aioartnet

aioartnet is a pure python asyncio connector for the royalty-free Art-Net protocol, which is a transport to transmit and recieve the DMX-512 lighting control protocol over Ethernet (UDP). The protocol is the modern standard for interconnecting complex lighting fixtures directly, and for branching out individual universes for a specific local area to control classic DMX-512 XLR-cable interlinked fixtures. Both open and propriety lighting control systems can emit Art-Net directly.

This library aims to be simple and robust, and can both input data into Art-Net, as well as output it from "artnet" to your user code. It builds a dynamic model of the network's Art-Net nodes, their ports and the universe(s) of DMX-512 that are being controlled. Fully type hinted to comply with PEP-561. No non-core dependancies, with a test suite that runs on Python 3.9 to 3.12.

It can also be used passively to build the network model without joining as an Art-Net Node, by listening to the broadcasts from other Art-Net devices. However it is not usually possible to monitor the DMX values passively.

We also have a propriety iOS/Xcode component available for commercial licensing that uses the same event driven API as this code, built on top of Apple's Networking framework NWProtocolUDP. This will be a significantly better starting point for new designs than libartnet for most projects.

Install & Demo

Use pip to install the package from pypi. Running aioartnet.main will show other nodes and universes discovered:

$ pip install aioartnet
$ python -m aioartnet.main
INFO:aioartnet:preferred interfaces: [(1, 'wlp4s0'), (10, 'br-ee82b9af434e'), (10, 'docker0'), (10, 'lo')]
INFO:aioartnet:using interface wlp4s0 with ip 192.168.1.205 broadcast ip 192.168.1.255
status:
  ArtNetNode<aioartnet,192.168.1.205:6454>                     [Port<Input,DMX,0:0:1>, Port<Output,DMX,0:0:5>]
  ArtNetNode<ODE Mk3,192.168.1.238:6454>                       [Port<Output,DMX,0:0:0>, Port<Output,DMX,0:0:1>]
  ArtNetNode<DMX Monitor,192.168.1.222:6454>                   []
 0:0:1 pubs:[ArtNetNode<aioartnet,192.168.1.205:6454>] subs:[ArtNetNode<ODE Mk3,192.168.1.238:6454>]
 0:0:5 pubs:[] subs:[ArtNetNode<aioartnet,192.168.1.205:6454>]
 0:0:0 pubs:[] subs:[ArtNetNode<ODE Mk3,192.168.1.238:6454>]

There is a built in command-line lighting console for quick Art-Net testing, which maps channels 1:1 to a single universe:

$ python -m aioartnet.console --universe 0:0:0
INFO:aioartnet:preferred interfaces: [(1, 'wlp4s0'), (10, 'br-ee82b9af434e'), (10, 'docker0'), (10, 'lo'), (10, 'lxdbr0')]
INFO:aioartnet:using interface wlp4s0 with ip 192.168.1.204 broadcast ip 192.168.1.255, listening on 0.0.0.0 our mac 4485009be628
> live on
> ch 1 at f
> record cue 1
> ch 2 at f
> record cue 2
> go
activated ActiveCue(cue=Cue(name='', fade_in=0, hold=1, fade_out=0, channels=[ChannelIntensity(channel=0, intensity=255)]), since=1750845334.3249245, state=<FadeState.FADE_IN: 2>)
finished ActiveCue(cue=Cue(name='', fade_in=0, hold=1, fade_out=0, channels=[ChannelIntensity(channel=0, intensity=255)]), since=1750845334.3249245, state=<FadeState.OFF: 1>)
> go
activated ActiveCue(cue=Cue(name='', fade_in=0, hold=1, fade_out=0, channels=[ChannelIntensity(channel=1, intensity=255)]), since=1750845336.5779428, state=<FadeState.FADE_IN: 2>)
finished ActiveCue(cue=Cue(name='', fade_in=0, hold=1, fade_out=0, channels=[ChannelIntensity(channel=1, intensity=255)]), since=1750845336.5779428, state=<FadeState.OFF: 1>)
> [ctl-C]

Getting Started - Code

1. Import the library and create an ArtNetClient

    from aioartnet import ArtNetClient
    client = ArtNetClient()

2. Configure the desired input/output ports on the client using set_port_config(), each call returns ArtNetUniverse objects. Use these to optionally set the initial published value of universes set as ArtNet inputs:

    u1 = client.set_port_config("0:0:1", is_input=True)
    u5 = client.set_port_config("0:0:5", is_output=True)
    u1.set_dmx(bytes(list(range(128))*4))

3. Await the co-routine client.connect() from an asyncio asynchronous context, e.g:

    async def main() -> None:
        client = ArtNetClient()
        u1 = client.set_port_config("0:0:1", is_input=True)
        u5 = client.set_port_config("0:0:5", is_output=True)
        u1.set_dmx(bytes(list(range(128))*4))
        await client.connect()

    if __name__ == "__main__":
        logging.basicConfig(level=logging.INFO)
        asyncio.run(main())
        asyncio.get_event_loop().run_forever()

4. Once the client is connected, you can add/configure further ports, read the universes of DMX data with universe.get_dmx() and for input ports, change the published values with universe.set_dmx(bytes).

5. The return value from client.connect() is an ayncio.DatagramTransport. If you call close() on this transport, the periodic task is also cancelled.

Networking

There are three ways to configure the networking:

Automatically

Creating an ArtNetClient with no arguments will perform discovery during connect() by iterating the interfaces available, checking if they have an IPV4 address (ie. AF_INET family) and prefering certain properties. Specifically we prefer 2.x.x.x subnets with the 255.0.0.0 subnet mask, then configured Ethernet (interface name enp*), then configured WiFi interfaces (wlp*), followed by any other remaining AF_INET interfaces. The priority ordering is set by aioartnet.PREFERED_INTERFACES_ORDER which can be monkey-patched if required.

Manually by specifying an interface

Passing an interface name to the constructor ArtNetClient(interface="en0"), or setting client.interface="en0" before calling connect() will force use of a specific interface, skipping the interface selection logic. The interface is then queried to determine the ip, broadcast and listening address to use using ioctl calls.

Manually by specifying unicast and broadcast IPs

If the two required IP addresses (ours, broadcast) are set manually on the client no discovery is performed. This will be the most portable method as we effectively supply the arguments directly to asyncio.

client = ArtNetClient()
client.unicast_ip = '192.168.1.15'
client.broadcast_ip = '192.168.1.255'
await client.connect()

Features

Implemented Messages

Art-Net

This application aims to be fully compatible with Art-Net devices. We have tested it with:

• QLC+ lighting controller
• Enttec ODE Mk3
• iOS DMX Monitor app
• Many other iOS apps

Art-Net™ Designed by and Copyright Artistic Licence Engineering Ltd.