An implementation of a socket.io client written in the rust programming language. This implementation currently supports revision 5 of the socket.io protocol and therefore revision 4 of the engine.io protocol. If you have any connection issues with this client, make sure the server uses at least revision 4 of the engine.io protocol.
Information on the async
version can be found below.
Add the following to your Cargo.toml
file:
rust_socketio = "0.4.1"
Then you're able to run the following example code:
use rust_socketio::{ClientBuilder, Payload, RawClient};
use serde_json::json;
use std::time::Duration;
// define a callback which is called when a payload is received
// this callback gets the payload as well as an instance of the
// socket to communicate with the server
let callback = |payload: Payload, socket: RawClient| {
match payload {
Payload::String(str) => println!("Received: {}", str),
Payload::Binary(bin_data) => println!("Received bytes: {:#?}", bin_data),
}
socket.emit("test", json!({"got ack": true})).expect("Server unreachable")
};
// get a socket that is connected to the admin namespace
let socket = ClientBuilder::new("http://localhost:4200")
.namespace("/admin")
.on("test", callback)
.on("error", |err, _| eprintln!("Error: {:#?}", err))
.connect()
.expect("Connection failed");
// emit to the "foo" event
let json_payload = json!({"token": 123});
socket.emit("foo", json_payload).expect("Server unreachable");
// define a callback, that's executed when the ack got acked
let ack_callback = |message: Payload, _| {
println!("Yehaa! My ack got acked?");
println!("Ack data: {:#?}", message);
};
let json_payload = json!({"myAckData": 123});
// emit with an ack
socket
.emit_with_ack("test", json_payload, Duration::from_secs(2), ack_callback)
.expect("Server unreachable");
socket.disconnect().expect("Disconnect failed")
The main entry point for using this crate is the ClientBuilder
which provides a way to easily configure a socket in the needed way. When the connect
method is called on the builder, it returns a connected client which then could be used to emit messages to certain events. One client can only be connected to one namespace. If you need to listen to the messages in different namespaces you need to allocate multiple sockets.
Documentation of this crate can be found up on docs.rs.
This implementation now supports all of the features of the socket.io protocol mentioned here. It generally tries to make use of websockets as often as possible. This means most times only the opening request uses http and as soon as the server mentions that he is able to upgrade to websockets, an upgrade is performed. But if this upgrade is not successful or the server does not mention an upgrade possibility, http-long polling is used (as specified in the protocol specs). Here's an overview of possible use-cases:
- connecting to a server.
- register callbacks for the following event types:
- open
- close
- error
- message
- custom events like "foo", "on_payment", etc.
- send JSON data to the server (via
serde_json
which provides safe handling). - send JSON data to the server and receive an
ack
. - send and handle Binary data.
This library provides an ability for being executed in an asynchronous context using tokio
as
the execution runtime.
Please note that the current async implementation is still experimental, the interface can be object to
changes at any time.
The async Client
and ClientBuilder
support a similar interface to the sync version and live
in the asynchronous
module. In order to enable the support, you need to enable the async
feature flag:
rust_socketio = { version = "0.4.1-alpha.1", features = ["async"] }
The following code shows the example above in async fashion:
use futures_util::FutureExt;
use rust_socketio::{
asynchronous::{Client, ClientBuilder},
Payload,
};
use serde_json::json;
use std::time::Duration;
#[tokio::main]
async fn main() {
// define a callback which is called when a payload is received
// this callback gets the payload as well as an instance of the
// socket to communicate with the server
let callback = |payload: Payload, socket: Client| {
async move {
match payload {
Payload::String(str) => println!("Received: {}", str),
Payload::Binary(bin_data) => println!("Received bytes: {:#?}", bin_data),
}
socket
.emit("test", json!({"got ack": true}))
.await
.expect("Server unreachable");
}
.boxed()
};
// get a socket that is connected to the admin namespace
let socket = ClientBuilder::new("http://localhost:4200/")
.namespace("/admin")
.on("test", callback)
.on("error", |err, _| {
async move { eprintln!("Error: {:#?}", err) }.boxed()
})
.connect()
.await
.expect("Connection failed");
// emit to the "foo" event
let json_payload = json!({"token": 123});
socket
.emit("foo", json_payload)
.await
.expect("Server unreachable");
// define a callback, that's executed when the ack got acked
let ack_callback = |message: Payload, _: Client| {
async move {
println!("Yehaa! My ack got acked?");
println!("Ack data: {:#?}", message);
}
.boxed()
};
let json_payload = json!({"myAckData": 123});
// emit with an ack
socket
.emit_with_ack("test", json_payload, Duration::from_secs(2), ack_callback)
.await
.expect("Server unreachable");
socket.disconnect().await.expect("Disconnect failed");
}
This repository contains a rust implementation of the socket.io protocol as well as the underlying engine.io protocol.
The details about the engine.io protocol can be found here:
The specification for the socket.io protocol here:
Looking at the component chart, the following parts are implemented (Source: https://socket.io/images/dependencies.jpg):
MIT