Clean Architecture & Design Patterns in Practice · レッスン

Dependency Ruleを解説

依存関係は内側にのみ向かうという基本的なDependency Ruleを理解し、コアのビジネスロジックを分離して保つ方法を学びます。

「Dependency Ruleを解説」はCoddyKit上の無料Clean Architecture & Design Patterns in Practiceレッスンです。これはレッスン3/4です。下記で完全なレッスンを無料で読むことができます。その後、ブラウザ内の組み込みコードエディタと24時間対応のAIチューターでハンズオン演習できます。これはClean Architecture & Design Patterns in Practice学習パスの一部であり、ウェブとCoddyKitアプリ全体で進捗が同期されます。 Clean Architecture & Design Patterns in Practiceコースには全4レッスンが含まれています。

このレッスンの一部はまだ翻訳されておらず、英語で表示されています。

The Core of Clean Architecture

Welcome to our lesson on the Dependency Rule! This principle is the cornerstone of Clean Architecture, ensuring your core business logic remains isolated and independent.

It's all about directing the flow of dependencies in your application in a very specific way.

Dependencies Flow Inwards

The fundamental idea is simple: dependencies can only flow inwards. This means outer layers of your architecture must depend on inner layers, but inner layers must never depend on outer layers.

Think of it as a one-way street, always leading towards the center of your application's logic.

Visualizing the Rule

Recall the concentric circles of Clean Architecture (Entities, Use Cases, Interface Adapters, Frameworks/Drivers). The Dependency Rule dictates that code in an outer circle can depend on code in an inner circle, but never the other way around.

Inner layers: Core business rules, independent.
Outer layers: UI, databases, web frameworks, external services.

Why This Rule Matters

Strictly following the Dependency Rule brings significant benefits:

Isolation: Your core business logic is protected from changes in UI, databases, or frameworks.
Testability: You can test your core logic without needing to set up a database or a web server.
Flexibility: You can swap out external components (e.g., change from SQL to NoSQL, or from one web framework to another) with minimal impact on your core.

Outer Layer Calling Inward

This is the natural and allowed flow. An outer layer, like a web controller, depends on an inner layer, like a Use Case, to perform an action. The controller knows about the use case.

Try running this example:

class CreateUserUseCase {
  public void execute(String username) {
    System.out.println("Logic to create user: " + username);
  }
}

// Outer layer (e.g., Web Controller)
class UserController {
  private final CreateUserUseCase createUserUseCase;

  public UserController(CreateUserUseCase useCase) {
    this.createUserUseCase = useCase;
  }

  public String register(String username) {
    createUserUseCase.execute(username);
    return "User registration initiated for " + username;
  }
}

public class Main {
  public static void main(String[] args) {
    CreateUserUseCase useCase = new CreateUserUseCase();
    UserController controller = new UserController(useCase);

    System.out.println(controller.register("Alice"));
  }
}

Inner Layer Independence

Now, what if our CreateUserUseCase (inner layer) needs to send a success message back to the UserController (outer layer)?

It cannot directly call a method on the UserController. That would mean the inner layer depends on the outer layer, breaking the Dependency Rule!

Inverting the Dependency

To solve this, we use Dependency Inversion. The inner layer (Use Case) defines an interface (often called a 'Port' or 'Output Port') that it expects to communicate through.

The outer layer (Controller/Presenter) then implements this interface and provides itself to the inner layer. This reverses the flow of dependency at compile time.

Code: Inverted Dependency

Observe how the CreateUserUseCase now depends only on its own interface, UserOutputPort. The UserController implements this port, allowing the Use Case to 'talk back' without knowing the specific UI details.

Try running this example:

// 1. Inner layer defines its 'output port' (interface)
interface UserOutputPort {
  void presentUserCreationResult(String message);
}

// 2. Inner layer (Use Case) depends on its own interface
class CreateUserUseCase {
  private final UserOutputPort outputPort;

  public CreateUserUseCase(UserOutputPort outputPort) {
    this.outputPort = outputPort;
  }

  public void execute(String username) {
    // Simulate user creation logic
    System.out.println("Processing creation for: " + username);
    outputPort.presentUserCreationResult("User '" + username + "' created!");
  }
}

// 3. Outer layer (e.g., Web Controller) implements the 'port'
class UserController implements UserOutputPort {
  private final CreateUserUseCase createUserUseCase;

  public UserController() {
    // Controller provides itself as the output port
    this.createUserUseCase = new CreateUserUseCase(this);
  }

  public void registerUser(String username) {
    createUserUseCase.execute(username);
  }

  @Override
  public void presentUserCreationResult(String message) {
    System.out.println("UI Update: " + message);
  }
}

public class Main {
  public static void main(String[] args) {
    UserController controller = new UserController();
    controller.registerUser("Bob");
  }
}

Consequences of Breaking the Rule

Violating the Dependency Rule leads to a fragile and rigid architecture:

Tight Coupling: Your core logic becomes tied to external details (UI, DB).
Difficult Testing: Testing business rules requires setting up external components.
Reduced Flexibility: Changing frameworks or databases becomes a massive undertaking.
Leaky Abstractions: Inner layers expose knowledge of outer layers, making the system harder to understand and maintain.

Quick Check

The Dependency Rule is crucial for maintaining a clean and flexible architecture. Let's test your understanding of how dependencies should flow.

Recap: The Dependency Rule

You've learned that the Dependency Rule is the core principle of Clean Architecture, dictating that dependencies must always flow inwards, from outer layers to inner layers.

It protects your core business logic.
It enhances testability and flexibility.
Dependency Inversion (using interfaces) is key to allowing inner layers to communicate outwards without direct dependencies.

Mastering this rule is essential for building robust and maintainable software systems!

よくある質問

「Dependency Ruleを解説」レッスンは無料ですか？

はい。「Dependency Ruleを解説」の完全なテキストはこのウェブで無料で読めます。インタラクティブに演習し（組み込みコードエディタと24時間対応のAIチューター）、Clean Architecture & Design Patterns in Practiceコースの残りをアンロックするには、CoddyKit PROにアップグレードしてください。 Clean Architecture & Design Patterns in Practiceコースには全4レッスンが含まれています。

「Dependency Ruleを解説」で何を学びますか？

依存関係は内側にのみ向かうという基本的なDependency Ruleを理解し、コアのビジネスロジックを分離して保つ方法を学びます。ブラウザで直接実行するハンズオンコードでClean Architecture & Design Patterns in Practiceを演習し、24時間対応のAIチューターがレッスンを進める中での質問に答えます。

Clean Architecture & Design Patterns in Practiceを始めるのに経験は必要ですか？

事前経験は必要ありません。CoddyKitのClean Architecture & Design Patterns in Practiceは初級者から上級者向けに構成されているため、ここから始めるか最初から始めて、自分のペースで進むことができます。これはレッスン3/4です。

「Dependency Ruleを解説」レッスンにはどのくらい時間がかかりますか？

ほとんどのCoddyKitレッスンは約5～10分かかります。各レッスンはコンパクトでインタラクティブなので、着実に進歩し、ウェブとアプリ全体で正確に前回の場所から再開できます。

このClean Architecture & Design Patterns in Practiceレッスンでコードを書いて実行できますか？

はい。すべてのClean Architecture & Design Patterns in Practiceレッスンに組み込みコードエディタが含まれているため、ブラウザでリアルコードを書いて実行し、即座のAIフィードバックを取得できます。ローカル設定は不要です。

このコースのすべてのレッスン

← Clean Architecture & Design Patterns in Practiceに戻る