Flaskonf

Configuration Driven API

Installation

Under your python environment

pip install flaskonf

Create API with configuration json files

Suppose you have configuration files like following

!ls ../tests/confs/

la.json       shanghai.json

!cat ../tests/confs/*.json

{
    "name":"Los Angeles",
    "country":"United States",
    "sites":["universal studio", "griffith observatory", "union station"]
}
{
    "name":"Shanghai",
    "country":"China",
    "sites":["french concession", "xin tiandi", "disney resort"]
}

from flaskonf.api import FlaskonfAPI
from typing import Dict

app = FlaskonfAPI("AnConfAPI")

Bind the app to the configuration directory

Example directory is for test inputs example, you can just leave it empety for now

app.build_on_config(
    confs_dir="../tests/confs/",
    examples_dir="../tests/examples/" )

Create API blueprint

The syntax looks like creating API, but this is a blueprint based on what we will create an API for each json file

@app.conf_route("/guide/", nobuild=True)
def guide_api(inputs: Dict):
    data = inputs['data']
    conf = data["conf"]
    user = data["user"]
    return {"city_data": conf, "user": user}

app.build_flaskonf()

Notice, the configuration data is accessible in the decorated function

The rest is flask usual you are familiared with, if you run like following, you can visit localhost:6060/guide/la/ or localhost:6060/guide/shanghai/ to test your API

app.run("0.0.0.0", port = 6060)

 * Serving Flask app "AnConfAPI" (lazy loading)
 * Environment: production
   WARNING: This is a development server. Do not use it in a production deployment.
   Use a production WSGI server instead.
 * Debug mode: off


 * Running on http://0.0.0.0:6060/ (Press CTRL+C to quit)
INFO:werkzeug: * Running on http://0.0.0.0:6060/ (Press CTRL+C to quit)

When loading API components takes effort

eg. when you have to load huge AI model into memory when starting the server

Set nobuild=True

@app.conf_route("/guide2/", nobuild=False)
def build_city_guide(conf_file: str, conf: Dict):
    logging.info(f"{conf}")
    # doing other things for building API here
    # like load huge model into memory with configuration
    def guide_api(inputs: Dict):
        data = inputs['data']
        user = data["user"]
        return {"city_data": conf, "user": user}
    return guide_api