»ö« Welcome to Perl 6! | perl6.org/ | evalbot usage: 'p6: say 3;' or rakudo:, or /msg camelia p6: ... | irclog: irc.perl6.org or colabti.org/irclogger/irclogger_logs/perl6 | UTF-8 is our friend!
Set by moritz on 22 December 2015.
00:18 w_richard_w1 joined 00:20 w_richard_w left 00:24 w_richard_w1 left 00:35 eliasr left 00:36 w_richard_w joined, kurahaupo_ left 00:37 kurahaupo joined 00:39 ribasushi left 00:40 kurahaupo_ joined 00:41 aborazmeh joined, aborazmeh left, aborazmeh joined 00:43 lichtkind left, lichtkind joined, kurahaupo left, lichtkind left, espadrine left 00:52 pierre_ joined 01:13 a3r0 joined 01:32 AlexDaniel left 01:36 markong left 01:40 foop joined 01:47 khisanth__ joined 01:51 foop left 01:54 aborazmeh left 02:10 w_richard_w left 02:27 perlawhirl joined
perlawhirl .tell jkramer Yes, I am in Australia 02:28
yoleaux perlawhirl: I'll pass your message to jkramer.
02:56 ilbot3 joined, ChanServ sets mode: +v ilbot3 03:00 wamba joined 03:05 dogbert11 joined 03:06 Seeder joined, Seeder left 03:44 ufobat___ joined
bocaneri perl 6 seems a bit slow. :\ 03:46
03:48 skids left
thowe It wil get faster. Speed optimizations haven't been a priority except maybe lately (I think I have heard that, I' 03:49
I'm not really in-the-know).
03:53 khw left
timotimo bocaneri: most of the time you can reach a dramatic speed-up with only tiny changes to the code. sometimes all that you need is to complain about a specific method or sub being slow here on the channel and the next day you'll find it to be between 1.5x and 100x faster 03:53
thowe and isn't there a profiler built in?
timotimo that's correct 03:55
bocaneri I have no idea how to use a profiler (yet) and it's not actually a complaint so much as an impression.
timotimo profile-m: (^100_000).grep(*.is-prime).head(100).tail(1).say 03:56
(just a totally random workload)
i'm not actually sure that's hooked up still
camelia: help
camelia timotimo: Usage: <(nqp-jvm|debug-cat|rakudo-moar|star-m|prof-m|rakudo-jvm|nqp-moarvm|nqp-js|p5-to-p6|sm|nqp-mvm|j|nqp-m|star|perl6|r-jvm|rakudo|rj|p6|m|r-m|r|master|nqp|p56|nqp-q|rm|r-j)(?^::\s(?!OUTPUT)) $perl6_program>
timotimo prof-m: (^100_000).grep(*.is-prime).head(100).tail(1).say
camelia No such file or directory
.. Prof: p.p6c.org/60ab517
timotimo well, that isn't working correctly it seems 03:57
camelia (timeout)Can't exec "./rakudo-inst/bin/perl6-m": No such file or directory at lib/EvalbotExecuter.pm line 206.
cat: /home/camelia/rakudo-inst/revision: No such file or directory

