Geth rakudo: 0e2485a83f | (Tom Browder)++ (committed using GitHub Web editor) | docs/Building-Rakudo.md
fix typo
 01:23
vrurg .tell lizmat blogs.perl.org/users/vadim_belman/2.../post.html – that's the best I can get for now. 04:37
tellable6 vrurg, I'll pass your message to lizmat
vrurg .tell lizmat there seem to be better version: blogs.perl.org/users/vadim_belman/2.../post.html 04:39
tellable6 vrurg, I'll pass your message to lizmat
nine Intriguingly the new csv-ip5xs.pl issue is much easier to reproduce with the JIT disabled and appears even with spesh disabled, despite appearing only after a lot of successful runs of the loop. Makes me suspect some GC related issue. 08:32
Alas, setting MVM_GC_DEBUG to 3 doesn't seem to provoke it either 08:36
|Tux| Rakudo version 2019.11-271-g0e2485a83 - MoarVM version 2019.11-93-g7a93b2897
csv-ip5xs-206.307 - 6.395
csv-parser22.734 - 22.934
csv-test-xs-200.417 - 0.424
test7.009 - 7.381
test-t1.740 - 1.754
test-t --race0.810 - 0.831
test-t-2029.951 - 30.080
test-t-20 --race9.637 - 9.747
 09:07
And this one:
Cannot iterate object with P6opaque representation (Scalar)
in sub no-args at /pro/3gl/CPAN/Text-CSV6/lib/.precomp/4B200A6BA6FA8B6D4F0968BCFE7815E1375DD6D3/3E/3ECC0B3D017463A31ED44AC02473634C96813BB1 line 1
in block <unit> at csv-ip5xs.pl line 13
and as yesterday 09:08
Too many positionals passed; expected 1 argument but got 2
in sub one-arg at /pro/3gl/CPAN/Text-CSV6/lib/.precomp/4B200A6BA6FA8B6D4F0968BCFE7815E1375DD6D3/3E/3ECC0B3D017463A31ED44AC02473634C96813BB1 line 1
in block <unit> at csv-ip5xs.pl line 13
lizmat .tell vrurg will feature it in the next RW 09:40
tellable6 lizmat, I'll pass your message to vrurg
lizmat . 09:41
tellable6 2019-12-17T04:37:51Z #raku-dev <vrurg> lizmat blogs.perl.org/users/vadim_belman/2.../post.html – that's the best I can get for now.
2019-12-17T04:39:57Z #raku-dev <vrurg> lizmat there seem to be better version: blogs.perl.org/users/vadim_belman/2.../post.html
patrickb tadzik: I brought up the rakudobrew -> rakubrew rename discussion recently: colabti.org/irclogger/irclogger_lo...12-12#l170 10:41
tellable6 patrickb, I'll pass your message to tadzik
patrickb tadzik: I don't want to just decide this over your head. Can you have another look? 10:42
tellable6 patrickb, I'll pass your message to tadzik
jmerelo I'm trying to wrap my head around this change in raku 6.d: "Native attributes can be used as bind targets in parameters rakudo" 17:38
If I understand it correctly, it means that you can bind parameters to a native attribute, for instance in a submethod like here:
m: class Foo { has num $.numillo; submethod BUILD( :$!numillo = 3.5e0 ) {}}; say Foo.new.perl 17:39
camelia Foo.new(numillo => 3.5e0)
jmerelo However, the problem is that that particular way of working has been there since 2016.10, way before 6.d gist.github.com/Whateverable/c3552...ee596cc90d
I am probably missing something here, but what is it? 17:40
jnthn jmerelo: But the previous language release was 6.c, in 2015.12, so in terms of language releases 6.c is the first one it appears in. 17:43
jmerelo jnthn: so that means that was simply implemented early and the example is correct 17:44
AlexDaniel I guess 6.c didn't have any tests for it one way or another, and 6.d now has tests for that feature 17:45
which is how rakudo got away implementing it early
jmerelo AlexDaniel: Oh, great. Thanks!
jnthn Not really sure what "implemented early" means; by definition, everything must be implemented before it can appear in a language release. :) 17:59
(Since a feature must have been proven implementable by at least one implementation.)
jmerelo jnthn: I kinda thought that features mentioned in 6.d would be available in 2018.11 or shortly before. I was wrong. 18:02
And here's a new one: I can't install Pod::To::Cached from zef even if it's in the ecosystem raw.githubusercontent.com/perl6/Po...META6.json 18:03
zef search returns only versions in the local cache, which are a newer version.
It's not in the ecosystem index: ecosystem-api.p6c.org/projects1.json 18:04
It's in here: github.com/ugexe/Perl6-ecosystems/...r/p6c.json 18:08
jmerelo m: say $?DISTRIBUTION.meta<version> 18:47
camelia Nil
vrurg AlexDaniel: do you have a quick way to connect nqp-configure repo to Geth? 19:07
tellable6 2019-12-17T09:40:56Z #raku-dev <lizmat> vrurg will feature it in the next RW
AlexDaniel Geth: help 19:08
tellable6 AlexDaniel, I'll pass your message to Geth
AlexDaniel vrurg: you just add a webhook
AlexDaniel vrurg: done 19:09
jmerelo AlexDaniel: while you're at it, can you connect Raku/doc too? 19:10
AlexDaniel jmerelo: it should be working 19:11
yeah, it was working just recently
Geth is simply offline, for some reason
vrurg AlexDaniel: "just" doesn't apply to me because I never used them (ashamed, yes) and not ready to spend time learning it.
AlexDaniel tyil: ping? ↑
jmerelo Ah, OK :-)
vrurg "them" – webhooks.
AlexDaniel vrurg: Settings/Webhook from any repo if you have enough privs, you just give it a link to send payloads to and that's it 19:13
vrurg Ah, ok. I googled wrong docs then. 19:14
