»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.
Zoffix psch: the reduce meta op is not allowed to have any whitespace in it. IIRC it's formally ambiguous with the Array constructor, so it's possible we can't even throw X::Cannot::InlineDefineReduce (if we could, why not just make it work then?) 00:02
timotimo oh! 00:08
that really shouldn't have been attributed to me
timotimo .tell lizmat almost all the work from that commit was actually done by nine, i assume he's jusf flattened the commits and it somehow kept me as the author? 00:09
yoleaux timotimo: I'll pass your message to lizmat.
sjn m: my @a = <a 1 b 2 c 3>; for (@a,) { .WHAT.say } 02:04
evalable6 (Array)
sjn m: my @a = <a 1 b 2 c 3>; for (%@a,) { .WHAT.say } 02:05
evalable6 (Hash)
sjn m: my @a = <a 1 b 2 c 3>; for (@%@a,) { .WHAT.say }
evalable6 (Seq)
sjn m: my @a = <a 1 b 2 c 3>; for (@%@a) { .WHAT.say } 02:05
evalable6 (Pair)
(Pair)
(Pair)
sjn m: my @a = <a 1 b 2 c 3>; for ($%@a) { .WHAT.say }
evalable6 (Hash)
sjn fun 02:06
m: my @a = <a 1 b 2 c 3>; for ($%$@%%%$$$$@@@$@$@@@@@@%$@a) { .WHAT.say } 02:07
evalable6 (Hash)
sjn that'll be an instant hit... :) 02:08
paul_ I'm trying to install mod_scgi on my computer to work with the http clients for perl6, has anybody successfully crossed that hurtle on a mac? 02:10
ls 02:11
lookatme Don't know MAC :) 02:26
paul_ switching to nginx 02:29
mspo scgi? 02:41
manchicken Hiya! 02:48
I have some progress on the POSIX stuff.
Took me a while, got sick for like four consecutive weeks.
(I don't recommend that)
I hope to have the POSIX starter revived soon. 02:49
manchicken Do we have someone who packages rakudo for homebrew? 02:58
manchicken I think `File::Which` is missing `Test::META` as a dependency in its `META6.json`. 04:11
Turns out closing the ^%@#ing file I just wrote is suuuuper important before using it in a child process. 04:20
ZzZombo m: class C {has Str() $.a=123};C.new 07:14
evalable6 (exit code 1) 04===SORRY!04=== Error while compiling /tmp/9UPtfsy_O0
Coercio…
ZzZombo, Full output: gist.github.com/d03f2111b902a4cbcd...f10e19b5ea 07:15
ZzZombo Uh, what does it try to tell me about the code? 07:15
psch coercers are not types 07:17
geekosaur in a sense, a coercion in a method parameter is an instruction to the parameter-passing code, not a type. variables don't have such parameter-passing code, so coercions can't work. 07:19
ZzZombo m: role R {has $.val is rw};class C {has Proxy $.a.=new(:STORE(method ($val){self.val=$val}),:FETCH(method (){self.val})) but R};$_=C.new;.a=3;.a.say 07:27
evalable6 (signal SIGHUP) «timed out after 10 seconds»
ZzZombo what did I break now? 07:28
m: role R {has $.val is rw};class C {has $.a = Proxy.new(:STORE(method ($val){self.val=$val}),:FETCH(method (){self.val})) but R} 07:29
evalable6
ZzZombo m: role R {has $.val is rw};class C {has $.a = Proxy.new(:STORE(method ($val){self.val=$val}),:FETCH(method (){self.val})) but R};$_=C.new 07:30
evalable6 (signal SIGHUP) «timed out after 10 seconds»
ZzZombo m: role R {has $.val is rw};class C {has $.a = Proxy.new(:STORE(method ($val){self.val=$val}),:FETCH(method (){self.val}))};$_=C.new
evalable6 (signal SIGHUP) «timed out after 10 seconds»
ZzZombo m: role R {has $.val is rw};class C {has $.a = Proxy.new(:STORE(method ($val){self.val=$val}),:FETCH(method (){self.val})) but R};$_=C.new(:a) 07:32
evalable6
ZzZombo m: role R {has $.val is rw};class C {has $.a};$_=C.new(:a(Proxy.new(:STORE(method ($val){self.val=$val}),:FETCH(method (){self.val})) but R)) 07:33
evalable6 (signal SIGHUP) «timed out after 10 seconds»
ufobat m: my $acp; dd split($acp, ",") 07:37
evalable6 (exit code 1) Cannot resolve caller split(Str: Any); none of these signatures match:
(…
ufobat, Full output: gist.github.com/201f5c5d1b55fffe92...1cb101839b
ufobat m: my Str $acp; dd split($acp, ",")
evalable6 (exit code 1) Cannot unbox a type object (Str) to a str.
