Scrapes the purpleair.com/json endpoint for a particular sensor and exposes the stats
via a Prometheus HTTP endpoint.
Calculates the instantaneous AQI from the PM 2.5 and PM 10.0 values. This is a little different from how PurpleAir or the EPA calculates the AQI, since those are calculated from a time-based average of PM values (10 minutes for PA, 1 hour for EPA). Doing an instantaneous calculation is much simpler, and yields a pretty close value to what a 10 minute PM average would yield assuming no giant jumps in PM (<20 ug/m^3).
Kudos to the EPA for publishing a doc on how AQI is calculated as well as the AQI level tables.
To use this script you'll need a sensor ID, which can be gotten from the PurpleAir map. Find your favorite sensor, click on it to bring up the sensor popup. Toward the bottom of the popup is a "Get This Widget" link, click it.
That will take you to a URL like https://www.purpleair.com/sensorlist?key=<key>&show=<id>.
The <id> is what you'll need to set as an environment variable to configure the
script.
This script is configurable via environment variables, which is handy when running as a Docker container.
Comma-separated list of IDs of the sensors to scrape, gotten from the PurpleAir /json URL for your sensor
This is the API token that you need to make requests to the API.
There's probably an official way to get one, but they also give you one when
you look at the map in a browser. If you open up Developer Tools and watch the
network requests, you'll see some requests that have a token query param.
That token is your API token. Don't worry about whether or not it's
URL-quoted, we'll take care of dealing with that.
Log level of the script, defaults to 'info', accepts any Python logging level name
The port to run the Prometheus HTTP server on, defaults to 9101