heracless 0.1.0

Yaml to Dataclass parser for Config files

heracless

Config Manager using yaml.

Description

heracless aims to make working with config files in python easy. It parses a config file into a dataclass and creates types as a python file with can be used for type hints. Generated types also make autocompletion a dreamy!

Usage

first create a config file in yaml format and put in your desired fields. to access the config information do like in following sample code.

from heracless.decorator import heracless
from config.types import Config

@heracless()
def main(cfg:Config) -> Any: 
"""
simply add a cfg parameter to any function at the first place
and access config attributes like you do any with any other object
"""
    some_sample_config_value = config.sample.some_value

its as simple as that!

Decorator arguments:

• cfg_path:Path yaml config file path -> Default ./config/config.yaml
• dump_dir:Path Path where types are going to be written to -> Default ./config/types.py
• dump_func:Callable control's the dumping behavior -> options: heracless.dump_in_file|heracless.dump_in_console| heracless.dump_dummy -> Default dump_in_file
• frozen:bool whether dataclass config objects are mutable or not -> Default True
• make_dir:bool make given paths or not if not existing -> Default True

installation

pip install heracless

Version

heracless 0.1
written in python 3.11

Future

add clitool
add config variants
add ray wrapper for hyperparameter optimization configs

Author

Felix Schelling
github: felixscode
Witten with <3 in Mexico