A music scraper that understands your metadata needs.

irs is a command-line application that downloads audio and metadata in order to package an mp3 with both. Extensible, the user can download individual songs, entire albums, or playlists from Spotify.

• Usage
  • Demo
• Installation
  • Pre-built
  • From source
  • Set up
• Config
• How it works
• Contributing

~ $ irs -h

Usage: irs [--help] [--version] [--install]
           [-s <song> -a <artist>]
           [-A <album> -a <artist>]
           [-p <playlist> -a <username>]

Arguments:
    -h, --help                  Show this help message and exit
    -v, --version               Show the program version and exit
    -i, --install               Download binaries to config location
    -c, --config                Show config file location
    -a, --artist <artist>       Specify artist name for downloading
    -s, --song <song>           Specify song name to download
    -A, --album <album>         Specify the album name to download
    -p, --playlist <playlist>   Specify the playlist name to download
    -u, --url <url>             Specify the youtube url to download from (for single songs only)
    -g, --give-url              Specify the youtube url sources while downloading (for albums or playlists only only)

Examples:
    $ irs --song "Bohemian Rhapsody" --artist "Queen"
    # => downloads the song "Bohemian Rhapsody" by "Queen"
    $ irs --album "Demon Days" --artist "Gorillaz"
    # => downloads the album "Demon Days" by "Gorillaz"
    $ irs --playlist "a different drummer" --artist "prakkillian"
    # => downloads the playlist "a different drummer" by the user prakkillian

Just download the latest release for your platform here.

Note that the binaries right now have only been tested on WSL. They should run on most linux distros, and OS X, but if they don't please make an issue above.

If you're one of those cool people who compiles from source

1. Install crystal-lang (https://crystal-lang.org/install/)
2. Clone it (git clone https://github.com/cooperhammond/irs)
3. CD it (cd irs)
4. Build it (shards build)

1. Create a .yaml config file somewhere on your system (usually ~/.irs/)
2. Copy the following into it

   binary_directory: ~/.irs/bin
   music_directory: ~/Music
   filename_pattern: "{track_number} - {title}"
   directory_pattern: "{artist}/{album}"
   client_key: XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX
   client_secret: XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX
   single_folder_playlist:
       enabled: true
       retain_playlist_order: true
       unify_into_album: false

3. Set the environment variable IRS_CONFIG_LOCATION pointing to that file
4. Go to https://developer.spotify.com/dashboard/
5. Log in or create an account
6. Click CREATE A CLIENT ID
7. Enter all necessary info, true or false, continue
8. Find your client key and client secret
9. Copy each respectively into the X's in your config file
10. Run irs --install and answer the prompts!

You should be good to go! Run the file from your command line to get more help on usage or keep reading!

You may have noticed that there's a config file with more than a few options. Here's what they do:

binary_directory: ~/.irs/bin
music_directory: ~/Music
search_terms: "lyrics"
filename_pattern: "{track_number} - {title}"
directory_pattern: "{artist}/{album}"
client_key: XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX
client_secret: XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX
single_folder_playlist:
    enabled: true
    retain_playlist_order: true
    unify_into_album: false

• binary_directory: a path specifying where the downloaded binaries should be placed
• music_directory: a path specifying where downloaded mp3s should be placed.
• search_terms: additional search terms to plug into youtube, which can be potentially useful for not grabbing erroneous audio.
• filename_pattern: a pattern for the output filename of the mp3
• directory_pattern: a pattern for the folder structure your mp3s are saved in
• client_key: a client key from your spotify API application
• client_secret: a client secret key from your spotify API application
• single_folder_playlist/enabled: if set to true, all mp3s from a downloaded playlist will be placed in the same folder.
• single_folder_playlist/retain_playlist_order: if set to true, the track numbers of the mp3s of the playlist will be overwritten to correspond to their place in the playlist
• single_folder_playlist/unify_into_album: if set to true, will overwrite the album name and album image of the mp3 with the title of your playlist and the image for your playlist respectively

In a pattern following keywords will be replaced:

Beware OS-restrictions when naming your mp3s.

Pattern Examples:

music_directory: ~/Music
filename_pattern: "{track_number} - {title}"
directory_pattern: "{artist}/{album}"

Outputs: ~/Music/Queen/Stone Cold Classics/9 - Bohemian Rhapsody.mp3

music_directory: ~/Music
filename_pattern: "{artist} - {title}"
directory_pattern: ""

Outputs: ~/Music/Queen - Bohemian Rhapsody.mp3

music_directory: ~/Music
filename_pattern: "{track_number} of {total_tracks} - {title}"
directory_pattern: "{year}/{artist}/{album}"

Outputs: ~/Music/2006/Queen/Stone Cold Classics/9 of 14 - Bohemian Rhapsody.mp3

music_directory: ~/Music
filename_pattern: "{track_number}. {title}"
directory_pattern: "irs/{artist} - {album}"

Outputs: ~/Music/irs/Queen - Stone Cold Classics/9. Bohemian Rhapsody.mp3

At it's core irs downloads individual songs. It does this by interfacing with the Spotify API, grabbing metadata, and then searching Youtube for a video containing the song's audio. It will download the video using youtube-dl, extract the audio using ffmpeg, and then pack the audio and metadata together into an MP3.

From the core, it has been extended to download the index of albums and playlists through the spotify API, and then iteratively use the method above for downloading each song.

It used to be in python, but

1. I wasn't a fan of python's limited ability to distribute standalone binaries
2. It was a charlie foxtrot of code that I made when I was little and I wanted to refine it
3. crystal-lang made some promises and I was interested in seeing how well it did (verdict: if you're building high-level tools you want to run quickly and distribute, it's perfect)

Any and all contributions are welcome. If you think of a cool feature, send a PR or shoot me an email. If you think something could be implemented better, please shoot me an email. If you like what I'm doing here, pretty please shoot me an email.

1. Fork it (https://github.com/your-github-user/irs/fork)
2. Create your feature branch (git checkout -b my-new-feature)
3. Commit your changes (git commit -am 'Add some feature')
4. Push to the branch (git push origin my-new-feature)
5. Create a new Pull Request