kong-incubator 0.4.3

Async Kong client and cli (forked from aio-kong)

Async Python Client for Kong (exp fork of aio-kong)

Forked from aio-kong by Luca Sbardella.

For managing Kong using declarative configuration (yaml), over manual bookeeping of curls and/or resource state handling bash scripts.

Changes to the original:

• Patch tests to pass with Kong 0.14.x and 1.0.0rc2
• Added docker-compose stack for tests
• Added make rules for flake8, test, build, publish_pypi, etc.
• Added separate .venvs for dev and release, handled by make rules
• Added (opinionated) pytest plugins for dev venv, to help myself
• Added --key-only to output only the consumer key, instead of whole JSON
• Added ./examples for Kong Admin API loopback and an example service via it
• Added KADMIN_APIKEY for using Kong Admin API via the loopback and key-auth

TODO:

• PR Kong 0.14.x to original if can make it compatible with 0.13
• Maybe add --init-loopback for creating the Kong Admin API loopback service

Installation

On Python >= 3.6:

pip install --upgrade kong-incubator

Usage

CLI

By default, target KONG_URL is http://127.0.0.1:8001.

Create or update the Kong resources according to configuration:

kong-incubator --yaml ./examples/mockbin.yaml

Create a key-auth for the consumer (if not already created) and output the key:

kong-incubator --key-auth mocker --key-only

Run kong-incubator for the list of options.

Python

from kong.client import Kong

async with Kong() as cli:
    services = await cli.services.get_list()
    print(json.dumps(services, indent=4))

Development

To create the docker-compose stack and run tests for it:

make test

Tests clean up the Kong resources they create. The docker volume for DB persists until make testdown or make clean is ran.

Run make for the list of rules.