Welcome the channel on the development of Cro, a set of libraries for building reactive distributed systems, lovingly crafted to take advantage of all the Raku Programming Language has to offer (cro.services). This channel is being logged for historical purposes.
Set by lizmat on 24 May 2021.
02:44 japhb left 02:46 japhb joined 08:05 sena_kun joined 10:52 xinming left
Geth cro-http/debug-actions-hang: 0320210641 | (Jonathan Worthington)++ | .github/workflows/ci.yml
Disable Mac CI

It's broken, again. Somebody who uses Mac can fix it.
cro-http/debug-actions-hang: 2e429a76e5 | (Jonathan Worthington)++ | .github/workflows/ci.yml
Try avoiding CBOR issue with a newer Rakudo
cro-http/debug-actions-hang: 5f9dfc473d | (Jonathan Worthington)++ | .github/workflows/ci.yml
Try avoiding CBOR issue with a newer Rakudo
cro-http/debug-actions-hang: da01e501b0 | (Jonathan Worthington)++ | .github/workflows/ci.yml
Turn on verbose output
17:29 Xliff left 17:52 sena_kun left 17:57 sena_kun joined 18:52 sena_kun left 18:53 sena_kun joined
japhb What went wrong with CBOR? 19:00
20:05 sena_kun left
jnthn japhb: It fails tests on an old enough Rakudo (some type check error) 20:18
japhb Huh. I wonder what I accidentally depended on .... 20:38
Was it Cro::CBOR or CBOR::Simple that failed? 20:40
japhb is going to file himself a bug report
21:17 ecocode_ left 21:20 ecocode joined
jnthn it was in the Cro::HTTP build, you can look at the Actions output for the commit just before the one where I did the change to see exactly how it failed 21:46
japhb WTH? That is a very strange error. And the complaint it gives seems to be guarded against directly above! Hmmmmm. 22:19
AH! I believe I see the problem. The actual compared values aren't *themselves* Mu, they *contain* Mu. I'm guessing older Test modules did not have any protection against that. 22:26
It's not Test that changed, it's Mu: 750c5d04a9d72656a42800ef73a911cc82af998c fixed infix:<eqv> to not blow up on non-Any Mu. 22:42
If I'm reading the git log properly, that means that 2021.06 is the horizon for working CBOR::Simple tests. :-( 22:43
jnthn :( 22:45
That's not that new, but also not particularly old.
japhb The broken CBOR::Simple test got added three months later. I bet I never had a time where it failed for me. 22:46
What horizon would you like to have? I'll directly test against that one explicitly.
jnthn It'd be nice to be able to leave 2021.03 working at least, since that's what we've tested against for a while 22:50
The more useful answer is "what is packaged by recent distros", but I dunno if that information is collected somewhere.
japhb I could probably find out for some of the open source ones. 22:51
I tried looking back in the Cro::HTTP actions logs, but the last time the tests all worked it was past the detailed log retention time. :-( 22:52
I guess there's some value in actually having that answer (about the distro packaging). Worth some time not thinking about my RakuConf presentation. ;-) 22:53
gist.github.com/japhb/88322d76326d...108c90e567 23:32
A friend gave me a couple other common upstreams to add; doing so. 23:38