now running scp...
lost connection
.. Prof: p.p6c.org/60ab52d
thowe oh, while I'm thinking of it.. The heck is "perl6-m"?
timotimo just "perl6 running on moarvm"
as opposed to perl6-j which is rakudo on the jvm
oh no, oh crap
camelia is feeding herself code?!
i might have misunderstood? 03:58
thowe I'm trying to figure out why it's explicitly called that way in this one example in Moritz' book.
timotimo anyway, you get the profiler by calling your code like perl6 --profile -e '(^100_000).grep(*.is-prime).head(100).tail(1).say'
it'll spit out an html file which is a little app thingy that'll let you go through all the data 03:59
i usually recommend going to the Routines tab and sorting by exclusive time
04:17 pierre_ left 04:35 isBEKaml left 04:47 releasable6 joined 05:05 lelf joined, wamba left 05:36 curan joined 06:05 andrzejku joined 06:08 wamba joined 06:25 w_richard_w joined, pierre joined, imcsk8_ joined 06:27 lelf` joined 06:29 imcsk8 left, lelf left 06:30 lelf`` joined 06:32 lelf` left 06:43 kaare__ is now known as kaare_ 06:44 wamba left 06:58 lalitmee joined
bocaneri anybody have any clue how to hook perl6 up to postgresql with something that has all the power of perl5's DBI? 06:59
lookatme you mean use Perl5 module in Perl6 ? 07:01
yoleaux 16 Mar 2018 10:52Z <Zoffix> lookatme: "I don't like this warning message :)" You should consider adding your voice to R#1356 github.com/rakudo/rakudo/issues/1356
lookatme .tell Zoffix Other guys told me the issue has reported. I will do that in next time :) 07:02
yoleaux lookatme: I'll pass your message to Zoffix.
07:04 domidumont joined 07:07 aindilis joined 07:09 domidumont left 07:10 domidumont joined 07:13 lelf``` joined 07:15 lelf`` left 07:17 lelf``` left 07:21 andrzejku left
buggable New CPAN upload: perl6/Algorithm-Evolutionary-Simple-0.0.1.tar.gz by JMERELO cpan.metacpan.org/authors/id/J/JM/...0.1.tar.gz 07:21
07:23 eythian joined, eythian left, eythian joined 07:24 darutoko joined 07:28 robertle joined 07:33 vike joined 07:34 ribasushi joined 07:53 wamba joined 07:54 vike left 07:59 vike joined
stmuk_ bocaneri: some of the power of DBI is found in DBIish 08:02
bocaneri I've not looked at that yet, looked a bit underwhelming on first glance. 08:03
08:03 neilb joined, neilb left 08:04 amalia___ joined 08:05 lowbro joined 08:18 pierre left 08:22 eythian left, eythian joined 08:23 pierre joined
El_Che bocaneri: there is also modules.perl6.org/dist/DB::Pg. I haven't tried it, though. Just saw it mentioned here. 08:26
bocaneri I just discovered zef, use it to list and found what you're talking about. Perl 5 has spoiled me rotten; all the versions to the packages I see listed are 0.something. 08:27
El_Che that's 28 years worth of modules indeed 08:28
bocaneri One of the things I'm planning on looking at as I continue to swear at perl6 is inlining some perl5. 08:29
El_Che I have ended writing perl 5 before when I wanted to write in perl 6 or go.
some things are trivial to roll your own
some aren't
some perl 6 modules are very nice 08:30
bocaneri Yup. Some aren't. I have no idea how to talk to Pg through a raw socket.
El_Che it depends on the usecase, I guess
08:31 ufobat___ is now known as ufobat
bocaneri I'm certainly not above using perl6 for the raw processing and setting up a perl5 server to act as a go-between. :D 08:31
El_Che once you use it, you'll maybe need something and create your own module for us to use :) 08:32
bocaneri If I succeed in cobbling something together that won't get me lynched, you're goddamn right I'll share! 08:34
08:35 markong joined
bocaneri I think mst could confirm, though, that my coding skills leave a LOT to be desired. 08:36
I wonder how well these inline modules work. 08:41
masak bocaneri: knowing one's coding skills leave a lot to be desired is a sign of programmer maturity ;) 08:49
bocaneri Oh, I'm "mature", alright. Couple years I'll be eating off the senior citizen's menus.
masak for example, I've been at this for decades. and I'm woefully aware of how inadequately I code. :) 08:50
bocaneri But... but... but... it can be so much FUN!
masak in fact, the only thing I'm getting better at is leaving important tasks to tools 08:51
08:53 pmurias joined
bocaneri Well... last thing I gid with perl5 was give up on trying to embed something in something else, so what I embedded instead was something that launches a subshell to call ncat to pipe output from the app to a 'broadcast' ncat. I hooked perl up to that same broadcaster to do my "real" work (DBI crap). Perl has no idea where the data are coming from, and the app generating them has no idea it's hooked up to a Pg DB. 08:53
pmurias github.com/rakudo/rakudo/issues/1626 - rakudo bug found by js backend work, got a plan for a fix, will attempt after a heary breakfast 08:54
yoleaux 07:00Z <nine> pmurias: The difference between :api and an extra name part is a very explicit and unified way to express "same goal, but incompatible API". Functionally it's the same as having Foo, Foo2 and Foo3. People are hesitant to tuck a number onto a name. So instead they use convention to assign additional semantics to a version number. :api is explicit and has a chance to become ubiquitous
El_Che releasable6: status 08:55
releasable6 El_Che, Next release will happen when it's ready. Blockers: github.com/rakudo/rakudo/issues?q=...%9A%A0%22. Changelog for this release was not started yet
El_Che, Details: gist.github.com/73f34f9d894aa93239...e6b1d7a817
masak "Next release will happen when it's ready"? :) 09:04
sounds like a message that used to say how many days there are left, but which then was changed in order not to stress people out...
09:08 scimon joined, darutoko- joined
lookatme :) when it's ready haha 09:09
masak it's like an API that used to say "The Empire State Building is 381 m tall" but that now says "The Empire State Building is as tall as itself" 09:11
09:11 darutoko left 09:18 neilb joined 09:22 dakkar joined
moritz .seen sergot 09:24
yoleaux I saw sergot 14 Mar 2018 21:46Z in #perl6: <sergot> ufobat_: ping
09:25 pierre left 09:26 pierre__ joined, rindolf joined 09:34 lalitmee left 09:38 w_richard_w left, DarthGandalf joined 09:39 AlexDaniel joined
Ulti has anyone upgraded macOS and have a vague checklist to get back to building 09:39
Im constantly getting the libc can't be found you should sign the XCode license agreement warning
but its definitely all sorted, so wondering if its just a path issue 09:40
El_Che I have only used rakudo on macos, not build it. Check with stmuk, he create macos binaries 09:46
stmuk_ yeah I don't think gcc generally works after a macOS upgrade 09:47
and there is a licence agreement if you try gcc -o hello hello.c you have to answer
09:48 Matthew[m] joined
Ulti its always been clang for me 09:48
and I've already sorted out the license agreement :/ 09:49
yeah I can compile my own stuff with clang or gcc 09:50
:<
same shell and environment
stmuk_ sorry I meant clang 09:51
Ulti ok so it looks like library paths
because Apple freaking love moving everything even with a minor OS change
stmuk_ is this a brand new macOS upgrade? I don't use Macs so much anymore 09:52
libc not being found sounds like headers since if it really wasn't there nothing would work :) 09:53
Ulti yeah exactly
its just headers it cant find 09:54
just running xcode-select --install as I suspect XCode totally lied about what it was doing
I really hate this corner of macOS
El_Che Ulti: Apple hates you back
stmuk_ I used to selectively install parts of macOS but gave up on that 09:55
macOS xcode I mean
El_Che you damn advanced user :)
Ulti I only use XCode to bootstrap a dev environment nothing else
stmuk_ my pet perve was not being able to dtrace anything with #!/bin/sh for "security reasons" 09:56
Ulti well yeah Apple need to secure their margin as much as the next mega corp with enough cash to purchase a small country 09:57
it annoys me they make exceedingly good computers other than this
stmuk_ hmm I probably haven't tried 10.13.3 but I'd be unsurprised if there weren't odd issues
they lost me with the 2mm keyboard travel 09:58
Ulti Windows 10 is also increasingly dev friendly with the new Ubuntu layer
stmuk_: I actually love the new keyboard type, I have really reduced sensitivity in my fingers from RSI the hard bottom out quickly is perfect 09:59
YAY that did it
stupid XCode 10:00
ok so if anyone else comes asking thats the answer: `xcode-select --install`
stmuk_ Ulti++
Ulti as well as the follow up command to sign the license agreement: `sudo xcodebuild -license accept` 10:01
weird how homebrew had zero issues
stmuk_ did the upgrade totally remove xcode? I'm sure I've seen that before 10:03
Ulti every time you do an OS update it nukes XCode and even if its installed it has to reinstall the CLI stuff and half the build system
its so stupid 10:04
I did this via the GUI but clearly it didnt properly setup the CLI environment
I cant see any changes to my user environment so really not sure what running it from a shell accomplished 10:05
it did download a tonne of additional software
10:06 El_Che joined
El_Che stmuk_: that bad keyboards? 10:07
stmuk_: at work we have windows laptops (I have a illegal Linux install to get work done), but they also looking into providing macs.
AlexDaniel clickbaits insights.stackoverflow.com/survey/2018 10:08
a lot of interesting stuff there
stmuk_ El_Che: the new butterfly mac kb has v low travel which many people hate 10:09
El_Che: they also removed many ports (HDMI, SD card etc.) from the Pro laptop
El_Che stmuk_: I have now an air to do some macos tests, which I use on the train to write. Keyboard on that generation is fine 10:11
only downside is the screen resolution, besides that great laptop
stmuk_ El_Che: yes that's fine ... also rumours of a coming upgrade to that line
El_Che If would be a hub step down if a laptop that costs 2,5x is less user friendly 10:12
stmuk_: no, the "real" work laptop would be macbook pro as an alternative for dell latitudes (which are fine nowadays)
10:12 andrzejku joined
AlexDaniel I am blind / have difficulty seeing 1.4% 10:13
stmuk_ I've moved to a Dell XPS13 which works great with linux
AlexDaniel that's very interesting
El_Che stmuk_: management: you can't have the xps 13 instead of the latitude 12 (even if we get both models at -60% price), but you are free to get a full stocked macbook pro 10:14
sigh
I am not buying dell inspirons (for home) anymore, they are pretty crappy 10:15
AlexDaniel no way! The majority of respondents have better lifestyle than me. “How Often Do Developers Skip Meals To Be Productive?” Never 63.6% 10:19
63% never do that! omg 10:20
Ulti yeah the Dell XPS are what I used to try and purchase as the non Apple option in my last work place 10:21
at least for engineers
its a shame they stopped doing the Linux only developer version for a couple hundred less in the UK
stmuk_ I got the windows one in the UK for less than £1000 10:35
El_Che around 500€ by now? :P 10:37
stmuk_: that's a great price! previous model?
stmuk_ it wasn't prev model at the time I bought it ... 7th gen I5 10:39
it didnt have the ultra high res display
10:46 pierre__ left 10:48 dalek joined, ChanServ sets mode: +v dalek, synopsebot joined, ChanServ sets mode: +v synopsebot, p6lert joined, Geth joined, ChanServ sets mode: +v Geth, araraloren joined, SourceBaby joined, ChanServ sets mode: +v SourceBaby 10:49 releasable6 left, releasable6 joined 10:52 scovit joined
Geth perl6.org: 6c80881650 | (Zoffix Znet)++ (committed using GitHub Web editor) | source/community/irc.html
Update URL for Undercover bot

