»ö« Welcome to Perl 6! | perl6.org/ | evalbot usage: 'p6: say 3;' or /msg camelia p6: ... | irclog: irc.perl6.org or colabti.org/irclogger/irclogger_log/perl6 | UTF-8 is our friend! 🦋
Set by Zoffix on 25 July 2018.
holyghost basically you have a 16-bit system at the minimum then so you char array is doubled 00:00
timotimo i remember some of that code! :)
holyghost :-)
timotimo so, where does the issue arise? from the decode method? or decode-part? 00:01
holyghost 0xff = 16*15+15 I meant to say 00:03
holyghost 0xffff = 15*16^3 + 15*16^2 + 15*16^1 + 15 00:03
Kaiepi both encode and decode 00:04
timotimo maybe store the original length and just cut it off after doing the calculations?
Kaiepi ohh 00:07
yeah i could do that
holyghost my system does not use a length, to be honest
you just encode a string as as number 00:08
timotimo if you explicitly transport the length, your code will also accept things like null bytes in the middle
holyghost right 00:09
Kaiepi yeah that works! 00:12
thanks guys
holyghost np 00:13
hythm_ Hello, I only know how to write cli scripts, what would be the way to go (module/framework) if I want to learn to write a desktop app, where i can align squares together, flip, reverse positions, swap, etc... may be this picture will explain better what I have in my mind goo.gl/images/wTZygK 00:28
The squares will have unicode characters 00:29
timotimo depending very much on how you want the input to go, it could be rather complicated 00:30
hythm_ The input will be set of symbols,,, basically i want to write a program to create the Zodiac cipher,, and be able to change, displace these symbols 00:31
timotimo the easiest is perhaps GTK, like with GTK::Simple; GTK::Plus is a new module that's in development since not too long and i'm not sure what percentage of it already works; i think a lot? 00:32
i don't know anything about that cypher 00:33
hythm_ This is a picture of the cipher www.zodiacciphers.com/uploads/4/9/7...r_orig.gif 00:35
timotimo you'd want to offer a bunch of pieces that the user could add to each character?
hythm_ Im not sure, what im thinking (for now) is for example to select a subset of the symbols say 4x4,, or 6x8,, the flip or reverse this dubset,,, then draw the whole tjing again with the new position 00:38
timotimo i know you can do this with SDL2, but then you'll have to roll your own UI stuff 00:39
hythm_ So i need to learn both GTK and SDL2 ? 00:40
timotimo no, it'd be one or the other
hythm_ Im sorry if the quetion does not make sense, this all new to me
timotimo theoretically they combine, but not quite easily 00:41
what does combine fine is GTK and Cairo
cairo gives you drawing functions like "draw a line from here to there" or "draw an arc" or "fill this circle" 00:42
and GTK gives you "put a row of buttons here" or "give me an input field" and "a menu bar at the top please"
and to combine them you can tell GTK "put a big rectangular space here" and then Cairo "draw some stuff on this space" 00:44
hythm- Sounds good, i will check both and see what works best for me 00:46
Thank you timo
timotimo good luck! 00:47
the SDL2 stuff is ... pretty raw. not very well documented on the perl6 side, but the SDL2 wiki translates very straight-forwardly if you know how NativeCall calls things from the C world 00:48
hythm- Looks like I will need to learn a lot, but i don't mind, as learning with perl6 is ofun 00:51
timotimo :) :) 00:55
holyghost hythm-: look at github.com/theholyghost, search for PaganVisions2 directory and look at the SDL2::Raw stuff, I've made a window with drawing loop there 00:59
inside one of the repos, Holly6Game I think 01:00
MasterDuke i would love a good Perl6QT binding...
