In an application using the DDD concepts I am in doubt about who could inject (dependencies) into the constructor of a given class if there is any standard for it.
For example, between the Application, Domain, and Repository layers.
1) A ClientAppService (Application layer) that needs to inject user, should I inject UserApplicationService and from it call UserService (Domain) or inject UserService directly in ClientApplicationService?
2) In the ClientService (domain) should I inject UserService and from it call UserRepository or could I inject UserRepository directly into ClientService?
I'm concerned about cyclic reference if I'm injecting peer classes.
But I also think that I should not inject the Repository of another Entity, because often the methods of the repository have a rule in the service that must be called previously.
Has anyone ever had this question, how do you usually handle it?
Thinking about separation of concerns and allocation of responsibilities, you should inject exactly what your artifact depends upon. This may sound a little obvious, but it goes a little deeper.
Considering your (2) example:
In the ClientService (domain) should I inject UserService and from it call UserRepository or could I inject UserRepository directly into ClientService?
You probably should first ask yourself which capability does your ClientService depend upon?
If it (ClientService) cares about being able to find user from information it (ClientService) currently possesses, it should probably receive the UserRepository directly and be able to find the user on its own.
If it (ClientService) needs a user, but it doesn't possess the information needed to find the user (this information is currently on application layer level), maybe ClientService should receive the User domain object directly, with the repository being used straight from application level.
If it (ClientService) needs some kind of domain-relevant functionality from UserService as part of its operation, then, in that case, the UserService might be directly injected into ClientService.
Other possible discussion on this topic might whether you really need all those Domain Services of if you would be better calling rule-rich Entities/Aggregates straight from the Application Layer, it might make your overall design, injection patterns and boundaries simpler.
Also, many times, you might want to inject factories for your artifact rather than the instantiated ones directly.
Another point might be made about:
But I also think that I should not inject the Repository of another Entity, because often the methods of the repository have a rule in the service that must be called previously.
This might be evidence of some confusion inside your domain. The role of a repository should be around the lines of "finding your domain entity from the universe of possible entities". In that sense, a UserRepository enables you to find users from the users existing in your universe so it should be a pretty isolated operation and shouldn't depend on services or other entities. If a user exists, it should be "findable" (and persistable, as it goes both ways) from the UserRepository.
In that case, you shouldn't worry about "injecting UserRepository in ClientService" from a dogmatic point of view. If the operation in your client service needs to find and use a User Entity, it should be alright for you to do so. What you might worry about is whether your entities/aggregates are well designed or if you have some kind of misplaced responsibilities that might be triggering this "feeling" of "I shouldn't be injecting this into that".
Domain Entities and Value Objects almost never use constructor injection.
This is motivated by separation of concerns; the responsibility of the objects in the domain model is to manage their own in memory representations.
Other capabilities that they may need to do their work are passed to them as arguments.
The typical mechanism for this is the "domain service", described by Evans in chapter 5 of the blue book.
To sketch an example - suppose my order aggregate needs to update its quote when the line items change. I might pass in as an argument an interface that accepts a SKU and returns a Price. As far as Order is concerned, that lookup is happening "somewhere else". It doesn't care about the details. The implementation might load up another aggregate to look up its current state, or send a message to some remote system, or hard code an answer.
Domain Service implementations will often have injected dependencies on capabilities provided by the application or infrastructure layers.
Related
I have started to go through Symfony2 code, studied a little bit small classes like Pimple and after few hours of examination have come to strange idea. The best start for this is to explain how I understand few terms, so:
Dependency
Something which is needed for another thing to work, like "engine" in "car"
Container
Object or class which is able to store a lot of another objects, like "engine","gearbox" or even "car"
Dependency Injection
Process in which every dependency is injected to object, so if I need "car" I know that I have to inject "engine","gearbox" and a lot of another stuff. Important is, that "car" don't create "engine", but "engine" is put inside "car"
Service Locator
Process in which object asks for another object, for example into the car there is inserted our container, and when car need to start it require from container "engine", so container return him "engine"
When i studied Symphony code, they start with dependency injection, but after some time I realize that when Controller is created, there is injected whole container and then you can use $this->get('serviceName') to get it, so it more looks like service locator, which according to few articles is anti-pattern.
Sow how is it? Is that line between DI and SL so small that it is sometimes broken? Or did I misunderstood something? If I use DI, do I need to insert every service into controller, so I know from outside what I use? Or can controller become in some cases container?
Your understanding of DI is pretty good. And yes, Symfony Controller does implement ContainerAwareInterface, and as you said, has a service locator role. But service locator isn't an anti-pattern. Each pattern has it's proper and improper uses.
Furthermore, Symfony doesn't enforce you in any way to use it's Controller. Your Controller can be a service. Hell, it can even be a function!
Here is one of the reasons why Controllers are implemented as service locators: Performance.
