Dropping

[markcsaving]

I'm new to Rust, but I'm curious as to why we don't implement drop for the first two lists like this:

fn drop(&mut self) {
while let Some(_) = self.pop() {}
}

This seems like a very obvious implementation, and it seems more elegant than the given implementations. I ran a test which creates a list of a million elements and lets said list go out of scope, and everything went fine (the test resulted in stack overflow when I ran it before implementing drop).

Is there some reason I'm not seeing to prefer the given implementations?

[Gankra]

I believe you are correct, it just didn't occur to me to do so when I wrote it. It doesn't really hurt to have some more motivation to write some more non-trivial code, though.

As a minor point in my favour, the pop-based implementation is semantically demanding that the payload of each node be copied out of the allocation and onto the stack to be destroyed. The manual implementation I provide avoids this, instead only copying around Boxes (pointers). This could be fairly important if the generic case ends up having a rather large type with an actual destructor to run. I wouldn't be surprised if this difference had a tendency to evaporate in optimized builds, though.

[Gankra]

(an implementation which wanted the best of both worlds would introduce a pop_node method that returns Option<Box<Node>>, from-which pop and drop are derived)

[Gankra]

Added https://rust-unofficial.github.io/too-many-lists/first-drop.html#bonus-section-for-premature-optimization

[markcsaving]

An excellent answer! Your book is a great resource, and I'm glad to have indirectly contributed to part of it.