Libraries

List of dependency injection libraries for Python, sorted by stars:

• dependency_injector
• dry-python/dependencies
• antidote
• diay

Tests

	dependencies	antidote	diay	dependency_injector
Easy to use	✔️	✔️	✔️	❌
Good docs	✔️	✔️	❌	❌
Explicit container	✔️	❌	❌	✔️
Type annotation	❌	✔️	✔️	❌
Lazy annotation	❌	✔️	✔️	❌
Name detection	✔️	❌	❌	❌
Function as target	❌	✔️	✔️	✔️
Partial injection	✔️	✔️	✔️	✔️
Overload deps	✔️	✔️	✔️	❌
Deps for deps	✔️	✔️	✔️	✔️

Test description:

• Easy to use -- you don't have to know about some difficult conceptions and write much of code.
• Good docs -- documentation exists and contains many examples and API reference.
• Explicit container -- you define set of dependencies explicitly that helps you to find out which dependencies was used here without debugging.
• Type annotation -- library can detect right dependency by type in annotations.
• Lazy annotation -- library can work with lazy type annotations (defined as string). In Python 4.0 all annotations will be lazy.
• Name detection -- library can detect right dependency by variable name.
• Function as target -- you can inject dependencies into function.
• Partial injection -- you can mix up dependencies with not injected arguments.
• Overload deps -- you can pass your own dependencies to override dependencies in container.
• Deps for deps -- dependencies can be injected into other dependencies.

Usage

See tests dir and docs for libraries for usage examples.

Run tests

pip3 install --user -r requirements.txt
pytest --tb=short

Not tested yet

• serum
• haps
• picobox
• injectable