This is a JavaScript library to "bake" ars-tracker songs into rendered samples, suitable for use as music or sound effects in the Web Audio APIs.

Example usage:

// uncomment if your package management system is anything fancier than
//     <script src="et209.js"></script><script src="baker.js"></script>
//let bakeArsTrackerSong = require("ars-tracker-baker");
let baked = bakeArsTrackerSong(my_module, 0, {"headphones":headphones});
let buffer = audioContext.createBuffer(2, baked.sampleCount, baked.sampleRate);
if(buffer.copyToChannel) {
  buffer.copyToChannel(baked.left, 0);
  buffer.copyToChannel(baked.right, 1);
}
else {
  let leftData = baked.getChannelData(0);
  let rightData = baked.getChannelData(1);
  for(var n = 0; n < baked.sampleCount; ++n) {
    leftData[n] = baked.left[n];
    rightData[n] = baked.right[n];
  }
}
// buffer can now be used with one or more `AudioBufferSourceNode`s

The first parameter to bakeArsTrackerSong is the ars-tracker module itself, after having gone through JSON.parse or otherwise been turned into an object.

The second parameter is the index of the song within the module: 0 for the first song, 1 for the second, and so forth.

The third parameter is an optional object containing options.

Options:

• loop (boolean, default true): Whether to try to loop the song.
• loopOverlapTime (seconds, default 2): If looping, the amount of extra time to add to make sure the loop is clean.
• loopFadeTime (seconds, default 5): If looping, the amount of extra time to add to the end, so it can fade out nicely if you don't use the looping information.
• startOrder (index, default 0): The order index to start at. When in doubt, leave unset.
• headphones (boolean, default false): Whether to enable the "headphones filter".

bakeArsTrackerSong returns an object with the following keys:

• sampleRate: For convenience, this is ET209.SAMPLE_RATE.
• loopLeft: The sample number of the beginning of the loop.
• loopRight: The sample number of the end of the loop.
• sampleCount: The number of samples that were rendered.
• left: A Float32Array containing sampleCount samples for the left stereo channel.
• right: A Float32Array containing sampleCount samples for the right stereo channel.

Unsupported things:

• Actual Waveform sequences. We only use the first element of the Waveform sequence. (tracklib does the same thing.)
• Pitchbend / arpeggio / slides on the noise channel. (tracklib doesn't support these either.)
• Proper Pxx (pan) effect handling unless a note on happens at the same time.
• Vxx (waveform) effect if a note on happens at the same time.