Let's drop car analogy and focus on real case that you'll encounter in 99% of projects: you need CRUD for a resource. Let's say you're building a Todo app and you need a RESTfulish controller to handle CRUD operations for Task Resource.
The least you need to have is a way to read all tasks and a way to add a new task, for that you need two actions: index (commonly named list too), and store (commonly named create too).
Common flow in Symfony would be this, in pseudo code:
indexAction -> getDoctrine -> getTaskRepository -> getAllTasks
storeAction -> getFormFactory -> createForm -> bindRequestDataToForm -> getDoctrine -> saveData
If Controller was a service locator
Index Action
When index action is executed, only service that will be resolved from the container will be ManagerRegistry (in this case Doctrine service). We will then ask it to give us task repository, and we'll do our operation with it.
Store Action
When store action is executed, we will do a bit more work: ask container to give us FormFactory, do some operations with it, and then ask it to give us Doctrine and do some operations with it too.
So summary: when index action is executed, only one service has to be constructed by service container, when update is executed, two will have to be constructed.
If Controller was a regular service
Let's see what our Controller needs. From the section above, we see that it needs FormFactory and Doctrine.
Now, when you just want to call index action to read all tasks from data storage, your controller will have to get instantiated by container. Before it can be instantiated, container needs to instantiate it's dependencies: FormFactory and Doctrine. Then instantiate controller while injecting those two into it.
So, you are calling index action which doesn't need FormFactory at all, but you still have overhead of creating it because it is needed for an action that will not be called at all in that request.
Lazy Services
To reduce this overhead, there is a thing called lazy service. It works by actually injecting a Proxy of your service into the controller. So, as far as controller is concerned, it got FormFactory. What it doesn't know is that isn't real FormFactory but instead a fake object which will delegate calls to real FormFactory code when you call some method on it.
Wrapping it up
Controller doesn't have to be a service locator, but can be. Making it a service locator can be a bit more performant and easier to bootstrap, but hides dependencies. Furthermore, it's a bit harder to test, since you'll need to mock dependency container. Whether you want to make your controllers services, functions or service locators is your choice and Symfony won't enforce you to use any of those ways.
In my experience, extending default Symfony Controller and having controllers be service locators is just fine as long as you don't write your business logic in them, but instead delegate all that work to services, which you get from container. That way, it's really unlikely that you'll have bugs in controller code (as methods will usually consist of 2-3 lines of code) and can get away without testing it.
I have a small app (Win Forms) where I have two different repositories (SQL Server and Redis) both implementing a interface IFilterRepo.
I also have a service class that depends on the IFilterRepo. The client (the Win Form) call the service to access filter data.
I want the client to have two radio buttons where a user can choose which repo to use. And here comes my dilemma. How should I tell the service which concrete class to instantiate as IFilterRepo? I mean, ALL Unity registrations and references to it shall be done in the composition root. Is that "rule" really possible in this case?
This is a common question, and the answer is generally to use an Abstract Factory.
Here is a good article on the subject (I link this all the time, but I didn't write it):
http://blog.ploeh.dk/2012/03/15/ImplementinganAbstractFactory/
As noted in the article, you can make the factory part of the composition root, so that calling container.Resolve() inside the factory doesn't violate that rule.
Edit
You would register different implementations of the service using a name (string):
http://msdn.microsoft.com/en-us/library/ff648211.aspx
myContainer.RegisterType<IMyService, CustomerService>("Customers");
And then your factory would resolve by that name:
public IFilterRepo Create(string myName)
{
return myContainer.Resolve<IFilterRepo>(myName);
}
Edit 2
The question you asked in your last comment is a bit much to answer here, but in brief: your factory itself would implement an interface, and would be resolved and registered via the container.
As a general matter, I would not recommend accessing a repository directly from the code behind--I would at least look at having a layered architecture (or better, an Onion architecture, which works very well with DI).
Finally, I have not done WinForms development in years, but I don't think it fits perfectly with using a container/Composition Root, since you don't have full control over the lifecycle of your objects (you can't inject services into your form constructors). The same is true of ASP.Net Webforms. So you may have to use property injection for your factory and other services needed in your form, or just resolve the factory directly via calling a static instance of the container (container.Resolve()). This is imperfect, and goes against the idea of having a Composition Root, and more toward service location.
You might google keywords "Unity WinForms" and/or "[OtherDIFramework] Winforms" to get some ideas of how to structure your code.
I'm quite new to Symfony 2 and I'm moving to advanced topics like services. When should an object be a service?
For example, say that you have a facade object for making a call to a REST service. This class needs a username and password. Would it be correct modeling that class as a global service? Even if it's used only in a portion of the whole project?
# app/config/config.yml
parameters:
my_proxy.username: username
my_proxy.password: password
services:
my_proxy:
class: Acme\TestBundle\MyProxy
arguments: [%my_proxy.username%, %my_proxy.password%]
Definition taken from the Symfony2 glossary:
A Service is a generic term for any PHP object that performs a specific task. A service is usually used "globally", such as a database connection object or an object that delivers email messages. In Symfony2, services are often configured and retrieved from the service container. An application that has many decoupled services is said to follow a service-oriented architecture.
