»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.
00:10 pmurias left 00:23 ely-se left
BenGoldberg m: my $foo = 42; say ++$foo xx 3; 00:28
camelia rakudo-moar 2e1782: OUTPUT«(43 44 45)␤»
BenGoldberg m: my $foo = 42; say $foo++ xx 3;
camelia rakudo-moar 2e1782: OUTPUT«(42 43 44)␤»
00:29 woolfy left
AlexDaniel BenGoldberg: didn't expect it to work 00:30
timotimo stackoverflow.com/questions/350275...6#35108936 00:31
AlexDaniel :) 00:32
skids Maybe that answer could be committed to perl6/user-experience. 00:35
timotimo oh, damn, maybe i've made a "little" mistake 00:40
i might have left in the "$word.ends-with('rl') && ..." 00:41
which dramatically speeds up things, too
skids m: my $f = "."; say do { $f ~= "." } X~ 1,2,3; my $foo = 42; say do { $foo++ } X~ 1,2,3 00:42
camelia rakudo-moar 2e1782: OUTPUT«(..1 ..2 ..3)␤(421 422 423)␤»
skids Good.
skids wonders why xx does that. 00:43
m: my $f = "."; say do { $f ~= "." } xx 3 00:44
camelia rakudo-moar 2e1782: OUTPUT«(.... .... ....)␤»
timotimo it creates a list of $f three times
skids m: my $foo = 42; say do { $foo++ } xx 3
camelia rakudo-moar 2e1782: OUTPUT«(42 43 44)␤»
timotimo m: my $f = "."; say do { ($f ~= ".")<> } xx 3
camelia rakudo-moar 2e1782: OUTPUT«(.. ... ....)␤»
timotimo postfix <> deconts for you
yeah, my answer on SO was wrong, it's not 10x 00:45
00:56 jack_rabbit left
timotimo composed a new answer 01:02
i hope it isn't just too rambly 01:03
01:12 jack_rabbit joined 01:19 raiph joined 01:20 leont left 01:23 revhippie left 01:30 pecastro left 01:34 mcmillhj left 01:37 revhippie joined
MadcapJake what does the associative indexing error mean? 01:42
01:42 xpen joined
AlexDaniel MadcapJake: which one exactly? 01:46
MadcapJake "Type Str does not support associative indexing."
01:46 xpen left
gfldex m: qq { $foo<html-tag> } 01:46
camelia rakudo-moar 2e1782: OUTPUT«5===SORRY!5=== Error while compiling /tmp/fHvxTH46LC␤Variable '$foo' is not declared␤at /tmp/fHvxTH46LC:1␤------> 3qq { 7⏏5$foo<html-tag> }␤»
gfldex m: my $foo; qq { $foo<html-tag> } 01:47
camelia rakudo-moar 2e1782: OUTPUT«WARNINGS for /tmp/GbkLA0Kq5z:␤Useless use of " $foo<html-tag> " in expression " $foo<html-tag> " in sink context (line 1)␤Use of uninitialized value $v of type Any in string context␤Any of .^name, .perl, .gist, or .say can stringify undefined things,…»
gfldex m: my $foo; dd qq { $foo<html-tag> }
camelia rakudo-moar 2e1782: OUTPUT«Use of uninitialized value $v of type Any in string context␤Any of .^name, .perl, .gist, or .say can stringify undefined things, if needed. in block <unit> at /tmp/fgluhjeP_Z line 1␤" "␤»
gfldex m: my $foo = "abc"; dd qq { $foo<html-tag> }
camelia rakudo-moar 2e1782: OUTPUT«Type Str does not support associative indexing.␤ in block <unit> at /tmp/5DsSOz4SMF line 1␤␤Actually thrown at:␤ in block <unit> at /tmp/5DsSOz4SMF line 1␤␤»
01:47 mattp_ left
gfldex MadcapJake: this one? ^^^ 01:47
MadcapJake yeah but what's dd? I'm not using that 01:48
gfldex doesn't matter, you got a quote problem
m: my $foo = "abc"; dd Q:c { {$foo}<html-tag> }
camelia rakudo-moar 2e1782: OUTPUT«" \{\$foo}<html-tag> "␤»
gfldex if found that works best when dealing with xml/html
01:49 lokien_ left
MadcapJake return Template::Mustache.render(slurp('./views/result.hbs'), { :$community, :message("Success! You were already invited.<br>Visit the <a href=\"$slack-url\">$community</a> Slack.") }, |%templates); 01:49
THat's the line causing the error 01:50
gfldex $community</a> <-- culprit
MadcapJake ahhhhh
i get it now, it thinks that's a <> postcircumfix 01:51
gfldex indeed
MadcapJake gfldex: thanks!
this is almost done! 01:54
01:54 yqt left
MadcapJake moritz: any chance I could host this on hack.p6c.org? It's an invite site for Perl6 slack (written in Perl 6) 01:54
01:59 Fleurety joined
AlexDaniel interestingly I don't see any way to disable “$var<foo>” behavior 02:01
m: my $x = ‘test’; say Q:s‘$x<b>’ # still we have a problem
camelia rakudo-moar 2e1782: OUTPUT«Type Str does not support associative indexing.␤ in block <unit> at /tmp/EZdXuHFoO9 line 1␤␤Actually thrown at:␤ in block <unit> at /tmp/EZdXuHFoO9 line 1␤␤»
AlexDaniel so I guess you have to use {}
gfldex AlexDaniel: there is a RT for that. I can find it if you need that. 02:02
AlexDaniel gfldex: out of curiosity – maybe. But other than that I don't really need it :)
m: my $x = ‘test’; say “{$x}<b>” # after all, this is not too bad 02:03
camelia rakudo-moar 2e1782: OUTPUT«test<b>␤»
gfldex AlexDaniel: rt.perl.org/Public/Bug/Display.html?id=127226 02:04
AlexDaniel ah that one. But it's a little bit different 02:05
or so it seems
02:07 jeek left, mattp_ joined
dalek c: 4b19544 | LLFourn++ | doc/Type/Signature.pod:
s/won't-work/wont-work/

and a typo		 02:12
MadcapJake how do you access environment variables again? I can't seem to recall or find it 02:13
02:14 josgraha_ left 02:15 lonewolf28 left
AlexDaniel MadcapJake: %*ENV<foo> ? 02:15
MadcapJake yeah, thanks
suprised that's not in the language/variables doc page 02:16
AlexDaniel you can kinda guess it :) 02:17
revhippie m: sub won't-work { say "but does" }; won't-work();
camelia rakudo-moar 2e1782: OUTPUT«but does␤»
llfourn REALLY?
yoleaux 30 Jan 2016 08:51Z <nine> llfourn: thanks for working on the require/import stuff :)
AlexDaniel it has to be a hash, * is there because it's kinda global, and ENV is kinda obvious :)
revhippie llfourn: ' has the same rules as - in identifiers 02:18
llfourn revhippie: right! well I guess I will fix that...
I actually did try that but it didn't work for me hrmph 02:19
MadcapJake AlexDaniel: yeah i couldn't remember if ENV was a subhash of some other dynamic (especially because it wasn't listed)
llfourn thanks!
02:19 jme_ joined
MadcapJake just moving the token to an ENV var so I can post this to github, then I'll start by at least hosting in on heroku, anyone done a heroku with perl 6 before? 02:20
02:21 mattp_ left
revhippie llfourn: you're welcome! 02:21
AlexDaniel m: my \ꞌ'ꞌ'ꞌ = 42; say ꞌ'ꞌ'ꞌ 02:22
camelia rakudo-moar 2e1782: OUTPUT«42␤»
dalek c: 67a4e12 | LLFourn++ | doc/Type/Signature.pod:
s/wont-work/won't-work/

I had no idea that that worked. Excuse me.
02:26 p_p_r joined
dalek c: 1f83b6c | raiph++ | doc/Language/variables.pod:
Add %*ENV
 02:27
AlexDaniel MadcapJake: interestingly ENV appears in so many places 02:28
e.g. doc.perl6.org/language/5to6-perlvar
oh wow, raiph++
MadcapJake oh, i've never even looked at the 5to6 stuff (never programmed in Perl 5)
raiph++
AlexDaniel MadcapJake: okay, what about this? doc.perl6.org/type/Hash
:)
.u ꞌ 02:29
yoleaux U+A78C LATIN SMALL LETTER SALTILLO [Ll] (ꞌ)
Juerd One of these days I'm going to suggest an overhaul to the way those documents are formatted. But first I need to think of how. :)
I'm really annoyed by the insanely long pages.
MadcapJake nice! That reminds me, I wish there was a link at the top of each type to the general type section, it's annoying to have to scroll through the list of methods or click the first entry and scroll back up. is this possible to add? 02:30
Juerd But at the same time, I like scrolling much better than hidden stuff or lazy loading.
MadcapJake Juerd: is that what you were talking about? ^
Juerd So it just has to take up less vertical space. Somehow.
AlexDaniel I don't mind these long pages. Long pages means I can ctrl+f without any problem
Juerd MadcapJake: My keyboard has a dedicated [Home] button 02:31
dalek c: 22ea0e0 | (Wenzel P. P. Peppmeyer)++ | doc/Language/faq.pod:
faq string interpolation and html
Juerd MadcapJake: I never find it problematic to scroll back to the top of any page.
MadcapJake no i don't mean the top of the page
AlexDaniel gfldex++
Juerd What do you mean then?
MadcapJake each type has a general section (like the one where you showed me has the env stuff in it)
but there is no way to move right to it
Juerd That wasn't me. I showed nothing to anyone :)
02:31 abaugher left
AlexDaniel hides 02:32
Juerd Oh, that section
MadcapJake AlexDaniel did I meant :P
Juerd Yes, that should go before the TOC
Or maybe the TOC should just be next to the content instead of above it
I dunno
MadcapJake either that or have a link to it so that you can just get right to it. currently the only routes are either scrolling through the TOC or clicking the first entry and scrolling back to the general section
02:33 abaugher joined
MadcapJake often the general section has some of the most useful info (to me) and it's the most difficult to reach! 02:33
AlexDaniel or maybe the first entry in TOC should be “Description” or something
so that you can click on it and go to that section immediately
MadcapJake AlexDaniel: yeah that's what I was thinkingt
Though personally, if it was possible to move the TOC to the side (for large screen widths) that would be ideal 02:34
Juerd juerd.nl/i/c09fb3f5ef6f536e2f7bfdd8151db36d.png
It's a start.
MadcapJake Juerd: it's gorgeous!
Juerd Anyhow, maybe I'll look into it a bit more later 02:35
Or by all means take this idea and make it beautiful while I'm asleep
But for now, good night #perl6!
MadcapJake Juerd: I'll see what I can do, what exactly did you do? just a css float:left?
llfourn o/
ugexe it should have a large hover over begging me to enter my email
Juerd MadcapJake: Yes. Which means it'll start ugly from where the TOC ends.
s/start/start to get/
02:36 jme_ left
MadcapJake Juerd: gotcha 02:36
Juerd Anyhow, good night :)
z &
MadcapJake nighty night!
AlexDaniel and you also had to make the font smaller :/
02:36 jme_ joined
MadcapJake here's a thought: postimg.org/image/wndyk9bch/ 02:40
Juerd Okay, a few things then before I really go to bed 02:41
AlexDaniel: Only because I couldn't figure out how to add some css to <li>, to reduce the huge indents 02:42
MadcapJake: IMHO the text is already too wide for comfortable reading and the extra column helps reduce that issue as well.
AlexDaniel Juerd: I don't see much available space there :)
Juerd MadcapJake: And I don't think the out-of-bounds layout would fly on portrait oriented devices
So, afk! & 02:43
MadcapJake no you'd have to set it as @media: screen (or however that's written, not a css wizard xD )
MadcapJake is gonna make a perl6-buildpack for heroku now 02:44
02:50 TimToady left 02:52 TimToady joined 02:55 p_p_r left 02:56 p_p_r joined
jme_ did anyone know why Pod::Coverage keeps failing testing with `panda install Linenoise` ? 02:59
03:00 jme_ left 03:01 jme_ joined
llfourn jme_: I just ran panda --force install Linenoise and it worked 03:06
jme_ llfourn: sweet! 03:09
03:17 jeek joined 03:18 Herby_ joined
Herby_ Evening, everyone! 03:18
\o
skids o/
AlexDaniel * o/
03:21 orbus left
revhippie i don't see those operators in the docs 03:21
03:24 jme_ left 03:35 noganex joined
AlexDaniel revhippie: at least * is spec-ed: design.perl6.org/S99.html#good_* 03:38
03:38 noganex_ left 03:39 raiph left, kid51_ joined
revhippie AlexDaniel: ha! 03:39
03:41 kid51 left 03:42 kid51_ is now known as kid51
MadcapJake What are some common perl6 web deployment patterns? app.pl6 in the main source dir? a RUN script in the main source dir? a bin/file referenced in META6.json? Are these considered standard and are there others? 03:53
AlexDaniel MadcapJake: I don't think that there are any common patterns yet 03:54
MadcapJake bummer xD 03:55
llfourn MadcapJake: when you figure them out let us know :) 03:57
MadcapJake I really miss «"main": "lib/main"» and «"scripts": {"start": "node app.js"}» from nodejs land
llfourn I never used it. What does it do? 03:58
lets you do 'node start' or something without referring to the actual script? 03:59
MadcapJake a lot of node frameworks hook into the main value as a way to find the entry point in an application. and the scripts hash was a list of shortcuts to scripts that you could run from the command line and start is a builtin one with special support
lets you do «npm run build; npm run test; npm start» to build, test and run 04:00
llfourn yeah sounds handy to me 04:01
MadcapJake also npm scripts had their own private environment that included everything in node_modules, so you didn't every have to fudge with your PATH
llfourn why would a node framework want to know the entrypoint out of interest?