  perl6.wtf is a ded domain. It's now at wtf.rakudo.party/		 10:59
11:03 kst` joined 11:31 scimon left
moritz I must say that I like the domain wtf.rakudo.party .-) 11:37
11:55 pierre_ joined 11:57 dogbert2_ joined 12:05 enheh joined
AlexDaniel “Who is Primarily Responsible for Considering the Ramifications of AI?” 12:16
well, uh… AI itself? :P 12:17
12:18 athenot joined 12:21 aindilis left
masak AlexDaniel: context? anyway, people like Yudkowsky would argue that at that point, it'd be too late. 12:27
moritz since we are on a bit of an AI tangent there, I'd like to share an interesting point I heard about it 12:31
so, context: might AI be dangerous? many people say we're too far away from that point to worry 12:32
so an interesting question is: when *should* we start to worry? what's the "fire alarm" for AI?
masak predicts moritz will mention exponentials soon
AlexDaniel masak: insights.stackoverflow.com/survey/2018
moritz like, if an AI takes a single day to become better at a game that humans spend centuries to improve at, would that be an AI fire alarm? 12:33
fun fact: that already happened.
masak: nope, no exponentials in that argument
masak :)
moritz: my prediction was based on the analogy with bacteria, and the thing that ends with "the jar is half-full at minute 59" 12:34
stmuk_ AlexDaniel: wrt rakudo/issues/1420 I think the changes which broke FreeBSD were reverted ... I can check on FreeBSD if it helps
12:35 xinming joined
AlexDaniel stmuk_: according to PufferBot and BeastieBot it builds just fine on freebsd and openbsd 12:35
moritz so, if somebody says that AI is no danger to humans, ask them: what event/achievement would convince you of the opposite?
</ai-rant>
AlexDaniel so can we just close that ticket?
stmuk_ yes 12:36
masak moritz: I find asking "what event/fact would convince you of [the opposite of what you believe]?" is a great way to discover open-minded/smart people, but that those are few and far between
and the by far most common answer is "no event/fact" or "that will never happen, since [I have confused my beliefs with reality]" 12:37
stmuk_ is there a bigendian bot? 12:38
AlexDaniel stmuk_: no, unfortunately not 12:39
stmuk_: well, apart from buildd: buildd.debian.org/status/package.p...;suite=sid
stmuk_ debian++ 12:40
AlexDaniel yes, debian++ indeed
also dod38fr++
see github.com/rakudo/rakudo/issues/1257 for more info
tbrowder_ hi #perl6 12:41
masak hi tbrowder_ 12:42
tbrowder_ hi, masak
i need some advice on concurrency
moritz avoid it!
tbrowder_ right! 12:43
masak wow, that was basically Threads 101
next up: monoids
tbrowder_ i’m playing with Zoffix’s Toaster and it’s leaving orphaned procs after an otherwise successfully completed run. 12:44
AlexDaniel because it doesn't kill the process tree 12:45
of stuff that is hanging
tbrowder_ it looks like all the await are in the right place.
yes, the hangers now get SIGKILL 12:46
AlexDaniel IIRC it's not a perl6-level problem. Some modules hang, it sends SIGKILL to the process
the top process dies but everything else remains alive
that's how I understand it
did we have a ticket asking for negated signals? 12:47
tbrowder_ i think my work around will be keep a list of all pids and handle them after all modules are toasted and before toaster exits. 12:48
AlexDaniel well, my workaround is to restart the VM
but keeping the list of all pids… what pids?
top process dies anyway if you sigkill it 12:49
tbrowder_ each one from the promises
Geth doc: zoffixznet++ created pull request #1852:
Post release 2018.03		 12:51
tbrowder_ i want the whole run to be cronable, no vm restarts, no top proc kill...
Geth doc/master: 7 commits pushed by (Zoffix Znet)++
tbrowder_ toaster should exit with 0 and leave no orphans. 12:53
12:53 aborazmeh joined, aborazmeh left, aborazmeh joined
tbrowder_ ideas welcome... 12:54
AlexDaniel tbrowder_: well my idea remains: implement process tree .kill in rakudo 12:56
maybe with a :tree param instead of negated signals
although I think it's OS-specific, but I really don't know
12:57 AlexDaniel left, AlexDaniel joined
tbrowder_ AlexDaniel: that’s what i was thinking, but it shouldn’t be any more system-specific than other low-level system stuff. jnthn could probably whip that out in no time! 12:59
12:59 mcmillhj joined
AlexDaniel you can whip that out in no time also, just allow negated signals :) 13:01
13:01 mcmillhj left
AlexDaniel IIRC the negative signal is handled by the sys_kill syscall 13:02
so there's nothing that should be done on rakudo side… maybe?
I'm really not qualified to discuss this :)
tbrowder_ ditto! 13:05
El_Che Rakudo seems to be out 13:06
files are there
kicking in the packages!
AlexDaniel yes!
tbrowder_ hm, some more info using promises to collect promises in the concurrency docs...
AlexDaniel tbrowder_: ok, see this: man7.org/linux/man-pages/man2/kill.2.html
tbrowder_ looks “promising” 13:07
AlexDaniel tbrowder_: “If pid is less than -1, then sig is sent to every process in the process group whose ID is -pid.”
tbrowder_: then also perl5 kill: perldoc.perl.org/functions/kill.html
tbrowder_: “A negative signal name is the same as a negative signal number, killing process groups instead of processes.”
13:07 cdg joined
AlexDaniel and I have *no idea* what it does on Windows 13:08
tbrowder_ good! i’ll try that
AlexDaniel but the point is that we don't have an alternative to that in perl 6?
or… can you actually pass a negative pid?
AH! I forgot 13:09
we have no &kill, whatsoever
tbrowder_: so the quick fix that you can try for your toaster experiments is to follow this: docs.perl6.org/language/5to6-perlfunc#kill
tbrowder_ but there is a kill method, in promise i think 13:10
AlexDaniel tbrowder_: and pass a negative pid instead
El_Che There are many languages where killing threads is considered bad practice
stmuk_ examples.perl6.org/categories/cookb...ignal.html
AlexDaniel we're not killing threads, we are killing processes
sending a SIGTERM or something like that to a process tree is a completely sane move in my opinion
El_Che death to the process!
AlexDaniel even more so if some process went nuts and you *have to* SIGKILL it 13:11
tbrowder_ ah, Proc::Async.kill: integer
going to see how tia 13:13
AlexDaniel coming to think about it, I guess SIGTERM-ing a process tree should not be done most of the time? 13:14
tbrowder_ ok, thnx for good ideas AlexDaniel and El_Che!
13:18 leah2 joined
El_Che tbrowder_: you mean stmuk :) 13:19
13:19 mahafyi joined 13:21 Unavowed joined
tbrowder_ yep, sorry, thnx stmuk! 13:21
13:28 wamba left, vcv joined
stmuk_ :) 13:29
vcv Given a directory, I'm trying to find all sub-directories (not recursive) where those sub-directories have a .git directory within them. This isn't working and I can't figure out why: 13:31
say $!repository-root.IO.dir.grep({ .d and .add(".git").d })
I know .d can throw a failure but i thought in boolean context it would resolve to True or False 13:32
MasterDuke evalable6: say ".".IO.dir.grep({ .d and .add(".git").d }) 13:37
evalable6 ()
MasterDuke evalable6: say ".".IO.dir.grep({ .d })
evalable6 ("bin".IO "t".IO "sandbox".IO "3rdparty".IO "maintenance".IO "services".IO "lib".IO "foo".IO "junk".IO ".git".IO "data".IO "logs".IO)
AlexDaniel e: say "../".IO.dir.grep({ .d and .add(".git").d })
evalable6 ("../rakudo_deleteme".IO "../force_bind".IO "../MoarVM".IO "../http-useragent".IO "../json…
AlexDaniel, Full output: gist.github.com/651183bb3aed276a61...a2af6879ff
13:37 wamba joined
AlexDaniel I think it works? 13:37
MasterDuke vcv: what version of perl6 do you have? 13:38
vcv i just reran my own code and now it works. it wasn't working all weekend. my apologies and thank you!
13:39 skids joined
MasterDuke knows Perl 6 touts being lazy as a feature, but taking the weekend off is a bit too much 13:40
AlexDaniel pffft :D :D
vcv Nice :)
13:41 mcmillhj joined
vcv .dir does return a lazy list so that checks out 13:41
13:47 greppable6 joined, ChanServ sets mode: +v greppable6 13:50 cdg left, cdg joined
Geth doc: efee510ced | (Will "Coke" Coleda)++ | t/pod-valid.t
test files can haz pod
 13:54
doc: de006c758b | (Will "Coke" Coleda)++ | xt/examples-compilation.t
add note about verbosity
13:54 cdg left 13:55 domidumont left
tbrowder_ ref kill procs again: if inside a script, e, 13:55
13:57 domidumont joined
El_Che 1 false positive (of 10 builds for the threading test) 13:57
tbrowder_ e.g., toaster, and i want to kill all its kid pids, shouldn’t i use “kill 0, $signal” instead of “kill -$*PID”?
13:58 mahafyi left
tbrowder_ er, instead of “kill -$*PID, $signal”? 13:59
13:59 mahafyi joined
tbrowder_ stmuk: ^^ ? 14:00
14:02 khw joined 14:09 aborazmeh left
Geth doc: 02594ed87a | (Will "Coke" Coleda)++ | doc/Type/X/Role/Initialization.pod6
Mark code

