Skip to content

Latest commit

 

History

History
145 lines (114 loc) · 5.37 KB

README.md

File metadata and controls

145 lines (114 loc) · 5.37 KB
Raw

PyPI PyPI - Python Version Code style: black

slats

slats saves local albums to Spotify.

Usage

The first thing you need to do is write (preferably in some automated way) a JSON file of albums you want to save to your Spotify account. For example,

[
  { "album_artist": "Biosphere", "album": "Seti Project" },
  { "album_artist": "Biosphere", "album": "Shenzhou" },
  { "album_artist": "Biosphere", "album": "Substrata" },
  { "album_artist": "Biosphere", "album": "Dropsonde" },
  {
    "album_artist": "Darcy James Argue's Secret Society",
    "album": "Infernal Machines"
  },
  { "album_artist": "Haruka Nakamura", "album": "Grace" },
  { "album_artist": "Haruka Nakamura", "album": "Twilight" },
  { "album_artist": "Jens Lekman", "album": "I Know What Love Isn't" },
  { "album_artist": "Jens Lekman", "album": "Night Falls Over Kortedala" }
]

Note that the JSON you provide needs to respect the following schema:

{
  "type": "array",
  "items": {
    "type": "object",
    "properties": {
      "album_artist": { "type": "string" },
      "album": { "type": "string" }
    }
  }
}

Once you have the JSON file ready, run slats with

slats --albums-json my_json_file.json

and slats will open up a Spotify authentication page in your favourite browser where you then need to give the okay to let slats add albums to your Spotify account. Once you've authenticated, slats will ask you to input the URL you were redirected to after authenticating the app in your browser.

slats will then attempt to find a corresponding album on Spotify's servers for each album in the JSON you provided, and if it can find the corresponding album on Spotify and if you don't already have that album saved, it will save the album to your Spotify account. slats will give you copious coloured output so you can see exactly what's going on.

What's the catch

Misfires

slats searches up albums by providing the album artist and album names that you provide to slats as JSON to the Spotify API. If the Spotify API returns a non-empty response, it will save album from the top hit of the Spotify response to your account (provided the album already isn't saved). This can occasionally backfire with obscure albums, where the response from the Spotify API might provide a different album than intended and slats would then save that album to your account. Again, slats will give you tons of output so you can tell what it's doing, so monitor its output as necessary to find out when misfires have occurred.

There might be something fancy I can do to protect against this—if you have any ideas let me know.

Spotify song limits

Spotify restricts the number of songs you can have saved to your account to 10,000 songs (at least as of the date of this writing, 2019-05-02). This means that if you want to import a monolithic music library to your Spotify account, then unfortunately slats will only end up saving a fraction of that library to your account. What's worse is that the Spotify API doesn't return any reasonable error for when this happens. When I imported my library, getting a 502 error was a pretty consistent indication that I was at my song limit. So if your slats run fails with 502s, you may have reached your Spotify account's song limit.

If you're worried about reaching Spotify's song limit, one good way of addressing this is to break up the albums you want to import into chunks and run slats on each of those chunks separately.

Spotify API rate limits

Spotify isn't transparent about its API's rate limits (number of requests per unit time). Because I don't know what the rate limit is, I haven't protected against it in the code. So if your run fails due to a rate limit error, wait a bit, and try again (and maybe remove the albums that have already been processed in your JSON file so that it uses less requests on the next iteration).

Installation

You can either install slats with pip using

sudo pip3 install slats

or just run it directly from source with the run_slats.py script.

Configuration

slats looks for a configuration file at two paths:

  1. $PROJECT_ROOT/config.yaml
  2. $XDG_CONFIG_HOME/slats/config.yaml

where $PROJECT_ROOT is the base of the slats project (which you generally only want to use if you're running from source), and $XDG_CONFIG_HOME defaults to $HOME/.config, if you don't have it defined.

To get started, copy the example configuration file config.yaml.example to one of the above locations (making sure to rename it to config.yaml).

The next thing you need to do is get your Spotify username and also register a Spotify app (which is necessary for slats to work). You can find your Spotify username in the account page on Spotify's main site; and see this page for details on registering a Spotify app.

In the config file you copied, fill in spotify-username with your Spotify username, spotify-client-id and spotify-client-secret with the values you get after registering your app, and set spotify-redirect-uri to any redirect URI you have whitelisted with your app.