Skip to main content

Decorator library to expose keyword arguments through config files

Project description


Decorator library to expose keyword arguments through config files

Configfy is a prototyping library that enables fast exposure of function parameters to the user in order modify program behavior without having to handle argument parsing or control the flow of config options through your program. It is implemented as a decorator library that makes use of the python configparser module.

It can be used to quick and dirty add new arguments to an function, or substitute the complete configuration management for an application.

Hostet on github


Configfy is best installed using pip

    pip install configfy

Simple example

Define a function that makes use of keyword arguments, and overwrites the default keyword argument with a config file setting.

import configfy
from configfy import configfy as cfy 

def hello(name, another_name='Pedro'):
    """Be nice and say hello 
    print(f'Hello {name}, I am {another_name}!')


With a config.ini containing

another_name = 'Suzan Flusan'


Hello Bob, I am Suzan Flusan!

Note: Be aware that the library expects the presence of a config file (./configfy.ini) containing at least the empty section [global].

Advanced Example

Show how to specify sections to use in config files and how the config file can be changed during runtime, or specified in the decorator.

From docs/

import configfy
from configfy import configfy as cfy 

def hello(name, another_name='Pedro'):
    """Be nice and say hello 
    print('Hello %s, I am %s!' % (name, another_name))

def greetings(name, language='english'):
    """Give a nice greeting ...
    if language == 'english':
        print('Hello %s! How are you doing?' % (name))
    elif language == 'spanish':
        print('Hola %s! Que tal?' % (name))
    elif language == 'german':
        print('Hallo %s! Wie gehts?' % (name))
    elif language == 'serbian':
        print('Zdravo %s! Kako si?' % (name))
        print('!nuqneH %s!' % (name))

def goodby(msg='Goodby!'):

print('# Use default configfy.ini file (missing greetings section)...')
print('# Current config: %s' % configfy.configfile.config)

print('\n# Changing config to "another_config.ini" ...')
print('After setting new config file, current config: %s' % configfy.configfile.config)

print('\n# Specifying kwargs, overwriting config settings...')
hello('Bob', another_name='Alfredo')
greetings('Tom', language='serbian')
goodby(msg='That\'s all Folks!')


# Use default configfy.ini file (missing greetings section)...
# Current config: OrderedDict([('global', {'another_name': 'Suzan Flusan', 'language': 'spanish'})])
Hello Bob, I am Suzan Flusan!
# 2018-08-01 17:45:53,834 - configfy - WARNING - Config section greetings_section not found!
Hello Tom! How are you doing?

# Changing config to "another_config.ini" ...
After setting new config file, current config: OrderedDict([('global', {'language': 'spanish', 'msg': 'A goodby message! Ignored'}), ('greetings_section', {'language': 'german'})])
Hello Bob, I am Pedro!
Hallo Tom! Wie gehts?

# Specifying kwargs, overwriting config settings...
Hello Bob, I am Alfredo!
Zdravo Tom! Kako si?
That's all Folks!

Complete config replacement

This is a complete example making use of configfy as a general tool to replace the configuration file for an application. The source code can be found in ./doc

import click
import configfy
from configfy import configfy as cfy

# Two user functions that get their kwargs from the current config file
def user_function1(kw1=23, name1=''):
    print('Calling user_function1(%s, %s) ...' % (kw1, name1))

def user_function2(kw2=None, name2=''):
    print('Calling user_function2(%s, %s) ...' % (kw2, name2))

# Use the excellent click library to handle program arguments
@click.option('-v', '--verbose', count=True)
@click.option('--config', default='configfy.ini', help='Specify the configfy file to use')
def example(verbose, config):
    Enabling sub commands
    # Load config file

    global config_file
    config_file = config

def ex1():

def ex2():

if __name__ == '__main__':

config file:

# Pass a number and a list
kw1 = 1001
name1 = ['l', 'i', 's', 't']

# Pass a dict and a set
kw2 = {'d': 42}
# Note: passing set(1, 2, 3) wont work. It will be passed as a string
name2 = {1, 2, 3}


> python --config complete_example.ini ex1
Calling user_function1(1001, ['l', 'i', 's', 't']) ...

> python --config complete_example.ini ex2
Calling user_function2({'d': 42}, {1, 2, 3}) ...


Debugging functions augmented with configfy can be tricky because instead of stepping directly into a user function modified by configfy, the debugger will enter the configfy library code first.

What one can do, is make the debugger "skip" the configfy library code.


import pdb
from configfy import configfy as cfy    

def fuu(kw_me=42):

if __name__ == '__main__':
    # load pdb, making use of the skip parameter


import pudb
from configfy import configfy as cfy

# Prevent pudb from stepping into the decorator library code

def fuu(kw_me=42):

if __name__ == '__main__':


Try it out yourself! Run

cd doc

I got

Comparing performance ... this can take a while.
native: 11.863 sec
configfy+config 11.953 sec
configfy 12.160 sec


I happy to receive any feedback or comments on github or privately to

Why I wrote this module

I wrote this library because I often have to write prototypes in a scientific settings, in which it is unclear upfront what goals to achieve.

It often happened to me that I had to introduce additional parameters and options to functions deep down in the application flow, in order to make its behavior alterable by the user. One can do this by either some global config file, or parse program arguments and pass them through multiple layers to the desired functions.

In order to make those steps easier and speed things up, I wrote this library which enables one to post-hoc add keyword arguments to any function in the codebase, and expose them to the user via a simple config file.



Project details

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Filename, size & hash SHA256 hash help File type Python version Upload date
configfy-1.0.10-py3-none-any.whl (7.7 kB) Copy SHA256 hash SHA256 Wheel py3

Supported by

Elastic Elastic Search Pingdom Pingdom Monitoring Google Google BigQuery Sentry Sentry Error logging AWS AWS Cloud computing DataDog DataDog Monitoring Fastly Fastly CDN SignalFx SignalFx Supporter DigiCert DigiCert EV certificate StatusPage StatusPage Status page