Is using static factories a valid pattern for dependency injection? - c#

I'm creating an internal framework, and at one point I would like to provide a good extensibility, supporting kind of dependecy injection.
Here is a simplified example of my filtering module in .NET/C#. Imagine there is a filter processor component, which takes a List of objects, each representing a filter condition. Then its task is "translate" these to Expressions and apply it to any IQueryable. There are trivial fitler types, for example a ColumnFilter, which refers to a fieldname, an operator, and some operands. However, I would like to provide a way to consumers to extend the filter processing mechanism with custom filter conditions. So I introduce an interface:
public interface IFilterProcessor {
Expression Process(FilterContext filter);
}
where FilterContext holds information like the currently processed filter, the root ParameterExpression, the Type of the root IQueryable, etc., not important for this example.
Then at some point one could define a new filter condition, and provide a corresponding IFilterProcessor to it, and implement the translation of the condition there.
And here comes my idea, I would provice the extensibility point as staticly registerable factories, something like this:
public class FilterProcessor {
public static readonly Dictionary<Type, Func<IFilterProcessor>> ProcessorFactories = new ...
{
{typeof(ColumnFilter), () => new ColumnFilterProcessor() }
};
...
public Expression Process(object filterCondition) {
if (filterCondition == null) return null;
var factory = ProcessorFactories.GetValueOfDefault(filterCondition.GetType()); // my custom Dictionary extension to avoid exceptions and return null if no key found
if (factory == null) return null;
using (var processor = factory()) {
return processor.Process(filterCondition);
}
}
...
}
Then at the entry point of the application I can "register" my custom filter condition processors like this:
FilterProcess.ProcessorFactories[typeof(MyCustomCondition)] = () => ... get from Ninject for example ...;
The idea behind using this pattern is that the core framework does not have to know anything about the extensions, similar to some "plugin" system.
Note that this is farly simplified and lot of things are omitted for clarity. In reality for example my filters can be grouped, nested, etc., whole lot complex.
And now my question:
I've been reading for a long time about design patterns, especially DI, but haven't found any example similar to this one. Is this a valid, and globally accepted pattern? If there is any, could anyone point out any contras about it? Finally, if this is a valid pattern, does it have a name?

Looks like you're asking and not telling (http://en.wikipedia.org/wiki/Hollywood_principle) by setting the processors using the property.
I think you can achieve the same outcome in a more 'classic' DI way: Make your FilterProcessor dependent on an enumeration of IFilterProcessor and then let the IoC framework do the resolution and injection for you by registering IFilterProcessors by convention.

Related

Use ExpandoObject to create a 'fake' implementation of an interface - adding methods dynamically