(I get why npm and other cli tools would) 04:02
04:02 SCHAAP137 left
MadcapJake if you import a module, main was considered the primary entry point for including that module, so anything you want publicly available to users must be accessible from there 04:03
llfourn perhaps we can build it into panda or some other cli tool. panda start etc which looks in META6.json.
MadcapJake llfourn: I'd definitely be interested in that 04:04
04:04 kid51 left
llfourn I'd certainly prefer it to makefiles 04:04
Herby_ isn't there some big perl 6 talks going on this weekend? 04:05
or a perl 6 meetup of some sort?
llfourn as I understand it FOSDEM is happening?
Herby_ hopefully some (most) of their talks find their way to youtube. Some of the topics look pretty interesting: fosdem.org/2016/schedule/track/perl/ 04:06
MadcapJake I hope so too 04:07
AlexDaniel there are also “official” recordings: video.fosdem.org/2015/ 04:11
in 2015 though half of them were lost
04:13 mr-foobar left
Herby_ sweet, i'll take a look 04:13
AlexDaniel half? No. Most of them were lost 04:14
MadcapJake how did that happen?
revhippie the bbc threw 'em all in a dumpster like they did with early doctor who :(
MadcapJake lol
04:18 Fleurety left 04:31 dolmen joined 04:34 espadrine left
ugexe those are hooks, and i've been bugging about them for awhile myself 04:36
nothing is stopping you from putting scripts in your resources and accessing them via %?RESOURCE though 04:39
MadcapJake ugexe: can you elaborate on that resources pattern? 04:41
llfourn m: say %?RESOURCE 04:42
camelia rakudo-moar 2e1782: OUTPUT«5===SORRY!5=== Error while compiling /tmp/BloATlHoEF␤Variable '%?RESOURCE' is not declared␤at /tmp/BloATlHoEF:1␤------> 3say 7⏏5%?RESOURCE␤»
ugexe instead of "start" : "node app.js" or whatever, it could be like 'EVALFILE(%?RESOURCE<app.pl6>)' 04:46
MadcapJake but where would that line go? 04:48
ugexe whats it for? starting a server for tests?
MadcapJake npm places the scripts in package.json which is equivalent to META6.json and then they are accessible from the cli
scripts are for anything, building, running tests, deploying, they are just mini bash scripts (in npm/node land) 04:49
however they contain a separate node-specific environment
ugexe right. in s22 those are called hooks
MadcapJake hooks in s22 is like one sentence xD 04:50
ugexe there is not much on them, but their intentions would appear the same :)
stand alone scripts to be ru during various phases of the install process
usually they have access to some sort of ENV var to keep track of where in the life cycle they are at 04:51
and act accordingly
MadcapJake yeah that sounds "pretty" close, npm provides a bunch of special hooks for preinstall, postinstall, prerelease,postrelease, and a few more i cant remember
but it also allows you to write your own for doing other things with your workflow
04:52 revhippie left, kaare_ joined
MadcapJake and you just do «npm run test» or «npm run version-bump» or «npm run upgrade-deps» 04:52
ugexe yeah. zef used to provide many of those. i would definately like to see something happen with them though 04:53
MadcapJake (and actually «npm run test» is equiv to «npm test» as it is a special script hook which gets run on install, release, etc
and also neat, they're composable, so write a few that have the bits and pieces of a build process and then compose into your main build script. 04:54
ugexe but, look under s22 under %?RESOURCE
"libraries" : { "libblah" : "build-time" } 04:55
04:55 Actualeyes joined
ugexe if the s22 spec was used then that would work nicely with those little commands... s/build-time/whatever cmd/ 04:55
MadcapJake yeah that'd be neat, it seems that the perlish tradition is to use a Configure.pl file and make for a lot of this stuff, is that accurate? 04:57
ugexe it wouldnt have to be a library... maybe like "hooks" : { "preinstall-script.pl6" : preinstall }
MadcapJake yeah that'd be more up my alley :)
honestly since pl6 is powerful enough to forego the need for bash scripts, and thus they can be lengthier, i'd say providing a series of hardcoded hooks would be all that is needed: pre-install, post-install, pre-run, run, post-run, server, watch, pre-build, build, post-build, pre-test, test, post-test 05:00
but then you'd need to provide a cli interface, some way to do «panda server» or «perl6 test» or «perl6 build», etc. 05:01
ugexe that works too, but the thing is those are only used by the installer
MadcapJake well that's my point, it would need to be more available to the user as a means to improving workflow 05:02
ugexe so the actual names may not be specced, and leaf nodes are meant to be arbitrary data for package managers
MadcapJake here's a good rundown of npm scripts blog.keithcirkel.co.uk/how-to-use-n...uild-tool/ 05:06
i think honestly part of why they're useful is how npm/node handles dependencies and running js files
i really liked prepublish but does that really fit with perl6? well there is no equivalent to npm site yet, and packages in perl6 aren't available inside "perl6_modules" in your source dir, they're precompiled inside a hashed folder hidden away from the user. 05:08
preinstall/postinstall are neat, but node devs rely on those because there really isn't any true install process (npm just dumps a tagged git repo into either your source's node_modules or a global node_modules 05:10
ugexe zef used to work like this before the CU refactor (everyone kept using Build.pm) with hard coded script names that were run automatically
05:10 AlexDaniel left
MadcapJake well that sounds cool! has anyone made a friendly api to all the CU stuff yet? 05:11
ugexe but you could do like you say and do like, cut a release version to github and auto increment your version
ive been proposing overhauling Distribution to serve just that very purpose
the s22 Distribution interfaces are what i would like (and spec'd %?RESOURCE) as they solve all of these problems 05:13
MadcapJake yeah that'd be really killer, an easy «zef publish minor» would be amazing.
ugexe in any case, i think Distribution will be updated soon to do some/all of this stuff 05:15
MadcapJake great, i'm looking forward to it! I think that's the biggest sore spot right now and also it has the most potential to strengthen the ecosystem. 05:16
ugexe you can mention any ideas/thoughts here if you want them to be recorded: github.com/perl6/toolchain-bikeshed
05:41 Cabanossi left 05:43 Cabanossi joined 05:56 Kogurr left 06:06 khw left
Herby_ when using "given...when" control, can you use <, as in less than? 06:08
06:08 ZoffixWin joined
Herby_ ZoffixWin! 06:08
\o
ZoffixWin hi 06:09
Herby_ m: my $var = 42; given $var { when < 50 { say "its less than 50"; } }
camelia rakudo-moar 2e1782: OUTPUT«5===SORRY!5=== Error while compiling /tmp/TctdY34s2e␤Unable to parse expression in quote words; couldn't find final '>'␤ ␤at /tmp/TctdY34s2e:1␤------> 3 when < 50 { say "its less than 50"; } }7⏏5<EOL>␤ expecting any of:␤ …»
Herby_ m: my $var = 42; given $var { when Int { say "its less than 50"; } } 06:10
camelia rakudo-moar 2e1782: OUTPUT«its less than 50␤»
llfourn Herby_: nah it's an infix
MadcapJake first attempt at a heroku buildpack *crosses fingers*
Herby_ m: my $var = 42; given $var { when $var < 50 { say "its less than 50"; } }
camelia rakudo-moar 2e1782: OUTPUT«its less than 50␤»
Herby_ llfourn, thanks
MadcapJake so far so good, it's installing nqp xD
ZoffixWin m: say split(';', "a;b;c", 2, :all).perl
camelia rakudo-moar 2e1782: OUTPUT«("a", "b;c").Seq␤»
ZoffixWin ^ that's not what the docs say the output should be :/ 06:11
("a", ";", "b;c").list
Missing ";"
llfourn m: .say for &split.candidates
camelia rakudo-moar 2e1782: OUTPUT«sub split ($pat, Cool $target, |c is raw) { #`(Sub|73808080) ... }␤»
llfourn so it should only split 2? 06:12
ZoffixWin it does split 2, but the delimiter is missing
Which sucks, because it breaks this Perl 5 code I'm trying to port :/
llfourn m: say split(';', "a;b;c", 1, :all).perl
camelia rakudo-moar 2e1782: OUTPUT«("a;b;c",)␤»
llfourn isn't that splitting 0? 06:13
ZoffixWin The "1" is the "LIMIT"
MadcapJake it leaves anything in the $limit's last element as a the rest of the text
llfourn oh.
so it's :all that isn't working when you put a limit? 06:14
ZoffixWin Limit is irrelevant 06:15
MadcapJake yeah all is definitely not working there, limit is strange though as you'd think 2 would mean you'd get two split out elements, right? or is that just me? xD 06:16
llfourn MadcapJake: not just you :\ 06:17
MadcapJake yeah, the way it's phrased it sounds like limit would mean "the limit of the number of split out elements" not "the limit of elems of the returned seq" 06:18
06:18 adu joined
ZoffixWin wtf is a "split out element"? 06:18
llfourn something that was matched as a delim
MadcapJake lol as in, split is splitting into elems, so if you specify a limit of 2, it means two elems should be "split away" from the core string 06:19
06:19 Fleurety joined
llfourn so limit => 1, would match and split on one ';' 06:19
adu llfourn: are you talking about math?
llfourn adu: no :)
ZoffixWin No, that's not at all what it means. You're splitting ON delimiters and you're asking for at most LIMIT elements.
llfourn ZoffixWin: it seems not :) 06:20
ZoffixWin llfourn, seems not what?
llfourn ZoffixWin: it seems that "that's not at all what it means" as you said 06:21
adu limit(f(x), x=1) # SageMath
Limit[F[x], x -> 1] # Mathematica
what's the Perl6 version?
ZoffixWin adu, we're not talking about those limits
I'm not seeing :all being used in Str.pm at all 06:22
llfourn maybe it's NYI? 06:23
m: .say for &split.candidates # it's only accepted to the sub because of the capture 06:24
camelia rakudo-moar 2e1782: OUTPUT«sub split ($pat, Cool $target, |c is raw) { #`(Sub|65256368) ... }␤»
ZoffixWin m: .say for "foo".split.candidates
camelia rakudo-moar 2e1782: OUTPUT«Cannot call split(Str: ); none of these signatures match:␤ (Cool $: Regex:D $pat, $limit = { ... };; :$all, *%_)␤ (Cool $: Cool:D $pat, $limit = { ... };; :$all, *%_)␤ (Str:D $: Regex:D $pat, $parts = { ... };; :$v is copy, :$k, :$kv, :$p, :…»
ZoffixWin m: .say for "foo".split(/./, :all).candidates 06:25
camelia rakudo-moar 2e1782: OUTPUT«Method 'candidates' not found for invocant of class 'List'␤ in block <unit> at /tmp/GAbT5t1biH line 1␤␤»
llfourn ZoffixWin: .can?
ZoffixWin m: .say for "foo".can('split').candidates
camelia rakudo-moar 2e1782: OUTPUT«Method 'candidates' not found for invocant of class 'List'␤ in block <unit> at /tmp/rx1imuglT9 line 1␤␤»
ZoffixWin m: .say for "foo".can('split')[0].candidates
camelia rakudo-moar 2e1782: OUTPUT«split␤split␤split␤split␤split␤split␤»
ZoffixWin heh
llfourn .can already gives you candidates
m: .say for "foo".can('split') 06:26
camelia rakudo-moar 2e1782: OUTPUT«split␤split␤split␤»
llfourn m: .gist.say for "foo".can('split')
camelia rakudo-moar 2e1782: OUTPUT«split␤split␤split␤»
llfourn m: .perl.say for "foo".can('split')
camelia rakudo-moar 2e1782: OUTPUT«method split (Mu $: | is raw) { #`(Method|71404376) ... }␤method split (Mu $: | is raw) { #`(Method|71406352) ... }␤method split (Mu $: | is raw) { #`(Method|71406504) ... }␤»
ZoffixWin Well, search in Rakudo source shows only Str implements it and it doesn't use :all
llfourn where did you find a mention of :all?