  (previously output as paragraph text)		 14:09
synopsebot Link: doc.perl6.org/type/X::Role::Initialization
doc: b895a847ca | (Will "Coke" Coleda)++ | 2 files
remove duplicate word
doc: 848e792c1a | (Will "Coke" Coleda)++ | doc/Language/unicode_entry.pod6
remove duplicate word
 14:10
synopsebot Link: doc.perl6.org/language/unicode_entry
AlexDaniel tbrowder_: ｢kill -$*PID, $signal｣ looks right 14:13
uhhh maybe not
not $*PID
unless you want to kill yourself
but… there's no way to get the pid of a Proc::Async, right? 14:16
moritz would be nice if Proc had a .pid 14:17
AlexDaniel I think it's spec-ed 14:18
but fudged on rakudo
OK, maybe not
but there is a ticket: RT#126380 14:19
synopsebot RT#126380 [open]: rt.perl.org/Ticket/Display.html?id=126380 [NYI] Proc.pid
14:23 ChoHag_ joined, ChoHag left 14:27 travis-ci joined
travis-ci Doc build failed. Will "Coke" Coleda 'remove duplicate word' 14:27
travis-ci.org/perl6/doc/builds/355377864 github.com/perl6/doc/compare/de006...95a847cadd
14:27 travis-ci left
buggable [travis build above] ☠ Did not recognize some failures. Check results manually. 14:27
14:29 travis-ci joined
travis-ci Doc build failed. Will "Coke" Coleda 'remove duplicate word' 14:29
travis-ci.org/perl6/doc/builds/355378554 github.com/perl6/doc/compare/b895a...8e792c1ae7
14:29 travis-ci left
buggable [travis build above] ☠ Did not recognize some failures. Check results manually. 14:29
14:32 vcv left 14:33 vcv joined 14:35 vcv left, vcv joined 14:44 mcmillhj left 14:45 wamba left 14:47 mcmillhj joined 14:51 wamba joined
El_Che New rakudo pkgs: github.com/nxadm/rakudo-pkg/releas...2018.03-01 14:51
14:51 curan left 14:53 wamba left, domidumont left
El_Che lizmat: for the weekly: "New Rakudo 2018003 Linux packages: github.com/nxadm/rakudo-pkg" 14:54
14:54 domidumont joined
masak likes the look of 2018003 14:55
"no-one has version numbers as large as ours!"
El_Che The hugest
masak but at the same time quite beautiful and esthetic 14:56
El_Che those puny passents with their 5 <-> 6 fights. We're at version 2018003!
masak I've been using Perl 2018003 for years now. it's great
El_Che peasants 14:57
masak actually, the main reason Perl 7 has not materialized in any form is that sorear and I are supposed to design it while on a generation ark through the galaxy
El_Che masak: is there a murder on board? 14:58
masak nope. it's all very civil and un-murder-y.
El_Che en.wikipedia.org/wiki/Ascension_(miniseries)
14:59 lelf joined
masak looks nice 15:00
anyway, murders are good for driving the plot of TV series, but less so when you're designing the next version of Perl
El_Che masak: I am not sure about that. I few well timed murders could have sped the process up 15:01
15:03 darutoko- left
AlexDaniel El_Che: why -01? :) 15:04
Geth doc: 0d060e781c | cfa++ | doc/Type/X/Role/Initialization.pod6
Unindent X::Role::Initialization example.
synopsebot Link: doc.perl6.org/type/X::Role::Initialization
masak .oO( "The language that wasn't -- what Perl 6 could've been, with a few well-timed murders" ) 15:05
El_Che it's the pkg revision 15:06
eg, 2018.02.1-01
starting revisions at 0 is weird :)
AlexDaniel El_Che: ↓ 15:07
weekly: "New Rakudo 2018003 Linux packages: github.com/nxadm/rakudo-pkg"
notable6 AlexDaniel, Noted!
15:07 cfa joined
tbrowder_ AlexDaniel: i’m going to try the “kill 0, $signal” because i think that kills all its child pids (fingers crossed). 15:07
El_Che eg if there is a needed zef update for the same release, I'll create 2018.03-02 packages, while if rakudo is bumped it would be 2018.03.1-01 15:08
AlexDaniel tbrowder_: hm, actually… if you do that when toaster exits, I guess it should work! 15:09
tbrowder_: so you're probably right!
El_Che: makes sense I guess
El_Che AlexDaniel: inventing stuff along the way
tbrowder_ yep, it will be at the end, the last statement 15:10
think i need an END block to be “safe”? 15:11
15:12 eliasr joined
Geth doc: 221c00f7b3 | cfa++ | xt/aspell.t
Explicitly set the aspell language to en_US.

