Asyncio support#

The asyncio module built into Python 3.4 and later can be used to write asynchronous code in a single thread. This library supports receiving messages asynchronously in an event loop using the can.Notifier class.

There will still be one thread per CAN bus but the user application will execute entirely in the event loop, allowing simpler concurrency without worrying about threading issues. Interfaces that have a valid file descriptor will however be supported natively without a thread.

You can also use the can.AsyncBufferedReader listener if you prefer to write coroutine based code instead of using callbacks.


Here is an example using both callback and coroutine based code:

#!/usr/bin/env python

This example demonstrates how to use async IO with python-can.

import asyncio
from typing import List

import can
from can.notifier import MessageRecipient

def print_message(msg: can.Message) -> None:
    """Regular callback function. Can also be a coroutine."""

async def main() -> None:
    """The main function that runs in the loop."""

    with can.Bus(
        interface="virtual", channel="my_channel_0", receive_own_messages=True
    ) as bus:
        reader = can.AsyncBufferedReader()
        logger = can.Logger("logfile.asc")

        listeners: List[MessageRecipient] = [
            print_message,  # Callback function
            reader,  # AsyncBufferedReader() listener
            logger,  # Regular Listener object
        # Create Notifier with an explicit loop to use for scheduling of callbacks
        loop = asyncio.get_running_loop()
        notifier = can.Notifier(bus, listeners, loop=loop)
        # Start sending first message

        print("Bouncing 10 messages...")
        for _ in range(10):
            # Wait for next message from AsyncBufferedReader
            msg = await reader.get_message()
            # Delay response
            await asyncio.sleep(0.5)
            msg.arbitration_id += 1

        # Wait for last message to arrive
        await reader.get_message()

        # Clean-up

if __name__ == "__main__":