ZoffixWin <ZoffixWin> ^ that's not what the docs say the output should be :/ 06:27
docs.perl6.org/routine/split
llfourn sounds like it was removed 06:28
ZoffixWin m: say 'Life, the Universe, and Everything'.WHY
camelia rakudo-moar 2e1782: OUTPUT«42␤»
MadcapJake lol
ugexe i know its purpose was being debated and changed before xmas
ZoffixWin Ugh, ugexe, you're right: github.com/rakudo/rakudo/commit/de...d60e3c4cc9 06:29
MadcapJake may have a working heroku-based bailador-built website *crosses fingers and toes*
llfourn I think it chaged to :v
oh too late :o 06:30
ZoffixWin m: say split ';', "a;b;c", :v
camelia rakudo-moar 2e1782: OUTPUT«(a ; b ; c)␤»
MadcapJake which looks like pacman eating "all" the delimiters
ZoffixWin m: say split ';', "a;b;c"
camelia rakudo-moar 2e1782: OUTPUT«(a b c)␤»
ZoffixWin m: say split /',' (';')/, "a,;b,;c", :v
camelia rakudo-moar 2e1782: OUTPUT«(a ｢,;｣␤ 0 => ｢;｣ b ｢,;｣␤ 0 => ｢;｣ c)␤»
MadcapJake heroku free tier is a lot faster at building rakudo than my rig :( 06:31
ZoffixWin Still not like Perl 5 :/
llfourn to me it makes sense to formalize what is and isn't accepted in &split
06:32 Herby_ left 06:34 dolmen left, xinming_ is now known as xinming
ZoffixWin m: my $i = 2; my @foo = ^10; say @foo[* - $i]; 06:35
camelia rakudo-moar 2e1782: OUTPUT«8␤»
ZoffixWin m: my $i = -2; my @foo = ^10; say @foo[* - $i];
camelia rakudo-moar 2e1782: OUTPUT«(Any)␤»
ZoffixWin :/ any way to have an $i that could be either positive or negative?
Man translating Perl 5 into Perl 6 is annoying as hell :|
06:37 M-Illandan joined
M-Illandan Oh huh... I can't build rakudo on git HEAD over some unrecognised version specifier. Has anyone else seen this before? 06:37
Unrecognized revision specifier '2016.01-RC1'
ugexe heh 06:38
skids m: my $i = -2; my @foo = ^10; say @foo[$i % *];
camelia rakudo-moar 2e1782: OUTPUT«8␤»
skids m: my $i = 2; my @foo = ^10; say @foo[$i % *]; 06:39
camelia rakudo-moar 2e1782: OUTPUT«2␤»
ZoffixWin huh
skids As long as you do not mind
m: my $i = -12; my @foo = ^10; say @foo[$i % *];
camelia rakudo-moar 2e1782: OUTPUT«8␤»
llfourn that is clever
ZoffixWin Yeah, skids++ 06:40
06:40 dixvagon joined
ugexe M-Illandan: github.com/rakudo/rakudo/pull/699 06:40
06:41 mattp_ joined 06:43 isBEKaml joined 06:50 CIAvash joined 06:54 cdg left
MadcapJake i've no idea if i'm doing heroku's buildpacks right :P github.com/MadcapJake/herokudo/blo...in/compile 06:56
i'm making one last effort with a `.profile.d/perl6.sh` file, then if that doesn't work, it's time for bed! 06:57
07:00 BenGoldberg left 07:01 ZoffixWin left 07:19 dixvagon left 07:20 go|dfish left 07:25 lizmat left 07:28 ELBeavers left 07:29 ELBeavers joined
M-Illandan ugexe (IRC): thanks 07:30
07:30 cibs left 07:32 cibs joined 07:33 go|dfish joined 07:35 Actualeyes left 07:48 skids left 07:55 lizmat joined
lizmat waves from the Perl DevRoom at FOSDEM 07:59
08:04 adu left
moritz m: 'abc'.split(/b/, :k) 08:05
camelia ( no output )
moritz m: say 'abc'.split(/b/, :k)
camelia rakudo-moar 2e1782: OUTPUT«(a 0 c)␤»
moritz waves back to lizmat and the devroom
m: say 'abc'.split(/b/) 08:06
camelia rakudo-moar 2e1782: OUTPUT«(a c)␤»
moritz m: say 'abc'.split(/b/, :v)
camelia rakudo-moar 2e1782: OUTPUT«(a ｢b｣ c)␤»
moritz m: say 'abc'.split(/b/, :kv)
camelia rakudo-moar 2e1782: OUTPUT«(a 0 ｢b｣ c)␤»
moritz m: say 'abc'.split(/b/, :p)
camelia rakudo-moar 2e1782: OUTPUT«(a 0 => ｢b｣ c)␤»
lizmat moritz: you have a question about those ? 08:08
moritz m: say 'abc'.split(/b/, :p, :k)
camelia rakudo-moar 2e1782: OUTPUT«Unsupported combination of adverbs (k p) passed to split on Str␤ in block <unit> at /tmp/Ezx2_U4qct line 1␤␤»
moritz lizmat: no; just getting a feel for what they do
08:08 go|dfish left
lizmat okidoki 08:08
moritz m: say 'abc'.split(/b/, :kv).elems 08:10
camelia rakudo-moar 2e1782: OUTPUT«4␤»
08:15 darutoko joined 08:21 go|dfish joined 08:25 rindolf joined
lizmat m: m: say 'abc'.split("b", :kv) # faster 08:25
camelia rakudo-moar 2e1782: OUTPUT«(a 0 b c)␤»
08:33 hankache joined
hankache .seen FROGGS 08:34
yoleaux I saw FROGGS 29 Jan 2016 18:28Z in #perl6: <FROGGS> [Tux]: META6.json though
hankache anyone using perl6 on windows ? 08:36
lizmat jnthn is, afaik
hankache hiya lizmat
lizmat hankache o/ 08:37
hankache i installed the RC1 MSI for windows to test it
panda is not working correctly. I wanted to make sure that it is not an issue on my machine before reporting it to FROGGS 08:38
lizmat hankache : sorry, can't check that myself 08:39
hankache anytime i try to install a module if always fails at the test stage with the same error messages 08:40
no such file in sub run-and-gather-output 08:41
lizmat that doesn't ring a bell with me... but that doesn't mean much :-) 08:42
hankache no worries 08:44
08:45 hankache left 08:50 domidumont joined 08:55 domidumont left 08:56 domidumont joined 09:02 jeek left
nine waves from 2 rwos behind lizmat 09:09
El_Che_ :) 09:10
you just came in before I closed the room
09:10 El_Che_ is now known as El_Che
nine And now I know your nickname on IRC :) 09:11
El_Che hehe 09:13
I tend to use a different nick on each platform
:)
nine: I leart yeasterday the pronunciation of your name/nick. I prnounced as a number in my head 09:16
09:21 eseyman joined 09:25 lnrdo_ joined
nine El_Che: most people do. That's why I used to go by "niner" in English speaking circles. 09:37
09:39 jeek joined 09:43 lnrdo_ left 09:44 lnrdo_ joined
nine .tell hankache panda requires a "prove.bat" command to be available for testing. This is a Perl 5 script and I guess strawberry perl is the easiest way to get it. 09:51
yoleaux nine: I'll pass your message to hankache.
dalek c: 72520a8 | moritz++ | doc/Type/Cool.pod:
split(:all) is gone, replaced by :v. Closes #375
 09:56
09:59 musiKk joined, kjs_ joined 10:07 musiKk left 10:08 SCHAAP137 joined 10:11 kanishka joined
[Tux] $ panda install CSV::Parser 10:14
==> Fetching CSV::Parser
Must specify something as a path: did you mean '.' for the current directory?
in method update-from-meta-file at /pro/3gl/CPAN/rakudobrew/moar-nom/install/share/perl6/site/sources/690D6834E4624C39C1A50928F4F02428C9A781E4 line 23
test 22.837
test-t 12.294
10:22 lnrdo_ left
lizmat m: subset Foo where *.chars < 7; use MONKEY-TYPING; augment class Str { method Foo() { self.chars < 8 ?? self !! self.substr(0,7) } }; sub aa(Foo(Str) $a ) { dd $a }; aa "abcdefghijklmnopqrstuvwxyz"' # example of a coercing subset Foo 10:29
camelia rakudo-moar 2e1782: OUTPUT«5===SORRY!5=== Error while compiling /tmp/LAp5DyBTOU␤Two terms in a row␤at /tmp/LAp5DyBTOU:1␤------> 3dd $a }; aa "abcdefghijklmnopqrstuvwxyz"7⏏5' # example of a coercing subset Foo␤ expecting any of:␤ infix␤ infix …»
lizmat m: subset Foo where *.chars < 7; use MONKEY-TYPING; augment class Str { method Foo() { self.chars < 8 ?? self !! self.substr(0,7) } }; sub aa(Foo(Str) $a ) { dd $a }; aa "abcdefghijklmnopqrstuvwxyz" # example of a coercing subset Foo 10:30
camelia rakudo-moar 2e1782: OUTPUT«"abcdefg"␤»
lizmat m: my $max = 42; subset Foo where * < $max; my Foo $a = 17; dd $a; $max = 3; my Foo $b = 17; dd $b # example of a subtype depending on a variable value 10:32
camelia rakudo-moar 2e1782: OUTPUT«Int $a = 17␤Type check failed in assignment to $b; expected Foo but got Int␤ in block <unit> at /tmp/ccy5JME9WD line 1␤␤»
lizmat (answers to questions that came up during Ovid's talk
)
El_Che eiro's quesiton 10:33
10:33 ChoHag left
lizmat the latterm yes 10:33
*latter,
CIAvash .tell Herby_ Is this what you want? my $var = 42; given $var { when * < 50 { say "it's less than 50" } } 10:37
yoleaux CIAvash: I'll pass your message to Herby_.
10:45 kjs_ left 10:53 dolmen joined, leont joined 10:54 flussence left 10:55 flussence joined 10:59 lizmat left, ChoHag joined 11:00 spider-mario joined 11:02 leont left 11:07 ely-se joined 11:10 RabidGravy joined 11:13 kjs_ joined
ely-se hey the FOSDEM streams actually work this year :D 11:16
not on the website though, but in VLC 11:17
11:21 TEttinger left
llfourn ely-se: do you get audio> 11:31
ely-se yes but only if I stream with VLC 11:32
11:32 musiKk joined
ely-se not on the website 11:32
llfourn I don't seem to get audio in the perl room 11:33
11:33 Zero_Dogg left, dolmen left
llfourn (throuhg vlc) 11:33
11:33 Zero_Dogg joined, Zero_Dogg left, Zero_Dogg joined 11:34 lizmat joined
llfourn watches Crust guy fiddle with mic in hopeful anticipation 11:36
11:39 brrt joined
brrt \o #perl6 11:39
anybody at FOSDEM today
that I haven't seen yet?
llfourn brrt: are you there?
ely-se llfourn: maybe it's broken for Perl :P 11:40
I'm watching Go talks
brrt yeah, i'm in Jansson currently
ely-se: wasn't the current go talk cancelled? 11:41
llfourn ely-se: yep, he's started talking and it aint workin
ely-se brrt: yeah but someone else jumped in
brrt cool, what about
ely-se JSON encoding and code generation
llfourn hmm Go doesn't work for me either 11:42
brrt hmmm 11:43
Jansson is a talk from facebook? not my cup of tea
llfourn ah the audio on the .flv works 11:46
even in the Perl room! 11:47
11:48 brrt left 11:53 kjs_ left 11:54 lnrdo_ joined
llfourn I think the Crust guy just said he used EVAL 'use lib ...' to get around no use lib in modules :S 11:59
lizmat llfourn: yeah, and needing to use use MONKEY-SEE-NO-EVAL
on top of that
llfourn that's a little scary!
