March 15, 2019 
< | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 | 16 | 17 | 18 | 19 | 20 | 21 | 22 | 23 | 24 | 25 | 26 | 27 | 28 | 29 | 30 | 31 | >
March 15, 2019 [07:24:34] <papertigers> nbaksalyar: I have another question for you whenever you see this. The Streams doc says to return Async::NotReady when your stream is not ready to emit a value yet. But then it also says "streams must not return Async::NotReady unless Async::NotReady was obtained by an inner stream or future" which is a bit confusing. So in my case my wrapper stream is buffering up chunks of data from[07:24:40] <papertigers> hyper::Client until I have at least one full payload that can be parsed. So if I don't have all the chunks of data that make up the payload should I return Async::NotReady? That seems to break in my testing, but then I am unsure what you do if you are not ready to emit said value yet[07:34:13] <papertigers> it seems that doing loop { try_ready! } works. But maybe I don't understand why thats not repeatedly calling poll[13:39:46] *** wiedi <wiedi!~wiedi@ip5b4096a6.dynamic.kabel-deutschland.de> has quit IRC (Quit: ^C)[14:17:05] *** sjorge <sjorge!~sjorge@unaffiliated/sjorge> has quit IRC (Ping timeout: 244 seconds)[14:22:03] *** sjorge <sjorge!~sjorge@unaffiliated/sjorge> has joined #illumos-rust[15:48:09] *** awj <awj!~awj@c-75-72-241-91.hsd1.mn.comcast.net> has joined #illumos-rust[17:50:00] <papertigers> nbaksalyar: you can ignore that btw I found a doc that answers my question. And it comes down to tokio and how the reactor/executor works[18:14:19] <nbaksalyar> papertigers: just got a second to look into this, it's an interesting problem. the solution is connected to the task system, right?[18:24:48] <papertigers> exactly[18:25:03] <papertigers> I should have read this page first https://tokio.rs/docs/internals/runtime-model/[18:25:26] <papertigers> I didn't know how things registered to be notified. Oops :)