After getting everything downloaded from Google Takeout, I unzipped everything and figured out what we have from the exported playlists.
In the unzipped Takeout folder, there is an archive_browser.html file that lists all of the stuff that was downloaded. To start, I’ll just get my Thumbs Up playlist (which had a nice property in Google Play Music where you could have all of your liked tracks downloaded to your device automatically - I don’t see that in Spotify).
The directory for the playlist has a whole bunch of small csv files. The each look like this:
Title,Album,Artist,Duration (ms),Rating,Play Count,Removed,Playlist Index
"1901","Wolfgang Amadeus Phoenix","Phoenix","193000","5","41","","123"
So, the first step getting the track id for each song. The search in the Spotify API looks like the ticket for that. I’m using Python to create the migration and Spotipy looks like the ticket for working with the API.
Since I don’t need to make this too sophisticated, I’ll create a simple command line tool (using click) that can run from a directory, read each exported csv file and add it to the playlist.
Before doing any of that, I cleaned up the exported files with detox to get rid of spaces and make them generally easier to work with (and look at).
The code is in Github. The README outlines how to use the project. It’s primitive but gets the job done.