lizmat: I didn't catch why it needs EVAL if you wanna ask that from me :D 12:00
lizmat: i mean use lib
lizmat: thanks! 12:01
RabidGravy I was looking at Crust last night for various reasons
12:03 ely-se left 12:11 araujo_ joined 12:12 araujo_ left 12:13 araujo_ joined 12:35 dg left, go|dfish left, p_p_r left, TimToady left, abaugher left, sftp left, pRiVi left, Guest95622 left, cfedde left, ilbot3 left, ashleydev left, ab5tract left, au left, garu_ left, awwaiid left, esh left, silug left, shadowpaste left, bakedb left, DrForr left, autogen_ left, isBEKaml left, Cabanossi left, alnewkirk left, anshin left, [particle] left, olinkl left, cognominal left, vendethiel left, drforr1 left, robinsmidsrod left, DrParis__ left, kshannon left, mithaldu_ left, vike left, diegok left, grondilu left, Grauwolf left, Kcop left, haircode left, noganex left, El_Che left, eyck left, gregf_ left, ruoso left, aigan_ left, drrho left, zhmylove left, Peter_R left, felher left, emdashcomma left, aindilis left, sjohnsen left, xiaomiao left, SHODAN left, lestrrat left, lnrdo_ left, kanishka left, jeek left, mattp_ left, jack_rabbit left, telex left, advwp left, autarch left, Psyche^_ left, ggoebel15 left, stmuk left, ambs left, boegel left, jevin left, dalek left, frew left, ugexe left, avuserow left, pierrot left, konobi left, mephinet left, mst left, mre left, bapa left, rvchangue left, sm0x left, tony-o_ left, petercommand left, mkz_ left, nemo left, raydiak left, ingy left, yeltzooo9 left, jferrero_ left, squain_ left, andrewalker left, hobbs left, jnap left, FatalNIX left, mspo left, pnu left, Spot__ left, Humbedooh left, krunen left, solarbunny left, domidumont left, burnersk left, nebuchadnezzar left, matt_ left, eiro left, jantore left, wtw left, sjohnson left, crux left, dpk left, avar left, literal left, jervo left, rarara left, babaj left, kaare_ left, lizmat left, musiKk left, ChoHag left, M-Illandan left, geekosaur left, cpage_ left, firstdayonthejob left, bpmedley left, gtodd left, sammers left, lsm-desktop left, [Tux] left, keix left, xnrand left, richi235 left, Juerd left, rodarmor left, MadcapJake left, domm__ left, breinbaas left, retupmoca left, tinita left, jdv79 left, mindos left, stux|RC-only left, ranguard left, DarthGandalf left, riatre left, cpage left, lucs left, jsimonet left, integral left, jnthn left, lnx left, sivoais left, Hotkeys left, arnsholt left, dindinx left, japhb left, SCHAAP137 left, ELBeavers left, xinming left, yurivish_ left, nine left, cxreg left, salv0 left, Upasaka left, mprelude left, sno left, cosarara left, Amnez777- left, tadzik left, luiz_lha left, risou left, shmibs left, M-matthew left, rudi_s left, notbenh left, Util left, rdleon left, agentzh left, ponbiki left, saaki left, yoleaux left, cibs left, Zero_Dogg left, RabidGravy left, darutoko left, Fleurety left, johan left, daxim left, pyrimidi_ left, mohae left, apathor left, nightfrog left, Sgeo left, araujo left, salva left, ribasushi left, keusej left, woodruffw left, charsbar__ left, |Tux| left, Celelibi left, eternaleye left, perlpilo1 left, cfloare_ left, leedo left, adrusi left, krakan left, xxpor left, maddingue left, profan left, polyfloyd left, f3ew left, Gothmog_ left, gypsydave5 left, flussence left, rindolf left, Lucas_One_ left, jast left, Bucciarati left, nchambers left, timotimo left, gfldex left, DrPete left, [ptc] left, samb1 left, zacts left, BooK left, vytas left, JimmyZ left, moritz left, khisanth_ left, dylanwh left, rafl left, decent_ left, atta_ left, bartolin left, erdic left, luis left, ggherdov_ left, corbyhaas_ left, kipd left, avalenn left, PotatoGim left, jcallen left, rhr_ left, b^_^d left, xdg left, ilbelkyr left, hacst left, dsp- left, sunnavy left, rjbs left, huf left, TeamBlast left, freeze left, sjn left, BinGOs left, gabiruh left, araujo_ left, kent\n left, nowan left, caasih left, sQuEE` left, eseyman left, jojotus left, jameslenz left, avenj left, roguelazer left, tth left, flaviusb left, zostay left, colomon left, gensym left, pochi left, bitmap left, Timbus left, huggable left, Rotwang left, mtj_ left, moznion left, cosimo_ left, Ulti left, bhm_ left, Mouq left, musca left, dustinm` left, mrsolo left, SmokeMachine___ left, chansen_ left, clkao left, BuildTheRobots left, hoelzro left, edenc left, Grrrr left, skarn left, ilmari left, Jonis left, perigrin left, simcop2387 left, kst left, broquaint left, camelia left, Amnez777 joined, Woodi left, Amnez777 left, Exodist left, brabo left, [Coke] left, pi4 left, llfourn left, xenu left, hahainternet left, a3r0 left, gypsydave5 joined, Gothmog_ joined, f3ew joined, profan joined, maddingue joined, polyfloyd joined, krakan joined, adrusi joined, leedo joined, cfloare_ joined, perlpilo1 joined, Celelibi joined, eternaleye joined, |Tux| joined, charsbar__ joined, woodruffw joined, keusej joined, salva joined, xxpor joined, Sgeo joined, nightfrog joined, apathor joined, mohae joined, pyrimidi_ joined, daxim joined, johan joined, Fleurety joined, darutoko joined, Zero_Dogg joined, RabidGravy joined, broquaint joined, kst joined, simcop2387 joined, perigrin joined, Jonis joined, ilmari joined, skarn joined, Grrrr joined, hoelzro joined, BuildTheRobots joined, clkao joined, chansen_ joined, SmokeMachine___ joined, mrsolo joined, dustinm` joined, musca joined, Mouq joined, bhm_ joined, Ulti joined, cosimo_ joined, moznion joined, mtj_ joined, Rotwang joined, huggable joined, Timbus joined, bitmap joined, pochi joined, gensym joined, colomon joined, zostay joined, flaviusb joined, edenc joined, tth joined, roguelazer joined, avenj joined, jameslenz joined, jojotus joined, sQuEE` joined, caasih joined, nowan joined, kent\n joined, eseyman joined, araujo_ joined, autogen_ joined, DrForr joined, bakedb joined, shadowpaste joined, silug joined, esh joined, awwaiid joined, garu_ joined, au joined, ab5tract joined, ashleydev joined, ilbot3 joined, cfedde joined, Guest95622 joined, pRiVi joined, sftp joined, abaugher joined, TimToady joined, p_p_r joined, go|dfish joined, ribasushi joined, xnrand joined, diakopter joined, yoleaux joined, Guest92014 joined, lnrdo_ joined, lizmat joined, musiKk joined, ChoHag joined, kanishka joined, jeek joined, domidumont joined, isBEKaml joined, mattp_ joined, M-Illandan joined, Cabanossi joined, noganex joined, jack_rabbit joined, [particle] joined, olinkl joined, geekosaur joined, cognominal joined, vendethiel joined, cpage_ joined, drforr1 joined, firstdayonthejob joined, telex joined, alnewkirk joined, bpmedley joined, gtodd joined, advwp joined, anshin joined, autarch joined, robinsmidsrod joined, DrParis__ joined, Psyche^_ joined, El_Che joined, eyck joined, ggoebel15 joined, stmuk joined, ambs joined, rarara joined, boegel joined, sammers joined, jevin joined, kshannon joined, mithaldu_ joined, vike joined, hitchcock.freenode.net sets mode: +v yoleaux, gregf_ joined, babaj joined, ruoso joined, dalek joined, lsm-desktop joined, diegok joined, [Tux] joined, grondilu joined, keix joined, frew joined, Grauwolf joined, ugexe joined, avuserow joined, pierrot joined, aigan_ joined, jantore joined, drrho joined, zhmylove joined, konobi joined, Peter_R joined, Kcop joined, felher joined, richi235 joined, emdashcomma joined, aindilis joined, mephinet joined, wtw joined, Juerd joined, mst joined, mre joined, rodarmor joined, MadcapJake joined, haircode joined, burnersk joined, bapa joined, rvchangue joined, sjohnsen joined, sm0x joined, xiaomiao joined, tony-o_ joined, petercommand joined, SHODAN joined, mkz_ joined, domm__ joined, nemo joined, breinbaas joined, lestrrat joined, hitchcock.freenode.net sets mode: +v dalek, raydiak joined, retupmoca joined, ingy joined, yeltzooo9 joined, sjohnson joined, jferrero_ joined, tinita joined, squain_ joined, jdv79 joined 12:36 andrewalker joined, crux joined, hobbs joined, jnap joined, mindos joined, dpk joined, stux|RC-only joined, FatalNIX joined, solarbunny joined, krunen joined, Humbedooh joined, Spot__ joined, pnu joined, mspo joined, eiro joined, matt_ joined, nebuchadnezzar joined, literal joined, avar joined, jervo joined, ranguard joined, DarthGandalf joined, riatre joined, cpage joined, lucs joined, jsimonet joined, integral joined, jnthn joined, lnx joined, dindinx joined, sivoais joined, Hotkeys joined, arnsholt joined, japhb joined, Praise left, kaare__ joined, keix left, flussence joined, rindolf joined, Lucas_One_ joined, jast joined, Bucciarati joined, timotimo joined, gfldex joined, DrPete joined, [ptc] joined, samb1 joined, BooK joined, vytas joined, JimmyZ joined, moritz joined, khisanth_ joined, dylanwh joined, rafl joined, decent_ joined, atta_ joined, bartolin joined, erdic joined, luis joined, corbyhaas_ joined, kipd joined, xdg joined, avalenn joined, PotatoGim joined, jcallen joined, rhr_ joined, b^_^d joined, ilbelkyr joined, hacst joined, dsp- joined, gabiruh joined, sunnavy joined, rjbs joined, huf joined, TeamBlast joined, freeze joined, sjn joined, BinGOs joined, dgl joined, ELBeavers joined, xinming joined, yurivish_ joined, nine joined, cxreg joined, salv0 joined, Upasaka joined, mprelude joined, sno joined, cosarara joined, tadzik joined, luiz_lha joined, risou joined, shmibs joined, M-matthew joined, rudi_s joined, notbenh joined, Util joined, rdleon joined, agentzh joined, ponbiki joined, saaki joined, MilkmanDan joined, ilbelkyr left, ggherdov_ joined, nchambers joined, ilbelkyr joined, obarb joined, [Coke]_ joined, llfourn_ joined, xenu_ joined, Amnez777 joined, exodist_ joined, Amnez777 left, hahainte1net joined, Woodi_ joined, a3r0_ joined, cibs joined, yoleaux left, diakopter left, M-Illandan left, Amnez777 joined, yoleaux joined, diakopter joined, hitchcock.freenode.net sets mode: +v yoleaux, M-matthew left, BenGoldberg joined, Praise- joined, M-Illandan joined, kjs_ joined, camelia joined, isBEKaml left, apejens_ left, apejens joined
nine I just submitted a pull request for Crust to get rid of the EVAL "use lib 12:36
12:36 ChanServ sets mode: +v camelia, eternaleye left
RabidGravy nine++ 12:36
nine lizmat: Getopt::Tiny actually breaks on the removal of IO::ArgFiles. It's amazing how fast stuff gets used in the ecosystem. 12:37
12:37 Skarsnik joined
Skarsnik hello 12:38
masak .oO( if I had known my stuff'd get *used*, I'd've never put it in the ecosystem )
12:38 M-Illandan left
lizmat but I reverted the IO::CatHandle stuff, didn't, I ? 12:39
12:39 ely-se joined
nine lizmat: yes, you did :) Just wanted to point out that our speculations about noone using IO::ArgFiles directly were wrong. Someone already found a use case... So no matter how hidden some API is, someone will use it. Soon. 12:41
El_Che I have met stuff on the ecosystem that doesn't work with christmas rakudobrewed rakudo. Eg.: github.com/retupmoca/P6-Net-SMTP 12:42
12:42 Guest95622 left
El_Che although the travis test pass, it does not work when you use it 12:42
12:42 M-matthew joined 12:43 bowtie joined
El_Che 6.c/d stuff will be part of the metacpan-like thing? 12:43
12:43 M-Illandan joined, bowtie is now known as Guest80033
RabidGravy that'll be bad tests then 12:43
lizmat El_Che: don't think so
llfourn_ nine: how does that method work with precomp? 12:44
will things in Crust require'd afterwards still be preocmped?
nine No, they won't. And I'd love to discuss this with Crust people.
llfourn_ nine: kk I thought so thanks! 12:45
nine: I would love to hear your solution
nine If I can find one :)
llfourn_ I was thinking you would just have to figure out all the things that need to be in inc and then spawn a new perl6 -I whatever 12:46
jnthn only half paid attention to recent backlog, but didn't spot an explanation of the problem they were solving? :)
yoleaux 29 Jan 2016 19:45Z <RabidGravy> jnthn: next time you're looking at asynchronous stuff you may want to cast an eye over rt.perl.org/Ticket/Display.html?id=127428 almost certainly unintended
30 Jan 2016 14:37Z <orbus> jnthn: okay - so if I create a Supply off a channel, and then I have two taps on that Supply, they'll compete over the values from the Channel? That seems to be how it behaves now. If so, I want to get that in the docs. I'll either open a ticket or do a pr
llfourn_ jnthn: I think they are reading in a conf file or something at runtime which could include libs they need to add and things they need to load (or something similar) 12:47
lizmat jnthn: the problem is that in Plack / Crust, you have a config file that determines at runtime which modules should be loaded
12:48 llfourn_ is now known as llfourn
lizmat e.g. "enable "ContentLength" 12:48
would load a ContentLength module, for adding Content-Length HTTP output header
12:49 lnrdo_ left
nine And the "config" file is really just a Perl 6 script 12:49
12:50 SCHAAP137 joined
lizmat well, yeah, the "enable" is just an exported sub 12:50
llfourn mm I think they need to introduce a web app compile stage prior to re-running with a static @INC once all the paths are known
lizmat
.oO( pre-compile the app file ? )
 12:51
llfourn why not :)?
