scenario

imagine you’re trying to make a playlist you can reach out to when you need music to work to. you then realise that you’ve got 5, 10 or even more playlists, that are small, & serve a specific purpose, but all those small & specific playlists are also appropriate to listen to when working.

why that’s a problem

It’s tedious & distracting to have to continuously shift across the many, smaller playlists. would be more efficient to create a big aggregated playlist.

compose playlists using playlists.

what I’m trying to build

the service should be able to take the state of many smaller playlists & propagate any actions to the main aggregated playlist

• I’m thinking comparisons of playlists' state, will be done with data that’s been stored in a non-relational DB
• meaning if I add a track to smaller playlist 1 & a few tracks to smaller playlist 2, & those two playlists are being watched by big aggregated playlist1, those tracks should be added to the big aggregated playlist
• delete is a little tricky, because what happens if the big aggregated playlist already had some tracks, that aren’t in any of the smaller playlists. I want the aggregated playlist to also be standalone, one I can add tracks to & manage individually.

tech stack I’d like to use

• Go (have already started the project in Go, integration with Spotify API is done)
• AWS (this is the area I need help with)
• Spotify API
  • working with playlists

notes

• Spotify has no event sourcing API.
• Polling for changes
• Batch processing
• how is sourcing of watcher playlist & source playlists going to be done?