»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.
travis-ci Doc build errored. Will "Coke" Coleda 'no trailing whitespace' 00:05
travis-ci.org/perl6/doc/builds/262044676 github.com/perl6/doc/compare/68057...dd129d35a7
buggable [travis build above] ☠ Did not recognize some failures. Check results manually. 00:05
Geth doc: 12ca890ab3 | (Will "Coke" Coleda)++ | doc/Language/functions.pod6
catch-all no longer exists
 00:15
doc: 0d0024f6a5 | (Will "Coke" Coleda)++ | 2 files
Split out EXAMPLES.md

Part of #1302
Geth doc: dc79809f04 | (Will "Coke" Coleda)++ | CONTRIBUTING.md
fix link
 00:16
Geth doc: 9ea4a12844 | (Will "Coke" Coleda)++ | EXAMPLES.md
fix formatting
 00:18
doc: 0a0d8463bf | (Will "Coke" Coleda)++ | doc/Type/PseudoStash.pod6
run test, show output
 00:23
Geth doc: 237aabfc29 | (Will "Coke" Coleda)++ | doc/Language/unicode_texas.pod6
remove since, #1236
 00:34
travis-ci Doc build errored. Will "Coke" Coleda 'fix formatting' 00:37
travis-ci.org/perl6/doc/builds/262052769 github.com/perl6/doc/compare/dc798...a4a12844e2
buggable [travis build above] ☠ Did not recognize some failures. Check results manually. 00:37
travis-ci Doc build errored. Will "Coke" Coleda 'fix link' 00:38
travis-ci.org/perl6/doc/builds/262052579 github.com/perl6/doc/compare/0d002...79809f04da
buggable [travis build above] ☠ Did not recognize some failures. Check results manually. 00:38
Geth doc: 9cdecdfc2f | (Will "Coke" Coleda)++ | doc/Type/IO/Pipe.pod6
recommend Proc::Async instead of Proc
 00:39
lookatme morning 00:42
ijneb lookatme: Good morning, again lol
lookatme ijneb, yeah, :P lol 00:43
Geth doc: 43ac7e2b85 | (Will "Coke" Coleda)++ | doc/Language/containers.pod6
Change the wording on self-referential structures

It's fine, you can use them.
Closes #1125		 00:50
travis-ci Doc build errored. Will "Coke" Coleda 'remove since, #1236' 00:55
travis-ci.org/perl6/doc/builds/262055474 github.com/perl6/doc/compare/0a0d8...7aabfc29bc
buggable [travis build above] ☠ Did not recognize some failures. Check results manually. 00:55
travis-ci Doc build errored. Will "Coke" Coleda 'recommend Proc::Async instead of Proc' 01:00
travis-ci.org/perl6/doc/builds/262056269 github.com/perl6/doc/compare/237aa...decdfc2faa
buggable [travis build above] ☠ Did not recognize some failures. Check results manually. 01:00
travis-ci Doc build errored. Will "Coke" Coleda 'Change the wording on self-referential structures 01:10
travis-ci.org/perl6/doc/builds/262057951 github.com/perl6/doc/compare/9cdec...ac7e2b85b3
buggable [travis build above] ☠ Did not recognize some failures. Check results manually. 01:10
[Coke] hoelzro: ===> Testing [FAIL]: IO::String:ver('0.1.1'):auth('Rob Hoelz') 01:11
Aborting due to test failure: IO::String:ver('0.1.1'):auth('Rob Hoelz') (use --force-test to override)
... We can actually speed up the doc travis if we don't install the optional modules required by xt/ 01:12
(of which IO::String is one)
timotimo isn't xt something that should be run especially on CI? 01:13
timotimo hm 01:15
perhaps only for the module author to run before a new commit is pushed or distro uploaded?
maybe the doc travis should be changed to run on docker and get rakudo that way? 01:16
if its rakudobrew build moar takes ~450 seconds ... 01:17
Geth whateverable: 025c05081a | (Aleks-Daniel Jakimenko-Aleksejev)++ | 35 files
bin/ and lib/ directories

Half resolves issue #194.
Everything is far from being perfect but at least this is a small step towards getting the tree properly organized.
... (10 more lines)		 01:25
Zoffix There haven't been any updates to IO::String since June. The newest failures in it are rakudo's fault. 02:11
Geth doc: d40bee7f43 | (Zoffix Znet)++ (committed using GitHub Web editor) | doc/Language/modules-core.pod6
Add doc links
 02:22
Zoffix Do we got docs for the `--doc=Text` feature? 02:22
Zoffix This is it, I guess: docs.perl6.org/language/pod#Text 02:23
Geth doc: 90ff6a5ea9 | (Zoffix Znet)++ (committed using GitHub Web editor) | doc/Language/pod.pod6
Document that you can pass `Text` value to --doc
 02:26
Zoffix TreyHarris! 02:31
travis-ci Doc build errored. Zoffix Znet 'Add doc links' 02:42
travis-ci.org/perl6/doc/builds/262073996 github.com/perl6/doc/compare/43ac7...0bee7f4372
buggable [travis build above] ☠ Did not recognize some failures. Check results manually. 02:42
travis-ci Doc build errored. Zoffix Znet 'Document that you can pass `Text` value to --doc' 02:46
travis-ci.org/perl6/doc/builds/262074714 github.com/perl6/doc/compare/d40be...ff6a5ea979
buggable [travis build above] ☠ Did not recognize some failures. Check results manually. 02:47
lookatme Last week report ? 03:15
Anyone know the link
?
Zoffix lookatme: p6weekly.wordpress.com/2017/08/07/...eekly-101/
lookatme Zoffix, thanks 03:16
Geth modules.perl6.org: 46bfeea222 | (Zoffix Znet)++ | 4 files
Include core modules in the listing

Since it may not be obvious where to look for their documentation to users unfamiliar with our site. Make them link to docs for the topic.
