A tiny kit of command line tools for getting tempo data for a Spotify playlist.
Python 3.8+ Poetry A suitable shell environment (Bash/Fish/Zsh/Powershell etc) A Spotify developer account (optional) NodeJS 16+ for running the test server
You'll need a Spotify API key and secret. Currently, passing these in at CLI level is not supported so create environment variables in your local environment as most suitable with the following names:
SPOTIFY_KEY=xxxxxxxx
SPOTIFY_SECRET=yyyyyyy
Then you'll need to integrate with your system however you like. Building a wheel and then symlinking into userspace is probably the best way. Once in place, the tool exposes 3 basic functions:
playlist_tempo tempo calculate_tempo
This takes a Spotify playlist ID and uses it to calculate the average BPM of the contained tracks
playlist_tempo tempo get-all-tempos
This takes a Spotify playlist ID and returns all the discrete BPMs of the contained tracks as a list
playlist_tempo tempo plot-tempos
This plots the BPMs of a Spotify playlist on a chart
All commands take a single argument -p or --playlist which is the ID string of a Spotify playlist (you can extract this from the client or web application)
This is a Poetry application so to set the application up for development, pull down the source and run
poetry install
followed by
poetry shell
You can invoke the individual application top level exposed commands via the app.py file directly.
Modify the config/env.py file to switch between local and production mode. In local mode, the application can load data from a local serI usually do this using Node json-server - running this command in the fakes folder will stand up a sample Spotify API response on port 3000:
cd ./fakes && json-server ./data.json
The application contains python-test-watcher; there are several test fixtures and the application will automatically build mocks of the Spotify API response from the fakes folder. Running ptw will invoke the test suite