timotimo wouldn't we all
holyghost That's a monster to work on
hythm- Thanks holyghost, will do 01:01
timotimo hythm-: there's also examples in the SDL2::Raw repository
buggable New CPAN upload: AI-Agent-0.1.8.tar.gz by HOLYGHOST modules.perl6.org/dist/AI::Agent:cpan:HOLYGHOST 01:01
hythm- Great. I also remember few days a go someone posted in the channel a spiral done with SDL2::Raw, it looked awesome 01:03
holyghost ^-- I've beautified it with some comments and get and ask methods for the MusicAgent, NESMusicAgent
you can lookup e.g. "Rock&Roll", with a db inside some agents 01:04
then it gives you something or nothing :-)
hythm- Sweet :)
holyghost :-) 01:05
holyghost If I need to use super.BUILD() on a multiple inherited class which method should I call then ? 01:24
I guess it'll be a mess of Class redirection for multiple BUILD method 01:27
never mind, I found it online 01:32
hythm_ Iirc BUILD is a submethod, it won't be inherited. And you don't need to call it as its called during object construction. 01:37
buggable New CPAN upload: AI-Agent-0.1.9.tar.gz by HOLYGHOST modules.perl6.org/dist/AI::Agent:cpan:HOLYGHOST 02:11
buggable New CPAN upload: AI-Agent-0.1.10.tar.gz by HOLYGHOST modules.perl6.org/dist/AI::Agent:cpan:HOLYGHOST 02:32
holyghost ^-- there's ask and get routines now which should make you any Agent except for agent-to-agent dispatching 02:33
that's for parsing parsers and shouldn't be too difficult 02:34
although agents have a default "agent" message where you double dispatch them and do your own code (parsing) on them 02:35
Agent is just an Actor with a dispatch routine for sending it messages (sync or async, depending on your further code on them 02:36
Now I need an agent network system, I'll do that later on 02:42
I can use a port such as 6969 :-)
I also need a binary string in a variable, IIRC from the 2nd edition perl book that should use pack 02:44
s/perl/camel
buggable New CPAN upload: AI-Agent-0.1.11.tar.gz by HOLYGHOST modules.perl6.org/dist/AI::Agent:cpan:HOLYGHOST 03:11
moony > die 03:59
hm
what's the thing again
moony p6: die; 03:59
camelia Died
in block <unit> at <tmp> line 1
moony p6: «iczero is a potato» 04:02
camelia WARNINGS for <tmp>:
Useless use of constant value iczero is a potato in sink context (lines 1, 1, 1, 1, 1, 1, 1, 1)
moony p6: say «iczero is a potato»; 04:03
camelia (iczero is a potato)
buggable New CPAN upload: AI-Agent-0.1.12.tar.gz by HOLYGHOST modules.perl6.org/dist/AI::Agent:cpan:HOLYGHOST 04:11
holyghost I've added 2 agents' databases with a microcode database (a hash of {"add.l" => opcode hex } for M68000 assembler and an empty one for NES assembler), if you parse with this inside an agent you can transform binaries with it e.g. automatic programming/genetic programming should write a binary file or another example you translate the game language inside the ROM's machine code, you can also dump rom offsets for e.g. ripping gra 04:13
phics from rom files
I still need to write the parsing methods for them, which takes a bit longer ;-) 04:15
Which is where my Bayes::Learn, Game::Markov and Mathx::Stat comes into play, if you want to make the agent learn something, needs an hidden markov model though 04:18
holyghost You can now deploy binaries and texts in several agents, it'll upload in a minute 04:52
buggable New CPAN upload: AI-Agent-0.1.13.tar.gz by HOLYGHOST modules.perl6.org/dist/AI::Agent:cpan:HOLYGHOST 05:01
holyghost Later on, I'll probably write a NES ROM graphics or Music ripper, with an Agent that searches for such offsets, not much learning in it I think, but it should be ok 05:03
now I am going to get a drink, sante!