A brainteaser for you!
I am developing a modular system, in such a way that module A could need module B and module B could also need module A. But if module B is disabled, it will simply not execute that code and do nothing / return null.
A little bit more into perspective:
Let's say InvoiceBusinessLogic is within module "Core". We also have a "Ecommerce" module which has a OrderBusinessLogic. The InvoiceBusinessLogic could then look like this:
public class InvoiceBusinessLogic : IInvoiceBusinessLogic
{
private readonly IOrderBusinessLogic _orderBusinessLogic;
public InvoiceBusinessLogic(IOrderBusinessLogic orderBusinessLogic)
{
_orderBusinessLogic = orderBusinessLogic;
}
public void UpdateInvoicePaymentStatus(InvoiceModel invoice)
{
_orderBusinessLogic.UpdateOrderStatus(invoice.OrderId);
}
}
So what I want is: When the module "Ecommerce" is enabled, it would actually do something at the OrderBusinessLogic. When not, it would simply not do anything. In this example it returns nothing so it can simply do nothing, in other examples where something would be returned, it would return null.
Notes:
As you can probably tell, I am using Dependency Injection, it is a ASP.NET Core application so the IServiceCollection takes care of defining the implementations.
Simply not defining the implementation for IOrderBusinessLogic will cause a runtime issue, logically.
From a lot of research done, I do not want to make calls to the container within my domain / logic of the app. Don't call the DI Container, it'll call you
These kind of interactions between modules are kept to a minimum, preferably done within the controller, but sometimes you cannot get around it (and also in the controller I would then need a way to inject them and use them or not).
So there are 3 options that I figured out so far:
I never make calls from module "Core" to module "Ecommerce", in theory this sounds the best way, but in practice it's more complicated for advanced scenarios. Not an option
I could create a lot of fake implementations, depending on the configuration decide on which one to implement. But that would of course result in double code and I would constantly have to update the fake class when a new method is introduced. So not perfectly.
I can build up a fake implementation by using reflection and ExpandoObject, and just do nothing or return null when the particular method is called.
And the last option is what I am now after:
private static void SetupEcommerceLogic(IServiceCollection services, bool enabled)
{
if (enabled)
{
services.AddTransient<IOrderBusinessLogic, OrderBusinessLogic>();
return;
}
dynamic expendo = new ExpandoObject();
IOrderBusinessLogic fakeBusinessLogic = Impromptu.ActLike(expendo);
services.AddTransient<IOrderBusinessLogic>(x => fakeBusinessLogic);
}
By using Impromptu Interface, I am able to successfully create a fake implementation. But what I now need to solve is that the dynamic object also contains all the methods (mostly properties not needed), but those ones are easy to add. So currently I am able to run the code and get up until the point it will call the OrderBusinessLogic, then it will, logically, throw an exception that the method does not exist.
By using reflection, I can iterate over all the methods within the interface, but how do I add them to the dynamic object?
dynamic expendo = new ExpandoObject();
var dictionary = (IDictionary<string, object>)expendo;
var methods = typeof(IOrderBusinessLogic).GetMethods(BindingFlags.Public);
foreach (MethodInfo method in methods)
{
var parameters = method.GetParameters();
//insert magic here
}
Note: For now directly calling typeof(IOrderBusinessLogic), but later I would iterate over all the interfaces within a certain assembly.
Impromptu has an example as follows:
expando.Meth1 = Return<bool>.Arguments<int>(it => it > 5);
But of course I want this to be dynamic so how do I dynamically insert the return type and the parameters.
I do understand that a interface acts like a contract, and that contract should be followed, I also understand that this is an anti-pattern, but extensive research and negotiations have been done prior to reaching this point, for the result system we want, we think this is the best option, just a little missing piece :).
I have looked at this question, I am not really planning on leaving .dll's out, because most likely I would not be able to have any form of IOrderBusinessLogic usable within InvoiceBusinessLogic.
I have looked at this question, but I did not really understand how TypeBuilder could be used in my scenario
I have also looked into Mocking the interfaces, but mostly you would then need to define the 'mocking implementation' for each method that you want to change, correct me if I am wrong.
Even tough the third approach (with ExpandoObject) looks like a holy grail, I foster you to not follow this path for the following reasons:
What guarantees you that this fancy logic will be error-free now and at every time in the future ? (think: in 1 year you add a property in IOrderBusinessLogic)
What are the consequences if not ? Maybe an unexpected message will pop to the user or cause some strange "a priori unrelated" behavior
I would definitely go down the second option (fake implementation, also called Null-Object) even though, yes it will require to write some boilerplate code but ey this would offer you a compile-time guarantee that nothing unexpected will happen at rutime !
So my advice would be to do something like this:
private static void SetupEcommerceLogic(IServiceCollection services, bool enabled)
{
if (enabled)
{
services.AddTransient<IOrderBusinessLogic, OrderBusinessLogic>();
}
else
{
services.AddTransient<IOrderBusinessLogic, EmptyOrderBusinessLogic>();
}
}
For as long as there is no other answer for the solution I am looking for, I came up with the following extension:
using ImpromptuInterface.Build;
public static TInterface IsModuleEnabled<TInterface>(this TInterface obj) where TInterface : class
{
if (obj is ActLikeProxy)
{
return default(TInterface);//returns null
}
return obj;
}
And then use it like:
public void UpdateInvoicePaymentStatus(InvoiceModel invoice)
{
_orderBusinessLogic.IsModuleEnabled()?.UpdateOrderStatus(invoice.OrderId);
//just example stuff
int? orderId = _orderBusinessLogic.IsModuleEnabled()?.GetOrderIdForInvoiceId(invoice.InvoiceId);
}
And actually it has the advantage that it is clear (in the code) that the return type can be null or the method won't be called when the module is disabled. The only thing that should be documented carefully, or in another way enforced, that is has to be clear which classes do not belong to the current module. The only thing I could think of right now is by not including the using automatically, but use the full namespace or add summaries to the included _orderBusinessLogic, so when someone is using it, it is clear this belongs to another module, and a null check should be performed.
For those that are interested, here is the code to correctly add all fake implementations:
private static void SetupEcommerceLogic(IServiceCollection services, bool enabled)
{
if (enabled)
{
services.AddTransient<IOrderBusinessLogic, OrderBusinessLogic>();
return;
}
//just pick one interface in the correct assembly.
var types = Assembly.GetAssembly(typeof(IOrderBusinessLogic)).GetExportedTypes();
AddFakeImplementations(services, types);
}
using ImpromptuInterface;
private static void AddFakeImplementations(IServiceCollection services, Type[] types)
{
//filtering on public interfaces and my folder structure / naming convention
types = types.Where(x =>
x.IsInterface && x.IsPublic &&
(x.Namespace.Contains("BusinessLogic") || x.Namespace.Contains("Repositories"))).ToArray();
foreach (Type type in types)
{
dynamic expendo = new ExpandoObject();
var fakeImplementation = Impromptu.DynamicActLike(expendo, type);
services.AddTransient(type, x => fakeImplementation);
}
}