Prevent false positives for machines that have non-US LANG settings
  (e.g. LANG=en_GB.UTF-8).		 15:13
AlexDaniel cfa++
cfa this was bugging me (guess what my LANG setting is :)) 15:14
AlexDaniel although… I have en_US.UTF-8 but it still complains about a bunch of stuff
cfa AlexDaniel: interesting
AlexDaniel or it did in the past, I should really check it
cfa at least (assuming the dictionary is installed) those problems should be more obviously reproducible now
also, good morning
moritz \o cfa 15:15
El_Che o/
cfa \o/
El_Che <o>
masak /o\
cfa 웃 15:16
araraloren _o_
El_Che (what did I do?)
that looks like operators Zoffix would use :)
cfa >-Ofun
araraloren prepare for sleep
night all :) 15:17
15:18 araraloren left
timotimo cfa: that is awesome, we need that as a real hat to put on our heads (and also some way to hide our arms) 15:18
cfa ha ha 15:19
oh, i came across an interesting (for some definition of interesting) question yesterday
given this,
m: my (@a, @b) := <one two three>, <four five six>; {a => @a, b => @b}.say;
camelia {a => (one two three), b => (four five six)}
cfa m: my ($a, $b) = <one two three>, <four five six>; {a => $a, b => $b}.say;
camelia {a => (one two three), b => (four five six)}
cfa we can either bind or assign items to destructure the list of lists on the rhs 15:20
but is there an elegant way to do non-binding assignment to @a and @b?
(without @a slurping up everything as in:)
m: my (@a, @b) = <one two three>, <four five six>; {a => @a, b => @b}.say;
camelia {a => [(one two three) (four five six)], b => []}
jnthn No, slurping up the rest is the intended semantics of an @foo in a list assignment 15:22
cfa yeah
jnthn That's why my ($head, @rest) = foo() works 15:23
The elegant way is to use binding :)
cfa yeah, i thought there might be a magical incantation i was missing
like, the slurpy case is as expected
jnthn It's only a : extra :)
cfa yeah
i wanted an array of scalars, so it ended being a bit more verbose 15:24
itemisation then @(...)
15:25 travis-ci joined
travis-ci Doc build failed. cfa 'Unindent X::Role::Initialization example.' 15:25
travis-ci.org/perl6/doc/builds/355405312 github.com/perl6/doc/compare/848e7...060e781c24
15:25 travis-ci left
buggable [travis build above] ☠ Did not recognize some failures. Check results manually. 15:25
cfa i also noticed there's no unzip (i.e., complement of Z or zip) builtin; so this was all re: using roundrobin to extract the two lists 15:26
AlexDaniel cfa: what the! I've never seen that one! 웃
u: 웃
unicodable6 AlexDaniel, U+C6C3 Hangul Syllable US [Lo] (웃)
cfa AlexDaniel: my work here is done
AlexDaniel unidump: 웃
unicodable6 AlexDaniel, gist.github.com/ef787acf48a628a79f...74086c4ce9 15:27
AlexDaniel Alpha – True
m: say \웃 = 42; say 웃 15:28
camelia 5===SORRY!5=== Error while compiling <tmp>
Preceding context expects a term, but found infix = instead.
Did you make a mistake in Pod syntax?
at <tmp>:1
------> 3say \웃 =7⏏5 42; say 웃
AlexDaniel ok maybe not
15:28 cdg joined
AlexDaniel I thought Lo can be used like that 15:28
cfa m: my \웃 = 42; say 웃 15:29
camelia Cannot allocate memory
cfa hunh
AlexDaniel e: my \웃 = 42; say 웃
evalable6 42
AlexDaniel ah, it's just me being blind
m: say 42
camelia 42
AlexDaniel m: my \웃 = 42; say 웃
camelia 42
AlexDaniel cfa: camelia was on a vacation
cfa ha, okay
15:29 cdg left
cfa lucky bot 15:29
15:30 cdg joined 15:31 travis-ci joined
travis-ci Doc build failed. cfa 'Explicitly set the aspell language to en_US. 15:31
travis-ci.org/perl6/doc/builds/355409896 github.com/perl6/doc/compare/0d060...1c00f7b304
15:31 travis-ci left
buggable [travis build above] ☠ Did not recognize some failures. Check results manually. 15:31
cfa i can't reproduce that pod-valid failure on my machine
but i don't think it's caused by the last two commits
15:32 athenot left
AlexDaniel restarts the build 15:32
cfa travis-ci, stop trolling me
15:32 cdg_ joined
DrForr o/ 15:33
Geth doc: 7c589cf0d3 | (Will "Coke" Coleda)++ | xt/duplicates.t
Check rendered docs, not raw pod

