| //@ edition: 2018 | |
| #![feature(arbitrary_self_types)] | |
| // tests that the referent type of a reference must be known to call methods on it | |
| struct SmartPtr<T>(T); | |
| impl<T> core::ops::Receiver for SmartPtr<T> { | |
| type Target = T; | |
| } | |
| impl<T> SmartPtr<T> { | |
| fn foo(&self) {} | |
| } | |
| fn main() { | |
| let val = 1_u32; | |
| let ptr = &val; | |
| let _a: i32 = (ptr as &_).read(); | |
| //~^ ERROR type annotations needed | |
| // Same again, but with a smart pointer type | |
| let val2 = 1_u32; | |
| let rc = std::rc::Rc::new(val2); | |
| let _b = (rc as std::rc::Rc<_>).read(); | |
| //~^ ERROR type annotations needed | |
| // Same again, but with a smart pointer type | |
| let ptr = SmartPtr(val); | |
| // We can call unambiguous outer-type methods on this | |
| (ptr as SmartPtr<_>).foo(); | |
| // ... but we can't follow the Receiver chain to the inner type | |
| // because we end up with _. | |
| // Because SmartPtr implements Receiver, it's arguable which of the | |
| // following two diagnostics we'd want in this case: | |
| // (a) "type annotations needed" (because the inner type is _) | |
| // (b) "no method named `read` found for struct `SmartPtr`" | |
| // (ignoring the fact that there might have been methods on the | |
| // inner type, had it not been _) | |
| // At present we produce error type (b), which is necessary because | |
| // our resolution logic needs to be able to call methods such as foo() | |
| // on the outer type even if the inner type is ambiguous. | |
| let _c = (ptr as SmartPtr<_>).read(); | |
| //~^ ERROR no method named `read` found for struct `SmartPtr<T>` | |
| } |