Taiki Endo
Taiki Endo
Seems Miri failure is due to detached threads in doctest: https://github.com/rust-lang/miri/issues/1371 A workaround that I sometimes use is adding sleep to the end of test: https://github.com/crossbeam-rs/crossbeam/pull/891
At first glance, I feel that generic metadata is a bit overkill/complicated, but I have no strong opinion on this.
@smol-rs/admins Any thoughts on this?
I'm not sure if we need to introduce such a breaking change to support WASM. Currently, async-io fails to compile on WASM anyway, so I think just disabling some APIs...
Why does this add both freestanding function and method?
I guess it is only happening with methods that take two or more futures/streams. (e.g., [catch_unwind](https://docs.rs/futures-lite/latest/futures_lite/future/trait.FutureExt.html#method.catch_unwind) is trait method only) There has been discussion in the past in `futures` that...
It seems you have removed the trait method, but I have the impression that `into_` is a prefix that is often used in the method (e.g., `String::into_bytes`, `{BufReader,BufWriter}::into_inner)`.
Maybe `stream::from_future`? That said, I have no strong opinion on this as I tend to prefer the `futures`'s APIs that take the future as an argument (i.e., I think `stream::once`...
@smol-rs/admins: Any thoughts on the API here?
Published a new version of async-lock including https://github.com/smol-rs/async-lock/pull/27.