bc-configs 0.4.1

Configuration module, that provides a way to manage configs structure, validation and casting to type inside your code. Based on pedantic models. Additionally it can automatically getting config from environment variables, .env, HashiCorp Vault, etc.

bc-config

Make configuring your application easier.

Installing

pip install bc-configs

Make your custom config class

import os
from bc_configs import BaseConfig, define

define()

class MyConfig(BaseConfig):
    some_int: int
    some_string: str
    some_bool: bool

my_config = MyConfig()  # type: ignore[call-arg]

assert int(os.getenv("MY_SOME_INT")) == my_config.some_int  # True
assert os.getenv("MY_SOME_STRING") == my_config.some_string  # True
assert bool(os.getenv("MY_SOME_BOOL")) == my_config.some_bool  # True

The name of the environment variable is formed based on the names of the class and field.

Features

• Load environment variables from .env files (using python-dotenv).
• Load environment variables from YAML files (e.g., .env.yml or custom path via YAML_CONFIG_FILE environment variable).
• Load secrets from HashiCorp Vault (using hvac).

Error handling

When a required configuration field is missing, the library raises a ValidationError with an enriched error message that includes the environment variable name:

ValidationError: 1 validation error for MyConfig
some_int
  Field required  →  env var: 'MY_SOME_INT' (description: Database port number) [type=missing, input_value={}, input_type=dict]

If the field has a description, it will be included in the error message:

class MyConfig(BaseConfig):
    some_int: int = Field(description="Database port number")

Error message:

ValidationError: 1 validation error for MyConfig
some_int
  Field required  →  env var: 'MY_SOME_INT' [type=missing, input_value={}, input_type=dict]