Skip to main content

betterconfig -- A Better Config for Python

Project description

betterconfig provides a more convenient and extensible configuration language, and a simpler interface built on python’s builtin ConfigParser format, along with a drastically simplified interface for loading configs.

Type Coercion Boilerplate Sucks

ConfigParser, like many config languages, treats all values as strings, meaning that when you have configs like this:

[config]
foo = 1
bar = a,list,of,strings
baz = just a plain old string

you end up with boilerplate that looks like this:

from ConfigParser import ConfigParser
MAP = {
    'foo': int,
    'bar': lambda x: x.split(','),
    'baz': str,
}
c = ConfigParser()
c.read('./foo.cfg')
config = {k:MAP[k](v) for k,v in c.items('config')}

Don’t you really wish you could just do this:

[config]
foo = 1
bar = ['a', 'list', 'of', 'strings']
baz = "just a plain old string"

and drop the map?

import betterconfig
config = betterconfig.load('./foo.cfg')['config']

betterconfig supports all literal types supported by ast.literal_eval: strings, numbers, tuples, lists, dicts, booleans, and None.

More Flexibility in Config, Less Config by Module

We wanted a config language that was as easy to use as a settings module in django or flask (and nearly as extensible), but less magical to initialize, and slightly safer than something like this:

import importlib
settings = importlib.import_module('settings')

So we want a config that can do stuff like this:

top_level  = 'variables defined outside of sections'
include    = ['./include.cfg', 'include.d/*.cfg']

[section]
namespaced = True

And we don’t want to have to iterate sections or items, we really just want to load it into a dict:

import betterconfig
settings = betterconfig.load('./fancy.cfg')

And if you’re really in love with . notation, you can always do something silly like make a module settings.py that does something magical like:

import betterconfig
globals().update(betterconfig.load('./fancy.cfg'))

Installing

The betterconfig project lives on github, and is available via pip.

Installing v0.2 From Pip

# 0.2 is not yet available via pip
sudo pip install betterconfig==0.2

Installing v0.2 From Source

curl https://github.com/axialmarket/betterconfig/archive/version_0.2.tar.gz | tar vzxf -
cd betterconfig
sudo python setup.py install

Running Tests

The betterconfig tests require py.test, which can be installed via pip.

sudo pip install pytest==2.5.2
py.test test_betterconfig.py

Authors and Contributors

Project details


Download files

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

Source Distribution

betterconfig-0.2.tar.gz (5.9 kB view details)

Uploaded Source

File details

Details for the file betterconfig-0.2.tar.gz.

File metadata

  • Download URL: betterconfig-0.2.tar.gz
  • Upload date:
  • Size: 5.9 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No

File hashes

Hashes for betterconfig-0.2.tar.gz
Algorithm Hash digest
SHA256 ad18ad3df6da2440784827aedd2f95a6fa8356ca408b1e802275bfcb2416ad8d
MD5 e189fcdb1011cba10e3d4e4b51fef94e
BLAKE2b-256 06814481c4d5e231ea3ba2bc14c8f99b11524c13e63cc5e4ba895b4997d47fea

See more details on using hashes here.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page