Roman Volosatovs (Aug 19 2025 at 18:05): Roman Volosatovs (Aug 19 2025 at 18:05):Sorry I had to drop earlier today, but what was the final conclusion on the discussion regarding the result/IO future associated with the requests and responses? Will consume-body take future<result<_, error-code>> as a non-optional parameter?
Alex Crichton (Aug 19 2025 at 18:06):no worries, and while I think we all lean towards non-optional I'd say it should be guided by perf/reality concerns of today
Alex Crichton (Aug 19 2025 at 18:07):e.g. with future.{splice,forward} intrinsics it should probably be non-optional, but without those it's unclear whether it's a pessimization to make it non-optional
Alex Crichton (Aug 19 2025 at 18:07):(I was also planning on writing up some issues to capture the results of our discussion to make sure they don't get lost)
Roman Volosatovs (Aug 19 2025 at 18:08):Awesome, thanks, Alex!
I will then just assume the result future will be passed into consume-body somehow in the implementation for now
Alex Crichton (Aug 19 2025 at 18:13):https://github.com/WebAssembly/wasi-http/issues/176
Last updated: Dec 06 2025 at 06:05 UTC