The Feelin'-Lucky search will direct to them on exact name match, otherwise, it'll search through them only when no other dists were found.		 03:51
Zoffix (that looks like this: i.imgur.com/wLhrKdX.png ) 03:52
Geth modules.perl6.org: 2c56c57681 | (Zoffix Znet)++ | templates/root/search.html.ep
Include core dists in search results count
Geth modules.perl6.org: 1dbc97f6ce | (Zoffix Znet)++ | lib/ModulesPerl6/Model/CoreModules.pm
Remove `is_core` key from CoreModules

Ended up never using it		 03:58
zengargoyle zomg 04:20
zengargoyle in begining of battle with AT&T over DSL being borked. :) on dual-routed LTE from phone and wired to home and watching DSL connection crash and burn. :P 04:23
zengargoyle there's a line card flapping out there and calling support really doesn't help. :P 04:24
TEttinger zengargoyle: I also have AT&T, sadface 04:25
TEttinger they should rebrand to AT&T&:( 04:25
zengargoyle i'm 99% sure that it's network issue, but conversation with support was LOL-tastic. i'll try some serious social engineering tomorrow if it doesn't magically go away. 04:27
TEttinger hey, it's dan verizon here... 04:29
alternately, get two calls going, one to verizon sales, one to AT&T support, both on speakerphone
zengargoyle lol...
zengargoyle i have problem that my original phone/dsl was a corporate paid deal and the mibration did not work. my normal-user stuff is terribly borked. 04:31
but i have a tech i can call tomorrow who will sort things out, or at least give me good info on what is happening. 04:32
outsourced call-center "reboot your router" folk in the middle of the night are no help at all no matter their good intentions and script following. :) 04:33
hehe, luckily or unluckily i'm not a bit cell phone user, do there's data to burn. :P 04:40
zengargoyle hah, i'm thinking they fixed it.... (crosses fingers and knocks on wood). 04:57
zengargoyle darn, maybe not. 400 seconds of uptime before boom... and now back up.... so frustrating, hopefully some tech is cursing at a piece of eqipment like i would. :) 05:07
zengargoyle kicked one of the DNS root servers in a fit of rage. 05:10
Resol p6: my $channel = Channel.new; 06:00
camelia ( no output )
Resol say $channel
evalable6 (exit code 1) 04===SORRY!04=== Error while compiling /tmp/k0lAdK6eT_
Variable '$channel' is not declared. Did you mean 'Channel'?
at /tmp/k0lAdK6eT_:1
------> 03say 08⏏04$channel
zengargoyle meh, seems like my tethering AP shuts off after some inactivity. 07:14
zengargoyle ha, i can feel the heat that using my laptop's wifi generates. 07:27
moritz typically feels the heat that the mobile phone generates while tethering is activated 07:33
yoleaux 00:21Z <AlexDaniel> moritz: FWIW, somehow I was under impression that only new unicody messages are broken, but everything seems to be affected: irclog.perlgeek.de/perl6/2016-09-05#i_13155366 . Maybe you know already, I just thought that I'll mention it just in case…
zengargoyle moritz: heh, it's the laptop where i feel the heat. the phone is normal. the laptop is generally wired to the 1G network (who can tolerate NFS and X over wireless... :P ) 07:48
nadim lizmat: morning, I see that you have digged a bit in NativeCall, there is an example here docs.perl6.org/language/nativecall...nd_CUnions that I need some help with. It show a CStruct in a CStruct, one via has and the other via HAS but no exaample of how to instanciate the structures. Could you show me or add such examples? 07:49
zengargoyle totally hasn't felt hot spots on this laptop before.
zengargoyle wireless is for the chromecast/roku/phone... plebian devices :) 07:51
zengargoyle my baby talks gig to her significant other. :) 07:52
zengargoyle gives laptop a love tap and says it will be all better soon. 07:53
zengargoyle totally hasn't felt hot spots on this laptop before.really wanting to go home. 07:55
llfourn o/ from Amsterdam #perl6! 07:56
yoleaux 29 Jul 2017 11:45Z <nadim> llfourn: Hi, I did not managed to make a RT since I could not golf it down, on the other hand I found what the problem was. Maybe this will help nopaste.linux-dev.org/?1160132
llfourn nadim: looks like that pasta expired? 07:58
zengargoyle and that last weird was <deleted> gnome doing the history thing when you touch the scroolpad, gah i hate gnome-terminal. 07:58
zengargoyle dsl is up for a good hour now, i imagine some tech really wanting to go home. 07:59
lizmat nadim: I'm really no expert on NativeCall, perhaps nine of Zoffix may be able to help out 08:17
yoleaux 03:11Z <Zoffix> lizmat: any idea why this gives only `"\n"` in the list instead of both `"\n"` and `"\r\n"`? Bisectbot points to the `is default` change in IO::Handle. I'm failing to golf it out of IO::Handle tho even after copying all the relevant bits. Code: class Z is IO::Handle { method x { dd $.nl-in } }.new.x;
lizmat clickbaits p6weekly.wordpress.com/2017/08/07/...eekly-101/ 08:18
nadim lizmat: thanks, I have asked Zoffix alread, I'll ask nine 08:24
nine: hi, any idea about the NativeCall question above? 08:25
nine nadim: you treat it pretty much like a normal class. The examples even contain the .new cll 08:55
nadim nine: this example docs.perl6.org/language/nativecall...d_CUnions, I could not set the embeded, or refered, Points. 09:19
nadim IE: $mystruct2.point = $point ; gave me error: Cannot modify an immutable Point ((Point)) 09:24
nine nadim: you'll probably have to initialize the fields individually: $mystruct2.point.x = 1; $mystruct2.point.y = 2; 09:33
lookatme nadim, you can write a read-only accessor 09:35
lookatme m: use NativeCall; class Point is repr("CStruct") { has num64 $.x; has num64 $.y; submethod TWEAK() { $!x = num64.new(10); $!y = num64.new(10);}; }; class MS is repr("CStruct") { has Point $.point; has int32 $.flags; submethod TWEAK() { $!point := Point.new; $!flags = 1; }; }; say MS.new 09:40
camelia MS.new(point => Point.new(x => 10e0, y => 10e0), flags => 1)
lookatme m: use NativeCall; class Point is repr("CStruct") { has num64 $.x; has num64 $.y; submethod TWEAK() { $!x = num64.new(10); $!y = num64.new(10);}; }; class MS is repr("CStruct") { has Point $.point; has int32 $.flags; submethod TWEAK() { $!point := Point.new; $!flags = 1; }; }; my $ms = MS.new; $ms.point.x = num64.new(32); say $ms; 09:41
camelia Cannot modify an immutable Num (10)
in block <unit> at <tmp> line 1
lookatme m: use NativeCall; class Point is repr("CStruct") { has num64 $.x is rw; has num64 $.y; submethod TWEAK() { $!x = num64.new(10); $!y = num64.new(10);}; }; class MS is repr("CStruct") { has Point $.point; has int32 $.flags; submethod TWEAK() { $!point := Point.new; $!flags = 1; }; }; my $ms = MS.new; $ms.point.x = num64.new(32); say $ms; 09:42
camelia MS.new(point => Point.new(x => 32e0, y => 10e0), flags => 1)
lookatme nadim, does ^^ helpful ?
