Skip to content
Switch branches/tags
Go to file

Latest commit


Git stats


Failed to load latest commit information.
Latest commit message
Commit time

Configly Python Library

The Python library for Configly: the modern config/static data key/value store.


Table of Contents

What is Configly?

Configly is the place software developers put their static / config data—like copy, styling, and minor configuration values. They can then update that data directly from without having to wait for a deploy process / app store review. Their app or webapp receives the data near instantly. Non-technical folks themselves can publish changes freeing developers to focus on hard software problems and not copy tweaks.

On the backend, Configly provides a read-optimized static-data key/value store built with the aim of being low-latency, and high-availability. The client libraries are made to be dead-simple, lean, and efficient (via enhancements like caching). There is a fancy web UI called the Configulator for setting and updating the configs as well as seeing things like change history. Configly is built for modern software development.

There are a host of other benefits to using Configly ( such as ensuring you do not have data duplicated across clients, reducing load on your primary DB, and better tolerance for traffic spikes), read more about the benefits at Configly.

Core Features

  • API to fetch Strings, JSON Blobs (arrays and objects), Booleans, and Numbers from the Configly backend
  • Web interface for modifying these values without having to deploy code (we call our beloved web interface the Configulator).
  • High availability, high-throughput, low-latency backend.
  • Smart caching on the client libraries to minimize server requests.
  • Client libraries available in an expanding amount of languages.

Concepts/ Data Model

  • A Configly account contains a set of Configs.
  • A Config is a key-value pair along with associated metadata (like TTL).
  • The keys are strings.
  • The values are one of the following types:


Type notes Example(s)
string "I <3 Configly!"
number Can be integers or decimal; be aware some clients require you to specify which when fetching 31337, 1.618
boolean only true or false true, false
jsonBlob A JSON5 (more relaxed JSON) array or object. ["one", 5, true], {"text": "Buy now!", color: "#0F0"}
More jsonBlob examples

You can make arbitrarily complex JSON structures -- as long as the top level is an object or array. This is incredibly powerful as you can send a host of data with a single config:

A more complex array for a store inventory. Note that because we're using JSON5, quotes are optional for single words.

  "Simple T-shirt",
  "Basic hoodie",
    item: "Complex T-shirt",
    sizes: ['S', 'M', 'L'],
    price_us_cents: [1099, 1499, 1599],

And a more complex object showing how you can internationalize and set style:

  "welcome_message": {
    copy: {
      'en': 'Welcome!',
      'es': "¡Bienvenidos!",
    }, style: {
      color: '#0F0',
      fontWeight: '700',
  "buy_button" : {
    copy: {
      'en': 'Buy',
      'es': "Comprar",
    }, style: {
      backgroundColor: "#F00",
      border: "border-radius 10px",

Getting Started

In four easy steps!

1. Get your API Key

You'll need a Configly account. Registration is lightning quick—you can register via visiting

After signing up, you can grab your API Key from You'll need your API Key to setup the API below.

2. Create your first Config

From, create a new Config via the "Add" button: image

Consider creating a simple JSON Object or Array Config called greetings and give it the value of: ['hello', 'hola', '你好', 'नमस्ते']: should look like this:


Be sure to save via clicking 'Send to Clients'. Now, we'll write client code to fetch this key.

3. Install the client library

pip install configly-python

4. Fetch the Config

Add the following code in the execution path of a Python application. It could be a Django or Flask app or a script.

import configly
from configly import errors

configly.api_key = 'YOUR_API_KEY'

    greetings_list = configly.get('greetings')
    if greetings_list is None:
        print('Can\'t find key')
        for saying in greetings_list:
            print(f'A greeting: {saying}')
except errors.InvalidApiKeyError as error:
    print(f'Invalid API Key: {error.message}')
except errors.ConfiglyRequestError as error:
    print(f'Generic error: {error.message}')
except errors.ConfiglyConnectionError as error:
    print(f'Connection error: {error.message}')
    # print(error.original_error)

Try changing some values on to confirm that the client is getting the updates.

