Huge News!Announcing our $40M Series B led by Abstract Ventures.Learn More
Socket
Sign inDemoInstall
Socket

@travetto/di

Package Overview
Dependencies
Maintainers
1
Versions
313
Alerts
File Explorer

Advanced tools

Socket logo

Install Socket

Detect and block malicious and high-risk dependencies

Install

@travetto/di

Dependency registration/management and injection support.

  • 0.3.22
  • Source
  • npm
  • Socket score

Version published
Weekly downloads
63
decreased by-84.05%
Maintainers
1
Weekly downloads
 
Created
Source

travetto: Dependency Injection

Dependency injection is a framework primitive. When used in conjunction with automatic file scanning, it provides for handling of application dependency wiring. Due to the nature of typescript and type erasure of interfaces, dependency injection only supports classes as type signafiers. The primary goal of dependency injection is to allow for separation of concerns of object creation and it's usage.

Declaration

The @Injectable and @InjectableFactory decorators provide the registration of dependencies. Dependency declaration revolves around exposing classes and subtypes thereof to provide necessary functionality. Additionally, the framework will utilize dependencies to satisfy contracts with various backends (e.g. ModelMongoSource provides itself as an injectable candidate for ModelSource).

  @Injectable()
  class CustomService {
    async coolOperation() {
      ... work ...
    }
  }

When declaring a dependency, you can also provide a token to allow for multiple instances of the dependency to be defined. This can be used in many situations:

  @Injectable()
  class CustomService {
    async coolOperation() {
      ... work ...
    }
  }

  const CUSTOM2 = Symbol('custom2');

  @Injectable({ target: CustomService, symbol: CUSTOM2 })
  class CustomService2 extends CustomService {
    async coolOperation() {
      await super.coolOperation();
      // Do some additional work
    }
  }

As you can see, the target field is also set, which indicates to the dependency registration process what class the injectable is compatible with. Additionally, when using abstract classes, the parent class is always considered as a valid candidate type.

  abstract class BaseService {
    abstract work():Promise<void>;
  }

  @Injectable()
  class SpecificService extends BaseService {
    async work() {
      // Do some additional work
    }
  }

In this scenario, SpecificService is a valid candidate for BaseService due to the abstract inheritance. Sometimes, you may want to provide a slight variation to a dependency without extending a class. To this end, the @InjectableFactory decorator denotes a static class method that produces an @Injectable.

  class Config {
    @InjectableFactory()
    static initService(): CoolService {
      return new CoolService();
    }
  }

Given the static method initService, the function will be provided as a valid candidate for CoolService. Instead of calling the constructor of the type directly, this function will work as a factory for producing the injectable.

NOTE Due to the lack of typechecker in the Compiler for performance reasons, the return type on the factory method is mandatory. Without it, the code will not know what the expected target type should be.

NOTE Other modules are able to provide aliases to @Injectable that also provide additional functionality. For example, the @Config or the @Controller decorator registers the associated class as an injectable element.

Injection

Once all of your necessary dependencies are defined, now is the time to provide those @Injectable instances to your code. There are three primary methods for injection:

The @Inject decorator, which denotes a desire to inject a value directly. These will be set post construction.

  @Injectable()
  class CustomService {
    @Inject()
    private dependentService: DependentService;
    
    async coolOperation() {
      await this.dependentService.doWork();
    }
  }

The @Injectable constructor params, which will be provided as the instance is being constructed.

  @Injectable()
  class CustomService {
    constructor (private dependentService: DependentService) {}
    
    async coolOperation() {
      await this.dependentService.doWork();
    }
  }

Via @InjectableFactory params, which are comparable to constructor params

  class Config {
    @InjectableFactory()
    static initService(dependentService: DependentService): CustomService {
      return new CustomService(dependentService);
    }
  }

Manual Invocation

Some times you will need to lookup a dependency dynamically, or you want to control the injection process at a more granular level. To achieve that you will need to directly access the DependencyRegistry. The registry allows for requesting a dependency by class reference:

@Injectable()
class Complex {}

class ManualLookup {
  async invoke() {
    const complex = await DependencyRegistry.getInstance(Complex);
  }
}

Application

Given that dependency injection is generally a pre-requisite for application execution, it stands as the primary entrypoint for application invocation. The Base provides a simplistic bootstrap to allow for the application to run, but that is not sufficient for more complex applications.

The module provides a decorator, @Application who's job is to register entry points into the application. For example:

import { Application, Inject, Injectable } from '../';

@Injectable()
class Server {
  name = 'roger';

  async launch() {
    ...
  }
}

@Application('simple')
class SimpleApp {

  @Inject()
  server: Server

  async run() {
    return this.server.launch();
  }
}

Will expose an entrypoint named 'simple'. This additionally allows for other entry points to be named for handling specific use cases. An additional entry point could look like:

import { Application, Inject, Injectable } from '../';

@Injectable()
class Server {
  name = 'roger';
  config = { ... };

  async launch() {
    ...
  }
}

@Application('config')
class ConfigApp {

  @Inject()
  server: Server

  async run() {
    console.log(server.configuration);
  }
}

Keywords

FAQs

Package last updated on 14 Oct 2018

Did you know?

Socket

Socket for GitHub automatically highlights issues in each pull request and monitors the health of all your open source dependencies. Discover the contents of your packages and block harmful activity before you install or update your dependencies.

Install

Related posts

SocketSocket SOC 2 Logo

Product

  • Package Alerts
  • Integrations
  • Docs
  • Pricing
  • FAQ
  • Roadmap
  • Changelog

Packages

npm

Stay in touch

Get open source security insights delivered straight into your inbox.


  • Terms
  • Privacy
  • Security

Made with ⚡️ by Socket Inc