This is a simple command line tool to help with music-related tasks. This is not a music player. It does not provide a TUI or GUI, and it uses VLC internally. There are some command line tools available for music, such as ncmpcpp or cmus, but they do not allow you to launch music before you get into the TUI. Here are the main features:
- Querying music quickly and playing it (on vlc)
- Setting up a watch server to watch for playing music, and scrobbling it to lastfm
- Tagging system for playlists and grouping of songs (stored in
$MUSIC_PATH/tags) and having sync ability with spotify
This program does not support piracy; you should have the rights to all your files.
- VLC (if you plan on listening to music or scrobbling)
- playerctl (if you plan on scrobbling to lastfm with vlc)
Install either from the release page or with go:
go install github.com/kitesi/music@latest
This program works off a music directory which defaults to $MUSIC_PATH or $HOME/Music. Commands that consider a music path will have a --music-path option. I recommend a folder structure of:
Music/
Artist1/
x.mp3
y.m4a
Category/
z.mp3
But this is not necessary, as any file in your music path will be considered. Files should follow some basic file naming rules: no new lines, no crazy characters, etc.
You can play music with the play command which will take in
any amount of positional arguments, these are called terms.
A term can have a "!" prefix, meaning it's a negation term, and anything that
matches that term fails. If no term is provided, the program will spawn VLC
with the music directory and the option --recursive=expand. Otherwise, a song
will have to match at least one of the terms and none of the negation terms. A
term can have required sections and one-of sections, specified with "#" and ","
respectively. When querying, the string that's tested is the lowercase full
path to the file minus your music path.
For example, ~/Music/Jaxson/Make Time For Me.m4a would use
jaxson/make time for me.m4a.
Example of usage:
music play tonight monday#mornings care,bear,say make#you,me#believe \!joeThere are four terms here:
tonight(song has to have the word "tonight" somewhere in the path)monday#mornings(song has to have both words "monday" and "mornings" (not necessarily next to each other))care,bear,say(song has to have one or more of "care", "bear", "say")make#you,me#believe(song has to have "make", either "you" or "me", and "believe")\!joe(the song can't have the term, notice the escaping of "!")
When combining these terms, the string is split by # first, and then ,.
You can use music play --live to get a live query search of your songs.
I personally bind this command to a keybinding of Ctrl+Alt+m. I also have an
i3 keybinding to Meta+m+x so that it spawns a terminal with just the program.
bindsym $mod+m mode "music"
mode "music" {
bindsym x exec alacritty --class floating -o window.padding.x=10 -o window.padding.y=10 -e music play --live; mode "default"
# ...
}
Tags are a way to group music. You can use it for playlists, genres or
whatever. Tags will be stored in $MUSIC_PATH/tags/ as a m3u file.
You can view your tags with music tags. If you want to see the songs in a tag
use music tags <tag>.
If you want to delete a tag use --delete or -d or edit with --edit or -e.
The intended way to add songs to a tag is to query the songs with music play
and then using --add-to-tag | -a <tag> or --set-to-tag | -s <tag>. If you
need to query but don't want to actually play the songs you can add --dry-run.
While VLC does have built in lastfm scrobbling, I could not get it to work (edit: I actually got it to work, but it doesn't scrobble certain tracks and I kinda already built this so whatever). You can run a watch server to watch for playing songs every x seconds (defaulted to 10). It uses playerctl (MPRIS) under the hood, so you will need to have that installed and be on linux (in the future I could add vlc tcp support so that non-linux users could also use). It also only checks for the VLC player so other media players or something like youtube will not be logged. An alternative would be multi-scrobbler which supports a lot more sources, and has a lot more functionality overall.
Currently, the scrobble detection is kind of poor. It follows the approach of minimizing false positives, so if you skip/seek around, it likely won't scrobble. Also, if you play the same song over and over it won't scrobble more than once (although this will be fixed in the future).
Lastly, it follows the lastfm standards:
- The track must be longer than 30 seconds.
- The track has been played for at least half its duration, or for 4 minutes (whichever occurs earlier.)
To get started first make an api account and application on last.fm. You then need to create a file (".lastfm-credentials") in your cache directory. In that file you should store your api_key and your api_secret:
api_key=xxxxx
api_secret=yyyy
Then just run the lastfm watch command, and it will automatically get the session_key:
music lastfm watch --interval 20 --debug
I personally have this command start on startup, and I redirect the output to /tmp/music-lastfm.log. This program can also
write the logs to a local sqlite database file. This is helpful for if you want to keep a local copy of your scrobbles
and in case a valid scrobble fails (network issues or lastfm downtime). You can enable this with the --log-db-file flag.
music lastfm watch --log-db-file /home/user/.local/state/lastfm-scrobbles.db
If a scrobble fails, it will not be retried automatically or on the next run, but you can use the music lastfm import to import
all the failed scrobbles from the database file.
