Fork me on GitHub
#onyx
<
2017-11-06
>
asolovyov14:11:49

(I'm still working on updating onyx-http, and finally got back to it, so I'm trying to understand what approach should I use)

lucasbradstreet17:11:50

@asolovyov one of those is async, the other isn’t

lucasbradstreet17:11:46

@asolovyov good to hear that onyx-http is getting some love again.

lucasbradstreet17:11:00

I’m happy to talk to help get it into shape.

lmergen18:11:21

@asolovyov typically you want to free up the onyx process as soon as possible, since plugin a write/read in the plugin means it's blocking everything in the peer lifecycle (including heartbeats, etc). however, as you noticed, this comes at the price of complexity.

asolovyov19:11:53

so it's better to make onyx-http async, right? I wonder though if it's possible to make it simpler

asolovyov19:11:04

or maybe it's easier to copy and forget πŸ˜„

asolovyov19:11:55

I'm trying to base this thing on Aleph and one thing that bothers me is that I have no idea how to make timeouts which are async there...

michaeldrogalis20:11:50

I believe the last version (pre 0.10) was async, yes

asolovyov20:11:15

it was async, but the code there blows my mind πŸ™‚

asolovyov20:11:30

plus it uses jet, which is unmaintained now

lucasbradstreet20:11:45

aleph would be great.

asolovyov20:11:17

yeah I guess so, it almost works, even with retries, except for timeout I'm doing Thread/sleep right now and I guess that's suboptimal πŸ˜„

lucasbradstreet20:11:01

Great. If you want to push up a prelim version for me to quickly review, let me know.

asolovyov20:11:25

ok, let me get tests working first and then I'll show up with stuff πŸ™‚

michaeldrogalis23:11:34

Thanks, will be really happy to have the HTTP plugin back into the fold.