Steven Fackler

Polling a sleep is background progress IMO! :) My initial use case was purely around detecting disconnects/timeouts and things like that, but Boats's post on poll_progress made me feel like...

I think that `Ready(Ok(())` would mean "I'm done making progress in the background". If we don't require `poll_progress` to be fused, the caller would need to remember that and not...

Added a few bits of docs on return values, and poll_progress implementations to http-body-util combinators.

What's the path forward here? conjure-typescript generates code that follows the spec, but won't be able to deserialize errors produced by conjure-java: https://github.com/palantir/conjure-typescript/blob/develop/src/commands/generate/__tests__/resources/errors/primitiveError.ts. I'm trying to figure out what to...

I thought the entire purpose of the new error system was *so that* clients could introspect into them and e.g. have localized templates of the various error conditions. If clients...

With some adjustments this could also handle #1814.

There doesn't seem to be super clear documentation about it anywhere, but from some googling, it seems like the command allows one client to kick another off of a stream...

The main API question I have (which is more about IoSlice::advance than this I guess) is if we're okay saying that the content of the `bufs` slice is unspecified after...

It could for example perform vectored writes when the current cursor position is at a slice boundry and switch to a normal write_all if it's in the middle of one...

You are only missing out on the efficienty of vectored writes some of the time.