Guest10 p6: say 3 05:22
camelia 3
pr0d1gy p6: say "Hello, World!" 05:49
camelia Hello, World!
holyghost I had to put in some legalese for the Perl6 crew and Perl6 users, you can ask Agent (derived) classes anything, but you have to know what you are deploying/disking. So always use the "ask" message on your Agent's dispatch method to see if you would deploy illegally. 07:06
"ask about *text*" and "ask" just list Agent's databases 07:07
So deploying "Faxanadu" Nintendo ROMs would be illegal unless you change it e.g. to support Nintendo 07:08
you can translate it to Spanish for example, which hepls everyone
Then send the agent to Nintendo ;-)
It's in the README.md now 07:10
Note that you can make your own NES game with tools on nesdev.com 07:11
you can distribute that freely
upload'll show in a minute, 0.1.14 versiob
s/versiob/version
buggable New CPAN upload: AI-Agent-0.1.14.tar.gz by HOLYGHOST modules.perl6.org/dist/AI::Agent:cpan:HOLYGHOST
holyghost NES roms are abandonware by Nintendo, SFY, they are a pain in the ass for emulator software not, but for current disk ROMs 07:14
holyghost Basically you have to watch out wat you "deploy" on your disk, as you can dissasemble M68000 code such as a Amiga kernel 07:38
with the M69kasmDB.pm6 07:39
s/M69/M68
xiaoyafeng hello, Could you tell me what type $in of Proc is？ 08:12
xiaoyafeng my $p1 = run 'echo', 'Hello, world', :out; 08:12
does it mean assign True to $out? 08:13
xiaoyafeng but I also see codes like :in($fh) and :in('_'); 08:14
it confuse me.
so what type $in is? bool, Str or IO? 08:15
lookatme_q xiaoyafeng, Anything can coerces to IO 08:19
maybe help => docs.perl6.org/type/IO
CIAvash xiaoyafeng: Have you read docs.perl6.org/type/Proc ? The second paragraph of docs.perl6.org/type/Proc#method_new explains the differences when in, out, err are set to '-', Bool or IO::Handle. 08:59
lookatme_q m: say Nil.xxxxxxxxxxxxxxxxxxxxxxxxxxx; 09:08
camelia Nil
lookatme_q It occurred to me that, why this not report any syntax error ? 09:10
andrzejku Hi!:) 09:17
xiaoyafeng Thanks! 09:18
"Anything can coerces to IO" is very helpful
holyghost I've put NES sound enable, sound disable search methods for a NES rom in AI::Agent, upload'll show in a minute 09:42
buggable New CPAN upload: AI-Agent-0.1.15.tar.gz by HOLYGHOST modules.perl6.org/dist/AI::Agent:cpan:HOLYGHOST 09:51
lizmat weekly: www.tyil.nl/post/2018/10/11/hacker...l6-part-2/ 10:35
notable6 lizmat, Noted!
lizmat .tell tyil do you know about docs.perl6.org/routine/substr-rw#(..._substr-rw ? 10:36
yoleaux lizmat: I'll pass your message to tyil.
tyil I believe I did consider that, yes 10:39
yoleaux 10:36Z <lizmat> tyil: do you know about docs.perl6.org/routine/substr-rw#(..._substr-rw ?
lizmat ok
tyil: also, re count-substring: $string.indices($substring,:overlap).elems 10:40
tyil so many ways to solve things!
lizmat hehe
tyil I presume .indices is faster as it doesnt do a regex?
lizmat indeed... probably orders of magnitude faster
tyil I like regexes, too bad they're still relatively slow in Perl 6 10:41
but good to know about .indices :)
lizmat sure.. ok if I mention them in comments on reddit ?
tyil I'll use that next time I need to solve something like this ^_^
of course!
