[Coke] /win 3 00:02
tellable6 2020-01-31T23:15:42Z #raku <uzl[m]> [Coke] re layout/css: Do you have any color scheme in mind? I myself find it too pale.
00:38 patrickb joined
patrickb o/ 00:38
tellable6 2020-01-31T23:28:29Z #raku-dev <rba> patrickb Please have a look at rakubrew.org - need your feedback.
patrickb .tell ugexe It seems as CircleCI is doing strange things with respect to credits usage for the rakudo org as well (circleci.com/gh/rakudo/rakudo/4183) Can you have a look at the credits page, if there is a hint to what might be wrong (circleci.com/gh/organizations/rakudo/settings)? I don't have permissions to see it. 00:43
tellable6 patrickb, I'll pass your message to ugexe
patrickb .tell rba Looking good! I don't see anything wrong. Thanks for getting it to work! What kind of feedback do you need? 00:45
tellable6 patrickb, I'll pass your message to rba
00:45 patrickb left, sena_kun left
rba patrickb: Still playing around a bit with having it installed on multiple servers... 00:45
tellable6 rba, I'll pass your message to patrickb
01:01 sena_kun joined 01:14 ufobat left
Xliff vrurg: You about to turn in? 02:20
vrurg Xliff: nah, I have own headache with console UI. Just keeping my eye on your work. 02:21
02:23 Ven_de_Thiel left 02:25 Ven`` joined
vrurg Just tried to share own experience. Anyway, wish you great luck on this! It's gonna be extremely useful. 02:25
02:28 patrickb joined
Xliff vrurg: Haha! Fair enough. Good luck with console. Wish I could figure out what's going on, here. 02:29
patrickb .tell rba I just pushed some more changes to the rakubrew.org repo. It's possible I get rakubrew to the point where I feel confident enough to invite others to test tomorrow. 02:30
tellable6 patrickb, I'll pass your message to rba
vrurg Xliff: try running tests with single test job
patrickb And now it's high time for me to go to bed. Night everyone! 02:31
tellable6 2020-02-01T00:45:42Z #raku-dev <rba> patrickb: Still playing around a bit with having it installed on multiple servers...
vrurg Normally they're ran in parallel and that's what might cause the problem.
02:31 patrickb left
Xliff That might be what it is. 02:31
Another thing is I am not seeing the "Multiple exceptions were thrown by LEAVE/POST phasers" in mine. 02:32
Hmm.... that gives me some idea of where it might be.
vrurg o/ :)
rather \o/ :D
Xliff vrurg: Any reason why I see better error messafes in AppVeyor than I do in my own build tree? 02:40
vrurg Xliff: no idea. I've been there, I felt that pain too...
Xliff Seriously? Never figured out the difference? 02:41
That's....seriously annoying.
vrurg Could it be different compiler version they use?
Xliff Well, I'm up to date on git master.... 02:42
vrurg In my case it wasn't my faulty code. Don't remember the details though. But yes, it was compiling/working for me but not under CI.
I mean, MSVC you use and the one in AppVeyor – are they the same? 02:43
Xliff Oh! No. I'm on Linux!
But on AppVeyor, I'm seeing "Multiple exceptions were thrown by LEAVE/POST phasers" which are RAKU level error messages. 02:44
I'm not seeing any of that, here.
Ahhh! 02:45
vrurg Windows, sir!
vrurg Ah, that's what you mean... You just didn't see the message? 02:46
I thought that the error they produce is not the one you get.
Xliff Yes.
02:46 sena_kun left
vrurg TEST_JOBS=1 also results in verbose output, BTW. 02:47
Xliff TEST_JOBS=1 HARNESS_VERBOSE=1 make test # Did the trick!