Inject dependency dynamically based on call chain using simple injector

In my application, I want to construct the following object graphs using my DI Container, Simple Injector:
new Mode1(
new CommonBuilder(
new Grouping(
new GroupingStrategy1())), // NOTE: Strategy 1
new CustomBuilder());
new Mode2(
new CommonBuilder(
new Grouping(
new GroupingStrategy2())), // NOTE: Strategy 2
new CustomBuilder());
The following classes represent the graphs above:
public class Mode1 : IMode
{
private readonly ICommonBuilder commonBuilder;
private readonly ICustomBuilder customBuilder;
public Mode1(ICommonBuilder commonBuilder, ICustomBuilder ICustomBuilder customBuilder)
{
this.commonBuilder = commonBuilder;
this.customBuilder = customBuilder;
}
public void Run()
{
this.commonBuilder.Build();
this.customBuilder.Build();
//some code specific to Mode1
}
}
public class Mode2 : IMode
{
//same code as in Mode1
public void Run()
{
this.commonBuilder.Build();
this.customBuilder.Build();
//some code specific to Mode2
}
}
With CommonBuilder and Grouping being:
public class CommonBuilder : ICommonBuilder
{
private readonly IGrouping grouping;
public CommonBuilder(IGrouping grouping)
{
this.grouping = grouping;
}
public void Build()
{
this.grouping.Group();
}
}
public class Grouping : IGrouping
{
//Grouping strategy should be binded based on mode it is running
private readonly IGroupingStrategy groupingStrategy;
public Grouping(IGroupingStrategy groupingStrategy)
{
this.groupingStrategy = groupingStrategy;
}
public void Group()
{
this.groupingStrategy.Execute();
}
}
I am using Simple Injector for DI in my project. As shown above, I've go 2 modes of code which are called as per user preference, I've got common code for each mode (which I don't want to duplicate), I want to bind my grouping strategy (I've go 2 grouping strategies, one for each mode) in my common code based on the mode of execution. I've come across solution to use factories and switch between bindings at run time, but I'don't want go with that solution as I've same scenario at multiple places in my code (I'll end up creating multiple factories).
Can anyone suggest how to do the binding in cleaner way
You can use Context-Based Injection. However, because the context-based injection based on consumer of the dependency's consumer (or parents of their parents) can lead to all sorts of complications and subtle bugs (especially when caching lifestyles such as Scoped and Singleton are involved), Simple Injector's API limits you to looking one level up.
There are several ways to work around this seeming limitation in Simple Injector. The first thing you should do is take a step back and see whether or not you can simplify your design, as these kinds of requirements often (but not always) come from design inefficiencies. One such issue is Liskov Substitution Principle (LSP) violations. From this perspective, it's good to ask yourself the question, would Mode1 break when it gets injected with a Grouping that contains the strategy for Mode2? If the answer is yes, you are likely violating the LSP and you should first and foremost try to fix that problem first. When fixed, you'll likely see your configuration problems go away as well.
If you determined the design doesn't violate LSP, the second-best option is to burn type information about the consumer's consumer directly into the graph. Here's a quick example:
var container = new Container();
container.Collection.Append<IMode, Mode1>();
container.Collection.Append<IMode, Mode2>();
container.RegisterConditional(
typeof(ICommonBuilder),
c => typeof(CommonBuilder<>).MakeGenericType(c.Consumer.ImplementationType),
Lifestyle.Transient,
c => true);
container.RegisterConditional(
typeof(IGrouping),
c => typeof(Grouping<>).MakeGenericType(c.Consumer.ImplementationType),
Lifestyle.Transient,
c => true);
container.RegisterConditional<IGroupingStrategy, Strategy1>(
c => typeof(Model1) == c.Consumer.ImplementationType
.GetGenericArguments().Single() // Consumer.Consumer
.GetGenericArguments().Single(); // Consumer.Consumer.Consumer
container.RegisterConditional<IGroupingStrategy, Strategy2>(
c => typeof(Mode2)) == c.Consumer.ImplementationType
.GetGenericArguments().Single()
.GetGenericArguments().Single();
In this example, instead of using a non-generic Grouping class, a new Grouping<T> class is created, and the same is done for CommonBuilder<T>. These classes can be a sub class of the non-generic Grouping and CommonBuilder classes, placed in your Composition Root, so you don't have to change your application code for this:
