[Feature Request][move-compiler-v2] allow unconstrained generic inline functions #12680
Labels
enhancement
New feature or request
stale-exempt
Prevents issues from being automatically marked and closed as stale
Comments
|
I don't think we want C++ semantics. At the team discussion last October, there was even a large opposition against duck typing, which is is more harmless than this. It also collides with traits. Its either traits or ad-hoc polymorhism (aka C++), which is independent of inline functions. We decided against ad-hoc overloading last October as well. Those decisions are reflected in the blog article which documents traits as in Rust and Haskell the way we plan to go. |
|
I agree that this is hacky and is as bad as duck typing. We should avoid it and wait for traits. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
馃殌 Feature Request
While it's unsafe to allow instantiation of generics in general without traits or other constraints, we can probably allow inlined functions to be used with arbitrary generic type parameters (a la traditional C++ templates) by re-doing type inference/checking on the inlined code during inlining, and complaining then if the type parameter doesn't have needed methods/operations.
Might be a stop-gap until we have traits.
The text was updated successfully, but these errors were encountered: