Harry Barber

There is also the opposite approach where test coverage doesn't cover anything outside the Smithy spec? For example, only check `Content-Length` in the case where `@requiresLength` is present. This approach...

Returning `UnknownOperationException` + 404 on routing failure sounds reasonable. Should this be something that is covered by Smithy documentation/protocol tests? I'm hesitant to commit to behavior on the `smithy-rs` side...

Smithy TypeScript generator has some relevant documented decisions: https://awslabs.github.io/smithy/2.0/ts-ssdk/error-handling.html#synthetic-errors

A corresponding change can be made to `TryStream`.

This would require an MSRV bump to `rustc` version 1.52. [Failing 1.51](https://rust.godbolt.org/z/MqY6GYhaM) [Passing 1.52](https://rust.godbolt.org/z/hh87aPojo)

The [smithy-rs](https://github.com/awslabs/smithy-rs) team have been playing with `metrics` for some time now and I'd like to contribute some points for consideration prior to stabilization: - Naming around `counter` and `increment_counter`...

I wasn't aware at the time, making a note of it now - if we _did_ split this trait in two it would become close to the suggestion by @olix0r...

Referencing related discussions: - https://github.com/tower-rs/tower/issues/626 - https://github.com/tower-rs/tower/issues/412 - https://github.com/tower-rs/tower/issues/408 - https://github.com/tower-rs/tower/issues/724

> How does this interact with drop and cancelation? Is it better or worse than the current model? I don't think the design here addresses the lower-level problems relating to...

I've now implemented a decent percentage of the existing `tower` middleware and [published it](https://docs.rs/burger/0.1.0-rc.0/burger/). Here are some obvious and subtle obstructions I've observed. Problems common to all `tower` "`async fn`...