class Grouping<T> : Grouping // inherit from base class
{
public Grouping(IGroupingStrategy strategy) : base(strategy) { }
}
class CommonBuilder<T> : CommonBuilder // inherit from base class
{
public CommonBuilder(IGrouping grouping) : base(grouping) { }
}
Using this generic CommonBuilder<T>,1 you make a registration where the T becomes the type of the consumer it is injected into. In other words, Mode1 will be injected with a CommonBuilder<Mode1> and Mode2 will get a CommonBuilder<Mode2>. This is identical to what is common when registering ILogger implementations as shown in the documentation. Because of the generic typing, however, CommonBuilder<Mode1> will be injected with an Grouping<CommonBuilder<Mode1>>.
These registrations aren't really conditional, rather contextual. The injected type changes based on its consumer. This construct, however, makes the type information of IGrouping's consumer available in the constructed object graph. This allows the conditional registrations for IGroupingStrategy to be applied based on that type information. This is what happens inside the registration's predicate:
c => typeof(Mode2)) == c.Consumer.ImplementationType // = Grouping<CommonBuilder<Mode2>>
.GetGenericArguments().Single() // = CommonBuilder<Mode2>
.GetGenericArguments().Single(); // = Mode2
In other words, if we can change the IGrouping implementation in such way that its implementation's type (Grouping<T>) provides information about its consumers (the IMode implementations). This way the conditional registration of IGroupingStrategy can use that information about it's consumer's consumer.
Here the registration requests the consumer's implementation type (which will be either Grouping<Mode1> or Grouping<Mode2>) and will grab the single generic argument from that implementation (which will be either Mode1 or Mode2). In other words this allows us to get the consumer's consumer. This can be matched with the expected type to return either true or false.
Although this seems a bit awkward and complex, advantage of this model is that the complete object graph is known to Simple Injector, which allows it to analyze and verify the object graph. It also allows Auto-Wiring to take place. In other words, if IGrouping or IGroupingStrategy implementations have (other) dependencies, Simple Injector will automatically inject them and verify their correctness. It also allows you to visualize the object graph without losing any information. For instance, this is the graph that Simple Injector will show if you hover over it in the Visual Studio debugger:
Mode1(
CommonBuilder<Mode1>(
Grouping<CommonBuilder<Mode1>>(
Strategy1())))
The apparent downside of this approach is that if either CommonBuilder<T> or Grouping<T> are registered as singleton, there will now be a single instance per closed-generic type. This means that CommonBuilder<Mode1> will be a different instance than CommonBuilder<Mode2>.
Alternatively, you can also just make the CommonBuilder registration conditional as follows:
var container = new Container();
container.Collection.Append<IMode, Mode1>();
container.Collection.Append<IMode, Mode2>();
container.RegisterConditional<ICommonBuilder>(
Lifestyle.Transient.CreateRegistration(
() => new CommonBuilder(new Grouping(new Strategy1())),
container),
c => c.Consumer.ImplementationType == typeof(Mode1));
container.RegisterConditional<ICommonBuilder>(
Lifestyle.Transient.CreateRegistration(
() => new CommonBuilder(new Grouping(new Strategy2())),
container),
c => c.Consumer.ImplementationType == typeof(Mode2));
This is a bit simpler than the previous approach, but it disables Auto-Wiring. In this case CommonBuilder and its dependencies are wired by hand. When the object graph is simple (doesn't contain many dependencies), this method can be good enough. When dependencies are added to either CommonBuilder, Grouping or the strategies, however, this can cause high maintenance and might hide bugs as Simple Injector will be unable to verify the dependency graph on your behalf.
Please do be aware of the following statement in the documentation about the RegisterConditional methods:
The predicates are only used during object graph compilation and the predicate’s result is burned in the structure of returned object graph. For a requested type, the exact same graph will be created on every subsequent call. This disallows changing the graph based on runtime conditions.

