This Spotify Flask API is pretty simple: 1) This repository takes data from the Spotipy API provided from Spotify, 2) utilizes data requested from Spotify via an email request, 3) transports these tables into an AWS RDS database, and 4) exposes database endpoints using Flask and Python. Thus, it provides a template for how you in you project could use Spotify data to analyze your own music tastes!
A tour through this repository:
A) The folder RDSPipeline includes a pre-mde API which connects Python to AWS RDS from existing CSVs found in data and from API calls using Spotipy B) The folder called "data" includes CSVs provided by Spotify via email. C) The app folder includes an init.py file which includes database connections and defines routes for the API, exposing endpoints from AWS. The schema of the database is very simple, since it only includes 5 tables, all in one schema. The API works via localhost, using a simple webrequest:
curl -Uri "http://localhost:5000/"
In sum, this project is relatively simple, but it includes a variety of components: this project combines previous knowledge of data engineering with API design, backend engineering. It also incorporate knowledge of Flask, use of existing APIs, and Github Secrets.