RabidGravy well if the model was slightly different, e.g. the app was a class that did a certain role then yeah 12:52
nine Well we are post-christmas, so we may as well try to figure out how to have precomp _and_ runtime repository chain changes. 12:54
Because it's really the same problem as the inability to use the precomp files generated at install time.
lizmat indeed
llfourn that would be lovely if it could be done! 12:55
nine That a Perl 6 PLACK may use more Perl 6 features for a nicer API is a separate issue :)
RabidGravy so, loading a module at runtime by name rather than path, is it possible right now? 12:57
12:58 M-matthew left
nine sure 12:58
RabidGravy (without using EVAL)
nine That's what require is for
llfourn m: require Test;
camelia ( no output )
nine Crust even does this correctly
RabidGravy no
m: require "Test"
camelia rakudo-moar 2e1782: OUTPUT«Could not find Test in:␤ /home/camelia/.perl6/2015.12-230-g2e17820␤ /home/camelia/rakudo-m-inst-1/share/perl6/site␤ /home/camelia/rakudo-m-inst-1/share/perl6/vendor␤ /home/camelia/rakudo-m-inst-1/share/perl6␤ CompUnit::Repository::…»
RabidGravy hence without EVAL
llfourn m: require ::("Test")
camelia ( no output )
nine m: my $module = "Test"; require ::($module); 12:59
camelia ( no output )
llfourn m: require ::("Test") <&ok>; ok 1;
camelia rakudo-moar 2e1782: OUTPUT«WARNINGS for /tmp/k6OKRJ6FZw:␤Useless use of constant string "Test" in sink context (lines 1, 1)␤ok 1 - ␤»
llfourn and that annoying warning will be gone once my PR is merged :D
nine llfourn: which PR is that? 13:00
13:00 M-Illandan left
llfourn nine: github.com/rakudo/rakudo/pull/694 13:00
the one you thanked me for using .tell earler :)
13:01 M-matthew joined
RabidGravy okay, what made me think that didn't work :-\ 13:01
nine Ah, yes :) Hopefully I'll find enough sleep in the days after FOSDEM to properly think it through
RabidGravy everyone++
llfourn nine: thanks! take your time :)
13:01 M-eternaleye joined
nine RabidGravy: while loading a module at runtime is easy, accessing the symbols in that module is hard, because they will not be there at compile time. 13:01
El_Che I have 2 days of conference after fosdem: there is a configuration management camp in Ghent mon and tue 13:02
lizmat that's a real long weekend for you
El_Che someone has to hassle all the ruby and python people there :) 13:03
lizmat :-)
13:03 M-Illandan joined
RabidGravy nine, for my use-case right now, I don't need the symbols just the type and the methods thereof 13:04
13:04 kid51 joined
llfourn RabidGravy: if you put it in brackets require should return whatever package has the same name as the target module 13:05
m: say (require Test);
camelia rakudo-moar 2e1782: OUTPUT«(Test)␤»
RabidGravy which saves the additional type lookup
llfourn yup
another tip with require is that it runs every time you hit it (unlike need and use) 13:07
so if you only want it to run the first time
m: for ^1000 { state $ = (require Test) }
camelia ( no output )
RabidGravy coolio, my actual use is rather niche, I have a bunch of methods that differ only by two pieces of information: a class and a "path" where the information for that class can be found 13:08
gfldex m: for ^1000 { once (require Test) }
camelia ( no output )
jnthn nine: One thought I had: the "only use the precomp store at the head of the chain" thing can be relaxed a little to "use the precomp store first encountered in the chain once skipping past things that have no modules"
nine: That'd mean that if a user-level repo has no installed modules, then we can use the system-wide installed precomps 13:09
llfourn gfldex: Thanks. That's my new feature of the day :)
RabidGravy so I figured a class trait that stashes the "path" and then a method trait that specifies the class by name, then the method is just load the class, get the "path" and so forth
so all the otherwise duplicated method bodies go away 13:10
llfourn where doe sit stash the path?
HOW?
does it*
RabidGravy yeah 13:11
llfourn RabidGravy: I did something which sounds similar to what you are saying github.com/LLFourn/p6-OO-Schema
(probably with different purposes)
in that you have like 'class lite' which has the same inheritence as the class but isn't loaded, but you can load it with a method 13:12
13:12 M-eternaleye left
nine jnthn: oh yes, I've had that idea at some point. As long as we handle the case correctly when the process is actually going to install modules into the head repo. 13:13
RabidGravy llfourn, yeah, that's similar, what it's actually for is a CouchDB thing where you have _stats, _config and so forth all represented by a separate class 13:14
nine Unfortunately I think we'll need a little API change to make that possible
llfourn RabidGravy: ah I think I get you :) 13:15
13:17 M-matthew left 13:18 CIAvash left 13:20 M-eternaleye joined, M-matthew joined
dalek osystem: 847f673 | (Sylvain Colinet)++ | META.list:
Rename GPTrixie meta file
 13:20
El_Che brrt in the second row? 13:23
lizmat denk dat ie buiten is 13:25
sorry, "think he's outside"
El_Che have you seen him at fosdem? 13:26
I don't know his face
I found his linkedin pic, but it's 2d :)
lizmat he wears glasses now 13:27
El_Che ok, looking for a white guy with glasses in fosdem
that can't be difficult :) 13:28
nine El_Che: he's kinda young, too ;)
Skarsnik are fosdem talk streamed?
nine yes
El_Che this one seem to work fine 13:29
I heard on #perl
13:30 M-eternaleye left 13:32 ely-se left 13:34 M-matthew left, M-eternaleye joined 13:35 Actualeyes joined, rindolf left 13:37 M-matthew joined
Skarsnik hm, it does not work for me :( 13:37
llfourn Skarsnik: which bit? 13:38
Skarsnik fosdem stream x)
llfourn stream-a.fosdem.org:8080/fosdem+h2214cam.flv # open with vlc
nine Actually...we could even take it a step further. We may note be able to serialize the whole DependencySpecification right now and therefore we cannot run the dependency resolution without actually loading a module's source code. 13:39
However we can serialize the short name. And if none of the repositories ahead in the chain contains a module with the same short name, we can be sure that it will not contain a module matching a DependencySpecification at all.
Skarsnik thx llfourn !
eiro hello everyone. i just connected. thanks for the example lizmat! awesome
(as always in perl6) 13:40
nine Therefore it can also not contain a newer version of said dependency and we're free to load the precompiled one from a repo down the chain.
Skarsnik damn the camera that does not show the board x) 13:41
llfourn mmm there is another stream for that
you will have to figure out how to open 2 vlcs
nine mplayer or ffplay? 13:42
Skarsnik I can open how many vlc I want x) 13:43
Just need to figure the url x)
llfourn stream-a.fosdem.org:8080/fosdem+h2214slides.flv 13:44
Skarsnik Ok thx :)
llfourn /Applications/VLC.app/Contents/MacOS/VLC # to open a second vlc on mac 13:45
13:46 M-eternaleye left
RabidGravy works with mplayer too if you're of that persuasion 13:47
13:51 dolmen joined 13:55 M-matthew left, Amnez777 left, Amnez777 joined 13:56 M-Illandan left
llfourn nice talk that was brrt on #perl6? 13:57
lizmat yup
13:59 AlexDaniel joined 14:01 hankache joined 14:03 brrt joined
lizmat brrt++ # nice presentation! 14:09
brrt thanks :-)
14:10 ely-se joined, brrt left, raiph joined 14:11 hahainte1net is now known as hahainternet 14:14 lnrdo_ joined 14:15 espadrine_ joined 14:16 xenu_ is now known as xenu
hankache computer, messages 14:18
yoleaux 09:51Z <nine> hankache: panda requires a "prove.bat" command to be available for testing. This is a Perl 5 script and I guess strawberry perl is the easiest way to get it.
hankache .tell nine thanks dear 14:19
yoleaux hankache: I'll pass your message to nine.
hankache .tell nine but this means that in order for it to work every person on Windows needs Perl 5 installed. While this is no problem on *nix, i find it counter intuitive to tell people to install Perl 5 in order for it to work. 14:21
yoleaux hankache: I'll pass your message to nine.
nine hankache: we know, a fix is in the works by leont++ 14:24
yoleaux 14:19Z <hankache> nine: thanks dear
14:21Z <hankache> nine: but this means that in order for it to work every person on Windows needs Perl 5 installed. While this is no problem on *nix, i find it counter intuitive to tell people to install Perl 5 in order for it to work.
14:26 pi4 joined
AlexDaniel separate streams for presentation and video! Great! 14:26
14:29 mscha joined
hankache nine: thanks 14:31
14:32 lichtkind joined 14:36 cdg joined 14:38 M-eternaleye joined 14:40 musiKk left, firstdayonthejob left 14:42 firstdayonthejob joined 14:44 kjs_ left
dalek c: 531bfd9 | (Zoffix Znet)++ | doc/Type/Cool.pod:
Document :skip-empty in .split
 14:49
c: 47c01e8 | (Zoffix Znet)++ | doc/Type/Cool.pod:
Remove accidentally added garbage
 14:50
14:50 muraiki joined
ely-se what does "dalek" mean? 14:50
AlexDaniel m: my $x = 1; sub foo($bar = $x++) { say $bar }; foo; foo; foo 14:51
camelia rakudo-moar 2e1782: OUTPUT«1␤2␤3␤»
masak ely-se: it's just a bot that watches some of our repos.
ely-se yes, I know, but where does the name come from?
masak Dr Who, I guess
en.wikipedia.org/wiki/Dalek
ely-se oh wait it's those robots from Dr Who isn't it
lizmat exterminate!
ely-se right :)
masak EX TER MIN ATE 14:52
some people will most likely enjoy www.youtube.com/watch?v=KB7HF82y9CI
RabidGravy and we all go and hide behind the sofa when they appear 14:53
14:58 dustinm` left
nine lizmat: :) 14:58
15:02 lizmat left
llfourn oh lizmat is up next 15:04
15:04 brrt joined
ely-se what happened to the Perl 6 to JS compiler? 15:04
llfourn ely-se: still a work in progress
nine it's in thecworks
moritz writing a compiler backend is slow work, especially if done by one person on the side 15:05
15:05 dustinm` joined
brrt go lizmat:-) 15:05
RabidGravy I'm ruined for POSIX regexes 15:06
brrt try emacs regexes
:-P
brrt is shutting up again due to presentation-starting
ely-se llfourn: oh ok :) 15:07
dalek osystem: 505ed52 | (Zoffix Znet)++ | META.list:
Text::CSV META.info -> META6.json
15:07 brrt left
moritz vim regexes are also "fun" 15:08
they come in at least three different flavours
15:12 leont joined
RabidGravy first thing I had forgotten was that back references start at \1 and \0 was the whole match 15:16
( was doing for file in t/*; do NEW=`echo $file | sed -e 's/\([0-9]\)-/\10-/'`; echo "$file -> $NEW"; git mv $file $NEW; done ) 15:17
15:17 lnrdo_ left 15:18 muraiki left 15:22 lnrdo_ joined 15:26 khw joined, dixvagon joined 15:31 a3r0_ left 15:35 kurahaupo joined
AlexDaniel mic! Miiic! :/ 15:37
Skarsnik yes what happened to the mic :( 15:38
15:38 leont left 15:40 zakharyas joined 15:46 perl6newbee joined 15:49 rindolf joined 15:51 CurtisOvidPoe joined
CurtisOvidPoe p6: await(^20).map: { start { sleep.rand; .print } } 15:52
camelia rakudo-moar 2e1782: OUTPUT«Must specify a Promise or Channel to await on (got a Int)␤ in block <unit> at /tmp/tmpfile line 1␤␤»
CurtisOvidPoe What have I missed?