C# how to "register" class "plug-ins" into a service class?

Update#2 as of year 2022
All these years have passed and still no good answer.
Decided to revive this question.
I'm trying to implement something like the idea I'm trying to show with the following diagram (end of the question).
Everything is coded from the abstract class Base till the DoSomething classes.
My "Service" needs to provide to the consumer "actions" of the type "DoSomethings" that the service has "registered", at this point I am seeing my self as repeating (copy/paste) the following logic on the service class:
public async Task<Obj1<XXXX>> DoSomething1(....params....)
{
var action = new DoSomething1(contructParams);
return await action.Go(....params....);
}
I would like to know if there is anyway in C# to "register" all the "DoSomething" I want in a different way? Something more dynamic and less "copy/paste" and at the same time provide me the "intellisense" in my consumer class? Somekind of "injecting" a list of accepted "DoSomething" for that service.
Update#1
After reading the sugestion that PanagiotisKanavos said about MEF and checking other options of IoC, I was not able to find exactly what I am looking for.
My objective is to have my Service1 class (and all similar ones) to behave like a DynamicObject but where the accepted methods are defined on its own constructor (where I specify exactly which DoSomethingX I am offering as a method call.
Example:
I have several actions (DoSomethingX) as "BuyCar", "SellCar", "ChangeOil", "StartEngine", etc....
Now, I want to create a service "CarService" that only should offer the actions "StartEngine" and "SellCar", while I might have other "Services" with other combination of "actions". I want to define this logic inside the constructor of each service. Then, in the consumer class, I just want to do something like:
var myCarService = new CarService(...paramsX...);
var res1 = myCarService.StartEngine(...paramsY...);
var res2 = myCarService.SellCar(...paramsZ...);
And I want to offer intellisense when I use the "CarService"....
In conclusion: The objective is how to "register" in each Service which methods are provided by him, by giving a list of "DoSomethingX", and automatically offer them as a "method"... I hope I was able to explain my objective/wish.
In other words: I just want to be able to say that my class Service1 is "offering" the actions DoSomething1, DoSomething2 and DoSomething3, but with the minimum lines as possible. Somehow the concept of the use of class attributes, where I could do something similar to this:
// THEORETICAL CODE
[RegisterAction(typeOf(DoSomething1))]
[RegisterAction(typeOf(DoSomething2))]
[RegisterAction(typeOf(DoSomething3))]
public class Service1{
// NO NEED OF EXTRA LINES....
}
For me, MEF/MAF are really something you might do last in a problem like this. First step is to work out your design. I would do the following:
Implement the decorator design pattern (or a similar structural pattern of your choice). I pick decorator as that looks like what you are going for by suplimenting certain classes with shared functionality that isn't defined in those clases (ie composition seems prefered in your example as opposed to inheritance). See here http://www.dofactory.com/net/decorator-design-pattern
Validate step 1 POC to work out if it would do what you want if it was added as a separate dll (ie by making a different CSProj baked in at build time).
Evaluate whether MEF or MAF is for right for you (depending on how heavy weight you want to go). Compare those against other techniques like microservices (which would philosophically change your current approach).
Implement your choice of hot swapping (MEF is probably the most logical based on the info you have provided).
You could use Reflection.
In class Service1 define a list of BaseAction types that you want to provide:
List<Type> providedActions = new List<Type>();
providedActions.Add(typeof(DoSomething1));
providedActions.Add(typeof(DoSomething2));
Then you can write a single DoSomething method which selects the correct BaseAction at run-time:
public async Task<Obj1<XXXX>> DoSomething(string actionName, ....params....)
{
Type t = providedActions.Find(x => x.Name == actionName);
if (t != null)
{
var action = (BaseAction)Activator.CreateInstance(t);
return await action.Go(....params....);
}
else
return null;
}
The drawback is that the Client doesn't know the actions provided by the service unless you don't implement an ad-hoc method like:
public List<string> ProvidedActions()
{
List<string> lst = new List<string>();
foreach(Type t in providedActions)
lst.Add(t.Name);
return lst;
}
Maybe RealProxy can help you? If you create ICarService interface which inherits IAction1 and IAction2, you can then create a proxy object which will:
Find all the interfaces ICarService inherits.
Finds realizations of these interfaces (using actions factory or reflection).
Creates action list for the service.
In Invoke method will delegate the call to one of the actions.
This way you will have intellisence as you want, and actions will be building blocks for the services. Some kind of multi-inheritance hack :)
At this point I am really tempted to do the following:
Make my own Class Attribute RegisterAction (just like I wrote on my "Theoretical" example)
Extend the Visual Studio Build Process
Then on my public class LazyProgrammerSolutionTask: Microsoft.Build.Utilities.Task try to find the service classes and identify the RegisterAction attributes.
Then per each one, I will inject using reflection my own method (the one that I am always copying paste)... and of course get the "signature" from the corresponding target "action" class.
In the end, compile everything again.
Then my "next project" that will consume this project (library) will have the intellisence that I am looking for....
One thing, that I am really not sure, it how the "debug" would work on this....
Since this is also still a theoretically (BUT POSSIBLE) solution, I do not have yet a source code to share.
Meanwhile, I will leave this question open for other possible approaches.
I must disclose, I've never attempted anything of sorts so this is a thought experiment. A couple of wild ideas I'd explore here.
extension methods
You could declare and implement all your actions as extension methods against base class. This I believe will cover your intellisense requirements. Then you have each implementation check if it's registered against calling type before proceeding (use attributes, interface hierarchy or other means you prefer). This will get a bit noisy in intellisense as every method will be displayed on base class. And this is where you can potentially opt to filter it down by custom intellisense plugin to filter the list.
custom intellisense plugin
You could write a plugin that would scan current code base (see Roslyn), analyze your current service method registrations (by means of attributes, interfaces or whatever you prefer) and build a list of autocomplete methods that apply in this particular case.
This way you don't have to install any special plugins into your Dev environment and still have everything functional. Custom VS plugin will be there purely for convenience.
If you have a set of actions in your project that you want to invoke, maybe you could look at it from CQS (Command Query Separation) perspective, where you can define a command and a handler from that command that actually performs the action. Then you can use a dispatcher to dispatch a command to a handler in a dynamic way. The code may look similar to:
public class StartEngine
{
public StartEngine(...params...)
{
}
}
public class StartEngineHandler : ICommandHandler<StartEngine>
{
public StartEngineHandler(...params...)
{
}
public async Task Handle(StartEngine command)
{
// Start engine logic
}
}
public class CommandDispatcher : ICommandDispatcher
{
private readonly Container container;
public CommandDispatcher(Container container) => this.container = container;
public async Task Dispatch<T>(T command) =>
await container.GetInstance<ICommandHandler<T>>().Handle(command);
}
// Client code
await dispatcher.Dispatch(new StartEngine(params, to, start, engine));
This two articles will give you more context on the approach: Meanwhile... on the command side of my architecture, Meanwhile... on the query side of my architecture.
There is also a MediatR library that solves similar task that you may want to check.
If the approaches from above does not fit the need and you want to "dynamically" inject actions into your services, Fody can be a good way to implement it. It instruments the assembly during the build after the IL is generated. So you could implement your own weaver to generate methods in the class decorated with your RegisterAction attribute.

Looking for a Design pattern which can create different instances of a class with different interface implementations

