Spawning

Spawning allows you to run a new asynchronous task in the background. This allows us to continue executing other code while it runs.

Say we have a web server that wants to accept connections without blocking the main thread. To achieve this, we can use the async_std::task::spawn function to create and run a new task that handles the connections. This function takes a future and returns a JoinHandle, which can be used to wait for the result of the task once it's completed.

use async_std::{task, net::TcpListener, net::TcpStream};
use futures::AsyncWriteExt;

async fn process_request(stream: &mut TcpStream) -> Result<(), std::io::Error>{
    stream.write_all(b"HTTP/1.1 200 OK\r\n\r\n").await?;
    stream.write_all(b"Hello World").await?;
    Ok(())
}

async fn main() {
    let listener = TcpListener::bind("127.0.0.1:8080").await.unwrap();
    loop {
        // Accept a new connection
        let (mut stream, _) = listener.accept().await.unwrap();
        // Now process this request without blocking the main loop
        task::spawn(async move {process_request(&mut stream).await});
    }
}

The JoinHandle returned by spawn implements the Future trait, so we can .await it to get the result of the task. This will block the current task until the spawned task completes. If the task is not awaited, your program will continue executing without waiting for the task, cancelling it if the function is completed before the task is finished.

#![allow(unused)]
fn main() {
use futures::future::join_all;
async fn task_spawner(){
    let tasks = vec![
        task::spawn(my_task(Duration::from_secs(1))),
        task::spawn(my_task(Duration::from_secs(2))),
        task::spawn(my_task(Duration::from_secs(3))),
    ];
    // If we do not await these tasks and the function finishes, they will be dropped
    join_all(tasks).await;
}
}

To communicate between the main task and the spawned task, we can use channels provided by the async runtime used.