confyml 2.1.5

Small package to apply YAML config to methods and functions

confyml

Project to allow easy application of config to python methods and functions.

The package allows the user to provide keyword arguments to any decorated function/method. If keyword arguments are passed to the function/method directly in the code this will overwrite any configured behaviour. By having this, one can easily support configurability in applications without the need for code changes or unwieldy config classes that also require code change to alter.

install

Install like any other python package, from pypi

pip install confyml

usage

At the definition of classes, methods and functions, import confyml and apply the decorator to allow the function to become configurable.

# example_module.py
from confyml import confyml


config = confyml.get_config()


class ExampleClass:

    @config.apply
    def __init__(self, kwarg=None):
        self.kwarg = kwarg

    @config.apply
    def example_method(self, arg1, kwarg_2=None):
        return arg1, self.kwarg, kwarg_2


@config.apply()
def example_function(arg2, kwarg_3=None):
    return arg2, kwarg_3

Create a yaml file to hold the definition. Default mode for confyml is mcf: modules, classes and functions; where the yaml file has a structure like the one below.

# example.yaml
example_module:
  ExampleClass:
    kwarg: 'class_kwarg'
    example_method:
      kwarg_2: 'method_kwarg'
  example_function:
    kwarg_3: 'function_kwarg'

In the calling script, the application that uses the classes and functions, set the config file to follow.

# main.py
from confyml import confyml
import example_module

confyml.set_config('example.yaml')

e = example_module.ExampleClass()

print(e.example_method('arg1'))
print(example_module.example_function('arg2'))

$ python main.py
arg1 class_kwarg method_kwarg
arg2 function_kwarg

The use should be one config per application.

Keyword arguments provided to the function/method directly will overwrite the config file's behaviour.

# main.py
from confyml import confyml
import example_module

confyml.set_config('example.yaml')

e = example_module.ExampleClass(kwarg='direct_kwarg')

print(e.example_method('arg1'))
print(example_module.example_function('arg2', 'direct_func_kwarg'))

$ python main.py
arg1 direct_kwarg method_kwarg
arg2 direct_func_kwarg

modes

The config can also be provided with a mode argument,

# main.py
from confyml import confyml


confyml.set_config('<yaml_filepath>', mode='<mode>')

to set the level of the config. By default the mode is mcf, which supports a yaml structure of

module:
  Class:
    <defined_kwargs>
    method:
      <defined_kwargs>
  function:
    <defined_kwargs>

mode cf - classes and methods and/or functions

Class:
  <defined_kwargs>
  method:
    <defined_kwargs>
function:
  <defined_kwargs>

mode mf - modules and functions

module:
  function:
    <defined_kwargs>

and finally mode f - functions only

function:
  <defined_kwargs>

Use the mode that best suits your application.