Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

supertrait methods should be callable on object types #4100

Closed
brson opened this issue Dec 3, 2012 · 13 comments
Closed

supertrait methods should be callable on object types #4100

brson opened this issue Dec 3, 2012 · 13 comments
Labels
A-traits Area: Trait system

Comments

@brson
Copy link
Contributor

brson commented Dec 3, 2012

Bound type parameters can call supertrait methods, objects should be able to as well.

trait A { fn f() }
trait B: A { }

fn foo(b: &B) {
    b.f()
}
@nikomatsakis
Copy link
Contributor

Bumping to 0.6

@catamorphism
Copy link
Contributor

Reproduced as of eed2ca6

@ghost ghost assigned catamorphism Feb 21, 2013
@catamorphism
Copy link
Contributor

Not critical for 0.6, de-milestoning

@Sodel-the-Vociferous
Copy link
Contributor

catamorphism, are you actively working on this one, or do you mind if I take a crack at it?

@catamorphism
Copy link
Contributor

@Sodel-the-Vociferous I'm not actively working on it, so please do! Ask on IRC if you have any questions :-)

@Sodel-the-Vociferous
Copy link
Contributor

Problems found so far:

  • LookupContext::push_inherent_impl_candidates_for_type() wasn't checking supertraits; seems to be fixed on my branch.
  • LookupContext::push_inherent_items_from_trait() wasn't checking supertraits, either. This is taking a bit longer because of the next problem.
  • ty::lookup_item_type() crashes the compiler when I give it a supertrait's def_id, and says there's no matching type in the tcache. I'm going to do a bit more testing to see what else does and does not aggravate this problem.

Also, by analogy to the method_origins "method_self" and "method_super", I'm going to split "method_trait" into "method_trait" and "method_trait_super". We might as well be explicit when we're trying to use trait inheritance with dynamic dispatch.

@Sodel-the-Vociferous
Copy link
Contributor

Objects need some refactoring first. Waiting on #5762.

@catamorphism
Copy link
Contributor

Nominating for milestone 5, production-ready

@graydon
Copy link
Contributor

graydon commented Jul 18, 2013

accepted for feature-complete milestone

@huonw
Copy link
Member

huonw commented Jul 28, 2013

@msullivan has been doing work in this area recently, maybe he has some light to shed?

@ghost ghost assigned msullivan Aug 13, 2013
@msullivan
Copy link
Contributor

Yeah, I'll take this.

@vi
Copy link
Contributor

vi commented Apr 10, 2015

It works, but casting b as &A not. It can be worked around:

struct ProxyObject<'q> (&'q B);

impl<'q> A for ProxyObject<'q> {
   fn f(&self) {let &ProxyObject(x) = self;  x.f();}
}

// &ProxyObject(b) as &A

Maybe the compiler can do something like this automatically?

@nikomatsakis
Copy link
Contributor

@vi indeed, adding more general upcasts is definitely something we'd like to do soon-ish. It's a touch tricky when you dig into the details.

bors pushed a commit to rust-lang-ci/rust that referenced this issue May 15, 2021
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
A-traits Area: Trait system
Projects
None yet
Development

No branches or pull requests

8 participants