lookatme off work now 09:46
nadim yes, should be in the example, may add it later 09:47
mp_ Can I compile perl6 scripts in moarvm bytecode and run the generated file? I would to use it in a production environment 10:53
thanks
Zoffix mp_: not yet 10:54
buggable: eco wix
buggable Zoffix, App::InstallerMaker::WiX 'Tool to make basic WiX installers for Perl 6 apps, bundling MoarVM/Rakudo.': github.com/jnthn/p6-app-installermaker-wix
Zoffix You can pack it up tho on Windows ^
mp_: note that modules are already precompiled. If you're just looking for a speed boost, shove everything into a module 10:55
mp_ ah ok. It was for hide my proprietary code 10:57
I'll try to use modules
Zoffix mp_: you can't hide that really. If someone really wants it, they'll disassemble your compiled binary. The cool kids use legal instruments for that purpose these days. 11:00
that purpose = ensuring your work isn't re-used and re-sold 11:01
mp_ Yes, I know, it's just to make it a bit more difficult to read 11:05
thanks a lot
nadim NativeCall support blog, thank you all for the help, blogs.perl.org/users/nadim_khemir/2...-side.html 12:11
Zoffix nadim++ # blogging 12:12
nadim takes hell of a time! 12:16
nadim seriously, kudos to those of you who do it often 12:33
jnthn Last time I blogged I was like "eh, I'll just finish this post quickly before bed" and then when I got to bed it was after 3am... 12:35
hoelzro o/ #perl6 12:39
yoleaux 26 May 2017 18:56Z <Zoffix> hoelzro: Sent you a PR fixing a couple of bugs; an old one in .get() and a new one in .lines due to latest Rakudo encoding refactor no longer using .get for .lines; github.com/hoelzro/p6-io-string/pull/8
hoelzro wow, that long, huh? :( 12:39
jnthn o/ hoelzro
hoelzro I saw some IO::String releated failures on here last night - is there anything I can do to fix that?
o/ jnthn
[Coke] hoelzro: the travis run of perl6/docs was dying trying to instll IO::String 12:40
hoelzro [Coke]: ah, ok - I just built a fresh rakudo and installed IO::String without incident - was it a hiccup, or something related to perl6/docs? or is there a problem with IO::String that needs fixing? 12:41
Zoffix hoelzro: it was a rakudo problem and is fixed on HEAD 12:43
[Coke] travis-ci.org/perl6/doc/ 12:44
Zoffix (or rather, the cause was reverted to avoid breakage; the fix is still need to be found; rt.perl.org/Ticket/Display.html?id...et-history )
[Coke] let me re-run that last one.
hoelzro Zoffix: ah, ok - thanks! 12:45
I don't do much these days, but the least I can do is keep my modules green =) 12:46
Zoffix buggable: eco author:hoelzro 12:46
buggable Zoffix, Nothing found
Zoffix buggable: eco author:hoelz
buggable Zoffix, Found 6 results: Linenoise, IO::String, Algorithm::LCS, PriorityQueue, Pod::EOD. See modules.perl6.org/s/author%3Ahoelz
Zoffix \o/
hm. the core modules thing needs an author.... 12:47
nadim Do we have a sorted hash? or better a trait that sorts hashes? 12:57
timotimo nothing built-in 12:58
nadim although I realize now that the two thing I named are different things
timotimo at one point someone wrote a role that you can apply to a hash that makes it remember the order of things added
nadim wonders what wold be the cleanest way to implement the
Ah, who was it? 12:59
timotimo i believe it was masak
Zoffix Search the modules site, dude :)
timotimo like 6 years ago or something
Zoffix buggable: eco turducken
buggable Zoffix, ArrayHash 'An array in a hash in an array... like a turducken': github.com/zostay/perl6-ArrayHash
nadim yeah right! of course a turducken, what was I thinking ;) 13:00
Zoffix :)
looks busted tho, if travis is to be believed
nadim Zoffix: I used to go to the module site like this modules.perl6.org/#q=data, IE it accepted a query but not any more. 13:01
will that come back?