I think your example is a perfect candidate for a service.
You don't want to copy construction code to all places you need your API client. It's better to delegate this task to the dependency injection container.
This way it's easier to maintain (as construction happens in one place and it's configurable).
It's also more flexible as you can easily change the API client class without affecting code which uses it (as long as it implements the same interface).
I don't think there's a golden rule. But basically all classes implementing a task are good candidates for a service. Entities on the other hand are not as they're most often just data holders.
I always recommend Fabien's series of articles on the subject: http://fabien.potencier.org/article/11/what-is-dependency-injection
Yes, because this will spare you the configuration part. You're not going to fetch the username and password and give it to the constructor each time you need this class.
Ok, so the problem is:
I've got some 'order' entity, and it has 'status' property. On changing status, i wanted some other objects to be informed of this event, so i've decided to use Observer pattern. One of the observers notifies clients via email. Now i want to render Email text's from some of the twig templates. As i get from the Book, rendering templates in controllers are done with 'templating' service.
So the question as it follows: How can i access 'templating' service in my Observer class?
Specification:
I was advised, to implement my Observer as a service, but i'm not sure 'bout that. I've tried to solve this problem, and here is my options:
Use Registry. Solution that is straight and hard as rail. I guess it misses the whole point of DI and Service Container. Huge plus of this solution, is that i can access all common services from any point of my application.
To pass needed services from the context via constructor, or via setters. This is more like in Sf2 spirit. There comes another list of problems, which are not related to this question field.
Use observers as a service. I'm not really sure 'bout this option 'cos, in the book it is written, that service is a common functionality, and i don't think that observing entity with number of discrete properties is a common task.
I'm looking for a Sf2 spirit solution, which will be spread over whole project, so all answers with an explanation are appreciated.
As with any other service in a Symfony2 project, you can access it from within other classes through the dependency injector container. Basically what you would do is register your observer class as a service, and then inject the templating service into your observer service. See the docs for injecting services.
If you're not familiar with how Symfony handles dependency injection, I'd suggest reading that entire chapter of the documentation - it's very helpful. Also, if you want to find all the services that are registered for application, you can use the console command container:debug. You can also append a service name after that to see detailed info about the service.
Edit
I read your changes to the question, but still recommend going down the DI route. That is the Symfony2 spirit :) You're worried that your observer isn't common enough to be used as a service, but there's no hard rule saying "You must use this piece of code in X locations in order for it to be 'common'".
Using the DIC comes with another huge benefit - it handles other dependencies for you. Let's say the templating service has 3 services injected into itself. When using the DIC, you don't need to worry about the templating service's dependencies - they are handled for you. All you care about is telling it "inject the templating service into this other service" and Symfony takes care of all the heavy lifting.
If you're really opposed to defining your observer as a service, you can use constructor or setter injection as long as you're within a container-aware context.
I've read through this article trying to understand why you want a session bean in between the client and entity bean. Is it because by letting the client access entity bean directly you would let the client know exactly all about the database?
So by having middleman (the session bean) you would only let the client know part of the database by implementing the business logic in some certain way. So only part of the database which is relevant to the client is only visible. Possibly also increase the security.
Is the above statement true?
Avoiding tight coupling between the client & the business objects, increasing manageability.
Reducing fine-grained method invocations, leads to minimize method invocation calls over the network, providing coarse-grained access to clients.
Can have centralized security & transaction constraints.
Greater flexibility & ability to cope with changes.
Exposing only required & providing simpler interface to the clients, hiding the underlying complexity and inner details, interdependencies between business components.
The article you cite is COMPLETELY out of date. Check the date, it's from 2002.
There is no such thing anymore as an entity bean in EJB (they are currently retained for backwards compatibility, but are on the verge of being purged completely). Entity beans where awkward things; a model object (e.g. Person) that lives completely in the container and where access to every property of it (e.g. getName, getAge) required a remote container call.
In this time and age, we have JPA entities that are POJOs and contain only data. Don't confuse a JPA entity with this ancient EJB entity bean. They sound similar but are completely different things. JPA entities can be safely send to a (remote) client. If you are really concerned that the names used in your entity reveal your DB structure, you could use XML mapping files instead of annotations and use completely different names.
That said, session beans can still perfectly be used to implement the Facade pattern if that's needed. This pattern is indeed used to give clients a simplified and often restricted view of your system. It's just that the idea of using session beans as a Facade for entity beans is completely outdated.
It is to simplify the work of the client. The Facade presents a simple interface and hides the complexity of the model from the client. It also makes it possible for the model to change without affecting the client, as long as the facade does not change its interface.
It decouples application logic with the business logic.
So the actual data structures and implementation can change without breaking existing code utilizing the APIs.
Of course it hides the data structure from "unknown" applications if you expose your beans to external networks