envarify 1.2.0

Environment variables parsing and validation using python type hints

Envarify

Environment variables parsing and validation using python type hints

Usage

Having some environment variables:

export TIMEOUT_S=2.5
export API_KEY=some_key
export ALLOWED_IDS=1,2,3
export ENABLE_FEATURE=true

We can create a config object in Python:

from envarify import BaseConfig, EnvVar, SecretString

class MyConfig(BaseConfig):
    timeout_s: float = EnvVar("TIMEOUT_S")
    api_key: SecretString = EnvVar("API_KEY")
    allowed_ids: set[int] = EnvVar("ALLOWED_IDS")
    enable_feature: bool = EnvVar("ENABLE_FEATURE", default=False)
    optional_arg: str | None = EnvVar("OPTIONAL_ARG", default=None)

config = MyConfig.fromenv()
print(config)
#> MyConfig(timeout_s=2.5, api_key='******', allowed_ids={1,2,3}, enable_feature=True, optional_arg=None)

Missing environment variables

If there are some required environment variables that are not set, error will be raised:

config = MyConfig.fromenv()
#> MissingEnvVarsError: TIMEOUT_S, API_KEY, ALLOWED_IDS

Testing

In tests for your application you don't have to worry about mocking the environment variables. Instead just create a mock config object:

mock_config = MyConfig(timeout_s=4.2, api_key="dummy", allowed_ids={1,2,3}, enable_feature=True)

Supported Types

• Primitive types

  • int
  • float
  • bool
  • str

• Dictionary

  dict type reads environmental variable as JSON

• Sequences (delimiter separated values)

  • list[T]

  • set[T]

  • tuple[T]

    where T is any primitive type

• Special types

  • SecretString

    Masks sensitive environment variables by displaying ****** when printed or logged. The actual value is accessible via the reveal() method, and memory is cleared when object is no longer needed.

    from envarify import BaseConfig, EnvVar, SecretString
    
    class MyConfig(BaseConfig):
        api_key: SecretString = EnvVar("API_KEY")
    
    config = MyConfig.fromenv()
    print(config.api_key)
    
    #> MyConfig(api_key='"******"')
    

• BaseConfig subtype itself

  With environment variables COMPONENT_TIMEOUT=5 and OTHER=dummy you can do:

  from envarify import BaseConfig, EnvVar
  
  class ComponentConfig(BaseConfig):
      timeout: int = EnvVar("COMPONENT_TIMEOUT")
  
  class ApplicationConfig(BaseConfig):
      component: ComponentConfig
      other: str = EnvVar("OTHER")
  
  config = ApplicationConfig.fromenv()
  print(config)
  #> ApplicationConfig(component=ComponentConfig(timeout=5), other='dummy')