It's sooo nice when you can get actual error messages from make test... :/ 02:48
\o/ 02:51
02:59 sena_kun joined
Xliff Aaaannd..... Travis (like my system) hangs on t/02-rakudo/14-revisions.t 03:02
I had to restart to get it to work. 03:05
That doesn't sound like my changes, since those are limited to precomp.... :/
03:08 Ven`` left 03:29 AlexDaniel left 04:26 Kaiepi left, Kaiepi joined 04:43 Kaiepi left, Kaiepi joined 04:46 sena_kun left 05:00 sena_kun joined 05:04 Kaiepi left 05:06 Kaiepi joined 05:38 Kaiepi left 05:44 Kaiepi joined 05:46 Kaeipi joined 05:49 Kaiepi left, Kaiepi joined, Kaeipi left 06:46 sena_kun left 07:00 sena_kun joined 07:26 dumarchie joined 08:41 domidumont joined 08:45 sena_kun left 09:01 sena_kun joined 09:11 AlexDaniel joined, AlexDaniel left, AlexDaniel joined 09:21 entonian left 09:22 entonian joined
bartolin a quick question: S32-io/socket-recv-vs-read.t explodes for me on FreeBSD (on 6c5615ecf1 from yesterday). It passes if I s/localhost/ here: github.com/perl6/roast/blob/d72a16...read.t#L11 09:22
linkable6 (2020-01-31) github.com/rakudo/rakudo/commit/6c5615ecf1 Merge pull request #3455 from rakudo/cci-precomp-build
bartolin is that a known problem or should I open an issue. (If I'm not mistaken there has been a merge of IO::Socket::INET related stuff recently.) 09:24
bbl &
nine bartolin: in what way does it explode? Could also be an issue with your local configuration 09:29
10:16 dumarchie left
lizmat Files=1302, Tests=109837, 211 wallclock secs (28.64 usr 8.21 sys + 2934.21 cusr 278.86 csys = 3249.92 CPU) 10:40
bartolin nine: well, it complains about: No such method 'recv' for invocant of type 'Any' in block <unit> at t/spec/S32-io/socket-recv-vs-read.t line 30 10:41
I think, I'll go and open an issue.
It used to pass and the test file still passes on the JVM backend.
the host is configured for IPv6 and IPv4 -- that could be a difference to travis hosts 10:42
m: my $hostname = "localhost"; my $port = 5002; my ($send-rest, $client); IO::Socket::Async.listen($hostname, $port).tap(-> $conn { $conn.print("first thing"); await $send-rest; $conn.print("another thing"); $conn.close; }); $send-rest = Promise.new; $client = IO::Socket::INET.new(:host("$hostname:$port")) 10:44
camelia ( no output )
bartolin ^^ that code fails with "Could not connect socket: Connection refused" 10:45
10:45 sena_kun left 11:00 Ven`` joined 11:01 sena_kun joined
nine bartolin: what does your localhost resolve to? 11:01
bartolin I have two entries in /etc/hosts 11:05
::1 localhost localhost.my.domain localhost localhost.my.domain
I don't think I did something special wrt the network configuration. 11:08
I've openend github.com/rakudo/rakudo/issues/3457 11:12
hmm, I think I've updated from 11.3-RELEASE-p3 to 11.3-RELEASE-p6 between the two mentioned spectest. To be sure, I'll re-compile with an older commit and see if the tests still pass. 11:15
rba .tell patrickb Sure I have rebuilt the container and restarted it on the two docker runhosts. We need to consider to automate the container build and push it directly into the github docker registry. And I will automate the pull+restart on the runhosts. 11:20
tellable6 rba, I'll pass your message to patrickb
nine bartolin: FWIW I've failed to reproduce the issue on Linux. So I guess it's FreeBSD specific. 11:29
bartolin nine: thanks. I've added that info the the issue. 11:40
Also, I've bisected it to bae5fc7d49
linkable6 (2020-01-28) github.com/rakudo/rakudo/commit/bae5fc7d49 Bump NQP to get Moar support for IP6NS changes
bartolin (which looks reasonable)
AlexDaniel is linkable6 behaving well? I wasn't paying attention in the last few days… 11:42
11:50 patrickb joined
sena_kun releasable6: status 11:50
releasable6 sena_kun, Next release in ≈21 days and ≈7 hours. 1 blocker. 0 out of 57 commits logged
sena_kun, Details: gist.github.com/45490b65094e845e7c...adbbb1a7ce
12:02 patrickb left 12:23 lizmat left 12:36 patrickb joined
patrickb o/ 12:36
tellable6 hey patrickb, you have a message: gist.github.com/ad68ac5cbeb7140ac1...0ad8ed7710
patrickb vrurg or lizmat: Can you tell me what the typical (XCode) toolchain on MacOS is? (the MoarVM Configure.pl --toolchain parameter) 12:37
I want to have the toolchain used as specific as possible in the precomp package creation process. Maybe `--toolchain=posix --compiler=clang` ? 12:39
12:46 sena_kun left
Geth rakudo: d01d4b2664 | (Patrick Böker)++ | 2 files
Fix errors in the binary-release Windows build script

I still had no chance to try this on CircleCI, but manually following the instructions revealed these errors. Also explicitly specify the compile toolchain on linux, so the script is not so much dependent on what shows up in the environment.