  * respect TEST_THREADS
  * ignore tables (requires a slightly recent rakudo)
  * ignore some single-char non-letters, e.g.: / < >
  (Whoops; part of this commit went in with b895a847)		 15:34
15:34 mcmillhj left
AlexDaniel ah… okay :) 15:35
15:35 pierre_ left
[Coke] ^^ Screwed up my commits here due to not paying attention after a git stash pop; but xt/duplicates is now checking the rendered pod, not the raw stuff, caught several more dupes that are also cleaned out 15:36
16:34 <+Geth> ¦ doc: * ignore some single-char non-letters, e.g.: / < >
16:34 <+Geth> ¦ doc:
16:34 <+Geth> ¦ doc: (Whoops; part of this commit went in with b895a847)
15:36 cdg left
[Coke] (crap, misclick) 15:36
cfa++
AlexDaniel [Coke]: fwiw I typically do ｢git diff --staged｣ before committing and ｢git show｣ after
sorta ensures that I'm not committing what I don't want 15:37
[Coke] Yes, I know how to avoid the issue, I merely failed to 15:38
thanks, though.
anyway, that avoids the C C<Str> issue (which I know was just skipped recently, but this avoids that entire class of issues) 15:39
cfa nice
thanks [Coke]
15:43 zakharyas joined 15:47 mcmillhj joined 15:50 travis-ci joined
travis-ci Doc build failed. cfa 'Explicitly set the aspell language to en_US. 15:50
travis-ci.org/perl6/doc/builds/355409896 github.com/perl6/doc/compare/0d060...1c00f7b304
15:50 travis-ci left
buggable [travis build above] ☠ Did not recognize some failures. Check results manually. 15:50
cfa sigh 15:52
Geth doc: 44e6081521 | (Zoffix Znet)++ | doc/Type/List.pod6
Update/fix/improve &combinations/&permutations

  - Document consistification update that makes sub forms
   be closer to method forms.
   Propspec: github.com/perl6/roast/commit/899d4d7858
   Rakudo impl: github.com/rakudo/rakudo/commit/29dcde1026
   Range $from still got an unresolved bug: R#1629
   github.com/rakudo/rakudo/issues/1629
  - Simplify examples (get rid of .join business and just show lists as is)
  - Don't use examples blocks to break up sentences in the middle		 15:53
synopsebot Link: doc.perl6.org/type/List
synopsebot R#1629 [open]: github.com/rakudo/rakudo/issues/1629 [audit entire codebase for such issues] [META] Treatment of non-Int ranges as a sequence of Int values
16:06 pierre_ joined, wamba joined 16:10 pierre_ left
timotimo nine: could you check out why prof-m doesn't work? 16:13
16:20 mcmillhj left 16:28 rindolf left 16:36 mcmillhj joined 16:37 athenot joined 16:38 aindilis joined 16:41 mcmillhj left 16:43 domidumont left 16:44 eckhardt_ joined, domidumont joined 16:45 eckhardt_ left 16:47 neilb left, mcmillhj joined, rindolf joined 16:49 domidumont left 16:57 lowbro left 17:00 cdg joined 17:02 cdg__ joined 17:03 cdg___ joined, cdg_ left 17:05 cdg left 17:06 cdg__ left
b2gills .tell thowe `state` variables (like `$++`) behave as if they were `my` declared in the next outer scope. Meaning they get reset when the outer scope does. 17:06
yoleaux b2gills: I'll pass your message to thowe.
17:09 pierre_ joined
lizmat reportable6: weekly New Rakudo 2018003 Linux packages: github.com/nxadm/rakudo-pkg 17:11
reportable6 lizmat, I cannot recognize this command. See wiki for some examples: github.com/perl6/whateverable/wiki/Reportable
lizmat notable6: weekly New Rakudo 2018003 Linux packages: github.com/nxadm/rakudo-pkg
notable6 lizmat, Noted!
17:12 unicodable6 left, unicodable6 joined 17:15 pierre_ left 17:16 robertle left 17:23 pharv joined
Geth doc: 045d14fc91 | (Zoffix Znet)++ | doc/Type/List.pod6
Clarify &combinations