Zoffix nadim: no
nadim: modules.perl6.org/q/data
nadim: modules.perl6.org/s/data 13:02
nadim I see, it's "search?q=" now
Zoffix /s/ endpoint works
zostay i don't think it's busted, let me kick travis and go for a retry
huh... "Unable to allocate an array of 18446744073709551615 elements" 13:03
nadim zostay: while you are lkicking travis, I have not made it pass once on Data::Dump::Tree, although tests are just passing fine, I haven't really got this and I'd appreciate a hand
zostay yeah, i've been getting a lot of transient errors in my tests lately on travis, but i don't why 13:04
i don't get them when i run the tests local
zostay yep, rerun of travis passed, so it was either some bug in rakudo that existed as of the commit it lasted tested or something weird with travis-ci 13:14
timotimo perlmonks.org/index.pl?node_id=1196984 - if someone has a perlmonks account 13:45
Zoffix reads but abstains from answering 13:49
Too many questions and many of them are "why did you design it like that"
timotimo mhm 13:50
ab5tract good *, #perl6
yoleaux 9 Jul 2017 22:19Z <lizmat> ab5tract: my solution to [(^)] : github.com/rakudo/rakudo/commit/0cdd6c2535
sena_kun arnsholt_, ping? 13:50
ab5tract i was wondering where i might find information on nqp::p6bindattrinvres 13:51
Zoffix ab5tract: github.com/rakudo/rakudo/blob/nom/...attrinvres 13:51
It's bind-this-value-to-this-attr-in-this-thing-and-return-the-thing 13:52
Zoffix m: class Foo { has $.bar }; use nqp; dd nqp::p6bindattrinvres(Foo.new, Foo, q|$!bar|, "meows") 13:53
camelia Foo.new(bar => "meows")
moritz answers 13:54
Zoffix moritz++ 13:55
timotimo answers++
moritz man, perlmonks really is a PITA to use
moritz quoting parts of a question that contains markup mean you have to recreate the markup you quoted 13:58
I know that when I used it a lot (around 2006-2009), I thought "it's not that bad"
stockholm syndrome at its finest
Zoffix Writing a GUI automation app in AutoIt ATM.... I'd trade Rakudo's features over rakudo's slowness any time. 14:16
Zoffix Can't even spread a function call onto multiple lines on this thing without putting ' _' as unspace on each line 14:16
moritz though I can kinda imagine that GUI automation is unfun in any language
Zoffix It's pretty neat in this language, I'd say. It's the lang itself that kinda eww. If I had Rakudo's features, it'd be pretty sweet. I think there's a C lib that comes with AutoIt. Might write bindings for it. 14:17
RockyBalboa Hello! Just a quick question. I have played with Perl 6 the past few days. I LOVE the language: the synthax, all the features, it's so good. I would love to switch to Perl 6 (coming from Ruby), but right now Perl 6 is sadly too slow (for a web app for example). I was wondering, where are now? When can we expect to have better performances? I would love to help, but I am just a web developer. Anyway, thanks for your work guys. :) 14:18
moritz RockyBalboa: Rakudo (the compiler) has been getting faster steadily over the last few years 14:19
Zoffix RockyBalboa: couple days ago we added new spesh that made our bench run 20% faster. And more work is being done. You can help sponsor the person most capable of performing the needed work: 6guts.wordpress.com/2017/05/12/loo...t-funding/
moritz and you could help by donating for www.perlfoundation.org/perl_6_core_...pment_fund :-)
Zoffix *new spesh changes
Zoffix RockyBalboa: also: help by reporting which constructs are unusually slow. 14:20
huggable: rakudobug 14:21
huggable Zoffix, Report bugs by emailing to [email@hidden.address]
Zoffix ^ by there
RockyBalboa: oh, and new expression JIT backend is about to be merged soon. That's 2 years worth of work right there. It'll let up write more JIT stuff. github.com/MoarVM/MoarVM/pull/625
RockyBalboa Ok thanks! :) Nice, I will help with funding as I can't code hardcore VM stuff :) 14:22
Zoffix \o/
Zoffix s/let up/let us/; -_- 14:23
RockyBalboa For example, just parsing a string variable seems somehow slow compared to Ruby. Don't get me wrong, I don't want to be rude, I know you are working hard :) 14:24
moritz RockyBalboa: what do you mean by "parsing"? matching a regex against it?
timotimo rakudo has a few "performance cliffs" in my experience; there's ways to do things that are definitely fast enough, but you can easily shoot your performance to hell with just a simple change or two 14:25
RockyBalboa For example, if I declare: my $string = 'long long string' and simply run this perl file, it's kind of slow (compared to other scripting languages). 14:26
moritz oh, that's mostly startup time
RockyBalboa Oh ok, startup/compile time? 14:27
moritz and yes, last I measured, rakudo was around 100ms, and ruby around 40ms
raschipi TimToady is working on making that faster, isn't he? 14:27
Zoffix RockyBalboa: nothing rude in that :) Ruby first release 22 years ago, we first released less than 2 years ago :) Ruby had a much larger head start :) 14:28
Last I heard TimToady was raking in huge amounts of air miles traveling :)
Zoffix But yeah, latest Cursor/Match stuff he worked on was to eventually make more optimizations in that area 14:28
RockyBalboa: what rakudo version are you on BTW? perl6 -v 14:30
RockyBalboa Zoffix: Ok nice! Yes, I imagine it's a huuuge task! Ruby was very slow too a few years ago
I was on rakudo-star-2017.07 14:31
Zoffix Ah. Good. That's the latest.
RockyBalboa In a web app, we need to do a lot of html template parsing and every ms count 14:32
timotimo you only need to parse html templates once, then a piece of code is generated for you that only does some string concatenation and loops and such 14:33
RockyBalboa But still, Perl 6 synthax is a magnificient! I love all these features. :)
Yes timotimo, I should do that. Ruby is fast enough I didn't even cache generated code, I just eval it on the fly 14:34
timotimo there's multiple benefits to not generating the code new every time you use the template 14:35
the dynamic optimizer starts investigating code after it's been run a few times
if you only run it once, then toss it, that's not going to help terribly much :)
RockyBalboa Ok thanks! Yes, I am maybe too used to 100% interpreted language (and Perl is compiled...) 14:36
mspo 100% interpreted?
