Everett is a Python configuration library for your app.

Goals of Everett:

1. flexible configuration from multiple configured environments
2. easy testing with configuration
3. easy automated documentation of configuration for users

From that, Everett has the following features:

• is flexible for your configuration environment needs and supports process environment, env files, dicts, INI files, YAML files, and writing your own configuration environments
• facilitates helpful error messages for users trying to configure your software
• has a Sphinx extension for documenting configuration including autocomponentconfig and automoduleconfig directives for automatically generating configuration documentation
• facilitates testing of configuration values
• supports parsing values of a variety of types like bool, int, lists of things, classes, and others and lets you write your own parsers
• supports key namespaces
• supports component architectures
• works with whatever you're writing--command line tools, web sites, system daemons, etc

Everett is inspired by python-decouple and configman.

Run:

$ pip install everett

Some configuration environments require additional dependencies:

# For INI support
$ pip install 'everett[ini]'

# for YAML support
$ pip install 'everett[yaml]'

Example:

# myserver.py

"""
Minimal example showing how to use configuration for a web app.
"""

from everett.manager import ConfigManager

config = ConfigManager.basic_config(
    doc="Check https://example.com/configuration for documentation."
)

host = config("host", default="localhost")
port = config("port", default="8000", parser=int)
debug_mode = config(
    "debug",
    default="False",
    parser=bool,
    doc="Set to True for debugmode; False for regular mode",
)

print(f"host: {host}")
print(f"port: {port}")
print(f"debug_mode: {debug_mode}")

Then you can run it:

$ python myserver.py
host: localhost
port: 8000
debug_mode: False

You can set environment variables to affect configuration:

$ PORT=7000 python myserver.py
host: localhost
port: 7000
debug_mode: False

It checks a .env file in the current directory:

$ echo "HOST=127.0.0.1" > .env
$ python myserver.py
host: 127.0.0.1
port: 8000
debug_mode: False

It spits out useful error information if configuration is wrong:

$ DEBUG=foo python myserver.py
<traceback>
everett.InvalidValueError: ValueError: 'foo' is not a valid bool value
DEBUG requires a value parseable by everett.manager.parse_bool
DEBUG docs: Set to True for debugmode; False for regular mode
Project docs: Check https://example.com/configuration for documentation.

You can test your code using config_override in your tests to test various configuration values:

# testdebug.py

"""
Minimal example showing how to override configuration values when testing.
"""

import unittest

from everett.manager import ConfigManager, config_override


class App:
    def __init__(self):
        config = ConfigManager.basic_config()
        self.debug = config("debug", default="False", parser=bool)


class TestDebug(unittest.TestCase):
    def test_debug_on(self):
        with config_override(DEBUG="on"):
            app = App()
            self.assertTrue(app.debug)

    def test_debug_off(self):
        with config_override(DEBUG="off"):
            app = App()
            self.assertFalse(app.debug)


if __name__ == "__main__":
    unittest.main()

Run that:

$ python testdebug.py
..
----------------------------------------------------------------------
Ran 2 tests in 0.000s

OK

That's perfectly fine for a 12-Factor app.

When you outgrow that or need different variations of it, you can switch to creating a ConfigManager instance that meets your needs.

Most other libraries I looked at had one or more of the following issues:

• were tied to a specific web app framework
• didn't allow you to specify configuration sources
• provided poor error messages when users configure things wrong
• had a global configuration object
• made it really hard to override specific configuration when writing tests
• had no facilities for autogenerating configuration documentation