Reword explanation in terms of method form and simplify signature, to make it easier to understand that it's essentially a front for the method form.		 17:25
synopsebot Link: doc.perl6.org/type/List
17:25 travis-ci joined
travis-ci Doc build failed. Zoffix Znet 'Update/fix/improve &combinations/&permutations 17:25
travis-ci.org/perl6/doc/builds/355429970 github.com/perl6/doc/compare/7c589...e6081521ca
17:25 travis-ci left
buggable [travis build above] ☠ Did not recognize some failures. Check results manually. 17:25
Geth doc: f68ef83399 | (Zoffix Znet)++ | t/pod-valid.t
Fix typo
 17:34
17:34 domidumont joined 17:35 bentham275 joined, bentham275 left 17:36 epony joined 17:37 dakkar left
pmurias blogs.perl.org/users/pawel_murias/2...fixes.html 17:43
El_Che reportable6: blogs.perl.org/users/pawel_murias/2...fixes.html 17:44
reportable6 El_Che, I cannot recognize this command. See wiki for some examples: github.com/perl6/whateverable/wiki/Reportable
El_Che reportable6: weekly blogs.perl.org/users/pawel_murias/2...fixes.html
reportable6 El_Che, Couldn't find a snapshot for weekly (try ｢list｣ command to see what's available).
El_Che notable6: weekly blogs.perl.org/users/pawel_murias/2...fixes.html
notable6 El_Che, Noted!
El_Che ouf 17:45
pmurias: your formatting is off. Lots of break lines
pmurias El_Che: thanks 17:47
El_Che at your service :) 17:49
17:50 pharv left 17:51 pharv joined
Geth doc: ab0a18a0cf | (Zoffix Znet)++ | t/pod-valid.t
Fix data race in t/pod-valid.t

Seem to produce floppy failures. No idea is this fixes them or not, but fairly sure there's a data race with multiple Proc's trying to create new keys in the same hash.		 17:52
17:52 AzaDO joined, travis-ci joined
travis-ci Doc build failed. Zoffix Znet 'Fix typo' 17:52
travis-ci.org/perl6/doc/builds/355481352 github.com/perl6/doc/compare/045d1...8ef8339937
17:52 travis-ci left
Geth doc: 05a77c6c7c | (Zoffix Znet)++ | 2 files
Add debug var for dumping STDERR of t/pod-valid.t failures

And simplify code a bit (the tap receives a Str not a buf and there's no point in assigning it to a param if we never use it);		 17:53
buggable [travis build above] ☠ Did not recognize some failures. Check results manually.
Geth doc: 421d2b07c5 | (Zoffix Znet)++ | t/pod-valid.t
Add ability to specify just 1 file for t/pod-valid.t
 17:55
[Coke] zoffix, that already did that.
17:57 pharv left, pharv joined
Geth doc: 7158320a4d | (Zoffix Znet)++ | 2 files
Follow convention for env var naming for test vars
 17:57
doc: adc83c923a | (Zoffix Znet)++ | README.md
Document P6_DOC_TEST_FUDGE env var
 17:58
17:58 cdg joined
Geth doc: 0389774865 | (Zoffix Znet)++ | .travis.yml
Enable P6_DOC_TEST_VERSBOSE during make test

To make travis tell us more about what it's not happy about with recent failures.		 17:59
doc: b190f963b8 | (Zoffix Znet)++ | t/pod-valid.t
Revert "Add ability to specify just 1 file for t/pod-valid.t"

This reverts commit 421d2b07c57ff97b5f9093426cdda1a4e52fe6f2.		 18:00
18:00 cdg left, cdg_ joined 18:01 cdg___ left 18:03 AzaDO left 18:09 cdg joined 18:13 comborico1611 joined, cdg_ left 18:14 Zoffix joined
Zoffix [Coke]: what's IO::String's purpose in xt/examples-compilation.t? Just to ignore output? 18:14
18:26 natrys joined
Zoffix m: "no worries; if False \{\nclass :: \{\n;\nsay q/a b/ ~~ /a b/; # OUTPUT: «False»\n\n}}".EVAL 18:28
camelia ( no output )
Zoffix weird; when run as part of xt/examplies-compilation.t that prints the warning :9
Geth doc: 45ce3b6a17 | (Zoffix Znet)++ | xt/examples-compilation.t
Get rid of IO::String usage

Fixes D#1827 github.com/perl6/doc/issues/1827		 18:29
synopsebot D#1827 [open]: github.com/perl6/doc/issues/1827 [build] travis build installing modules it doesn't need
18:32 mahafyi left
Geth doc: 3a0eadcd50 | (Zoffix Znet)++ | xt/examples-compilation.t
no worries

Here is a little song I wrote You might want to sing it note for note Don't worry be happy In every life we have some trouble When you worry you make it double Don't worry, be happy......		 18:32
18:33 mahafyi joined
AlexDaniel xD 18:34
18:34 espadrine joined 18:35 thowe joined
Zoffix Ahhhh 18:42
m: "no worries; if False \{\nclass :: \{\n;\n/a b/\n\n}}".EVAL;
camelia ( no output )
Zoffix m: /./; "no worries; if False \{\nclass :: \{\n;\n/a b/\n\n}}".EVAL;
camelia Potential difficulties:
Space is not significant here; please use quotes or :s (:sigspace) modifier (or, to suppress this warning, omit the space, or otherwise change the spacing)
at /home/camelia/EVAL_0:4
------> 3/a7⏏5 b/
Zoffix The language braid caching strikes again, I guess.
Geth doc: c78c8e6de6 | cfa++ | 3 files
P6_DOC_TEST_VERSBOSE => P6_DOC_TEST_VERBOSE
cfa Zoffix: ^ 18:43
Zoffix m: no worries; /./; "no worries; if False \{\nclass :: \{\n;\n/a b/\n\n}}".EVAL;
camelia ( no output )
Zoffix cfa: thanks
cfa np
Geth doc: 6725d6975e | (Zoffix Znet)++ | xt/examples-compilation.t
Workaround for xt/examples-compilation spewing warning

