Awaiting the future

In the last section, we saw how tasks composed of multiple futures are often difficult to write and debug. One attempt at remedying this is using a crate that wraps futures and associated error handling, yielding a more linear flow of code. This crate is called futures-await and is under active development.

This crate provides two primary mechanisms of dealing with futures:

• The #[async] attribute that can be applied to functions, marking them as asynchronous. These functions must return the Result of their computation as a future.
• The await! macro that can be used with async functions to consume the future, returning a Result.

Given these constructions, our earlier example download will look like this:

#[async]
fn download(url: &str) -> io::Result<Data> {
    ...
}

#[async]
fn parse_html(data: &Data) -> io::Result<Links> {
    ...
}

#[async]
fn process_links(links: &Links) -> io::Result<bool> {
    ...
}

#[async]
fn download_parse...