SuperCollider 3 class library Python 3 port
Port of core features from sclang to Python 3, 3.6 for sure maybe >= 3.5 (not tested). It is intended to be the same library in a different language and to keep sclang elegance in a pythonic way (if possible).
So far SynthDef compiles and Routine and TempoClock are barely working, for that I started form AbstractFunction. Basics are not finished, I’m still studding the code. Also I’m not being linear in feature transcription, more like jumping and iterating (revisiting and completing) from place to place so to have the whole picture, sclang library has features integrated from bottom up and between classes. Everything is in a very early stage.
Code is still full of study comments and references (very unprofessional, I
know) most of them useless, don’t try understand, those are just reminders to
myself. I will be erasing them gradually. Some original sclang comments were
kept in code with two slash within python comments (
# //) and each file has a
reference to the source (many are obvious but some are not and some names were
changed for different reasons).
The main reason for this port is Python's capacity of interaction with other libraries applicable to composition, sonic-art and alike, and to be able to compile synth definitions and send them to the server is very handy.
I still don't know what would be the scope regarding features, what I will get finished for sure are server abstractions and interaction. The first idea was to transcribe patterns too, sounds easy, but patterns and events are a big deal for many reasons.
Python is not a real time language, it has much more jitter than sclang. While in my laptop sclang maintains an average 0.1ms time drift peak between calls with Python I have peaks of 5ms which is quite a lot. A pure Python implementation will not solve that and I don’t have intentions to go low level by now. It is not that it is impossible but that it is out of my scope.
It is not really a deal breaker for many use cases because SuperCollider algorithms sync routines time no matter what but if you use routines to sequence musical events non real time code will produce rhythmic irregularities (soft real time degradation) if the system is a bit loaded and specially if code is complex and make many calls, and even less than 1ms is too much when it starts to add.
Having said that, the idea is that you can write the same in Python as in sclang, with the same logic regarding multichannel expansion, arguments conversion to Control ugens, etc., it should be the same result. For example:
from sc3.all import * # interactive shell run... s = Server.local s.boot() # wait or error... @synthdef.add() def sine(freq=440, amp=0.1, gate=1): sig = SinOsc.ar(freq) * amp sig *= EnvGen.kr(Env.adsr(), gate, done_action=2) Out.ar(0, sig.dup()) sine.dump_ugens() n = Synth('sine') n.set('amp', 0.05) n.set('freq', 220) s.query_all_nodes(True) n.release() s.quit()
That's a working example but many things are not finished or tested and some are no decided, things may change or move.
Install (in develop mode)
python3 setup.py develop --user
The sc3 library holds the same license as SuperCollider: sc3 is free software available under Version 3 of the GNU General Public License. See COPYING for details.
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
|Filename, size||File type||Python version||Upload date||Hashes|
|Filename, size sc3-0.0.5-py3-none-any.whl (190.5 kB)||File type Wheel||Python version py3||Upload date||Hashes View hashes|
|Filename, size sc3-0.0.5.tar.gz (158.5 kB)||File type Source||Python version None||Upload date||Hashes View hashes|