I have a class which contains a view dependencies (all interfaces). Basically the behavior of the class is defined through the implementation of those interfaces. I want to be able to have a "builder" which can create instances of this class with different implementations of the interfaces(or parts of it). Something like this:
public class API
{
private readonly ISomeInterface _someInterface;
private readonly ISomeOtherInterface _someOtherInterface;
private readonly ISomeAnotherInterface _someAnotherInterface;
API(ISomeInterface someInterface,ISomeOtherInterface someOtherInterface,ISomeAnotherInterface someAnotherInterface)
{*/implementation ommitted*/}
//Example method
public void DoSomethingWhichDependsOnOneOrMoreInterfaces()
{
//somecode
id(_someInterface != null)
_someInterface.SomeMethode();
}
public class MyApiBuilder()
{
// implementation ommitted
API CreateAPI(someEnum type)
{
switch(type)
{
case SpecificAPI32:
var speficImplementationOfSomeInterface = new ImplementsISomeInterface();
speficImplementationOfSomeInterface .Setup("someSetup");
var specificImplementationOfOtherInterface = new ImplementsISomeOtherInterface();
returns new API(speficImplementationOfSomeInterface,specificImplementationOfOtherInterface ,null);
}
}
}
What is the most elegant way of implementing this (if this makes sense at all)? I was first thinking of the Builder Design Patterns but as far as I understood it, its slightly different.
[Edit]
As pointed out, the way I am implementing it is a factory method but I am not fully satisfied with it. The API can contain a varity of different interfaces which can be totally independent of each other but some may depend on others.(but not mandatory) I would like to give the user (the developer using this "API") as much freedom as possible in creating the API he wants to use. Lets try to explain what I am basically up to:
Let's say I am developing a plugin for a game engine which can post achievments and other stuff to various social media channels. So basically there could be a Interface which implements the access to twitter,facebook,youtube,whathever or some custom server. This custom server could need some kind of authentification process. The user should be able to build at start the API in a nice (hmm fluent is nice..) way. So basically something like this:
var myTotallyForMyNeedsBuildAPI = API.CreateCustomApi().With(Api.Twitter).And(Api.Facebook).And(Api.Youtube).And(Api.CustomServer).With(Security.Authentification);
I actually do not know how to make that fluent but something like this would be nice.
It's a good practice to use Dependency Injection as you want to give the programmer the ability to compose the object with desired configuration.
Check MEF and Unity frameworks which are great for this job.
For example in Unity you can write this:
// Introducing an implementation for ISomeInterface
container.Register<ISomeInterface, SomeImplementation>();
// Introducing an implementation for ISomeOtherInterface
container.Register<ISomeOtherInterface, SomeOtherImplementation>();
// Introducing an implementation for ISomeAnotherInterface
container.Register<ISomeAnotherInterface, SomeAnotherImplemenation>();
container.Register<API, API>();
// and finally unity will compose it for you with desired configurations:
var api = container.Resolve<API>();
In this scenario the api will be composed with desired implementations.
What you have implemented is the Factory method pattern.
It's perfectly fine for what you are trying to do, but you could have a look at the other factory patterns (i.e. here) based on your context and how you think you're code will evolve in the future.
Anyway, I will also consider to not tie this three interface together in a single factory. If they are really so tighten together to be consumed together and built together, maybe they should not be three different interfaces in the first place, or at least all three implemented by the same class, so your factory will build the appropriate class with the proper implementation of these.
Probably what you are after is the Decorator pattern.
In your API class you invoke each interface if they have been provided to the API instance, which is the behaviour of the Decorator pattern.
With this pattern you obtain a modular implementation that allow you to add multiple behaviours to your API.

