golang-dependency-injection

Dependency injection in Go improves service testability by enforcing explicit constructor injection, eliminating hidden coupling from globals and init() so dependencies can be easily mocked and refactored.

The best way to structure Go constructors is through explicit constructor injection, passing dependencies as parameters rather than relying on globals, init() functions, or service-locator patterns.

Refactoring tightly coupled Go code to use dependency injection involves replacing globals and init() with constructor parameters, defining consumer interfaces, and establishing a composition-root container.

Choose manual wiring for small Go services, or select a dependency injection library like google/wire, uber-go/dig+fx, or samber/do for larger systems requiring automated wiring and service lifecycle management.

Define interfaces at the consumer boundary when implementing dependency injection in Go, ensuring implementations can vary without breaking consumers, which maintains loose coupling and enhances testability.

A common anti-pattern when using dependency injection containers in Go is passing the container itself into services, which recreates hidden coupling and defeats the purpose of explicit constructor injection.