David Barsky

While I don't have a solution for bootstrapping a compiler for compiling rules to WebAssembly (but if I were to make a guess, it would involve Starlark at _some_ level),...

> I like that usecase! Could you approximate it today by wrapping buck2 as a lib and writing your own "frontend" that calls your logic in rust (conceptually replacing https://github.com/facebook/buck2/tree/main/app/buck2_bxl...

It might be related to casing or non-ASCII characters in the returned headers. To debug this, I'd suggest trying to place all the headers you get from `curl` or Postman...

> Efforts for LSP/rust-analyzer support is unknown 🤷🏻‍♂️ It works today with a bit of configuration, but once I land [this branch](https://github.com/rust-lang/rust-analyzer/compare/master...davidbarsky:rust-analyzer:david/move-rust-project-generation-to-server) into rust-analyzer, the overall experience should be much...

> There is certain functionality in foundations that can't be implemented with `tracing`'s design. Also, there are certain unresolved bugs in `tracing` that prevented it's adoption for certain scenarios that...

(Sorry, I should've commented sooner.) In terms of approaches, I really the idea of introducing Valuable into tracing 0.1 under a feature flag. As for the actual mechanism, I think...

> I wouldn't like that to be the default, it would add a lot of extra printouts and I personally do use the default for my default logging. Adding it...

> To clarify why default would be useful for me: I don't use events at all, I use exclusively spans (as they give you anything events do, plus a timing...

I've confirmed that `#[async_trait]` doesn't have this issue. `cargo rustc --profile=check -- -Zunpretty=expanded` does show that the `#[instrument]` macro expands into something a bit more complex than `#[async_trait]`.

This might be caused by https://github.com/rust-lang/rust/issues/110486.