Dependency Injection and the Strategy Pattern

There is an enormous amount of discussion on this topic, but everyone seems to miss an obvious answer. I'd like help vetting this "obvious" IOC container solution. The various conversations assume run-time selection of strategies and the use of an IOC container. I will continue with these assumptions.
I also want to add the assumption that it is not a single strategy that must be selected. Rather, I might need to retrieve an object-graph that has several strategies found throughout the nodes of the graph.
I will first quickly outline the two commonly proposed solutions, and then I will present the "obvious" alternative that I'd like to see an IOC container support. I will be using Unity as the example syntax, though my question is not specific to Unity.
Named Bindings
This approach requires that every new strategy has a binding manually added:
Container.RegisterType<IDataAccess, DefaultAccessor>();
Container.RegisterType<IDataAccess, AlphaAccessor>("Alpha");
Container.RegisterType<IDataAccess, BetaAccessor>("Beta");
...and then the correct strategy is explicitly requested:
var strategy = Container.Resolve<IDataAccess>("Alpha");
Pros: Simple, and supported by all IOC Containers
Cons:
Typically binds the caller to the IOC Container, and certainly requires the caller to know something about the strategy (such as the name "Alpha").
Every new strategy must be manually added to the list of bindings.
This approach is not suitable for handling multiple strategies in an object graph. In short, it does not meet requirements.
Abstract Factory
To illustrate this approach, assume the following classes:
public class DataAccessFactory{
public IDataAccess Create(string strategy){
return //insert appropriate creation logic here.
}
public IDataAccess Create(){
return //Choose strategy through ambient context, such as thread-local-storage.
}
}
public class Consumer
{
public Consumer(DataAccessFactory datafactory)
{
//variation #1. Not sufficient to meet requirements.
var myDataStrategy = datafactory.Create("Alpha");
//variation #2. This is sufficient for requirements.
var myDataStrategy = datafactory.Create();
}
}
The IOC Container then has the following binding:
Container.RegisterType<DataAccessFactory>();
Pros:
The IOC Container is hidden from consumers
The "ambient context" is closer to the desired result but...
Cons:
The constructors of each strategy might have different needs. But now the responsibility of constructor injection has been transferred to the abstract factory from the container. In other words, every time a new strategy is added it may be necessary to modify the corresponding abstract factory.
Heavy use of strategies means heavy amounts of creating abstract factories. It would be nice if the IOC container simply gave a little more help.
If this is a multi-threaded application and the "ambient context" is indeed provided by thread-local-storage, then by the time an object is using an injected abstract-factory to create the type it needs, it may be operating on a different thread which no longer has access to the necessary thread-local-storage value.
Type Switching / Dynamic Binding
This is the approach that I want to use instead of the above two approaches. It involves providing a delegate as part of the IOC container binding. Most all IOC Containers already have this ability, but this specific approach has an important subtle difference.
The syntax would be something like this:
Container.RegisterType(typeof(IDataAccess),
new InjectionStrategy((c) =>
{
//Access ambient context (perhaps thread-local-storage) to determine
//the type of the strategy...
Type selectedStrategy = ...;
return selectedStrategy;
})
);
Notice that the InjectionStrategy is not returning an instance of IDataAccess. Instead it is returning a type description that implements IDataAccess. The IOC Container would then perform the usual creation and "build up" of that type, which might include other strategies being selected.
This is in contrast to the standard type-to-delegate binding which, in the case of Unity, is coded like this:
Container.RegisterType(typeof(IDataAccess),
new InjectionFactory((c) =>
{
//Access ambient context (perhaps thread-local-storage) to determine
//the type of the strategy...
IDataAccess instanceOfSelectedStrategy = ...;
return instanceOfSelectedStrategy;
})
);
The above actually comes close to satisfying the overall need, but definitely falls short of the hypothetical Unity InjectionStrategy.
Focusing on the first sample (which used a hypothetical Unity InjectionStrategy):
Pros:
Hides the container
No need either to create endless abstract factories, or have consumers fiddle with them.
No need to manually adjust IOC container bindings whenever a new strategy is available.
Allows the container to retain lifetime management controls.
Supports a pure DI story, which means that a multi-threaded app can create the entire object-graph on a thread with the proper thread-local-storage settings.
Cons:
Because the Type returned by the strategy was not available when the initial IOC container bindings were created, it means there may be a tiny performance hit the first time that type is returned. In other words, the container must on-the-spot reflect the type to discover what constructors it has, so that it knows how to inject it. All subsequent occurrences of that type should be fast, because the container can cache the results it found from the first time. This is hardly a "con" worth mentioning, but I'm trying for full-disclosure.
???
Is there an existing IOC container that can behave this way? Anyone have a Unity custom injection class that achieves this effect?
As far as I can tell, this question is about run-time selection or mapping of one of several candidate Strategies.
There's no reason to rely on a DI Container to do this, as there are at least three ways to do this in a container-agnostic way:
Use a Metadata Role Hint
Use a Role Interface Role Hint
Use a Partial Type Name Role Hint
My personal preference is the Partial Type Name Role Hint.
I have achieved this requirement in many forms over the last couple of years. Firstly let's pull the main points I can see in your post
assume run-time selection of strategies and the use of an IOC container ... add the assumption that it is not a single strategy that must be selected. Rather, I might need to retrieve an object-graph that has several strategies ... [must not] binds the caller to the IOC Container ... Every new strategy must [not need to] be manually added to the list of bindings ... It would be nice if the IOC container simply gave a little more help.
I have used Simple Injector as my container of choice for some time now and one of the drivers for this decision is that it has extensive support for generics. It is through this feature that we will implement your requirements.
I'm a firm believer that the code should speak for itself so I'll jump right in ...
I have defined an extra class ContainerResolvedClass<T> to demonstrate that Simple Injector finds the right implementation(s) and successfully injects them into a constructor. That is the only reason for the class ContainerResolvedClass<T>. (This class exposes the handlers that are injected into it for test purposes via result.Handlers.)
