could even dispose objects for you after you're done using with them, it makes object disposal just too implicit. The IDisposable interface is created because there was the need of deterministic clean-up of resources. Therefore, in the context of DI, I personally like to make this clean-up very explicit. When you make it explicit, you've got basically two options: 1. Configure the DI to return transient objects and dispose these objects yourself. 2. Configure a factory and instruct the factory to create new instances.
I favor the second approach over the first, because especially when doing Dependency Injection, your code isn't as clean as it could be. Look for instance at this code:
public sealed class Client : IDisposable
{
private readonly IDependency dependency;
public Client(IDependency dependency)
{
this. dependency = dependency;
}
public void Do()
{
this.dependency.DoSomething();
}
public Dispose()
{
this.dependency.Dispose();
}
}
While this code explicitly disposes the dependency, it could raise some eyebrows to readers, because resources should normally only be disposed by the owner of the resource. Apparently, the Client became the owner of the resource, when it was injected.
Because of this, I favor the use of a factory. Look for instance at this example:
public sealed class Client
{
private readonly IDependencyFactory factory;
public Client(IDependencyFactory factory)
{
this.factory = factory;
}
public void Do()
{
using (var dependency = this.factory.CreateNew())
{
dependency.DoSomething();
}
}
}
This example has the exact same behavior as the previous example, but see how the Client class doesn't have to implement IDisposable anymore, because it creates and disposes the resource within the Do method.
Injecting a factory is the most explicit way (the path of least surprise) to do this. That's why I prefer this style. Downside of this is that you often need to define more classes (for your factories), but I personally don't mind.
RPM1984 asked for a more concrete example.
I would not have the repository implement IDisposable, but have a Unit of Work that implements IDisposable, controls/contains repositories and have a factory that knows how to create new unit of works. With that in mind, the above code would look like this:
public sealed class Client
{
private readonly INorthwindUnitOfWorkFactory factory;
public Client(INorthwindUnitOfWorkFactory factory)
{
this.factory = factory;
}
public void Do()
{
using (NorthwindUnitOfWork db =
this.factory.CreateNew())
{
// 'Customers' is a repository.
var customer = db.Customers.GetById(1);
customer.Name = ".NET Junkie";
db.SubmitChanges();
}
}
}
In the design I use, and have described here, I use a concrete NorthwindUnitOfWork class that wraps an IDataMapper that is the gateway to the underlying LINQ provider (such as LINQ to SQL or Entity Framework). In sumary, the design is as follows:
An INorthwindUnitOfWorkFactory is injected in a client.
The particular implementation of that factory creates a concrete NorthwindUnitOfWork class and injects a O/RM specific IDataMapper class into it.
The NorthwindUnitOfWork is in fact a type-safe wrapper around the IDataMapper and the NorthwindUnitOfWork requests the IDataMapper for repositories and forwards requests to submit changes and dispose to the mapper.
The IDataMapper returns Repository<T> classes and a repository implements IQueryable<T> to allow the client to use LINQ over the repository.
The specific implementation of the IDataMapper holds a reference to the O/RM specific unit of work (for instance EF's ObjectContext). For that reason the IDataMapper must implement IDisposable.
This results in the following design:
public interface INorthwindUnitOfWorkFactory
{
NorthwindUnitOfWork CreateNew();
}
public interface IDataMapper : IDisposable
{
Repository<T> GetRepository<T>() where T : class;
void Save();
}
public abstract class Repository<T> : IQueryable<T>
where T : class
{
private readonly IQueryable<T> query;
protected Repository(IQueryable<T> query)
{
this.query = query;
}
public abstract void InsertOnSubmit(T entity);
public abstract void DeleteOnSubmit(T entity);
// IQueryable<T> members omitted.