my main reason to write these posts is to learn more about the languages 10:42
lizmat tyil: will do
Zoffix lookatme_q: because that's a valid syntax for a method call and Nil has FALLBACK defined, so any non-existent methods still exist and return a Nil 11:52
Geth doc: 8d1061c4c0 | (Zoffix Znet)++ (committed using GitHub Web editor) | doc/Type/Nil.pod6
Document Nil.FALLBACK

Spec: github.com/perl6/roast/blob/2fa2ad.../nil.t#L29		 12:10
synopsebot Link: doc.perl6.org/type/Nil
Zoffix "Perl 6 version 6.d" looks really weird. "Perl 6 6.d" even more so 12:20
Why do people use "ANNOUNCE" instead of "ANNOUNCEMENT" as subject in announcement emails? 12:21
Zoffix or ANNOUNCING 12:22
moritz because that's what pmichaud wrote years ago 12:23
masak wasn't there an email group perl6-announce ? 12:26
moritz there was
masak then that wordage probably predates pmichaud's use, even 12:27
maybe the names of the groups (which I think were formed around the time of the RFC process) were inspired by some other language community 12:29
m: my &MAIN = -> {}; say "alive" 12:32
camelia alive
No such method 'cando' for invocant of type 'Callable'
in block <unit> at <tmp> line 1
masak I'd like to better understand this error message.
isBEKaml masak: wordage, heh
Zoffix masak: the MAIN handler tries to figure out whether the defined MAIN can be called with the given arguments, but Block ain't got a .cando method 12:33
masak: and the handling of MAIN commences after the mainline has run, hence why alive is printed
masak isBEKaml: hi! long time 不见了! 12:34
Zoffix Should be here: github.com/rakudo/rakudo/blob/mast...n.pm6#L268 12:34
masak Zoffix: thanks for the explanation
masak Zoffix: why don't Block instances have a .cando method? 12:34
isBEKaml masak: hi! Been a long time, yes. Unfortunately, I can't read Chinese anymore. How are things?
holyghost needs to transform 8bit strings to 16bit hexes in AI::Agent::NESMusicAgent 12:35
masak isBEKaml: the Chinese just said "no see" :)
isBEKaml: thinks are perpetually interesting, thanks. and you?
things*
isBEKaml masak: ah, been away for a while. Boredom likes to keep coming back, right? 12:36
masak isBEKaml: hey, don't undersell yourself like that! :P
isBEKaml Haha 12:37
Zoffix masak: no idea. R#2420 12:38
synopsebot R#2420 [open]: github.com/rakudo/rakudo/issues/2420 [RFC] Block can't do .cando
masak Zoffix: thank you kindly.
this happens to be a case where, 'cept for the lack of a .cando attitude on the part of the Block, that &MAIN would actually work since I passed in 0 arguments
I'm pretty sure I know how blocks differ from routines, and I don't see any of those differences being in the way of giving blocks a .cando 12:40
also, the method name "cando" is very silly in a way that feels almost compulsively Perl 6ish at this point, and by the way it looks like two words without a separator, like a hyphen or something 12:41
but I'm not really out to clamor for changing it; making light fun of it is enough for my purposes ;) 12:42
Zoffix That's karmic reparations for making fun of PHP's method naming a decade ago.
.isNaN vs .is-prime too 12:43
masak .oO( Q: Could Perl 6 naming go any more overboard at this point, silly-wise? A: Mu ) 12:43
moritz cando reminds me of the CANDU nuclear reactor 12:43
masak like CANDU from a baby 12:44
masak .oO( cantspell )
lizmat masak Zoffix now testing a Code.cando method 12:45
Zoffix \o/
masak yay, lizmat++ for doing stuff instead of joining in the hot air about naming 12:45
masak .oO( Q: Why did Perl 6 adoption fail? A: No idea, but at my company we had a policy against languages whose top type was the sound of cattle ) 12:46
tadzik . o O ( Q: Any idea to improve the startup time of our app? A: We tried this tiny replacement for Moose, but it failed our policy ) 12:48
masak tadzik: it's a harsh policy, but nobody ever got fired for calling their top type "Object" 12:57
"it's a Mu" -- "it's a *what*" -- "a Mu, a Most Undefined" -- *slams door* 12:58
"but why didn't you just call it Object" -- "what's so hilarious about that? I don't get it" 12:59
I mean, what's even the point of making a whole language if I can't express my undying love for bovines through the naming of the least derived type 13:00
(heh, it's been years since I last vented about this one) 13:01
chalk it up to an unusual degree of constraint. also, I've been busy
Ven`: by the way, did you manage to catch github.com/masak/007/issues/410 ? 13:04
Ven` I read it, but it's written there you'll publish it as a blogpost :P 13:05
Geth doc: f28da97a01 | (Elizabeth Mattijsen)++ | doc/Type/Parameter.pod6
Document Parameter.coerce_type
synopsebot Link: doc.perl6.org/type/Parameter
Geth doc: 7af9d67f04 | (Elizabeth Mattijsen)++ | doc/Type/Code.pod6
Document Code.cando
 13:19
synopsebot Link: doc.perl6.org/type/Code
masak I think I might publish it as a blog post when I have it actually working in 007 13:29
basically, having that working also means having the infix:<ff> macro working
tadzik "nobody ever got fired for calling their top type "Object"", eh?
