Registry / http-networking / asyncio-dgram

asyncio-dgram

library3.0.0pypypi✓ verified 85d ago

asyncio-dgram provides a higher-level, more convenient API for working with UDP datagram sockets in Python's asyncio framework. It simplifies common UDP operations like binding, connecting, sending, and receiving data with an async/await interface. The current version is 3.0.0, and it generally follows a release cadence driven by feature additions, bug fixes, and API improvements.

http-networking

pip install asyncio-dgram

github ↗package ↗changelog ↗

INSTALL

IMPORT

SIG · ASYNCIO-DGRAM

asyncio-dgram

http-networkingpythonv3.0.0

Install

1.5s avg

Import

211ms

Disk

16MB

Pass rate

10/ 10

Env Coverage10 / 10

glibc

3.9–3.13

musl

3.9–3.13

Install & Compatibility

Where this runs

tested against v3.0.0 · pip install

no network on importno background threads

Install × environment matrix

Each cell = how many times install + import succeeded across repeated harness runs. Partial = flaky.
glibc = Debian/Ubuntu slim · musl = Alpine Linux

musl

py 3.10–3.910 runs

installs and imports cleanly · install 0.0s · import 0.225s · 17.8MB

glibc

py 3.10–3.910 runs

installs and imports cleanly · install 1.5s · import 0.197s · 18MB

16MB installed

● package 16MB

Code

Verified usage

Verified import paths — ran on the pinned version, not inferred.

bind

✓ from asyncio_dgram import bind

connect

✓ from asyncio_dgram import connect

This example demonstrates a simple UDP server and client using `asyncio-dgram`. The server binds to a port, receives a message, and echoes it back. The client connects to the server, sends a message, and receives the echo. Both use the `async with` syntax for managing the datagram stream.

import asyncio from asyncio_dgram import bind, connect async def udp_server(host, port): async with bind((host, port)) as stream: print(f"Server: Listening on {host}:{port}") data, remote_addr = await stream.recv() message = data.decode() print(f"Server: Received '{message}' from {remote_addr}") response = f"Hello from server! You sent: '{message}'".encode() await stream.send(response, remote_addr) print(f"Server: Sent '{response.decode()}' back to {remote_addr}") print("Server: Stream closed.") async def udp_client(host, port, message): async with connect((host, port)) as stream: print(f"Client: Sending '{message}' to {host}:{port}") await stream.send(message.encode()) data, remote_addr = await stream.recv() response = data.decode() print(f"Client: Received '{response}' from {remote_addr}") print("Client: Stream closed.") async def main(): host = '127.0.0.1' port = 9999 server_task = asyncio.create_task(udp_server(host, port)) # Give the server a moment to start listening await asyncio.sleep(0.1) client_task = asyncio.create_task(udp_client(host, port, "Hello UDP World!")) await asyncio.gather(server_task, client_task) if __name__ == "__main__": asyncio.run(main())

Debug

Known issues

breakingAs of version 3.0.0, the `bind()` and `connect()` functions no longer return the datagram stream directly. Instead, they return an asynchronous context manager that must be used with `async with`.

fix

Change `stream = await asyncio_dgram.bind(...)` to `async with asyncio_dgram.bind(...) as stream:` (and similarly for `connect`).

affects: >=3.0.0

gotchaUDP is a connectionless protocol. While `asyncio_dgram.connect()` 'connects' the local socket to a specific remote address, it primarily filters outgoing and incoming packets on the local socket. It does not establish a persistent connection in the TCP sense, and `recv()` can still theoretically receive from other addresses if not explicitly handled or if the 'connected' state is not enforced by the OS.

fix

Always be mindful of UDP's connectionless nature. Verify remote addresses in `recv()` results if strict sender validation is needed, especially when using `bind()`.

affects: All

gotchaUDP offers no reliability guarantees; packets may be lost, duplicated, or arrive out of order. `asyncio-dgram` does not implement any form of reliability on top of UDP. Applications must handle these possibilities or use a higher-level protocol.

fix

Implement application-level acknowledgements, retransmissions, or sequence numbering if reliability is required, or consider using TCP-based solutions (like `asyncio.start_server` / `asyncio.open_connection`) if strict reliability is paramount.

affects: All

Errors

Common errors & fixes

TypeError: 'coroutine' object is not an async context manager

Attempting to `await` the result of `bind()` or `connect()` directly instead of using `async with` (a breaking change in v3.0.0).

fix

Use `async with bind(...) as stream:` or `async with connect(...) as stream:` instead of `stream = await bind(...)`.

OSError: [Errno 98] Address already in use

Another process is already using the specified UDP port, or the previous socket was not properly closed and the OS hasn't released it yet (common after a crash).

fix

Ensure no other application is listening on the port. If running immediately after stopping a previous instance, wait a moment or choose a different port. On Linux, `sudo netstat -apun | grep <port>` can help identify the rogue process.

TimeoutError: [Errno 110] Connection timed out (when sending/receiving to a remote address)

When using `connect()` and sending, the remote server might not be running, or a firewall is blocking packets, preventing any response from being received within a reasonable timeframe (though `asyncio-dgram`'s `send` itself doesn't typically timeout for UDP, `recv` can).

fix

Verify the remote host and port are correct and reachable. Check firewall settings on both sender and receiver. Ensure the remote server is actively listening for UDP packets.

Upgrade

Version history

3.0.0latest on PyPI · released Jan 21, 2026

Audit

Dependencies

No dependency data recorded yet.

Agent activity

23 hits · last 30 days

node

OpenAI (training)

Resources

githubhttps://github.com/jsbronder/asyncio-dgram ↗

packagehttps://pypi.org/project/asyncio-dgram/ ↗

changeloghttps://github.com/jsbronder/asyncio-dgram/blob/master/ChangeLog ↗