}
The NorthwindUnitOfWork is a concrete class that contains properties to specific repositories, such as Customers, Orders, etc:
public sealed class NorthwindUnitOfWork : IDisposable
{
private readonly IDataMapper mapper;
public NorthwindUnitOfWork(IDataMapper mapper)
{
this.mapper = mapper;
}
// Repository properties here:
public Repository<Customer> Customers
{
get { return this.mapper.GetRepository<Customer>(); }
}
public void Dispose()
{
this.mapper.Dispose();
}
}
What's left is an concrete implementation of the INorthwindUnitOfWorkFactory and a concrete implementation of the IDataMapper. Here's one for Entity Framework:
public class EntityFrameworkNorthwindUnitOfWorkFactory
: INorthwindUnitOfWorkFactory
{
public NorthwindUnitOfWork CreateNew()
{
var db = new ObjectContext("name=NorthwindEntities");
db.DefaultContainerName = "NorthwindEntities";
var mapper = new EntityFrameworkDataMapper(db);
return new NorthwindUnitOfWork(mapper);
}
}
And the EntityFrameworkDataMapper:
public sealed class EntityFrameworkDataMapper : IDataMapper
{
private readonly ObjectContext context;
public EntityFrameworkDataMapper(ObjectContext context)
{
this.context = context;
}
public void Save()
{
this.context.SaveChanges();
}
public void Dispose()
{
this.context.Dispose();
}
public Repository<T> GetRepository<T>() where T : class
{
string setName = this.GetEntitySetName<T>();
var query = this.context.CreateQuery<T>(setName);
return new EntityRepository<T>(query, setName);
}
private string GetEntitySetName<T>()
{
EntityContainer container =
this.context.MetadataWorkspace.GetEntityContainer(
this.context.DefaultContainerName, DataSpace.CSpace);
return (
from item in container.BaseEntitySets
where item.ElementType.Name == typeof(T).Name
select item.Name).First();
}
private sealed class EntityRepository<T>
: Repository<T> where T : class
{
private readonly ObjectQuery<T> query;
private readonly string entitySetName;
public EntityRepository(ObjectQuery<T> query,
string entitySetName) : base(query)
{
this.query = query;
this.entitySetName = entitySetName;
}
public override void InsertOnSubmit(T entity)
{
this.query.Context.AddObject(entitySetName, entity);
}
public override void DeleteOnSubmit(T entity)
{
this.query.Context.DeleteObject(entity);
}
}
}
You can find more information about this model here.
UPDATE December 2012
This an an update written two years after my original answer. The last two years much has changed in the way I try to design the systems I'm working on. Although it has suited me in the past, I don't like to use the factory approach anymore when dealing with the Unit of Work pattern. Instead I simply inject a Unit of Work instance into consumers directly. Whether this design is feasibly for you however, depends a lot on the way your system is designed. If you want to read more about this, please take a look at this newer Stackoverflow answer of mine: One DbContext per web request…why?
If you want to get it right, i'd advise on a couple of changes:
1 - Don't have private instances of the data context in the repository. If your working with multiple repositories then you'll end up with multiple contexts.
2 - To solve the above - wrap the context in a Unit of Work. Pass the unit of work to the Repositories via the ctor: public MyRepository(IUnitOfWork uow)
3 - Make the Unit of Work implement IDisposable. The Unit of Work should be "newed up" when a request begins, and therefore should be disposed when the request finishes. The Repository should not implement IDisposable, as it is not directly working with resources - it is simply mitigating them. The DataContext / Unit of Work should implement IDispoable.
4 - Assuming you are using a web application, you do not need to explicitly call dispose - i repeat, you do not need to explicitly call your dispose method. StructureMap has a method called HttpContextBuildPolicy.DisposeAndClearAll();. What this does is invoke the "Dispose" method on any HTTP-scoped objects that implement IDisposable. Stick this call in Application_EndRequest (Global.asax). Also - i believe there is an updated method, called ReleaseAllHttpScopedObjects or something - can't remember the name.
Instead of adding Dispose to IMyRepository, you could declare IMyRepository like this:
public interface IMyRepository: IDisposable
{
SomeClass GetById(int id);
}
This way, you ensure all repository will call Dispose sometimes, and you can use the C# "using" pattern on a Repository object:
using (IMyRepository rep = GetMyRepository(...))
{
... do some work with rep
}