callrua/spotipyutils

Automate common Spotify interaction with Python

spotipyutils's Issues

Garbage being picked up for playlist name

Picking up the "Edit setlist" options from the HTML it seems;

Most likely the underlying HTML of setlist.fm has changed

Create a web server to serve requests

• Create a flask/django/whatever server in a docker image
• set up an endpoint for each feature

I got this working locally in flask, but not in a docker image. The main issue is the auth, specifically the redirect URI stuff. Basically, when you can't interact with the app live and input the response code, it doesn't work.

Need to figure out a different authentication method.

Broaden custom setlist creation options

We have a genre_to_playlist function which can convert a given list of genres to a playlist, but why stop there?

With a view to creating a website, we should create a flexible engine that takes a list of parameters to craft more specific playlists. These parameters should include the spotify track audio features listed here

We should also consider track details, for example the year or decade the track was made, the language the song is in etc etc.

Add feature to create a playlist of artist's top played songs

eg: https://www.setlist.fm/stats/tool-2bd6d836.html

Better CLI functionality

I'm working on better cli functionality so that instead of a bunch of isolated scripts with reused code (eg: auth) we will have one entry into the app with behaviour dictated by args.

So instead of:

./setlist_to_spotify.py --id --setlist

We will be able to do stuff like:

spotipyutils setlist_to_playlist --id --setlist
spotipyutils genre_to_playlist --id --genre

Should help with containerisation/running locally.

Port to python 3

With python 2 being deprecated, would be nice to have the utils ported to python 3.

Use setlist.fm API rather than parsing raw HTML?

I was digging around and found setlist.fm have an API - https://api.setlist.fm/docs/1.0/index.html

@mattmooree thoughts on refactoring to use this instead of bs4 for parsing the HTML?

My initial reaction for using it is "no" because it's another login/auth set to manage. I don't want users having to sign up for setlist.fm for example. (I don't even have a setlist.fm account and I've been using it for years!) Also seems overkill for our current usage, perhaps there's potential for other features by leveraging it though.