Waill wait for Geth to get back online then. 19:15
tyil AlexDaniel: ack 19:20
tyil perhaps fallout from the netsplit earlier today, geth doesn't seem to reconnect on those (neither do my other IRC::Client based bots) 19:21
vrurg: I present to you, Geth 19:22
vrurg tyil, Geth: my pleasure! :D 19:23
Geth: help
Geth vrurg, Source at github.com/perl6/geth To add repo, add an 'application/json' webhook on GitHub pointing it to geth.svc.tyil.net/?chan=%23perl6 and choose 'Send me everything' for events to send | use `ver URL to commit` to fetch version bump changes
tellable6 2019-12-17T19:08:24Z #raku-dev <AlexDaniel> Geth: help
vrurg AlexDaniel: interestingly enough, I don't have access to nqp-configure settings. 19:24
vrurg tyil: BTW, with regard to disconnecting Geth, perhaps it makes sense to auto-restart it each hour? 19:47
tyil vrurg: I could do that, but it sounds like a dirty workaround that also generates connect/disconnect spam needlessly 19:54
I'd prefer to have a way to check if its doing its job and only kill it if needed, or just have IRC::Client do autoreconnecting after netsplits
Geth rakudo: kalkin++ created pull request #3360:
Pod::To::Text display enums		 19:59
Geth rakudo: kalkin++ created pull request #3361:
Pod::To::Text display Subsets		 20:03
vrurg tyil: Sure, that'd be better. I was proposing a quick workaround until real solution gets in place. 20:12
tyil for that it'd be fair 20:13
I hope netsplits arent common enough to warrant it anytime soon, though :p
patrickb rba: Can you update stage-rakudo.rakulang.site/ once more please? 20:30
rba patrickb: updated & restarted, please check 20:32
patrickb rba: Looks good. Thanks! 20:33
rba patrickb: somewhat the "binary" all release is now pointing to github, where there are no binaryies. 20:36
patrickb: may we ajust the redirects in the nginx.conf as well? 20:38
patrickb rba: There are no binaries yet at all! We still have to create and upload some. 20:40
rba patrickb: so where to put the binaries? on github?
patrickb We could do both. Then we keep the backup aspect of github. 20:41
rba patrickb: If I click on the MacOS tab on "All releases" under "Binary Releases" I would expect to find the latest (or all) prebuilt dmg archives I can just download. 20:43
patrickb rba: They would show up right there if there were any releases on the server.
rba Ok and where I need to put them? 20:44
Let's do it.
patrickb same folder the src files are in
MasterDuke pmurias: think it would make sense to try and have rakudo listed here? github.com/neomatrix369/awesome-graal 20:46
tellable6 MasterDuke, I'll pass your message to pmurias
rba patrickb: Sorry, I see the issue. We have only rakudo star as .dmg and .msi. 20:47
patrickb rba: Rakudo gained the ability to have relocatable builds only in the latest release. So there is no process of creating the binary packages established yet. 20:48
El_Che is working on automating that task. 20:49
rba: I can create archives for Windows and Linux. I have no access to a MacOS machine though. 20:50
rba patrickb: MacOS is comming :-) I see the point.
[Coke] there are some folks in channel with macs would would probably volunteer to test something. 20:52
rba patrickb: With the idea in mind that we are on the way to glue the rakudo.org website into a docker container as well, we might quickly think how to map the folder with the downloadable binaries into it.
patrickb .tell El_Che The rakudo-pkg releases are missing the files in rakudo-repo/tools/binary-release/Linux. 20:53
tellable6 patrickb, I'll pass your message to El_Che
rba [Coke]: I have MacMini resting around. But I need to put it into a separate network before giving access to community members.
rba Idea would be to involve it somehow into a pipline to do automated builds for MacOS as well. 20:55
patrickb .tell El_Che The rakudo-pkg files are now named "rakudo-moarvm-2019.11-..." instead of "rakudo-moar-2019.11-..." different from what I answered you. Was that accidental? (I don't really prefer one over the other, we just have to keep it consistent) 20:56
tellable6 patrickb, I'll pass your message to El_Che
patrickb rba: El_Che is currently utilizing Travis as the platform to create the packages automatically.
Geth nqp-configure: 1a9539a8f6 | (Vadim Belman)++ | lib/NQP/Config.pm
Test for Microsoft Compiler version to be no less than 19.0

We don't actually compile with older versions.		 21:06
nqp-configure: 78837f445b | (Vadim Belman)++ | lib/NQP/Config.pm
Little improvement for error message

Always show what we really expect.
patrickb I'll boot my Windows and see if I can revive my MSVC again...
vrurg patrickb: could you have a look at github.com/rakudo/rakudo/pull/3359? if you can. 21:19
patrickb vrurg: I'm having a look 21:42
patrickb vrurg: Review added. 22:46
Are test failures on Windows in 04-nativecall/11-cpp.t and 13-cpp-mangling.t known? 22:51
They succeed on CircleCI. :-( 22:52
patrickb Wait CircleCI is only Linux and Mac... 22:54

Powered by Cro and the Raku® Programming Language.

Please report any issues / comments / feature requests as an issue on App::Raku::Log.

Thank you!