timotimo there aren't many of those around any more 14:38
mspo dash
RockyBalboa Ok :) 14:39
Zoffix AutoIt :)
Zoffix is writing it ATM :)
Nothing like getting a syntax error half way through execution ^_^ 14:40
RockyBalboa Ohhh and I was wondering why we have array.elems (which I expected to return some elements) and not array.size or array.count? As a noob, that's the only tiny think I found counter-intuitive diving in Perl 6 :) 14:42
Geth_ perl6-most-wanted: 55cf48a9b3 | (Itsuki Toyota)++ (committed using GitHub Web editor) | most-wanted/bindings.md
Delete gnuplot WIP

I recently made a gnuplot bindings, however it is not a native one but a naive one. If there are anyone who want to use a native one, feel free to add gnuplot again.		 14:44
moritz RockyBalboa: in Perl 6, we try to encode the unit of the result in the method name 14:45
RockyBalboa: .size -- could be size in bytes 14:46
RockyBalboa: likewise, string lenghts are measured in .chars and .codes (and .bytes on buffers), there is no .length
geekosaur people like to think strings are simple, but they are anything but and pretending leads to e.g. web pages containing œ style gobbledygook 14:47
RockyBalboa Ok thx, it makes sense :) 14:48
In Ruby, we can make string.chars (returns an array of chars) and chain with string.chars.size, but it's not efficient, so there is string.size as a shorthand 14:49
raschipi .comb.elems does the same thing in Perl6 as .chars 14:50
ugexe it probably returns the same result, but does it really do the same thing? 14:51
raschipi m: "Hello World!".comb.say
camelia (H e l l o W o r l d !)
geekosaur I thinkt hat is only true if all your "characters" are < 0x80 14:52
RockyBalboa Ok nice! In the end, it's just about embrassing the change, and not trying to do Ruby in Perl :)
geekosaur but poeeple like to pretend that's the world of characters
raschipi I think I got confused in my anser, .comb.elems does the same thing in Perl6 as .chars.size in Ruby. 14:53
Or does it?
RockyBalboa But what I was really saying is that "elems" feels like I will get an array or a collection of elements back, not an integer :) 14:55
geekosaur the real problem is, what is a character? how many characters in Æ̈̊ ?
RockyBalboa But it's a minor thing yes :)
A character = a unicode character?
raschipi RockyBalboa: Yes a "grapheme cluster". 14:56
timotimo it could very well be that when you comb a string and ask for the .elems that it'll use count-only to not have to create an array of characters 14:57
andreoss is it possible to specify varience for parametric role? 14:58
raschipi m: "🤷😂👨‍👩‍👦‍👦".comb
camelia ( no output )
moritz andreoss: what does that mean?
raschipi m: "🤷😂👨‍👩‍👦‍👦".comb.say
camelia (🤷 😂 👨‍👩‍👦‍👦)
timotimo don't we have multi roles? :) 14:59
moritz we do 14:59
timotimo is that perhaps what andreoss is asking for?
andreoss role Container[::A] { has A $.v }; class Foo {}; class Bar is Foo {}; Container[Foo] is a super class of Container[Bar] 15:00
timotimo ah, now i get it 15:01
i.e. do we smartmatch (for example) the parameters when we smartmatch roles
andreoss m: role Container[::A] { has A $.v }; class Foo {}; class Bar is Foo {}; say Container[Foo] ~~ Container[Bar] 15:04
camelia False
timotimo looks like we don't 15:05
of course role parameters can be more than just types
moritz m: role Container[::A] { has A $.v }; class Foo {}; class Bar is Foo {}; say Container[Bar] ~~ Container[Foo] 15:07
camelia True
moritz we do ^^ 15:07
timotimo oh, huh, the ~~ was just the other way around
it's not even late, but i feel like i'm tired 15:08
andreoss my bad 15:08
andreoss so it's covariant by default 15:08
RockyBalboa timotimo: insall redshift on your computer, it's like power saving mode for the eyes :) 15:09
RockyBalboa * install 15:09
timotimo i have that 15:10
RockyBalboa Ok lol :) 15:10
andreoss what is the motivation behind not having parametic classes? 15:12
timotimo when you use a role as a type it'll create a class for you, called a "pun"
Dis3buted Rakudo 15:15
timotimo that's right 15:16
andreoss is there a differece betwee a role type and a class type? 15:20
timotimo there's differences, sure 15:21
jnthn andreoss: Classes are open (can be augmented) would be one notable problem, but in general roles are units of re-use and classes are units of instance management.
andreoss m: constant infix:<+++> = * + *; say 2 +++ 1 15:26
camelia 5===SORRY!5=== Error while compiling <tmp>
Variable '&infix:<+++>' is not declared
at <tmp>:1
------> 3constant infix:<+++> = * + *; say 2 7⏏5+++ 1
andreoss m: constant infix:<+++> = * + *; say infix:<+++>
camelia { ... }
timotimo m: constant &infix:<+++> = * + *; say 1 +++ 3
camelia 4
timotimo it looks for things starting with &
which you get by default with a sub declaration, but not with a constant declaration 15:27
Geth_ perl6-most-wanted: 22b5025aa4 | Altai-man++ (committed using GitHub Web editor) | most-wanted/bindings.md
Remove bzip2 bindings

The library was here since forever, it implements its features set, is tested, etc, so no reason to keep it here.		 16:12
pmurias is now at dutch design hotel artemis 16:17
Geth modules.perl6.org: d51f007333 | (Zoffix Znet)++ | lib/ModulesPerl6/Controller/Root.pm
Do not show core dists in result if author filter is used
 16:37
Geth modules.perl6.org: a99540fb1f | (Zoffix Znet)++ | update-modules.perl6.org
Defend against potential commit trigger conflict

If a commit to REAPP goes in in the same batch as NEWDB, we'll be restarting the app over a db it can't handle, leading to a 500 error. It's might be safe to REAPP when REBUILD is asked for, but I'm avoiding REAPPING in that case as well, just in case.		 16:59
ChristopherBotto Hello everyone \o/ 17:00
gfldex m: say "ohai ChristopherBotto!"; 17:01
camelia ohai ChristopherBotto!