13:01 sena_kun joined 13:06 patrickb left 13:23 lizmat joined 13:28 lucasb joined 13:29 Ven`` left 13:42 patrickb joined 13:48 patrickb left 13:51 patrickb joined
patrickb .tell rba I have created binary release archives for the 2020.01 release. They can be downloaded here: patszim.volans.uberspace.de/patclo...EJLQDZcjbf You can put them on rakudo.org 14:27
tellable6 patrickb, I'll pass your message to rba
sena_kun patrickb++ 14:28
patrickb The MacOS and Linux builds were done automatically using the new CircleCI pipeline. Sadly I had to do the Windows build manually as CircleCI currently refuses to do Windows builds because something with how it calculates credits is off. 14:31
But hey! I was able to do a MacOS build without owning a Mac machine. That's quite a thing!
sena_kun That's quite a thing indeed. Great job on improving these things. \o/ 14:33
Geth rakudo: aaf9a93e46 | (Patrick Böker)++ | docs/release-guide-binary.md
Update the binary release guide

Now it includes instructions on how to use the CircleCI build pipeline.
14:45 sena_kun left
lizmat bisectable6: $_ = "a1xa2"; say m/a./; say $/.to; say m/a./ 14:45
bisectable6 lizmat, On both starting points (old=2015.12 new=aaf9a93) the exit code is 0 and the output is identical as well
lizmat, Output on both points: «「a1」␤2␤「a1」␤»
lizmat docs.raku.org/language/regexes#Continue states that the default for :continue is $/.to 14:46
$_ = "a1xa2"; say m/a./; say $/.to; say m:c(2)/a./
evalable6 「a1」
lizmat apparently, this does not work, and hasn't since the dawn of time, so the documentation is wrong 14:47
$_ = "a1xa2"; say m/a./; say m:c($/.to)/a./
evalable6 「a1」
AlexDaniel lizmat: huh? 14:49
but it does seem to work correctly? 14:50
lizmat $_ = "a1xa2"; say m/a./; say m/a./
evalable6 「a1」
lizmat that should read: «「a1」␤「a2」␤»
according to my interpretation of the documentation 14:51
m: $_ = "a1xa2"; say m/a./; say m:c/a./ 14:52
camelia 「a1」
lizmat aaah... ok
and m:c/./ codegens the $/.to call 14:53
15:02 sena_kun joined
rba patrickb: Your new binary builds are uploaded and ready to use on rakudo.org 15:12
tellable6 2020-02-01T00:45:00Z #raku-dev <patrickb> rba Looking good! I don't see anything wrong. Thanks for getting it to work! What kind of feedback do you need? 15:13
2020-02-01T02:30:36Z #raku-dev <patrickb> rba I just pushed some more changes to the rakubrew.org repo. It's possible I get rakubrew to the point where I feel confident enough to invite others to test tomorrow.
