Definition of Option, and enum ordering

[lucas-deangelis]

In chapter 6.1, defining an enum, Option is described with Some(T) first and None later, while in Rust it's defined with None first and Some(T) later. This can break intuition when comparing enums. Here's the definition of Option in chapter 6.1:

This enum is Option, and it is defined by the standard library as follows:

enum Option<T> {
   Some(T),
   None,
}

However, this is not exact. The standard library defines Option this way:

#[derive(Copy, PartialEq, PartialOrd, Eq, Ord, Debug, Hash)]
#[rustc_diagnostic_item = "option_type"]
#[stable(feature = "rust1", since = "1.0.0")]
pub enum Option<T> {
    /// No value
    #[lang = "None"]
    #[stable(feature = "rust1", since = "1.0.0")]
    None,
    /// Some value `T`
    #[lang = "Some"]
    #[stable(feature = "rust1", since = "1.0.0")]
    Some(#[stable(feature = "rust1", since = "1.0.0")] T),
}

As you can see, None comes before Some(T). This has a consequence:

#[derive(Clone, Copy, PartialEq, PartialOrd, Eq, Ord, Debug, Hash)]
enum MyOption<T> {
    Some(T),
    None,
}
fn main() {
    println!("{}", Option::None < Option::Some(0));
    println!("{}", MyOption::None < MyOption::Some(0));  
}
// => true
// => false

So with this, I would suggest reversing the current order of Option in the book to match with the one used in Rust.

I can make a PR for this, but I'd like to see if people are okay with the change first. I may be missing something.

[Lokathor]

There's an observable difference, but I would not say a significant difference. Still, if it helps things match the standard library it probably wouldn't hurt.

[lucas-deangelis]

There's an observable difference, but I would not say a significant difference.

I agree, maybe I wasn't clear enough about it but I don't say it's something big. Still, I wanted to document it as much as I can. The book is often how new people (me included!) learn Rust, and this change could look like trivial bikeshedding without a logical explanation behind it.