ChristopherBotto I was wondering why unused parameters don't invoke a warning. This silently swallows its parameters, for example. p6: Hash.new(A => 1, B =>2);
p6: Hash.new(A => 1, B =>2)
camelia ( no output )
gfldex ChristopherBotto: it can be quite handy of you overload or wrap methods 17:02
ChristopherBotto gfldex: Hi
gfldex ChristopherBotto: if you don't like it, there is a module for you github.com/nxadm/StrictNamedArguments 17:04
ChristopherBotto Thanks! gfldex++ 17:05
sena_kun timotimo, ping? 17:19
ugexe strange, many of my perl6 repos are now showing up as unknown on github 17:51
github.com/github/linguist/pull/3735/files looks suspect 17:52
timotimo sena_kun: pong 17:59
sena_kun timotimo, hello. There are some changes from me to Net::ZMQ, I've pinged arnsholt, but no success, so I thought you can help me with a pull request review/merge, especially since you seem to use that library. I'd like rather not fork it. 18:05
sena_kun timotimo, of course, if it doesn't interfere too much with your plans. 18:08
timotimo i don't actually use it, but i know someone else does :) 18:09
timotimo huh, this is frustrating, i can't find the relevant emails any more 18:11
user3 regex. what is the difference in regard to capture between (\w+)? and ([\w+]?) 18:15
timotimo the first capture is allowed to "be missing" while the second is allowed to match an empty string 18:17
user3 ok, so the first would produce Nil while the 2nd would produce an empty string? 18:17
user3 i'll run a few tests... 18:18
ugexe well if it uses positional matches anywhere it could change that 18:22
user3 i think i get the idea. (\w+)? is probably what the cleaner one to use most often 18:23
ugexe maybe not
user3 *is probably the cleaner one... 18:24
AlexDaniel gfldex: hey ^^ 18:26
nice to see you around
sena_kun arnsholt, ping? 18:27
arnsholt sena_kun: Yeah? 18:28
ugexe ah I guess some Perl 6 repos will show up as Unknown language on github until the repo is updated / cache somehow reset since they changed the name from Perl6 to Perl 6
sena_kun arnsholt, hi! I am interested in Net::ZMQ a bit and sent a PR today, but it brings some changes and I think it needs to be reviewed. If you are still interested in this project, I will be very grateful for any help from you. 18:30
user3 what is the official name of the language, Perl6 or Perl 6?
[Coke] Perl 6 18:32
(with a nbsp if you can manage it, thanks)
user3 ok, so the official name is the name of a version 18:33
doesn't that lack identity?
just saying... 18:34
mspo perl 6000 18:35
user3 maybe in 6000 years, i.e. another 6000 since creation 18:36
jk 18:37
tbrowder m: sub foo { say &?ROUTINE.name }
camelia ( no output )
arnsholt sena_kun: It's been ages since I did anything with that stuff
And AFAIK noone else is really working on it
So if you want, I could give you a commitbit, really 18:38
sena_kun the last PR was merged by Timo and I assumed he has write access too, so pinged him, yes.
tbrowder m: sub f{ my $s=&?ROUTINE.name; say "DEBUG: $s} foo 18:39
camelia 5===SORRY!5=== Error while compiling <tmp>
Unable to parse expression in double quotes; couldn't find final '"'
at <tmp>:1
------> 3y $s=&?ROUTINE.name; say "DEBUG: $s} foo7⏏5<EOL>
expecting any of:
postfix
sena_kun arnsholt, are there any known to you users of the library? Because I sort of need to update it to current ZMQ version and things went deprecated, etc, it can break stuff. 18:40
ugexe m: sub foo { &?ROUTINE.name }; say foo() 18:40
camelia foo
arnsholt sena_kun: AFAIK, only iperl6kernel 18:41
Which isn't exactly bursting with life either 18:42
mspo the number should be higher 18:43
tbrowder ugexe: thanks, too hard to recover on anbiphone. i'm trying to add a debugging statement that says the name of the sib and haven't had much luck getting something like: say "DEBUG: entering sub &?ROUTINE.name"
mspo to communicate the total lack of compatibility
raschipi Perl 6000000000000000000, otherwise the version name doesn't fit in 64-bit integer. 18:52
jnthn Please, if you write a test using Test::META or so, put it in xt/, *not* in t/. Again, I have trouble installing a module because of this. It passes all its tests except this one. :/ 18:53
ugexe we really need to spec an author-depends
nadim I get this error when running Test::META, I rported it and the author said it was fixed and wondered if it was zef that would not update. error: Could not find META6:ver<0.0.4..*> at line 66 , bug report: github.com/jonathanstowe/Test-META...-319090770
jnthn oh heh, that's exactly the error I just got installing Pluggable 18:54
nadim jnthn: read what JStowe write in his answer 18:54
jnthn $ zef install Test::META 18:56
All candidates are currently installed
nadim same here
jnthn And I already did a zef update
nadim I forced it and still the same error
ugexe since it was fixed in .10, it might be that your zef version doesn't have the fix where it was doing numerical comparison
gfldex did you try `zef upgrade Test::META`?
jnthn Let's add --force
nadim I also updated to the latest zef 18:56
jnthn And that complains about JSON::Class 18:57
ugexe they installo fine for me, so one of two things should fix it
nadim Could not find JSON::Class:ver<0.0.5..*>
jnthn Same here
ugexe rm -rf ~/.perl6, or rm -rf ~/.zef
jnthn OK, I'll give this 2 mroe minutes and then I'll copy-paste :)
jnthn Pluggable depends on two further modules for its tests anyway, which I don't care to depend on for what I'm doing 18:58
jnthn just nabs the idea at the heart of it :) 18:59
nadim rm -rf ~/.zef did not help, trying the other
nadim rm -rf ~/.perl6, still same error 19:00
ugexe: any test one could run for you to help find what it is? 19:02
ugexe i would try to find the file on disk to see if that line of code is as it says, and see if you could correlate that to a stale version somehow 19:04
[Coke] (xt vs. t) - perl6/docs also needs this separation. 19:05
ugexe nadim: maybe need to `zef nuke site home` if the module with that code is already installed (so it can be reinstalled) 19:06
that will delete all installed modules
nadim ohhh noooo! 19:08
can I get a list that I can give back to zef to reinstall all?