masak tadzik: it's true
tadzik you may obey, but I Object!
masak ok, I had that one coming 13:30
masak .oO( no need to sound so biernik ) 13:31
tadzik :D
I didn't want to just biernie accept it
pmurias masak: I wouldn't be so sure, I can imagine some poor sod getting into trouble for calling the top of their (user level) class hierarchy Object 13:35
tadzik someone not knowing what we're talking about may think it's about a certain Passive US Senator
pmurias masak: do you still have doubts about the loading time/compile time destinction? 13:40
kurahaupo tadzik: cf Senators, perhaps the topmost (bottommost?) class should be "Sand" and the instances should be "Grains" 13:45
uzl .tell tyil Great Hackerrank (part 2) write-up! BTW, in What's your name? - "... it's .Str value..." -> "... its .Str value..." 14:57
yoleaux uzl: I'll pass your message to tyil.
uzl .tell tyil in Designer Door Mat - "A thing I learned is that the .. operator that generates a sequence...". I think the ... operator could've done the job but I'm not sure if it'd have some downside. 14:58
yoleaux uzl: I'll pass your message to tyil.
jdv79 how many perl offshoots are there now? 15:01
today is the first time i've heard of perl11 15:02
Zoffix there's also cperl
tadzik the hacker news effect, eh
Zoffix and perl7 :)
github.com/perl7/perl7/
moritz don't forget perlito, and fanlang 15:05
jdv79 yeah, saw it on hn. cperl, rperl, perl11, etc...
Zoffix Also, Your™ :)
Zoffix The minimal compiler able to pass 6.d roast 15:06
uzl any link to Your^{TM}? 15:07
Zoffix I think it exists only in head ATM 15:08
Ah, there's a repo with two lines of code: github.com/zoffixznet/your 15:09
It was harder than a couple of lines of code to make and I had more urgent things to take care of first.
jdv79 how will Your be different from Rakudo? 15:23
SmokeMachine m: sub postcircumfix:<.times( )>($on, [Int $times, &func]) { my $a = $on; $a = func($a) for ^$times; $a }; say "".times(3, * ~ ++$) # why does this warns? 15:25
camelia WARNINGS for <tmp>:
123
Useless use of "~" in expression "* ~ ++" in sink context (line 1)		 15:26
SmokeMachine m: sub postcircumfix:<.times( )>($on, [Int $times, &func]) { }; say "".times(3, * ~ ++$)
camelia WARNINGS for <tmp>:
Nil
Useless use of "~" in expression "* ~ ++" in sink context (line 1)
Zoffix jdv79: rakudo isn't a minimally passing compiler. It's also used by users. So Your™ will basically demonstrate how well the Roast specs a usable language 15:44
Zoffix SmokeMachine: looks like a bug to me. R#2422 R#1571 15:47
synopsebot R#2422 [open]: github.com/rakudo/rakudo/issues/2422 [useless useless] Useless useless use in WhateverCode given as arg to postcircumfix.
