Skip to main content

Python port of PeerJS client library.

Project description

Gitpod ready-to-code

peerjs-python

Python port of PeerJS client.

  • Tracked in issue #160 of the official PeerJS project.

Enables Progressive Web Apps to discover and pair directly with Python apps using secure, browser supported WebRTC protocol.

Additional features:

  • HTTP Proxy over WebRTC data channel. It allows PeerJS browser clients to make remote REST requests over WebRTC to a local REST API running behind a firewall.
  • Plug-and-play functionality that allows seamless Airdrop-like pairing between peers running on the same local network.

See Ambianic UI PNP module for a real-world example how PeerJS Python is used with PnP and HTTP Proxy.

Dependencies

Uses aiortc as Python WebRTC provider. This requires installing a few native dependencies for audio/video media processing.

On Debian/Ubuntu run:

apt install libavdevice-dev libavfilter-dev libopus-dev libvpx-dev pkg-config libsrtp2-dev

On OS X run:

brew install ffmpeg opus libvpx pkg-config

Motivation

This project was originally motivated while searching for a way to:

  • Connect a Progressive Web App (Ambianic UI) directly and securely to an edge device (Ambianic Edge) running Python app on a Raspberry Pi behind a home Internet router.

This article in WebRTCHacks provides more insight into the background of this project.

Main requirements

  • Easy Airdrop-like plug and play discovery and pairing between web app and edge devices.
  • Direct peer-to-peer connectivity to minimize:
    • latency
    • architecture complexity
    • costs associated with traffic and hosting of tunneling servers
    • exposure to public server security attacks
  • Support for:
    • Secure connections
    • Bi-directional data-channel
    • Low latency audio/video media streaming
    • Bi-directional live audio/video media
  • Rely only on standard broadly supported web browser features.
    • Stable mobile device support (iOS, Android, Raspberry Pi)
    • Stable desktop OS support (Windows, Mac OS, Linux)
    • No need for browser plug-ins
  • No intermediary cloud service providers to store and sync user data.
  • No tedious and complicated NAT setups of dynamic DNS with SSH tunnels between public IP servers and edge devices behind firewall.
  • High throughput and scalability via lightweight signaling service without a persistence layer.

Project Status

Initial working prototype completed. PeerJS Python is now able to connect over WebRTC DataChannel to PeerJS in the browser and exchange messages.

  • Complete and test connectivity with signaling server (peerjs-server).
  • Complete and test data channel connectivity with web app peers.
  • Release initial version to PyPi.
  • support for python 3.7 & python 3.8
  • support for python 3.9 (see dependency issue)
  • >90% code coverage with CI tests.
  • Port media support.

Code Examples

A typical p2p session takes these steps:

  1. Establish signaling server session that enables peers to discover each other.
  2. Discover remote peer ID (either via signaling server room affinity or other means)
  3. Request connection to remote peer via signaling server
  4. Connect to remote peer via WebRTC ICE protocol.
  5. Exchange data or media with remote peer over p2p WebRTC connection.

The following code snippet shows the initial part of establishing a signaling server connection.

    options = PeerOptions(
        host=config['host'],
        port=config['port'],
        secure=config['secure'],
        token=new_token,
        config=RTCConfiguration(
            iceServers=[RTCIceServer(**srv) for srv in config['ice_servers']]
        )
    )
    peer = Peer(id=savedPeerId, peer_options=options)
    await peer.start()
    log.info('peer activated')
    _setPnPServiceConnectionHandlers(peer)

Once a signaling server connection is established, a peer can request connection to another peer or listen for requests from a remote peer. The example snippet bellow shows the latter:

    @peer.on(PeerEventType.Connection)
    async def peer_connection(peerConnection):
        log.info('Remote peer trying to establish connection')
        _setPeerConnectionHandlers(peerConnection)

After a p2p connection is established, a peer can receive and send application messages. The following snippet shows how a peer receives a message:

    @peerConnection.on(ConnectionEventType.Data)
    async def pc_data(data):
        log.debug('data received from remote peer \n%r', data)

For a complete working example see this file.

Other Related Open Source projects

There are several great projects that solve the problem of accessing IoT devices behind firewall via tunneling servers.

  • Python Proxy: Asynchronous tunnel proxy implemented in Python 3 asyncio.
  • Proxy.py: HTTP proxy server written in Python.
  • Inlets: Reverse proxy and service tunnel written in Go.
  • Macchina.io: IoT tunneling proxy written in C++.

A few popular p2p projects that use WebRTC:

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

peerjs-1.5.1.tar.gz (27.9 kB view details)

Uploaded Source

Built Distribution

peerjs-1.5.1-py3-none-any.whl (28.9 kB view details)

Uploaded Python 3

File details

Details for the file peerjs-1.5.1.tar.gz.

File metadata

  • Download URL: peerjs-1.5.1.tar.gz
  • Upload date:
  • Size: 27.9 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/3.4.2 importlib_metadata/4.8.1 pkginfo/1.7.1 requests/2.26.0 requests-toolbelt/0.9.1 tqdm/4.62.3 CPython/3.9.7

File hashes

Hashes for peerjs-1.5.1.tar.gz
Algorithm Hash digest
SHA256 aac4c43ce382d7a9d3a3f34fb09cc1f70fb24a57f2fb66f89e3019be1d3fc639
MD5 a793bd3a87347b4eb87c8b0623d550c8
BLAKE2b-256 8fe78647940c378f3664e51394ddc018dea44bec0d79a01b9871928682fff5f9

See more details on using hashes here.

File details

Details for the file peerjs-1.5.1-py3-none-any.whl.

File metadata

  • Download URL: peerjs-1.5.1-py3-none-any.whl
  • Upload date:
  • Size: 28.9 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/3.4.2 importlib_metadata/4.8.1 pkginfo/1.7.1 requests/2.26.0 requests-toolbelt/0.9.1 tqdm/4.62.3 CPython/3.9.7

File hashes

Hashes for peerjs-1.5.1-py3-none-any.whl
Algorithm Hash digest
SHA256 7214d7da5e6802d515dae20b7971fa55ec2f5ea9edaf3ee7f0c5018732731559
MD5 2103dba03d3d8cde57d609aedc527f0f
BLAKE2b-256 313b35b31054c3b1de0afca7700f62a4ec5ca1610c4377db36818968c51a640e

See more details on using hashes here.

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page