This first test requires that we get one implementation back for the fictional class Type1:
[Test]
public void CompositeHandlerForType1_Resolves_WithAlphaHandler()
{
var container = this.ContainerFactory();
var result = container.GetInstance<ContainerResolvedClass<Type1>>();
var handlers = result.Handlers.Select(x => x.GetType());
Assert.That(handlers.Count(), Is.EqualTo(1));
Assert.That(handlers.Contains(typeof(AlphaHandler<Type1>)), Is.True);
}
This second test requires that we get one implementation back for the fictional class Type2:
[Test]
public void CompositeHandlerForType2_Resolves_WithAlphaHandler()
{
var container = this.ContainerFactory();
var result = container.GetInstance<ContainerResolvedClass<Type2>>();
var handlers = result.Handlers.Select(x => x.GetType());
Assert.That(handlers.Count(), Is.EqualTo(1));
Assert.That(handlers.Contains(typeof(BetaHandler<Type2>)), Is.True);
}
This third test requires that we get two implementations back for the fictional class Type3:
[Test]
public void CompositeHandlerForType3_Resolves_WithAlphaAndBetaHandlers()
{
var container = this.ContainerFactory();
var result = container.GetInstance<ContainerResolvedClass<Type3>>();
var handlers = result.Handlers.Select(x => x.GetType());
Assert.That(handlers.Count(), Is.EqualTo(2));
Assert.That(handlers.Contains(typeof(AlphaHandler<Type3>)), Is.True);
Assert.That(handlers.Contains(typeof(BetaHandler<Type3>)), Is.True);
}
These tests seems to meet your requirements and best of all no containers are harmed in the solution.
The trick is to use a combination of parameter objects and marker interfaces. The parameter objects contain the data for the behaviour (i.e. the IHandler's) and the marker interfaces govern which behaviours act on which parameter objects.
Here are the marker interfaces and parameter objects - you'll note that Type3 is marked with both marker interfaces:
private interface IAlpha { }
private interface IBeta { }
private class Type1 : IAlpha { }
private class Type2 : IBeta { }
private class Type3 : IAlpha, IBeta { }
Here are the behaviours (IHandler<T>'s):
private interface IHandler<T> { }
private class AlphaHandler<TAlpha> : IHandler<TAlpha> where TAlpha : IAlpha { }
private class BetaHandler<TBeta> : IHandler<TBeta> where TBeta : IBeta { }
This is the single method that will find all implementations of an open generic:
public IEnumerable<Type> GetLoadedOpenGenericImplementations(Type type)
{
var types =
from assembly in AppDomain.CurrentDomain.GetAssemblies()
from t in assembly.GetTypes()
where !t.IsAbstract
from i in t.GetInterfaces()
where i.IsGenericType
where i.GetGenericTypeDefinition() == type
select t;
return types;
}
And this is the code that configures the container for our tests:
private Container ContainerFactory()
{
var container = new Container();
var types = this.GetLoadedOpenGenericImplementations(typeof(IHandler<>));
container.RegisterAllOpenGeneric(typeof(IHandler<>), types);
container.RegisterOpenGeneric(
typeof(ContainerResolvedClass<>),
typeof(ContainerResolvedClass<>));
return container;
}
And finally, the test class ContainerResolvedClass<>
private class ContainerResolvedClass<T>
{
public readonly IEnumerable<IHandler<T>> Handlers;
public ContainerResolvedClass(IEnumerable<IHandler<T>> handlers)
{
this.Handlers = handlers;
}
}
I realise this post is a quite long, but I hope it clearly demonstrates a possible solution to your problem ...
This is a late response but maybe it will help others.
I have a pretty simple approach. I simply create a StrategyResolver to not be directly depending on Unity.
public class StrategyResolver : IStrategyResolver
{
private IUnityContainer container;
public StrategyResolver(IUnityContainer unityContainer)
{
this.container = unityContainer;
}
public T Resolve<T>(string namedStrategy)
{
return this.container.Resolve<T>(namedStrategy);
}
}
Usage:
public class SomeClass: ISomeInterface
{
private IStrategyResolver strategyResolver;
public SomeClass(IStrategyResolver stratResolver)
{
this.strategyResolver = stratResolver;
}
public void Process(SomeDto dto)
{
IActionHandler actionHanlder = this.strategyResolver.Resolve<IActionHandler>(dto.SomeProperty);
actionHanlder.Handle(dto);
}
}
Registration:
container.RegisterType<IActionHandler, ActionOne>("One");
container.RegisterType<IActionHandler, ActionTwo>("Two");
container.RegisterType<IStrategyResolver, StrategyResolver>();
container.RegisterType<ISomeInterface, SomeClass>();
Now, the nice thing about this is that I will never have to touch the StrategyResolver ever again when adding new strategies in the future.
It's very simple. Very clean and I kept the dependency on Unity to a strict minimum. The only time I would have touch the StrategyResolver is if I decide to change container technology which is very unlikely to happen.
Hope this helps!
I generally use a combination of your Abstract Factory and Named Bindings options. After trying many different approaches, I find this approach to be a decent balance.
What I do is create a factory that essentially wraps the instance of the container. See the section in Mark's article called Container-based Factory. As he suggests, I make this factory part of the composition root.
To make my code a little cleaner and less "magic string" based, I use an enum to denote the different possible strategies, and use the .ToString() method to register and resolve.
From your Cons of these approaches:
Typically binds the caller to the IOC Container
In this approach, the container is referenced in the factory, which is part of the Composition Root, so this is no longer an issue (in my opinion).
. . . and certainly requires the caller to know something about the strategy (such as the
name "Alpha").
Every new strategy must be manually added to the list
of bindings. This approach is not suitable for handling multiple
strategies in an object graph. In short, it does not meet
requirements.
At some point, code needs to be written to acknowledge the mapping between the structure that provides the implementation (container, provider, factory, etc.) and the code that requires it. I don't think you can get around this unless you want to use something that is purely convention-based.
The constructors of each strategy might have different needs. But now the responsibility of constructor injection has been transferred to the abstract factory from the container. In other words, every time a new strategy is added it may be necessary to modify the corresponding abstract factory.
This approach solves this concern completely.
Heavy use of strategies means heavy amounts of creating abstract factories.[...]
Yes, you will need one abstract factory for each set of strategies.
If this is a multi-threaded application and the "ambient context" is indeed provided by thread-local-storage, then by the time an object is using an injected abstract-factory to create the type it needs, it may be operating on a different thread which no longer has access to the necessary thread-local-storage value.
This will no longer be an issue since TLC will not be used.
I don't feel that there is a perfect solution, but this approach has worked well for me.

Categories

Resources