Exercise 2.4.1.J Deref and DerefMut not working as said

[fubupc]

The question states:

Replacing the implementation of deref with self.as_ref() results in a stack overflow when running an unoptimized version. Why? (Hint: deref coercion)

But actually self.as_ref() does not cause stack overflow: https://play.rust-lang.org/?version=stable&mode=debug&edition=2021&gist=9ce74e7edda0542dbf20c04f521b2879.

Not sure if I misunderstand something here.

PS: I cannot give a link to above quote directly because the book's website is broken: https://teach-rs.trifectatech.org/full .

[squell]

The AsRef impl makes this work; it doesn't call the Deref coercion. I think the exercise is also missing entirely the fact that you need to provide an AsRef impl. My guess is that in an earlier iteration self.as_ref() would cause a Deref-inspired self-loop but due to various modifications it didn't -- I also can't get it to loop on me (but I have seen such loops so it's good to have it in an exercise somewhere).

We should probably import the Vector math exercise from https://training.tweede.golf/traits-and-generics.html since that is maybe a gentler illustration of traits and generics.

[squell]

The overflow is triggered with &*self (kind of obviously) and maybe less intuitively, self; maybe the comment was referring to that. But as is, I don't see how the obvious solution would trigger such as self-loop if we just supply the Deref impl's.

I've also checked the earlier Rust editions (2015, 2018) but I just think this is a mistake in the exercise.