El_Che await expects a promise 16:01
16:01 ZoffixWin joined, ZoffixWin left, ZoffixWin joined
hankache m: ^20.map: { await start { sleep.rand; .print } } 16:01
ZoffixWin Any way to make a sub operate on $_?
m: sub foo {say $_::OUTER}; given "foobar" { foo }
camelia rakudo-moar 2e1782: OUTPUT«(timeout)Potential difficulties:␤ Precedence of ^ is looser than method call; please parenthesize␤ at /tmp/n8etCWQCP6:1␤ ------> 3^207⏏5.map: { await start { sleep.rand; .print␤WARNINGS for /tmp/n8etCWQCP6:␤Useless use of "^" in ex…»
rakudo-moar 2e1782: OUTPUT«(Any)␤»
hankache m: (^20).map: { await start { sleep.rand; .print } }
camelia rakudo-moar 2e1782: OUTPUT«(timeout)»
hankache m: ^3.map: { await start { sleep.rand; .print } }
camelia rakudo-moar 2e1782: OUTPUT«(timeout)Potential difficulties:␤ Precedence of ^ is looser than method call; please parenthesize␤ at /tmp/lYHt2e7FkV:1␤ ------> 3^37⏏5.map: { await start { sleep.rand; .print␤WARNINGS for /tmp/lYHt2e7FkV:␤Useless use of "^" in exp…»
hankache m: (^3).map: { await start { sleep.rand; .print } }
ZoffixWin What are you doing?
camelia rakudo-moar 2e1782: OUTPUT«(timeout)»
llfourn what is sleep.rand
hankache see CurtisOvidPoe ^^
perl6newbee hi all
hankache i am trying to make his example work
llfourn m: say sleep.rand.^name
ZoffixWin hankache, sleep.rand sleeps forever and even if it returned it'd call .rand on Nil
camelia rakudo-moar 2e1782: OUTPUT«(timeout)»
llfourn m: say sleep.^name;
ah sleep just sleeps forever
hankache m: (^3).map: { await start { sleep(3); .print } }
16:01 CurtisOvidPoe left
camelia rakudo-moar 2e1782: OUTPUT«(timeout)» 16:01
rakudo-moar 2e1782: OUTPUT«012»
16:01 CurtisOvidPoe joined
ZoffixWin p6: await ( ^20 .map: { start { sleep.rand; .print } } ) 16:01
camelia rakudo-moar 2e1782: OUTPUT«Memory allocation failed; could not allocate 14064 bytes␤»
llfourn m: ^20.map: { await start { sleep rand; .print } }
camelia rakudo-moar 2e1782: OUTPUT«Potential difficulties:␤ Precedence of ^ is looser than method call; please parenthesize␤ at /tmp/DKAcTWX3SI:1␤ ------> 3^207⏏5.map: { await start { sleep rand; .print␤WARNINGS for /tmp/DKAcTWX3SI:␤Useless use of "^" in expression …»
ZoffixWin CurtisOvidPoe, perhaps you were looking for that ^?
hankache m: (^20).map: { await start { sleep(1); .print } }
llfourn m: (^20).map: { await start { sleep rand; .print } }
CurtisOvidPoe p6: await (^20).map: { start { sleep rand; .print } }
camelia rakudo-moar 2e1782: OUTPUT«012345678910Memory allocation failed; could not allocate 14000 bytes␤»
rakudo-moar 2e1782: OUTPUT«012345678910Memory allocation failed; could not allocate 14256 bytes␤»
rakudo-moar 2e1782: OUTPUT«Memory allocation failed; could not allocate 14312 bytes␤»
llfourn lol
hankache m: (^5).map: { await start { sleep(1); .print } }
camelia rakudo-moar 2e1782: OUTPUT«01234»
hankache CurtisOvidPoe (^5).map: { await start { sleep(1); .print } }
llfourn m: await (^20).map: { start { sleep rand; .print } }
camelia rakudo-moar 2e1782: OUTPUT«Memory allocation failed; could not allocate 14312 bytes␤»
ZoffixWin hankache, that'll await each promise before proceeding to the next one
May as well not use any Promises at all
llfourn m: await (^10).map: { start { sleep rand; .print } } 16:02
camelia rakudo-moar 2e1782: OUTPUT«2704165839»
hankache gives up
llfourn m: await (^15).map: { start { sleep rand; .print } }
camelia rakudo-moar 2e1782: OUTPUT«Memory allocation failed; could not allocate 14312 bytes␤»
ZoffixWin m: (^5).map: { await start { sleep(1) } }; say now - INIT now
camelia rakudo-moar 2e1782: OUTPUT«5.0069193␤»
ZoffixWin m: await (^5).map: { start { sleep(1) } }; say now - INIT now
camelia rakudo-moar 2e1782: OUTPUT«1.0054976␤»
ZoffixWin Pretty cool that 5 1-second sleeps on a 4-core box still complete in 1s :)
llfourn so camelia's memory limit is just super low?
ZoffixWin Yeah
hankache m: await (^5).map: { start { sleep(1); .print } }
camelia rakudo-moar 2e1782: OUTPUT«01234»
hankache so this is the correct one ^^ 16:03
hi perl6newbee
ZoffixWin m: use NativeCall; sub system is native {}; say system 'ulimit'
llfourn well the rand one is coolor imo but yes
camelia rakudo-moar 2e1782: OUTPUT«5===SORRY!5=== Error while compiling /tmp/Dakw169KOt␤Calling system(Str) will never work with declared signature ()␤at /tmp/Dakw169KOt:1␤------> 3ativeCall; sub system is native {}; say 7⏏5system 'ulimit'␤»
ZoffixWin m: use NativeCall; sub system(Str) is native {}; say system 'ulimit'
camelia rakudo-moar 2e1782: OUTPUT«30720␤(Mu)␤»
ZoffixWin hankache, yup 16:04
So... any idea for my thing? I want a sub that operates on $_
m: sub foo {say OUTER::CALLER::{$_} }; given "foobar" { foo }
camelia rakudo-moar 2e1782: OUTPUT«Use of uninitialized value $_ of type Any in string context␤Any of .^name, .perl, .gist, or .say can stringify undefined things, if needed. in sub foo at /tmp/QqsFXw2QCD line 1␤Nil␤»
16:04 cdg left 16:05 CurtisOvidPoe left
llfourn ZoffixWin: why are you using OUTER::CALLER and not just CALLER? 16:05
16:05 spider-mario left
hankache no idea 16:05
llfourn m: sub foo {say CALLER::{$_} }; given "foobar" { foo }
camelia rakudo-moar 2e1782: OUTPUT«Use of uninitialized value $_ of type Any in string context␤Any of .^name, .perl, .gist, or .say can stringify undefined things, if needed. in sub foo at /tmp/aK7OioMIIC line 1␤Nil␤»
ZoffixWin llfourn, saw it in rakudo source. Neither works though
llfourn m: sub foo {say CALLERS::{$_} }; given "foobar" { foo }
camelia rakudo-moar 2e1782: OUTPUT«Use of uninitialized value $_ of type Any in string context␤Any of .^name, .perl, .gist, or .say can stringify undefined things, if needed. in sub foo at /tmp/boeWVwDcMT line 1␤Nil␤»
llfourn m: sub foo {say CALLERS::<$_> }; given "foobar" { foo }
camelia rakudo-moar 2e1782: OUTPUT«foobar␤»
llfourn m: sub foo {say CALLER::<$_> }; given "foobar" { foo } 16:06
camelia rakudo-moar 2e1782: OUTPUT«foobar␤»
llfourn amidoinrite?
ZoffixWin Thanks
llfourn nw :D
16:11 hankache left, kmel joined
ZoffixWin hm 16:13
m: sub foo {say CALLER::<$_> }; foo for @*ARGV
camelia rakudo-moar 2e1782: OUTPUT«(HANDLED) Dynamic variable @*ARGV not found␤ in block <unit> at /tmp/MQy6LQDHLH line 1␤␤»
ZoffixWin llfourn, ^ any idea for how to take care of that stuff?
moritz use @*ARGS instead? 16:14
ZoffixWin moritz, that's not the issue, the sub cases the failure to be handled.
s/cases/causes/;
moritz oh
llfourn I don't even get it what's happening? 16:15
16:15 mcmillhj joined
ZoffixWin m: sub foo { CALLER::<$_> }; foo for @*ARGV; say "Weee"; 16:15
camelia rakudo-moar 2e1782: OUTPUT«Weee␤»
ZoffixWin Here it's clearer.
llfourn oh
ZoffixWin m: sub foo { CALLER::<$_> }; foo for @*ARGVZOMFGGG; say "Weee";
camelia rakudo-moar 2e1782: OUTPUT«Weee␤»
moritz ZoffixWin: possibly something in sub DYNAMIC
16:15 mcmillhj left
llfourn m: .say for @*NOTEXIST; 16:15
camelia rakudo-moar 2e1782: OUTPUT«Dynamic variable @*NOTEXIST not found␤ in block <unit> at /tmp/CAGi6hGXzn line 1␤␤Actually thrown at:␤ in block <unit> at /tmp/CAGi6hGXzn line 1␤␤»
llfourn m: say "wut" for @*NOTEXIST; 16:16
camelia rakudo-moar 2e1782: OUTPUT«wut␤»
llfourn m: say .^name for @*NOTEXIST;
camelia rakudo-moar 2e1782: OUTPUT«Failure␤»
llfourn I see.
ZoffixWin m: say "wut" for @*NOTEXIST; say "Weee"
16:16 lnrdo_ left
camelia rakudo-moar 2e1782: OUTPUT«wut␤Weee␤» 16:16
ZoffixWin Methinks this is a bug. 16:17
llfourn it's definetly bugish
16:17 lnrdo_ joined
llfourn imo a non-existent @* should just be empty 16:17
16:17 mr-foobar joined 16:19 musiKk joined
llfourn m: my @*a := 3; say @*a.WHAT # there is also this I noticed the other day 16:19
camelia rakudo-moar 2e1782: OUTPUT«(Int)␤»
16:20 kmel left, ely-se left 16:21 SCHAAP137 left, lnrdo_ left
ZoffixWin m: say (0 .. 0xFFFF)».chr.join 16:23
camelia rakudo-moar 2e1782: OUTPUT«Error encoding UTF-8 string: could not encode codepoint 55296␤ in block <unit> at /tmp/oNTIkNNjyj line 1␤␤»
ZoffixWin :/
16:24 yqt joined
ZoffixWin m: say sprintf '%x', 55296 16:25
camelia rakudo-moar 2e1782: OUTPUT«d800␤»
ZoffixWin .u 0xd800
yoleaux No characters found
16:26 mscha left
geekosaur hm. isn't that range reserved for surrogate pairs? 16:27
unicode makes no promise that all codepoints are valid
16:28 skids joined
ZoffixWin k 16:28
geekosaur perl 6 shouldn't need surrogate pairs, and encountering a codepoint int hat range means something was incorrectly decoded at some point, so an exception is appropriate 16:30
(they're a hack for utf16 to handle non-BMP codepoints) 16:31
16:32 SCHAAP137 joined 16:36 zakharyas left 16:44 musiKk left
AlexDaniel m: $_ = 25; sub foo($x is rw) { $x++ }; foo($_); say $_ 16:44
camelia rakudo-moar 2e1782: OUTPUT«26␤»
AlexDaniel ZoffixWin: ↑ ?? 16:45
ZoffixWin Hm?
AlexDaniel “So... any idea for my thing? I want a sub that operates on $_” 16:46
ah 16:47
ZoffixWin: you just don't want to pass it explicitly?
ZoffixWin Right
CALLERS::<$_> works
AlexDaniel indeed
geekosaur is not sure that implicit stuff like that is a great idea, tbh 16:50
vendethiel ^
please don't..
ZoffixWin is replicating a Perl 5 sub
geekosaur perl 5 did it a lot... but it's also part of why people decry spooky-action-at-a-distance in p5
vendethiel "let's try making different mistakes"
geekosaur making different mistakes can lead to you learning better ways to do things. making the same mistake over and over usually doesn't >.> 16:51
ZoffixWin Making different mistakes will make my port of Perl 5 code behave differently. 16:52
16:52 ZoffixWin left 16:58 ely-se joined 17:08 Guest92014 left 17:10 musiKk joined, zacts joined 17:18 pdl joined 17:19 hankache joined 17:21 ZoffixWin joined
ZoffixWin Is this a bug? 17:21
m: say so '÷' ~~ /<[ \x[f7] ]>/
camelia rakudo-moar 2e1782: OUTPUT«True␤»
ZoffixWin m: say so '÷' ~~ /<[ \x[f7] \x[34f] ]>/
camelia rakudo-moar 2e1782: OUTPUT«False␤»
ZoffixWin :/
.u U+034F 17:22
yoleaux U+034F COMBINING GRAPHEME JOINER [Mn] (◌͏)
ZoffixWin Is the char class actually combining things :/
geekosaur ? 17:23
ZoffixWin ?? 17:24
geekosaur I don;t quite understand what you're asking about
ZoffixWin Why is the match failing if I add another char into my character class
I want to match a char that is either ÷ or ◌, but the character class fails.
