flexigurator 0.3.2

Python Configuration solution.

Flexigurator

Flexigurator is a collection of useful tools when working with configuration. Flexigurator builds on top of ConfZ and Pydantic while adding some extra functionality to make your life easier when testing and deploying.

Examples

config_patch

When using ConfZ to define a configuration singleton using a .yaml file,

# config.yaml
some_string: default
some_int: 42

from confz import ConfZ, ConfZFileSource

class Config(ConfZ):
    some_string: str
    some_int: int

    CONFIG_SOURCES = ConfZFileSource("config.yaml")

configuration variables can be easily hotswapped using patch_config:

from flexigurator import patch_config


data = dict(some_int=9001)


with patch_config(Config, data):
    Config().some_int     # 9001

    
Config().some_int     # 42

Importantly, the new data does not need to be complete.

placeholder

When having nested, optional BaseModels in your Config,

class Config(ConfZ):
    ui: UIConfig | None
    server: ServerConfig | None    

it would be nice if trying to load configuration automatically throws an exception if it is not configured. The placeholder method provides such functionality, and it can be used as follows:

from flexigurator import placeholder, patch_config


class Config(ConfZ):
    ui: UIConfig = placeholder(UIConfig)
    server: ServerConfig = placeholder(UIConfig)
    

Config().server.ip_address  # NotConfiguredException!

This removes the need for None-checking as exception handling is done behind the scenes.

ConfigForm

ConfigForm allows for the easy creation of forms for ConfZ or BaseModel classes. Forms are generated using Json Editor and are served using FastAPI. Having instantiated ConfigForm,

# form.py
from flexigurator import ConfigForm


app = ConfigForm(Config, "save/path", "templates/path")

The server can be started using (for instance) Uvicorn:

uvicorn form:app

Installation

Flexigurator is available on PyPi and can be installed using pip:

pip install flexigurator