There are a few alternatives to this scrobbling functionality. In particular, one thing this program does not include that many others do, is some way to edit/filter artist/titles/albums before scrobbling. I personally don't need it, as I only use this program for local files, and if there is an error with my metadata I fix it at the source using atomicparsley. Similarly, I don't include any other MPRIS player other than VLC as that's the only one I use (and I'm not sure anyone else uses this). However, I understand that these are useful features for many people.
I have not tried any of the following, so I don't necessarily endorse them, but they are worth checking out:
- YodaEmbedding/scrobblez last update was 6 years ago, so actually, maybe not this one
- Ladbaby/lastfm-scrobbler
- InputUsername/rescobbled
You can also get lastfm suggestions (on any OS, without authentication) with the suggest command.
music lastfm suggest username --limit 20
If you would like to automatically install the music add the --install flag which will install to $MUSIC_PATH/Suggestions.
You can also sync Spotify playlists and albums with local tags. To get started first create a Spotify developer account and make an application. Then create a file ".music-spotify-credentials" in your cache directory.
Your cache directory is determined by Go below:
- On Unix systems, it returns
$XDG_CACHE_HOMEas specified by https://specifications.freedesktop.org/basedir-spec/basedir-spec-latest.html if non-empty, else $HOME/.cache. - On Darwin, it returns
$HOME/Library/Caches - On Windows, it returns
%LocalAppData% - On Plan 9, it returns
$home/lib/cache
In your new file you should store your client_id and client_secret:
client_id=xxxxxx
client_secret=yyyyyy
Now you can just run music spotify import <tag> [playlist] like so:
# Spotify user playlist
music spotify import https://open.spotify.com/playlist/hjklajskdlfj my-tag
# Spotify artist album
music spotify import https://open.spotify.com/album/3I2KkX13lHXuYqfBjSOopo mitski-creekThis will try to find all the songs in the given playlist/album, try to match it with songs from your local library, and tag them if they aren't already tagged with the provided tag. How it matches songs is through metadata rather than filenames.
If you have a tag you want to associate with an album you can set a relationship like so:
music spotify set-origin my-tag https://open.spotify.com/playlist/hjklajskdlfjNow you can just run the import command without specifying the url:
music spotify import my-tagTo delete a relationship between a tag and a Spotify playlist you can just run the same command without a second argument:
music spotify set-origin my-tagThis tool uses cobra which provides a completion command you can use to generate completions.
It works fine, but it doesn't have reactive completion to a few things: tags, music subdirectory install, format, and sort-type.
The cobra provided completion is also a bit more descriptive than I personally
like, which is why I use my own personal bash completion. It can be found in
./completion.bash
- Note: I have
mas an alias formusicandmxas an alias formusic play - Note: It's also more static/hard-coded, so a bit more error-prone/inaccurate.
If you would like to use the one of main functionalities of querying on android, you can do so using Termux. You will likely have to install go first and then follow the instructions in #installation.
After you have the music command installed, you want to download the wrapper script
in ./android-termux-mx. This script calls the play command with the given query
in a dry run so that no program is called. It then sets the results to a new tag (query.m3u).
You should copy the file to a directory in your Termux's $PATH:
# assuming your $PATH is just one folder (like default)
curl https://raw.githubusercontent.com/kitesi/music/main/android-termux-mx > $PATH/mxOnce you have the script mx, you can now run it like you would music play:
mx jaxson#tonight. Now you can open up vlc and see that in your playlists tab
there should be a playlist called "query."
You are able to install vlc/cvlc/nvlc on termux, and in turn music play will
work like normal, but I would not suggest doing so because termux will be
hanging, and you won't have vlc in your notification tray.
The configuration file is located in <$CONFIG_DIR>/go-music-kitesi/config.json. An example can be found in assets/default-config.jsonc. <$CONFIG_DIR> is specified as below:
- On Unix systems, it returns $XDG_CONFIG_HOME as specified by https://specifications.freedesktop.org/basedir-spec/basedir-spec-latest.html if non-empty, else $HOME/.config.
- On Darwin, it returns
$HOME/Library/Application Support. - On Windows, it returns
%AppData%. - On Plan 9, it returns
$home/lib.