Congratulations you have Configly working end-to-end! Now, feel free to use Configly with all your projects!


The golden rule of Configly library use is: do NOT assign the result of a get() to a long-lived variable; in order to check for new values from the server, you must call get().

The package needs to be configured with your account's API key, which is available in the Configly Configulator

// This value is stored on the servers.
   has_sale: true,
   discount: 0.8,
   items: ['T Shirt', 'Hoodie', 'Ferrari'],
   price: [100, 250,  200000],

On the Python client:

# You can try this example out by setting the `CONFIGLY_API_KEY` environmental variable to our demo account: 'Dem0apiKEY'
    catalog = configly.get('store_catalog')

    # `get` returns None if the key is not found.
    if catalog is None:
        print('The key was not found')
        items = catalog['items']
        prices = catalog['prices']

        for index in range(0, len(items)):
            item = items[index]
            price = prices[index]
            if catalog['has_sale']:
                price = catalog['discount']*price
            print(f'{item}: {price:.2f} USD')
except Exception as error:
    # configly.errors.InvalidApiKeyError is thrown when there's an unknown API Key
    # configly.errors.ConfiglyRequestError is thrown on a timeout or failure to connection to the server
    # configly.errors.ConfiglyrequestError is thrown when the server responds with an error

API Reference


The module must be imported and configured prior to any call to get.

You can set global options directly on the module. API Key is a required value and the rest are optional.

import configly

configly.api_key = 'YOUR_API_KEY' # Required

# Global Defaults - these are optional and can be overriden per-request.

# Seconds to timeout on requests to the servers
configly.timeout = 3

# Disabling caching will cause the client to ignore server-set TTLs and make a
# a request to Configly's servers on every call to `get` - use this only if
# it is essential that your data is always fresh and you are not worried about
# latency/bandwidth usage.
configly.enable_cache = True

get(key, options?)

get() exposes the core function of the library; is to request values stored in Configly.

get() accepts a string as its first argument‐a key. Configly will fetch the corresponding value from the Configly servers (or look it up in the local library cache).

hero = configly.get('the_best_superhero')

Unknown keys

get returns None if the key is not recognized

hero = configly.get('the_best_superhero')
if hero is None:
  print('There is no best superhero!')


You can pass per-request values for timeout and enable_cache that will override any global settings set on module first load.

# In this example, it's crucial we have up-to-date data and we're willing wait
# for it (in the case that something goes wrong). Note these overrides are for
# _only_ this call to get()
options = { "timeout": 5, "enable_cache": False }
feature_flag = configly.get("feature1_enabled", options)

if feature_flag:
    print("Feature is enabled")
    print("Feature is disabled!")

Both options are optional.


get can throw any of several errors. The errors are in the configly.errors namespace:

Key Explanation
InvalidApiKeyError Configly's server returned a 401. This likely means the API Key supplied in init() is incorrect. You can see your API Key in the
ConfiglyConnectionError There was a problem communicating with the backend. This could be due to a network fault, bad internet connection, or timeout. You can see the original error in the original_error member. Try again later. If the problem persists let us know!
ConfiglyRequestError A miscellaneous error. Take a look at the request value inside the returned object to see the response from Configly's servers. This could indicate a problem with the library; if so, you can create a Github issue and we'll look into it.

Each error has a message variable with further details.

Example Error Handling
    copy = configly.get('en_copy')
    if copy is None:
        print("No copy was found")
except errors.InvalidApiKeyError as error:
    print(f"Invalid API Key: {error.message}")
except errors.ConfiglyRequestError as error:
    print(f"Generic error: {error.message}")
    # You can access the response via error.response
except errors.ConfiglyConnectionError as error:
    print(f"Connection error: {error.message}")
    # You can access the internal error via original_error


This repository is published under the MIT license.


No description, website, or topics provided.




No releases published


No packages published