2020-02-01T14:27:45Z #raku-dev <patrickb> rba I have created binary release archives for the 2020.01 release. They can be downloaded here: patszim.volans.uberspace.de/patclo...EJLQDZcjbf You can put them on rakudo.org
Geth rakudo/releaseable-tweaks: c854b1278c | Altai-man++ | tools/releasable/Sakefile
Make release SSH configurable via envvars
rakudo/releaseable-tweaks: b247fc2a28 | Altai-man++ | tools/releasable/Sakefile
Update tarball distributors
patrickb rba: Thanks! 15:19
rba patrickb: jw
patrickb rba: I'm currently preparing a test release version 1 of rakubrew. How should I upload the files to the server? Just as we just did with the rakudo archives? You mentioned ssh access some time ago. 15:20
Geth rakudo: Altai-man++ created pull request #3458:
Releaseable tweaks
rba patrickb: Hmm. No. I decied to use rakubrew.org as the first site to only dockerized and there are restricted access currently to the RancherOS docker runhosts. 15:21
patrickb: That's why I started thinking about how to handle the persistent data part of the containers... 15:22
patrickb: So far I just logged in the the only user using ssh keys and put the volumes in the home of the ranger user manully on the two runhosts. 15:23
patrickb are the releases part of the containers or mounted in?
rba patrickb: This means I'm open to discuss how we like to handle this...
patrickb If they are mounted in they can in principle also reside on a different host
rba patrickb: I use your containerfile. So they are mounted in.
Geth rakudo/releaseable-tweaks: 2a50b65d2b | Altai-man++ | tools/releasable/Sakefile
Specify scp port to use
rba patrickb: One host is in New York and the other one in Singapore.
patrickb: I use docker-compose to get it run with traefik. And I consider doing something like binfalse.de/2017/01/24/automatical...er-images/ 15:25
patrickb Ah! There are multiple deployments. So the upload needs to happen on both...
rba: This is very interesting. :-)
rba patrickb: One task on my list is to make sure we have things running redundant.
patrickb: At the moment I use just multiple A / AAAA DNS records for the hosts. 15:26
patrickb: try: $ host rakubrew.org
patrickb Two hosts show up. I didn't know this was possible... 15:27
rba patrickb: But maybe I configure the current webhost (which is Ubuntu) to be a docker runhost as well and there we could login and then like remote control the two addtional RancherOS runhost remotely...
patrickb: Two time https is not perfect, as it might lead to ssl handshake issues when one is down. Yet better than going offline totaly... 15:28
patrickb: I might try to use rsync to keep the volumes of the containers in sync... 15:33
patrickb rba: So then there is a master server for the synching. Is a plausible approach.
afk for a bit... 15:35
back 16:14
16:17 zostay left, zostay joined 16:46 sena_kun left 17:00 sena_kun joined
patrickb .tell rba The same URL as above now contains a rakubrew-1.tgz with a new release. Can you put them on rakubrew.org? 17:11
tellable6 patrickb, I'll pass your message to rba
17:20 patrickb left 17:25 patrickb joined 18:07 lucasb left 18:08 lucasb joined 18:11 patrickb left 18:12 patrickb joined 18:41 kawaii left 18:42 kawaii joined 18:45 sena_kun left 18:47 Kaeipi joined, Kaiepi left 18:48 Kaeipi left, Kaeipi joined 18:55 patrickz joined, domidumont left 18:56 domidumont joined 19:00 sena_kun joined 19:26 Kaeipi left, Kaiepi joined 19:45 domidumont left 19:56 Kaiepi left, Kaiepi joined 20:00 Kaiepi left 20:01 Kaiepi joined 20:25 Kaiepi left, Kaiepi joined 20:27 lucasb left 20:40 Kaiepi left 20:43 Kaiepi joined 20:46 sena_kun left 20:59 sena_kun joined 22:05 Kaiepi left 22:06 Kaiepi joined 22:15 Kaiepi left 22:17 Kaiepi joined 22:27 patrickz left 22:28 patrickb left 22:46 sena_kun left 23:00 sena_kun joined 23:09 Kaiepi left 23:10 Kaiepi joined 23:22 Kaiepi left, Kaiepi joined 23:37 Xliff left