R#1630 github.com/rakudo/rakudo/issues/1630 is the cause; add a hack pre-caching `no worries` pragma in regexes, to prevent spewing warnings from regexes when we're compiling examples.		 18:48
synopsebot R#1630 [open]: github.com/rakudo/rakudo/issues/1630 `no worries` pragma gets cached by regex braid
18:50 itaipu joined, travis-ci joined
travis-ci Doc build passed. Zoffix Znet 'no worries 18:50
travis-ci.org/perl6/doc/builds/355507012 github.com/perl6/doc/compare/45ce3...0eadcd5002
18:50 travis-ci left
Zoffix yey 18:51
18:51 Zoffix left
cfa huzzah 18:51
buggable New CPAN upload: ANTLR4-Grammar-0.0.1.tar.gz by JGOFF cpan.metacpan.org/authors/id/J/JG/...0.1.tar.gz
18:59 travis-ci joined
travis-ci Doc build passed. cfa 'P6_DOC_TEST_VERSBOSE => P6_DOC_TEST_VERBOSE' 18:59
travis-ci.org/perl6/doc/builds/355510682 github.com/perl6/doc/compare/3a0ea...8c8e6de6ed
18:59 travis-ci left 19:06 zakharyas left 19:09 travis-ci joined
travis-ci Doc build passed. Zoffix Znet 'Workaround for xt/examples-compilation spewing warning 19:09
travis-ci.org/perl6/doc/builds/355513275 github.com/perl6/doc/compare/c78c8...25d6975e75
19:09 travis-ci left, FROGGS joined 19:10 itaipu left 19:11 itaipu joined, athenot left 19:25 cdg_ joined
Geth doc: a4c124ca66 | (JJ Merelo)++ | htmlify.p6
Dealing with ? in URLs

Which means it fixes #1677 but it actually does nothing to #1851, which has completely disappeared.		 19:26
doc: 07fd1ac367 | (JJ Merelo)++ | xt/examples-compilation.t
Merge branch 'master' of github.com:perl6/doc
19:28 cdg left 19:30 cdg_ left
buggable New CPAN upload: ANTLR4-Grammar-0.0.2.tar.gz by JGOFF cpan.metacpan.org/authors/id/J/JG/...0.2.tar.gz 19:31
19:32 pharv left 19:34 pharv joined
Geth whateverable: 5eb618dcca | (Aleks-Daniel Jakimenko-Aleksejev)++ | 2 files
Make bisectable DWIM

Resolves #282. The user is expected to use `old=…, new=…` syntax, but sometimes people forget. This commit makes bisectable understand committable-like syntax. Note that bisectable can only detect this by checking the existence of suspected revisions, meaning that it won't be able to produce helpful “Did you mean …” messages if you misspell one ... (6 more lines)		 19:40
19:40 domidumont left
19:41 bisectable6 joined
AlexDaniel Zoffix: I can't believe you'll find a way to make bisectable not understand what you wn 19:41
want now ↑
bisect: 2016.02 2017.05 say 42 19:42
bisectable6 AlexDaniel, Using old=2016.02 new=2017.05 in an attempt to DWIM
AlexDaniel, On both starting points (old=2016.02 new=2017.05) the exit code is 0 and the output is identical as well
AlexDaniel, Output on both points: «42␤»
19:43 cdg joined, amalia___ left 19:47 cdg left 19:50 pharv left, pharv joined 19:51 Zoffix joined
Zoffix AlexDaniel++ thanks. 19:51
19:51 Zoffix left 20:02 athenot joined 20:04 mahafyi left
thowe Oh. Perl Conference in Salt Lake again :) 20:04
yoleaux 17:06Z <b2gills> thowe: `state` variables (like `$++`) behave as if they were `my` declared in the next outer scope. Meaning they get reset when the outer scope does.
thowe b2gills, Your message through time is appreciated. My day is brightened and enlightened. 20:05
20:07 wamba left, wamba joined 20:13 pharv left, pharv joined
[Coke] zoffix, yes, IO::String is to capture and ignore all output. 20:22
... and that fix looks fine. Thanks.
20:40 cdg joined 20:42 cdg left, cdg joined 20:48 wamba left 20:53 pecastro joined 20:59 rindolf left, MilkmanDan left 21:01 MilkmanDan joined 21:04 skids left 21:05 andrzejk_ joined 21:07 mcmillhj left, lelf` joined 21:09 lelf left 21:10 athenot_ joined, vcv left, athenot left 21:12 ChoHag_ left 21:13 ChoHag joined 21:24 lelf`` joined, pharv left 21:25 lelf` left 21:26 pharv joined 21:29 pharv left, pharv joined 21:43 andrzejk_ left 22:03 FROGGS left 22:28 natrys left 22:29 lelf`` left 22:30 dct joined
lizmat And another Perl 6 Weekly hits the Net: p6weekly.wordpress.com/2018/03/19/...y-edument/ 22:30
AlexDaniel weekly: clear
notable6 AlexDaniel, Moved existing notes to “weekly_2018-03-19T22:30:44Z”
lizmat AlexDaniel: thanks :-) 22:31
AlexDaniel lizmat++ 22:34
22:35 sjn_ joined 22:38 sjn_ left 22:42 itaipu left 22:45 mcmillhj joined 22:47 notable6 left, notable6 joined 22:50 sjn joined, mcmillhj left 22:59 skids joined 23:01 cdg left 23:22 Kaiepi joined 23:25 comborico1611 left 23:27 pierre joined 23:32 pierre left 23:34 pmurias left 23:36 MilkmanDan left 23:39 MilkmanDan joined 23:42 markong left 23:52 cdg joined, pharv left 23:54 stmuk_ left 23:55 AlexDaniel left 23:56 cdg left 23:59 cdg joined

Powered by Cro and the Raku® Programming Language.

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

Thank you!