ugexe zef list --installed (but its not something you can just feed back in) 19:09
otherwise you could just do `zef uninstall Blah` for anything by name
nadim I try that with META, otherwise I'll nuke all
the list looks easy enough to parse 19:10
gfldex Pluggable installs fine for me
ugexe nadim: hell you could just rename the site and home module directories to test 19:10
rename them back after 19:11
nadim it used to work and I am a stubborn SOB 19:16
raschipi stubborn is an understatement
mspo what about vendor? 19:16
nadim raschipi: I take that as a compliment ;) 19:18
raschipi Yes it is. 19:18
ugexe it only worked by coincidence 19:19
nadim zef uninstall JSON::Class JSON::Marshal META6 Test::META then zef install Test::META 19:20
jnthn: ^^
ugexe m: say Version.new("0.0.5..*"); say v0.0.5..*
camelia v0.0.5.*
v0.0.5..Inf
jnthn nadim: ah, thanks. Already ended up with about 10 lines of code to do what I needed, so not worth the dep anyway. But will fix things so I'm not in bother next time :) 19:23
nadim Yu're welcome this has bugged me for a week, I hate seing tests numbers go dowm. 19:25
ugexe i think `zef upgrade` would have worked too actually
jnthn nadim: It worked. Yay. :)
nadim++
nadim ugex++ 19:26
raschipi Who's ugex? 19:27
nadim guy who makes zef among other things 19:28
Geth modules.perl6.org: 4d648afcf0 | (Zoffix Znet)++ | 2 files
Fix corruption of same-named dists

The code erroneously finds cached data for the dist using its name only, which causes it to find the wrong dist when two dists with the same name exists in the ecosystem.
Use META url as the key to find the cached data by.
smls bisectable6: sub f { say @_; say %_; }; f |(a => 1, b => 2); 19:30
bisectable6 smls, Bisecting by output (old=2015.12 new=6745517) because on both starting points the exit code is 0
smls, bisect log: gist.github.com/82807ad44152bb3d96...099b68ac5b
smls, (2016-04-05) github.com/rakudo/rakudo/commit/6a...822dac090b
smls bisectable6: sub f { say @_.elems; say %_.elems; }; f |(a => 1, b => 2); 19:31
bisectable6 smls, On both starting points (old=2015.12 new=6745517) the exit code is 0 and the output is identical as well
smls, Output on both points: «2␤0»
jnthn |% 19:32
smls m: say (a => 1, b => 2).Capture;
camelia \(:a(1), :b(2))
smls jnthn: Does | in argument list context not use .Capture? 19:33
Zoffix No, that's for unpacking 19:34
smls ok
nadim Meh! now that works and I have updated the repo ... I have nothing to do
Zoffix | calls a secret method
jnthn smls: No 19:35
Zoffix m: dd [.FLATTENABLE_LIST, .FLATTENABLE_HASH] with (a => 1, b => 2);
camelia [(:a(1), :b(2)), {}]
jnthn Shh, it's secret!!!
smls I'm not using f |%(...) because one of the named argument values is an Array, and the %(...) creates a Hash which itemizes the Array, and that messes up the assignment to the object attribute.
Zoffix :D
Yeah, they're secret methods. Don't use them directly
jnthn m: sub f { say @_.elems; say %_.elems; }; f |\(a => 1, b => 2); 19:36
camelia 0
2
Zoffix :)
smls jnthn: Thanks!
jnthn walk, bbiab :)
Zoffix So for CPAN dists, I plan to ignore PSIXDISTS and anything that doesn't have a cpan URL as a source url 19:37
for modules.perl6.orgh
gfldex Capture literals seam not to be indexed or something is confused by it's backslashy nature 19:38
it needs a \\ to search for a \ to start with
gfldex files an issue 19:39
raschipi Gotta love UNIX. 19:41
Zoffix Why? 19:43
raschipi "it's backslashy nature" 19:45
Zoffix doesn't see what a JS piece of code with an error in it has to do with UNIX or the love of it :P
Zoffix will fix in 1hr, unless someone beats to it 19:48
Zoffix ponders taking an axe to htmlify.p6 19:49
Blow it up. Write from scratch. Properly. 19:50
Zoffix relocates
Zoffix Blow it up and rewrite it and make it generate an SQLite db instead of HTML. Then have a mojo web app serve from that db. Solve all the problems with URLs, improve search, and shorten build times 20:05
???
Profit!
and have a bot use the docs from the API 20:06
and have p6doc just use the DB
jdv79 a web service? 20:08
mspo mm mojo
Zoffix jdv79: well, with mojo it kinda comes for free 20:10
Zoffix p6doc would just use a locally stored db, if that's what you meant 20:10
jdv79 where is this freeness? 20:11
Zoffix jdv79: right there: github.com/perl6/modules.perl6.org...oot.pm#L93 20:12
jdv79: is Meta6PAN ded? 20:13
jdv79 oh, i thought you were referring to some mojo restish pluging thing i didn't know about 20:14
Zoffix Ah. dunno maybe it's there
jdv79 it seems it. nobody has contributed. i ran out of tuis when i was porting it to their new version. 20:15
Zoffix ok
good, even 20:16
jdv79 you seem to be building up modules.perl6.org and i haven't heard of any clamoring for metacpan6 so...
