Home

Awesome

tokio-tungstenite

Asynchronous WebSockets for Tokio stack.

MIT licensed Crates.io Build Status

Documentation

Usage

Add this in your Cargo.toml:

[dependencies]
tokio-tungstenite = "*"

Take a look at the examples/ directory for client and server examples. You may also want to get familiar with Tokio if you don't have any experience with it.

What is tokio-tungstenite?

This crate is based on tungstenite-rs Rust WebSocket library and provides Tokio bindings and wrappers for it, so you can use it with non-blocking/asynchronous TcpStreams from and couple it together with other crates from Tokio stack.

Features

As with tungstenite-rs TLS is supported on all platforms using native-tls or rustls through feature flags: native-tls, rustls-tls-native-roots or rustls-tls-webpki-roots feature flags. Neither is enabled by default. See the Cargo.toml for more information. If you require support for secure WebSockets (wss://) enable one of them.

Note, that if you're using rustls features with tokio-tungstenite version 0.23.0 or higher, you might observe a panic that is easy to fix (see the issue). Check the discussion for more details.

Is it performant?

In essence, tokio-tungstenite is a wrapper for tungstenite, so the performance is capped by the performance of tungstenite. tungstenite has a decent performance (it has been used in production for real-time communication software, video conferencing, etc), but it's definitely not the fastest WebSocket library in the world at the moment of writing this note.

If performance is of a paramount importance for you (especially if you send large messages), then you might want to check other libraries that have been designed to be performant or you could file a PR against tungstenite to improve the performance!

We are aware of changes that both tungstenite and tokio-tungstenite need in order to fill the gap of ~30% performance difference between tungstenite and more performant libraries like fastwebsockets, but we have not worked on that yet as it was not required for the use case that original authors designed the library for. In the course of past years we have merged several performance improvements submitted by the awesome community of Rust users who helped to improve the library! For a quick summary of the pending performance problems/improvements, see the comment.