R#1571 [open]: github.com/rakudo/rakudo/issues/1571 Flaws in implied sinkage / `&unwanted` helper
SmokeMachine Zoffix: thanks 15:52
atroxaper Hello, #perl6 ! 16:41
Please help me to understand why wrapped sub has strange signature? :)
p6: sub wr(:$a) {say $a}; say %(:a(3)) ~~ &wr.signature; &wr.wrap(-> |c {callsame(c)}); say %(:a(3)) ~~ &wr.signature; 16:42
camelia True
False
atroxaper p6: sub wr(:$a) {say $a}; say &wr.signature; &wr.wrap(-> |c {callsame(c)}); say &wr.signature; 16:43
camelia (:$a)
(:$a)
atroxaper As I see, the signature is the same after wrapping the sub. But ACCEPTS works differently. Why?
Moreover, signatures has the same WHICH... 16:47
p6: sub wr(:$a) {say $a}; say &wr.signature.WHICH; &wr.wrap(-> |c {callsame(c)}); say &wr.signature.WHICH;
camelia Signature|58354488
Signature|58354488
Geth marketing: 207506fbfe | (Zoffix Znet)++ | 17 files
Add "Set Ops" 6.d Teaser Flyer / ID#1540393287
 16:54
Zoffix Perl 6 'Diwali' 6.d major language release will see much more well-defined Set Operators, which in Rakudo compiler are now also a lot faster than they were when 6.c was released: marketing.perl6.org/id/1540393287/pdf_digital 16:58
atroxaper Zoffix: nice! 17:02
Zoffix :)
jdv79 so "Your" will be a forked rakudo that's pared down? 17:43
AlexDaniel jdv79: ? 18:13
Zoffix jdv79: no, rakudo is not involved in any part of the You™ process. 18:21
*Your™
jdv79 good luck 18:28
jdv79 sounds insane 18:29
Zoffix Thanks?
I doubt it's gonna be more than 500 lines of code.
jdv79 of p5?
Zoffix nqp 18:30
But doesn't really matter of what.
jdv79 but then what runs the nqp
Zoffix MoarVM
JS
JVM... anything
Oh and Truffle 18:31
jdv79 ah, so it is uding parts of the existing stuff
what is the line length of these 500 lines?
Zoffix I think you're misunderstanding what's being built. 18:32
jdv79 isnt it a "perl6 compiler"? 18:33
Zoffix The premise of Roast is anything that can pass it can claim to be a Perl 6 compiler. That begs the question: how little can that compiler do in order to pass the test. Ignoring the most nonsensical answers (just output a single `1 - ok`), the base requirement is it must be capable of producing valid TAP output and for that TAP output to contain the correct number of tests present in roast. The rest is all 18:34
optional
So—presumably—and all it needs to do is able to figure out sub calls for a select number of subs and print " - ok" for all of them, with a test counter. 18:35
And the further that is from the truth, the stronger is the spec.
lichtkind lizmat, what i have to do so you recognize math::matraix udates in your posts? 18:36
lizmat lichtkind: did you upload your module to CPAN? 18:36
if not, that would help a lot
lichtkind lizmat, not yet
lizmat otherwise you will have to ping me "manually" e.g. by adding a "weekly: " notice on the channel 18:37
chrom hello, anyone here succesfully running perl6 through openbsds httpd/slowcgi and has written a short howto on that? :) 18:42
Zoffix \o 18:42
Never heard of openbsds httpd/slowcgi. Sounds like a web thing and "Cro" is what the cool kids use for Web in Perl 6. Maybe its docs mention that slowcgi stuff: cro.services/ 18:43
chrom mhm..thats not the question but its worth a try to get an answer 18:48
jdv79 Zoffix: oh. thats boring. 18:50
AlexDaniel chrom: sure! What's the statement? :) 18:51
AlexDaniel ahhh I just noticed your previous messages 18:51
whoops! :)
Zoffix jdv79: it's just a small part of testing our system plays well with multi-compiler paradigm. And it's hardly high on priorities list :)
jdv79 cool 18:53
buggable New CPAN upload: Algorithm-SkewHeap-0.0.1.tar.gz by JEFFOBER cpan.metacpan.org/authors/id/J/JE/...0.1.tar.gz 19:02
chrom p6: say 5; 19:02
camelia 5
lizmat twitter.com/tipjar/status/1055174718326620160 # perhaps someone needs to tell him about Failures in Perl 6 19:12
El_Che lizmat: done 19:19
lizmat El_Che++ 19:20
jdv79 what's an easy way to say "string exists in a key of hash" 19:29
kinda like $str ~~ any(%h.keys) ?