Zoffix heh
jdv79 i would say stalled but not dead yet 20:17
ugexe grep.metacpan6 would be great
Zoffix ugexe: tell AlexDaniel to slap a web interface to his bot :) 20:18
Zoffix greppable6: Rakudo::Internal 20:18
greppable6 Zoffix, gist.github.com/3d5f9f157d3d6bc824...c6a86ffa77
Zoffix tsk tsk so many offenders :) 20:19
greppable6: use nqp
greppable6 Zoffix, gist.github.com/96eff7b3821f154e3a...b057a6d88b
AlexDaniel haha
/o\
I'm planning to make whateverables easily runnable locally 20:20
Zoffix wow :) that's a lot
cool
AlexDaniel so that people can bisect without talking to the bot at all
no plans for a web interface yet, but contributions are welcome :) 20:21
AlexDaniel NeuralAnomaly: status 20:23
NeuralAnomaly AlexDaniel, [✘] Next release will be in 1 week and 3 days. Since last release, there are 40 new still-open tickets (7 unreviewed and 1 blockers) and 138 unreviewed commits. See perl6.fail/release/stats for details
AlexDaniel fwiw I've marked RT #131857 as a blocker
synopsebot6 Link: rt.perl.org/rt3/Public/Bug/Display...?id=131857
AlexDaniel greppable6: storage_test333 20:30
greppable6 AlexDaniel, gist.github.com/0cfe4ad671fd71697c...76f9efcf45
Zoffix NeuralAnomaly: blockers 20:31
NeuralAnomaly Zoffix, There is 1 release blocker. See perl6.fail/release/blockers
Zoffix, perl6.fail/131857 : Aliased named parameters broke in a recent MoarVM commit
Zoffix ah, yeah
Zoffix the `is default` stuff is suspicious too. I think dogbert is now having socket bug that looks similar that I had with handle before is default was reverted from it 20:33
or rather, I think it's hitting somw other bug
buggable: toast 20:34
buggable Zoffix, Between 2017.06-241-gb3f679b and 2017.06: 13 (1.56%) modules got burnt; 14 (1.68%) got unsucced; 203 (24.37%) out of 833 modules appear unusable. See toast.perl6.party/ for details.
AlexDaniel that's a bit outdated, no? 20:36
Zoffix yup
AlexDaniel Zoffix: are you planning to rerun this stuff this month?
or, this week rather :P
Zoffix Ok 20:37
AlexDaniel I wish I finally implented safe eval in whateverable, so things like this were not so scary 20:37
Zoffix in ~20hr
AlexDaniel awesome, thanks! 20:38
Geth doc: 38846bcd26 | (Zoffix Znet)++ | template/search_template.js
Escape regex meta chars

Fixes github.com/perl6/doc/issues/1442		 21:39
El_Che so, who's at TPCiA? lizmat, nine already here I think, anyone else?
Zoffix TimToady? 21:40
llfourn 21:41
AlexDaniel MasterDuke: so what's current the status of memory usage issue? 21:48
lizmat cono 21:54
MasterDuke i think i saw that jnthn++ fixed a leak today or yesterday, but i don't believe it's the one we're seeing 22:01
figuring out what's going on is difficult, because one of the recent changes increases memory churn a lot 22:02
which makes some numbers look much much bigger than before, but aren't necessarily indicative of a leak or truly increased consumption 22:03
AlexDaniel MasterDuke: can you submit a ticket or should I do it? I find it hard to track the status otherwise
timotimo MasterDuke: i don't properly recall, did we try reverting that one commit that turned memset into free + calloc 22:04
?
MasterDuke i can't do it for a while, about to afk for a bit
timotimo: don't think so
El_Che lizmat: how was the pre-social? I went to dinner and it was to late to make it 22:06
lizmat well, it wssn't on the roof because of the rain / wetness after the rain
it was a nice gathering :-)
people had to be thrown out at 22:30 :-) 22:07
El_Che lots of people?
lizmat I would say about 100
El_Che we considerirng walking to the venue at 23h, but we reckon that 23:30 would be too late
MasterDuke timotimo: think just that commit can be reverted? 22:08
El_Che are you having breakfast in the hotel tomorrow morning?
timotimo MasterDuke: i think so 22:10
MasterDuke timotimo: i.imgur.com/1iucZ0k.png 22:13
timotimo why is gc_collect being listed as a memory leak? what does it do? 22:15
MasterDuke heh, i know almost nothing about how the moarvm gc works, don't ask me! 22:17
MasterDuke building 2017.07 now to compare with 22:20
MasterDuke i.imgur.com/F63DUkB.png 22:24
now afk for a bit
El_Che is there an OS-independant way to create a tar/zip/whatever archive file with rakudo with using C libraries? (looking for something portable, can't find anything on modules.p6) 22:27
ugexe well, if you can use base perl 5... 22:38
but maybe thats only for extraction
El_Che mmm
I wanted a way to tar modules directories to package dependencies. Dependency on perl 5 would be difficult on windows 22:39
MasterDuke timotimo: anything jump out at you? 22:48
timotimo the line in collect.c is what does the calloc for the nursery 22:50
jnthn El_Che: I used github.com/frithnanth/perl6-Archive-Libarchive in an app on both Windows and Linux.
El_Che: To produce tarballs 22:51
And extract them
It's set up on Windows to automatically download the DLLs
So they ship and work if made into an installer with github.com/jnthn/p6-app-installermaker-wix 22:52
El_Che jnthn: thx, I will look into it. I was thinking on a way to bundle all perl6 only deps so an app can be packages. Was reading stuff to understand if it's better to create a versioned pkg to be unpackaged and run, or to do something like Fatpacker on perl5 22:55
that's why I would prefer no C deps so it's really portable
ugexe install 7zip standalone on windows with Build.pm 23:03
El_Che going to bed, so I can have breakfast before the conf :) 23:04
thx for the pointers ugexe and jnthn
ugexe windows can unzip through powershell as well 23:05
ugexe zef will try powershell, gnu or bsd tar (arguments normalized to work between them), and finally perl -MArchive::Extract ... to achieve portability 23:07
ugexe i kept trying to figure out a way to rig `git` to do it but I never managed 23:09
MasterDuke El_Che: samcv created appimages of rakudo, but i don't know how portable they are to windows, or if it's easy to include modules 23:27
zengargoyle have fun everybody at TPC::EU 23:48
MasterDuke timotimo: and for the full comparison, this is HEAD (with nothing reverted) i.imgur.com/GVxPsUk.png
zengargoyle my DSL now has FEC and CRC errors that climb so fast it's totally toast. :) 23:49
BenGoldberg m: role R {}; constant T = R[:()] 23:58
camelia ===SORRY!===
QAST::Block with cuid 3 has not appeared
ryu0 El_Che: avoid C so it's more portable? lol. C is usually the most portable language there is. 23:59

Powered by Cro and the Raku® Programming Language.

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

Thank you!