Add a new method_super origin for supertrait methods. Also make
coherence create a table that maps pairs of trait IDs and self types
to impl IDs, so that it's possible to check a supertrait method
knowing only its index in its trait's methods (without knowing all
supertraits for a given trait).
r=nmatsakis and graydon -- with hope, we'll revamp all of this code as
per #4678, but for now this fixes the bug.
Closes#3979
"Dual impls" are impls that are both type implementations and trait
implementations. They can lead to ambiguity and so this patch removes them
from the language.
This also enforces coherence rules. Without this patch, records can implement
traits not defined in the current crate. This patch fixes this, and updates
all of rustc to adhere to the new enforcement. Most of this patch is fixing
rustc to obey the coherence rules, which involves converting a bunch of records
to structs.
1. In the first case, the previous code was failing during type inference
due to mismatched structure. Fix is to use the X structure at both
points in the code.
2. In the second case, a naive transcription that subsitutes *nothing*
in for the omitted statements signified by "..." will actually
compile without an error. Furthermore, any pure code could also be
substituted for the ellipsis and the code would compile (as the
text already states). So to make the example more illustrative, it
would be better to include an impure callback, which makes the
potential for aliasing immediately obvious to the reader.
1. The section on trait definitions of static methods should include
a trait with a static method in the generated document.
2. The section on trait inheritance had a expression that appears
nonsensical ("let mycircle = @mycircle") in the generated document.
The text would be clearer (IMO) if we continued with the running
example of CircleStruct.