jdv79 is htat right? 19:29
but that's whole matches. i want substring. 19:30
jnthn I'd probably so something like ?%h.keys.grep(*.contains($str)) 19:31
*do
jnthn heh, `so` woulda worked there :) 19:31
jdv79 jnthn: thanks 19:33
lizmat m: my %h = a => 42; dd "a" (elem) %h; dd "b" (elem) %h 19:34
camelia Bool::True
Bool::False
lizmat jdv79: ^^^
timotimo no, that's whole matches again 19:39
lizmat ah, I misinterpreted "$str ~~ any(%h.keys)" 19:43
jdv79: then probably something like "with %h.keys.first(*.contains($str))" 19:46
so it will short-circuit on the first match
jnthn lizmat: grep in boolean context also will, no? :) 19:53
jdv79 actually its the opposite. its the keys that are the substrings 19:56
lizmat m: say so ^10 .grep: *.say # jnthn, looks like :) 19:57
camelia 0
True
jdv79 "The use of hash variables in regexes is preserved." 20:01
smells like a typo in the docs
jnthn likes pickles :)
Should be reserved :) 20:02
m: my %h; "foo" ~~ /%h/
camelia 5===SORRY!5=== Error while compiling <tmp>
The use of hash variables in regexes is reserved
at <tmp>:1
------> 3my %h; "foo" ~~ /%h7⏏5/
jnthn At least it's correct there
jdv79 pickle reserves?
so the var in the regex has to be explictily be an array (starts with @) to be interpolated as a alternation?
jnthn Yes, but note that you can do @(%h.keys) 20:03
jdv79 ah. i tried @{}. been doing a lot of p5 lately. 20:04
thakns
lizmat twitter.com/petdance/status/1055188193698570240 # lack of taint is now at #1 20:05
moritz would be nice to know how much Andy has used Perl 6 20:06
El_Che I was under the impression he wasn't impressed 20:07
(and do people of taint on perl5?)
moritz I don't use taint in p5 myself 20:08
El_Che I did when I was learning it :) 20:09
AlexDaniel hmmm repology.org/metapackage/rakudo/versions 20:13
hah there's a Debian-based distro called Parrot 20:14
and it has rakudo 2018.09 FP 20:15
:P
Geth doc: 5c5815699f | (Zoffix Znet)++ | doc/Language/regexes.pod6
🥝🥝🥝 Eat the preserves

Also, the restriction isn't restricted to variables, rather just hashes, even if they're given via a $-sigiled var.		 21:36
synopsebot Link: doc.perl6.org/language/regexes
Xliff Can someone tell me why I am getting the following error message: "Cannot assign to a readonly variable or value" from this line: github.com/Xliff/p6-GtkPlus/blob/m...or.pl6#L49 22:07
Is that due to a fact that the action class method is using ($/) as a parameter? What's the best way to work around that?
TimToady ($/ is copy) is one way 22:24
leont Yeah 22:25
TimToady or pick a different var, and then $var<foo> rather than $<foo> 22:27
Geth doc: chsanch++ created pull request #2417:
Fix 2412		 22:47
timotimo you can also use $_ and then .<foo> instead of $<foo> 23:00
Xliff Is this noted in the docs? 23:55
geekosaur it's just a method call wit implied topic, should be documented with $_ 23:57
and use of "indexing" is implied by postcircumfixes being methods 23:58

Powered by Cro and the Raku® Programming Language.

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

Thank you!