Enfig - Environment Config

Lean, zero dependency environment parsing library based on annotations, inspired by pydantic.

Environmental variables are being parsed and casted automatically to the corresponding type when class variables are being accessed.

Examples:

import os
import enfig

# Just for demonstration, of course env vars should already be set outside the application.
os.environ["COLOR"] = "RED"
os.environ["TEMPERATURE"] = "50"
os.environ["DISABLE_AUTOTUNE"] = "y"


class Config(enfig.BaseConfig):
    COLOR: str
    TEMPERATURE: int
    DISABLE_AUTOTUNE: bool


assert Config.COLOR == "RED"
assert Config.TEMPERATURE == 50
assert Config.DISABLE_AUTOTUNE is True

The Config class casts the values automatically. Moreover, you can test if all mandatory variables have been set and have the correct type.

import os
import enfig

os.environ["COLOR"] = "RED"
os.environ["DISABLE_AUTOTUNE"] = "I can't sing but I pretend to be a singer"
os.environ["WEIGHT"] = "haha invalid"


class Config(enfig.BaseConfig):
    COLOR: str
    TEMPERATURE: int
    WEIGHT: float
    AMOUNT: int | None = None
    DISABLE_AUTOTUNE: bool | None = None


try:
    Config.validate()
except enfig.ValidationError as error:
    for attribute_error in error.errors:
        print(attribute_error)

Output:

DISABLE_AUTOTUNE: Invalid value, required type: `bool`
TEMPERATURE: Not set, required type: `int`
WEIGHT: Invalid value, required type: `float`