17:25 nebuchadnezzar left
ZoffixWin And fails only when the 34f follows the f7 17:25
m: say so '÷' ~~ /<[ \x[34f] \x[f7] ]>/
camelia rakudo-moar 2e1782: OUTPUT«True␤»
ZoffixWin m: say so '÷' ~~ /<[ \x[f7] \x[34f] ]>/
camelia rakudo-moar 2e1782: OUTPUT«False␤»
ZoffixWin This works, but is 3 times slower :/ 17:26
m: say so '÷' ~~ /[ \x[f7] | \x[34f] ]/
camelia rakudo-moar 2e1782: OUTPUT«True␤»
hankache if i have a string of 1000 characters, how can i split it into multiple lines? having each line hold 10 characters?
ZoffixWin hankache, .comb(10)
geekosaur oh, I see
ZoffixWin hankache, .comb(10).join: "\n" I suppose
17:26 isBEKaml joined
hankache thanks ZoffixWin will try it now 17:27
geekosaur that said, iirc u+034f is weird and I would not be surprised if it acts up
ZoffixWin Yeah, it's a "joiner", so I'm guessing it joins up with the division
m: say "\x[f7]\x[34f]"
camelia rakudo-moar 2e1782: OUTPUT«÷͏␤»
ZoffixWin :/
17:27 musiKk left
geekosaur it's not really a joiner 17:28
that's the weird 17:29
it's more a separator, in fact (it prevents a combining character from combining)
ZoffixWin ah
pdl m: say so "÷\x[34f]" ~~ /<[ \x[f7] \x[34f] ]>/
camelia rakudo-moar 2e1782: OUTPUT«True␤»
geekosaur so it's special cased, and I suspect it's the special case that's biting
pdl m: say so "÷\x[34f]" ~~ /^<[ \x[f7] \x[34f] ]>$/ 17:30
camelia rakudo-moar 2e1782: OUTPUT«True␤»
geekosaur also possibly some places that incorrectly treat it as a combining char still need to be fixed...
17:38 araujo_ left, araujo_ joined
ZoffixWin :S "chars requires a concrete string, but got null" 17:39
Not even using .chars
17:43 yqt left 17:49 Guest80033 left 17:50 molaf joined 17:52 eyck left 17:53 eyck joined
ZoffixWin Planned time: 20 minutes; Spent time so far: 2 hours 17:55
Programming is hard :(
diakopter Planning is hard!
ZoffixWin .u U+2adc 18:01
yoleaux U+2ADC FORKING [Sm] (􏿽xE2􏿽xAB􏿽x9C)
ZoffixWin :o
geekosaur what's that saying about the first half of a project takes 90% of the time and the second half takes the other 90%? >.> 18:02
ZoffixWin heh 18:03
18:04 sevvie joined
MadcapJake ok so i totally don't understand how to do heroku buildpacks, so. much. bash. 18:07
i tried to make a rakudo buildpack, but failed 18:08
unfortunately any time rakudo moves locations, hard coded paths to moar fail 18:09
actually, now that i think about it more, i'm guessing i'd have to build nqp and moar separately 18:10
18:10 dolmen left 18:12 eyck left, eyck joined
dalek osystem: f0e307a | (Zoffix Znet)++ | META.list:
Add String::Quotemeta to ecosystem

Implementation of Perl 5's quotemeta subroutine: github.com/zoffixznet/perl6-String-Quotemeta		 18:17
18:17 mr-foobar left 18:24 nebuchadnezzar joined 18:29 Actualeyes left 18:36 mr-foobar joined 18:37 eyck left, Gido_back joined 18:38 Gido_back left 18:43 p_p_r left 18:46 p_p_r joined 18:49 dixvagon left 18:52 mr-foobar left 18:54 eyck joined 18:55 mr-foobar joined 18:56 rindolf left, ely-se left 18:57 ely-se joined 18:59 rindolf joined 19:07 espadrine_ left 19:09 johan left
ely-se m: class C { has Str $.x; }; C.new(x => 'a').say; C.new(x => Nil).say; 19:11
camelia rakudo-moar 2e1782: OUTPUT«C.new(x => "a")␤C.new(x => Str)␤»
ely-se m: class C { has Str:D $.x; }; C.new(x => 'a').say; C.new(x => Nil).say; 19:13
camelia rakudo-moar 2e1782: OUTPUT«5===SORRY!5=== Error while compiling /tmp/Lt8_YiBrLq␤Variable definition of type Str:D requires an initializer␤at /tmp/Lt8_YiBrLq:1␤------> 3class C { has Str:D $.x7⏏5; }; C.new(x => 'a').say; C.new(x => Nil␤ expecting any of:␤ …»
ely-se :(
19:14 johan joined 19:17 b2gills left 19:20 perl6newbee left
dalek kudo-star-daily: 9024448 | coke++ | log/ (10 files):
today (automated commit)
 19:23
kudo-star-daily: 293f395 | coke++ | log/ (8 files):
today (automated commit)
kudo-star-daily: 5fcc6f3 | coke++ | log/ (2 files):
today (automated commit)
rl6-roast-data: cfc31df | coke++ | / (6 files):
today (automated commit)
19:26 kjs_ joined 19:27 musiKk joined 19:30 b2gills joined 19:33 ely-se left 19:36 b2gills left 19:39 ely-se joined 19:40 yurivish joined
hankache class A {}; class B is A {}; class C is A {}; is there a way using the MOP to do something like: say A.children ? and the return value would be B,C 19:45
?
19:45 yqt joined
ugexe a blog post about implementing s22's Distribution interface: ugexe.com/perl6-distribution-though...proposals/ 19:45
19:47 damnlie joined
hankache ugexe++ 19:47
19:47 darutoko left 19:49 b2gills joined 19:51 orbus joined 19:57 kjs_ left
musiKk Is it possible to add '.' to the module search path by default? 19:57
ugexe add it to PERL6LIB env var maybe 19:58
MadcapJake moarvm's make doesn't respect the «--silent» flag 19:59
hankache m: class D { }; class C1 is D { }; class C2 is D { }; say C2.^parents 20:00
camelia rakudo-moar 2e1782: OUTPUT«((D))␤»
hankache i need something like this: D.^children 20:01
any ideas?
musiKk ugexe: Thanks!
20:11 kjs_ joined
MadcapJake when creating a 2015.12 build of perl6 and panda, should i use panda's 2015.12 tag tar.gz or just «git clone --recursive» as the readme states? 20:15
timotimo o/ 20:16
any reason to backlog today?
hankache hola timo
ZoffixWin timotimo, yes, someone came up with cure for cancer 20:18
hankache, no, there isn't a ^children. You may find some of this useful, but it won't work for anonymous classes and stuff like that: github.com/zoffixznet/perl6-Evil/b...b/Evil.pm6 20:19
20:19 domidumont left
hankache ZoffixWin I'll take a look at i. Thanks 20:20
20:22 chienjo joined
chienjo p6: say |('A'..'Z') 20:24
camelia rakudo-moar 2e1782: OUTPUT«ABCDEFGHIJKLMNOPQRSTUVWXYZ␤»
20:28 spider-mario joined 20:29 rindolf left
chienjo p6: my @lower = 'a'..'z'; say @lower.comb 20:31
camelia rakudo-moar 2e1782: OUTPUT«(a b c d e f g h i j k l m n o p q r s t u v w x y z)␤»
chienjo p6: my @lower = 'a'..'z'; say @lower.pick(*)
camelia rakudo-moar 2e1782: OUTPUT«(i c z q r o u l h f m t p y a d e w s n j x v k g b)␤»
chienjo p6: my @lower = 'a'..'z'; say @lower.pick(^5)
camelia rakudo-moar 2e1782: OUTPUT«(g j b y o)␤» 20:32
chienjo p6: my @lower = 'a'..'z'; say @lower.pick(5)
camelia rakudo-moar 2e1782: OUTPUT«(u g c w n)␤»
20:33 Praise- is now known as Praise 20:35 kaare__ left
chienjo p6: my @lower = 'a'..'z'; say @lower.combinations(4) 20:36
camelia rakudo-moar 2e1782: OUTPUT«((a b c d) (a b c e) (a b c f) (a b c g) (a b c h) (a b c i) (a b c j) (a b c k) (a b c l) (a b c m) (a b c n) (a b c o) (a b c p) (a b c q) (a b c r) (a b c s) (a b c t) (a b c u) (a b c v) (a b c w) (a b c x) (a b c y) (a b c z) (a b d e) (a b d f) (a b …»
20:37 lokien_ joined
chienjo p6: my @lower = 'a'..'z'; say @lower.combinations(4).elems 20:38
camelia rakudo-moar 2e1782: OUTPUT«14950␤»
20:38 pmurias joined
pmurias ely-se: re Perl 6 -> JavaScript compiler, it's still at the NQP stage 20:39
timotimo the fosdem totally took me by surprise 20:41
20:41 renatolrr joined
MadcapJake timotimo: how so? 20:41
20:42 renatolrr left, chienjo left
vendethiel by surprise 20:42
Juerd How was fosdem?
I had to miss it 20:43
Spent most of my days in bed.
20:44 chienjo joined
MadcapJake vendethiel: -_- 20:45
vendethiel I hope more videos will be available this year
MadcapJake I spent the weekend battling heroku (and losing) 20:46
moritz MadcapJake: try hosting it on hack.p6c.org? 20:47
MadcapJake gladly!
i asked earlier but never received a reply
20:48 yurivish left
timotimo MadcapJake: it just happened without me realizing it'd happen at that time 20:48
MadcapJake ahh i see
moritz MadcapJake: what username do you want? 20:49
20:50 renatolrr joined
moritz MadcapJake: oh wait, are you jrusso on there? 20:50
MadcapJake yeah
moritz MadcapJake: so, what do you need?
20:50 chienjo left
MadcapJake well i created a free domain «perl6.bestforever.com» so i'll need to direct that to a web server, i already have the website built: github.com/MadcapJake/perl6-slack-inviter 20:51
moritz MadcapJake: can you control the DNS entries of perl6.bestforever.com? 20:53
MadcapJake yeah
moritz MadcapJake: I'd propose we do the following: you start a web server listening on localhost on a free port, I add a virtualhost for that domain, and set up a reverse proxy that connects to your port; then you point the DNS to hack's public IP (213.95.82.54) 20:55
MadcapJake: and when that works for a while, you write a systemd service file that keeps the Perl 6 process running
20:56 wamba joined 21:05 BenGoldberg left
MadcapJake ok, i may have forgotten my password :P 21:07
moritz MadcapJake: should I reset it?
MadcapJake yeah, is the network unreachable because i tried too many times xD 21:09
21:11 lichtkind_ joined
hankache what is the difference between module and package ? 21:12
RabidGravy I'm confusled
21:12 lichtkind left
moritz hankache: a package can do less stuff basically 21:13
RabidGravy m: package F {}; package F{}
camelia ( no output )
RabidGravy m: module F {}; module F{}
camelia rakudo-moar 2e1782: OUTPUT«5===SORRY!5=== Error while compiling /tmp/IQ4ed0IYTm␤Redeclaration of symbol F␤at /tmp/IQ4ed0IYTm:1␤------> 3module F {}; module F7⏏5{}␤ expecting any of:␤ generic role␤»
moritz m: package A { ... }; class A { }
camelia rakudo-moar 2e1782: OUTPUT«5===SORRY!5=== Error while compiling /tmp/N8unmOv6pH␤The following packages were stubbed but not defined:␤ A␤at /tmp/N8unmOv6pH:1␤------> 3package A { ... }; class A { }7⏏5<EOL>␤ expecting any of:␤ postfix␤ state…»
RabidGravy m: module F {}; module F {}
camelia rakudo-moar 2e1782: OUTPUT«5===SORRY!5=== Error while compiling /tmp/A9WefOgXd2␤Redeclaration of symbol F␤at /tmp/A9WefOgXd2:1␤------> 3module F {}; module F7⏏5 {}␤ expecting any of:␤ generic role␤»
Skarsnik lol 21:14
that really a weird error
RabidGravy but yeah the most user visible difference is that you can have the same package more than one, not so with a module
Skarsnik Package was not to have one module on multiple files? 21:15
or to have like Java package
hankache aha 21:16
thanks
Skarsnik I like java package stuff, because you can say "This class is public for the package" and you can access the class freely in your package without exposing it to the outside world that use your package
RabidGravy well, it's what you get if have A::B::C if you don't specify otherwise
m: module A::B { class C { } }; say A.HOW; say A::B.HOW; say A::B::C.HOW; 21:17
camelia rakudo-moar 2e1782: OUTPUT«Perl6::Metamodel::PackageHOW.new␤Perl6::Metamodel::ModuleHOW.new␤Perl6::Metamodel::ClassHOW.new␤»
21:18 musiKk left
AlexDaniel m: say [~] (0..0x1FFFF ==> grep { .uniprop ~~ m/‘Me’/ })».chr 21:20
camelia rakudo-moar 2e1782: OUTPUT«҈҉᪾⃝⃞⃟⃠⃢⃣⃤꙰꙱꙲␤» 21:21
21:22 geraud joined 21:24 labster joined 21:26 pecastro joined
masak Skarsnik: I think I like lexical scoping for the reason you give. you declare a variable in a scope, and you can access it in exactly that scope. doesn't even need a special mechanism or keyword. 21:26
21:28 pdl left
AlexDaniel m: say [~] (0..0x1FFFF ==> grep { .uniname ~~ m/‘FACE’/ })».chr 21:28
camelia rakudo-moar 2e1782: OUTPUT«፦∯☹☺☻⚀⚁⚂⚃⚄⚅⾯〠龜𝤔𝤕𝨇𝨈𝨉🌚🌛🌜🌝🌞🌬🐭🐮🐯🐰🐱🐲🐴🐵🐶🐷🐸🐹🐺🐻🐼💆🕐🕑🕒🕓🕔🕕🕖🕗🕘🕙🕚🕛🕜🕝🕞🕟🕠🕡🕢🕣🕤🕥🕦🕧😀😁…»
21:28 Guest9988 joined
ZoffixWin .u 🐹 21:28
yoleaux U+1F439 HAMSTER FACE [So] (🐹)
ZoffixWin m: say [~] (0..0x1FFFF ==> grep { .uniname ~~ m/‘DONG’/ })».chr 21:31
camelia rakudo-moar 2e1782: OUTPUT«༙᭡₫␤» 21:32
ZoffixWin :}
21:34 domidumont joined
masak .u ᭡ 21:35
yoleaux U+1B61 BALINESE MUSICAL SYMBOL DONG [So] (᭡)
21:35 BenGoldberg joined 21:37 Guest9988 left 21:38 bowtie_ joined, hankache left
moritz \o/ perl6.bestforever.com/ runs now 21:46
hosted on hack, powered by Perl 6 (and Bailador and Apache and Linux and ... )
MadcapJake moritz++ 21:47
21:47 kjs_ left, mr-foobar left
RabidGravy cool 21:48
MadcapJake waits in Perl 6 slack for anyone to actually join :P
moritz: would it be alright to post to reddit?
