Skip to content

An easy way to provide environ backed config in your projects.

Notifications You must be signed in to change notification settings

funkybob/confucius

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

16 Commits
 
 
 
 
 
 
 
 

Repository files navigation

Confucius

A simpler, clearer approach to configuration.

Quick Start

from confucius import BaseConfig

class Config(BaseConfig)
    HOST = '127.0.0.1'
    PORT : int = 8000

    DEBUG : bool = False
from myapp import Server
from config import Config


server = Server(Config.HOST, Config.PORT)
$ python app.py
- Starting server: 127.0.0.1:8000

$ PORT=80 python app.py
- Starting server: 127.0.0.1:80

$ DEBUG=y python app.py
- Starting debug server: 127.0.0.1:80

Types

Any ANGRY_SNAKE_CASE attributes of a BaseConfig sub-class will be intercepted by the meta-class, and checked for in the environment using os.getenv.

Their type will be determined by their annotation in the class, or fall back to str.

Methods will automatically behave like a property, with access to self.

Handling of type casting can be overridden [as it is for bool] by adding it to the __types__ dict:

class Config(BaseConfig):
    __types__ = {
        json: lambda v: json.loads(v) if isinstance(v, str) else v,
    }

    LOGGING : json = {'version': 1 ...}

All types defined on parent Config classes will be merged with this dict.

Inheritance

Classes, as usual, inherit from their parents. If the type of an attribute is changed, it will raise an AssertionError.

Methods

Method in all-caps will be invoked, and can access self as usual:

class Config(BaseConfig):
   DB_ENGINE = 'postgresql'
   DB_HOST = 'localhost'
   DB_PORT : int = 5432
   DB_USER = 'test_user'
   DB_PASS = 'secret'
   DB_NAME = 'test-db'

   def CONNECTION_STRING(self):
       return f'{self.DB_ENGINE}://{self.DB_USER}:{self.DB_PASS}@{self.DB_HOST}/{self.DB_NAME}'

Using in Django

In your settings.py, put your settings class (or classes), then use the following code to select one to use:

import os
MODE = os.getenv('DJANGO_MODE', 'Local')
globals().update(globals()[f'{ MODE.title() }Settings'].as_dict())

With Python 3.7

In Python 3.7, a new feature was added which allowed you to define __getattr__ for a module (See PEP 562).

The BaseConfig metaclass provides a module_getattr_factory factory method to provide a __getattr__ that will look up the Config object.

from confucius import BaseConfig

class Config(BaseConfig):
    DB_HOST = 'localhost'
    DB_PORT : int = 5432

__getattr__ = Config.module_getattr_factory()

After importing this module, attempts to access attributes will resolve normally and, if they're not found, call __getattr__, just like on an object.

About

An easy way to provide environ backed config in your projects.

Resources

Stars

Watchers

Forks

Packages

No packages published

Languages