A python 3.6+ script intended to be run as a daemon used to sap spotify songs into your local library, slowly freeing you from the proprietary hinges. The idea here is not to crack spotify protection algorithm or anything is just a daemon intended to be the most 'set it and forget it' possible, and will slowly sap all the songs you hear into your local disk, until one day, you won't need spotify anymore.

But if you don't wan't to be bothered with the moral aspect, don't have enough local storage space, or don't care, you still can use the ad-skipping functionally stand-alone with the -a flag.

• We support both spotify free and premium - The only difference is that the bitrate on spotify free (128kbps VBR) is lower than spotify premium. See the next point.
• Ignores advertisements automatically - Even in spotify free we wont record advertisement to clutter your library. We will even try to skip them, since spotify is allowing some ads to be skipped. And you can choose to do only this too, freedom also in choice.
• JACK audio server - The sky is the limit here, sapfy got the left and right ends connected, to whatever source of audio, you do you, have fun. Personally I left spotify permanently playing in the background, whether I want to listen to music or not I just (un)plug spotify's outputs to the actual physical output.
• Supports pauses - You can pause and resume whenever, the final song will be intact, no silence, as if it were never paused.
• We can detect incomplete tracks - And even delete them if you want.

The song recording part it's a extremely simple proof of concept and assumes a lot of things about your sound setup, I intend to make it more seamless to more setups, but here is what you need to get it working:

• A running jack audio server. Mainly for the 'connecting' ability that allows us to isolate spotify's output. There is no real need for all that real-time kernels, limits, etc.
• An exclusive/dedicated jack sink named spotify, on my system I created a pulse-jack sink to represent spotify since on linux spotify outputs to pulseaudio e.g.:
  • On linux it needs an ALSA -> Pulse -> Jack (Plugin) bridged setup. This is mostly because spotify only supports pulseaudio, and this is the only way you are able to create a dedicated output and not to mix spotify output with the rest of the computer sounds. Arch wiki has a good tutorial on how to setup the two to work together. E.g.:
  • pactl load-module module-jack-sink client_name=spotify connect=yes
• Redirect spotify's (only spotify's) output to the newly created, exclusive jack source

Given such setup, the script should be able to create a jack output and automatically connect it to spotify's jack-pulseaudio sink.

The required dependency are all on requirements.txt you can use pip to install them all. So to sum it all up, this should get you up and running:

git clone https://github.com/samosaara/sapfy ~/.sapfy
pip install --user -r ~/.sapfy/Requirements.txt 
mkdir -p  ~/.local/bin/
ln -s $HOME/.sapfy/sapfyd ~/.local/bin/sapfy

There should be an executable symbolic link to named sapfy inside your ~/.local/bin that you can use to execute the daemon. That folder might not be on your PATH, you may move the executable, or see here.

If you want you can also setup a service in your init system to get the ultimate seamless sapping experience. I pretend on providing this on the future.

The way I structured the imports, you need to run the program as a python module if want to run it directly through python. Or you can just run the provided executable sapfyd.

python3 -m sapfy
# Is the same as
./sapfyd

You can use --help to see the available command line options.

The --output flag receives a string, that will be formatted with the song's current metadata. It must not have the file extension, it will be added automatically. The output string can have the following place-holders that will be replaced based on the track metadata:

For an example, the default output string is

"{albumArtist[0]}/{album}/{trackNumber} {title}'"