moritz MadcapJake: sure 21:49
pmurias what's the advantage of slack over irc?
moritz the chances of it attrackting much traffic isn't big anyway :-)
MadcapJake well, that's a big topic xD
moritz: yeah agreed :)
moritz anyway, time for sleep here
21:49 domidumont left
MadcapJake pmurias: lots of zealous opinions on slack vs irc, my whole purpose behind setting this up is *just* about adoption (especially amongst the young hip crowd) 21:50
MadcapJake loves that he found a fun domain at freedns.afraid.org 21:52
21:53 renatolrr left 21:54 ZoffixWin left
pmurias nqp-js beats nqp-m on the most silly benchmark ever (paste.debian.net/377859) 21:57
21:58 CurtisOvidPoe joined
pmurias 1.29s vs 5.38s (and for perl5 without type annotation 14s) ;) 22:00
nine So we should clearly abandon MoarVM
pmurias duh, looping and multiplication over doubles is the only thing that counts 22:01
;)
RabidGravy well yeah 22:03
geekosaur reminds me of ghc's llvm backend sucks for everything except floating point ops, where it's way better than ncg >.> 22:09
22:11 dolmen joined
MadcapJake ofc i'm getting downvotes from all the irc fans :( oh well 22:13
jast repeat after me: reddit isn't for unpopular opinions 22:14
MadcapJake lol
22:15 kjs_ joined
geekosaur will note that he doesn't mind Slack as an addiitonal "channel" --- just don't expect him to switch to it 22:15
MadcapJake geekosaur: I completely understand that, and that's exactly what this is, an adoption channel, just as the perl6 chanel in irc.perl is funofficial, same goes for this 22:16
AlexDaniel “Success! Check [email@hidden.address] for an invite from Slack.” ok… no sorry 22:17
MadcapJake: perhaps we should let the young crowd try irc :) 22:19
22:20 molaf left
geekosaur the young crowd seems to think irc is passe and needs to be replaced with something more inscrutable >.> 22:21
skids wonders why all the fads never seem to include ditchin centralized services run by unaccountable providers. 22:22
MadcapJake AlexDaniel: What's the issue with the wording there? Do you not like checking your email?
AlexDaniel MadcapJake: that's not my email. I just want to get in
MadcapJake Oh you do need an email :P 22:23
AlexDaniel MadcapJake: anyway, is it planned to have a bot here that will be showing what this young crowd is saying there? 22:24
MadcapJake The young crowd will always do whatever it wants, it pays little attention to principles. It's more important to meet them where they stand than to just expect them to come to you. Again, this is an adoption channel
AlexDaniel: I can actually do a full irc bridge that just mirrors all the chatter here there.
and vice versa
22:25 kjs_ left
AlexDaniel perhaps it is a meaningful thing to do? Otherwise I don't see how this could possibly be helpful 22:25
MadcapJake I really want to see if I can get a perl6 bot like camelia too
22:25 captain-adequate joined
AlexDaniel e.g. young guy joins that shiny thing, no one is there, young guy quits 22:26
MadcapJake AlexDaniel: I'm not so sure actually, it's okay for people to meet and discuss things in seperate places than the official channel btw
AlexDaniel: yes it does need to be used for it to be usable
AlexDaniel yeah except that lots of smart folks who are present here will not be using this shiny thing
jast there's lots of features IRC could stand to gain... there was an unnatural level of attention on the IRCv3 post a day or so ago
geekosaur there could be a #perl6-slack linked and those who want to can join it?
MadcapJake I placed a memo at the top of the general channel discussing what the official channel is and that this is *not* official by any means
jast so I do understand why stuff like slack gets the attention it's getting 22:27
MadcapJake geekosaur: yeah I like that idea
geekosaur jast, there are lots of things tat IRC could stand to have --- but it looks to me like slack's architecture is a disaster
hipchat's somewhat better
jast IRC's architecture is a disaster, too... but everyone who came up with something better never got anywhere worth mentioning
22:27 Skarsnik left
MadcapJake Again, adoption is less about architecture and principles and more about what is currently popular. 22:27
geekosaur (but only somewhat. let's not replace irc with something merely *differently* sucky) 22:28
(I mean, if we'd followed that with perl 6, it'd be python or something >.> )
jast both slack and IRC have a single point of failure issue
(I'm guessing about slack... it's pretty difficult to do resilient and fault-tolerant chat, though, so I'm optimistic about my guess) 22:30
MadcapJake hope to see some of you on slack anyways! I'm off to dinner. 22:31
AlexDaniel yeah, adoption is about marketing
too bad it doesn't work on conscious crowd… 22:32
22:32 yurivish joined
jast marketing can work on anyone, you just need to understand your "target" 22:32
MadcapJake hehe yep, it's more about the glitz and I think the language itself has tons of that, so just connecting with them will help!
AlexDaniel jast: actually, you're probably right. Gitter is doing great just by focusing on developers 22:33
jast for example, Camelia is an ingenious targeting device
MadcapJake is off to dinner 22:34
jast gitter could not send me a billion e-mails by default, that'd be great
AlexDaniel jast: just leave it. That's what I did. For exactly the same reason
jast and do blocking work in their javascript, misinterpret my click to create a chatroom for an org with many people in it, send them all e-mail, and send a pull request to the project :} 22:35
22:35 labster left
jast AlexDaniel: you're right... there isn't even an option to configure notifications, so you've successfully convinced me 22:36
geekosaur tbh my main issue with slack isn't actually with slack. it's with the realization a couple years ago when I was being pushed to be on 15 different message services each with its own client, and several of them didn't like the idea of anyone using anything but their designated client 22:38
and, well, no. I dont care how special your messaging is, I;m not dealing with 15 clients. 22:39
22:39 lokien_ left
jast yeah, one client for all IM-type messaging seems reasonable 22:39
geekosaur (skype was actually the straw that broke this camel's back) 22:40
jast I don't know any client that does both group chat/IRC and IM properly
so even with services that make that possible I'm still forced to use multiple clients
geekosaur sure. Im willing to do one of each of those, having also noticed that there doesn't seem to be a good way to unify what turn out to be different messaging models 22:41
22:41 kurahaupo_ joined
dalek kudo/nom: 8c406bf | (Andy Weidenbaum)++ | tools/lib/NQP/Configure.pm:
fix parse_revision subroutine regex

  - compare release candidate versions, e.g. 2016.01-RC1
  - use extended p5 regex for readability
  - quiet warnings when regex captures undefined		 22:41
kudo/nom: a5fe347 | niner++ | tools/lib/NQP/Configure.pm:
Merge pull request #700 from atweiden/tools-parse-revision

fix parse_revision subroutine regex
jast it doesn't feel like it *has* to be impossible... but yeah, same thing for me
22:42 wamba left
AlexDaniel jast: pidgin? 22:42
jast then again, after all these years there still isn't any decent mail client ;)
geekosaur but only *one*. when I found myself trying to fit a separate skype in around the hexchat and the pidgin/adium, after finding that skype integration had been broken and wasn't really fixable, I drew the line
jast AlexDaniel: for a frame of reference... I use irssi for IRC 22:43
22:43 kurahaupo left
jast not because I prefer console over widget toolkits, but because that level of customizable and advanced features doesn't seem to exist in any other shape 22:43
geekosaur zephyr actually made that clear, there were clients that acted like IM and clients that acted like IRC and the latter worked far better for the way I was using it
AlexDaniel well, there is Tox that is quite promising… it's just that it will take another 5 years for it to get ready for normal use. 22:45
it has different clients, it's foss, it has reasonably good groupchats (the ones that are not merged yet)… 22:47
jast your description fits XMPP, too 22:48
AlexDaniel it's also distributed. If slack or anything else were promising these things in the future, then yeah, that'd be great. Otherwise not so much, I'll wait for better things to come.
jast quite a few years ago I spent a lot of time designing an alternative protocol. well, it turned out design doesn't magically turn into implementation... 22:52
22:56 RabidGravy left
geekosaur and sometimes what looks like a good design doesn't work quite as well in implementation (hello XMPP...) 22:59
23:02 salv0 left, lizmat joined 23:03 kurahaupo_ left 23:04 kurahaupo joined, kurahaupo left 23:05 CurtisOvidPoe left 23:10 BenGoldberg left
nine geekosaur: I'd argue that XMPP is not exactly an example for a good design (streaming XML!?) 23:12
23:12 yurivish left 23:15 TEttinger joined 23:20 caasih left, Spot__ left, chansen_ left, clkao left, SmokeMachine___ left, BuildTheRobots left, mrsolo left 23:21 chansen_ joined 23:22 BuildTheRobots joined, SmokeMachine___ joined 23:23 caasih joined, bapa left, pmurias left 23:24 bapa joined 23:25 clkao joined, bitmap left, mattp_ left, raydiak left 23:26 mattp_ joined, krakan left, Timbus left, apejens left, M-eternaleye left, mrsolo joined 23:27 obarb left, eseyman left, tth left, ingy left 23:28 firstdayonthejob left, eseyman joined, tth joined, Spot__ joined 23:29 firstdayonthejob joined 23:31 dolmen left, bitmap joined, leont joined, Timbus joined 23:33 M-eternaleye joined 23:34 brabo joined 23:35 apejens joined 23:37 sevvie left 23:39 raydiak joined
avar Are there some less trivial examples of concurrency in perl6? 23:39
I went to lizmat++'s talk at FOSDEM today and while it's completely fair to show easier concurrency examples in a "what's new in p6" talk I'm looking for something that shows how the hairy edge cases are covered. 23:40
23:40 ZoffixWin joined
avar Been to a few demos of p6 concurrency, and it's mostly easy stuff without side-effects 23:40
ZoffixWin What sort of hairy edge, side-effects stuff are you looking for?
avar Looking for e.g. something that sends off 100 concurrent DB requests, the db connection for each thing can die at any time and will need to be restarted, the whole thing needs a global timeout and stopping/killing/ignoring offending threads 23:41
23:42 raiph left
avar Or e.g. how is something where you have an underlying object (e.g. a network connection) managed with either mutexes or spawning lots of clones.. 23:42
23:43 ingy joined
skids The concurrency stuff was more or less a beta API right up until close to the 6.c release so not many people went that far out on a limb yet I think. 23:44
23:44 labster joined
skids (There were some late changes ("serial supplies") before 6.c, and there are still some issues to work out, mostly documetation, but some behavioral) 23:45
23:47 krakan joined
leont My TAP harness is a non-trivial example, but currently it's also exposing bugs :-/ 23:48
masak 'night, #perl6
ZoffixWin night 23:51
skids I would say odds are that at this point, you are unlikely to rely on a major feature and get the rug pulled completely out from underneath you, but relying on undocumented behavior is still a bit iffy which is why doc fixes are needed. 23:54
23:57 firstdayonthejob left

Powered by Cro and the Raku® Programming Language.

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

Thank you!