in block <unit> at /tmp/n0B_mfFXOZ line 1
ufobat :-(
ZzZombo m: role R {has $.asd='asd'};class C {has $.a but R=1};C.new.a 07:56
evalable6 (exit code 1) Cannot modify an immutable Str (Nil)
in block <unit> at /tmp/wmBuSkCUkh line 1
ZzZombo m: role R {has $.asd};class C {has $.a but R=1};C.new.a 07:57
evalable6 (exit code 1) Cannot modify an immutable Str (Nil)
in block <unit> at /tmp/848HMiomDp line 1
ZzZombo m: role R {has $.asd};class C {has $.a but R};C.new.a
evalable6
lookatme m: role R {has $.asd='asd'};class C {has $.a does R=1};C.new.a
evalable6 (exit code 1) 04===SORRY!04=== Error while compiling /tmp/NmMuznaty2
Cannot …
lookatme, Full output: gist.github.com/0989c254edf32de7a4...6e9b7879fc
lookatme m: role R {has $.asd='asd'};class C {has $.a =1 but R};C.new.a
evalable6
ZzZombo m: role R {has $.asd='asd'};class C {has $.a but R};C.new.a
evalable6 07:58
lookatme m: role R {has $.asd='asd'};class C {has $.a is rw = 1 };my $c = C.new; $c.a does R; say $c.a.asd; 08:00
evalable6 asd
lookatme m: role R {has $.asd='asd'};class C {has $.a =1 but R};say C.new.a.asd
evalable6 asd
ZzZombo m: role R {has $.asd='asd'};class C {has $.a is rw=1 but R};C.new.a.asd=123;C.new.a.asd.say 08:01
evalable6 (exit code 1) Cannot modify an immutable Str (asd)
in block <unit> at /tmp/BuH51kDWFJ line 1
ZzZombo m: role R {has $.asd is rw='asd'};class C {has $.a is rw=1 but R};C.new.a.asd=123;C.new.a.asd.say 08:02
evalable6 123
ZzZombo shit, I need it to be per per object.
lookatme `but` is only using in declaration sentence of variable, I think
timotimo is that a bug in our optimization i wonder 08:20
at some point we got an opt in if the RHS of the default value "assignment" is a constant
but it should still have cloned it 08:21
ZzZombo m: class C {has $a='wut???'};dd C.new 08:32
evalable6 C.new
ZzZombo Why is that allowed?
timotimo it's shorthand for a private attribute i believe? 08:33
ZzZombo m: class C {has $a='wut???'};say C.new.^attrs
evalable6 (exit code 1) No such method 'attrs' for invocant of type 'Perl6::Metamodel::ClassHOW'
in block <unit> at /tmp/0c1d0lE6lQ line 1
ZzZombo m: class C {has $a='wut???'};say C.new.^attributes
evalable6 (Mu $!a)
timotimo m: class C {has $a='wut???'; method wut { say $a } }; C.new.wut 08:34
evalable6 wut???
timotimo we even install a lexical alias for it
ZzZombo So it's not private? 08:37
ah
timotimo no, it is
ZzZombo you changed the class
lookatme :| 08:38
ZzZombo m: grammar G {has $a} 08:51
evalable6
ZzZombo m: grammar G {has $a=1};G.new 08:52
evalable6
ZzZombo m: grammar G {has $a=1};G.new.say
evalable6 ｢｣
ZzZombo m: grammar G {has $a=1};G.new.a.say
evalable6 (exit code 1) No such method 'a' for invocant of type 'G'. Did you mean 'at'?
in block <unit> at /tmp/VJevxCVUB2 line 1
ZzZombo m: grammar G {has $.a=1};G.new.a.say
evalable6 1
ZzZombo huh
where can I find the latest docs regarding slang creation? 08:53
lookatme no documents currently :) 08:54
timotimo there was a draft document somewhere. it had the same style as docs.perl6.org, but it was hosted on someone's private server 08:57
mouq.github.io/slangs.html
lizmat there's some code in Slang:: modules that can be cargo-culted
Slang::Roman comes to mind
ZzZombo I see there is `define_slang`, but no info on it, what is that? 08:58
lizmat meanwhile I'm still hoping for help with github.com/lizmat/P5wantarray/issues/1
lizmat $*LANG.define_slang( 08:59
'MAIN',
$*LANG.slang_grammar('MAIN').^mixin(Scalar::Grammar),
$*LANG.slang_actions('MAIN').^mixin(Scalar::Actions),
);
is basically the trick
(supposedly, anyway)
scimon I really need to control that voice in the back of my head that says. "How hard can it be?" Currently it's pointing at the Fork button on the SWIG Github... 09:00
Because yeah, I've written more than 10 lines of c in the last 5 years.
I might give it a quick look though.
scimon (Yeah. This may be somewhat beyond me ATM) 09:03
El_Che lizmat: isn't wantarray on Perl5's chopping block? 09:05
lizmat well, potentially, yes 09:06
El_Che ah, not yet, then
lizmat it was more an exercise for me creating a slang
El_Che I see an upcoming CCC talks by a clown trashing perl6 becasue of p5wantarray :) 09:07
ZzZombo m: multi trait_mod:<has>(Any:D $s,:$thing){say "\"$s\" has $thing."};my $a has thing<foo>;
evalable6 (exit code 1) 04===SORRY!04=== Error while compiling /tmp/L9w0wzzipj
Two ter…
lizmat hehe
evalable6 ZzZombo, Full output: gist.github.com/12c3ecb5fb585fbaab...becb8ce0d3
ZzZombo Why is this not allowed? 09:07
lizmat anyways, I need the same slanginess for handling P5exists() and P5delete() :-(
Zoffix ZzZombo: there are no docs on slang creation because there's no supported interface yet. It's an entirely experimental feature that can change at any time.
ZzZombo: there's no 'has' trait. 09:09
lookatme yeah, like the macro, slang is experimental feature. I am expect the macro feature :)
ZzZombo How so? I just declared it!
lizmat clickbaits p6weekly.wordpress.com/2018/01/29/...s-modules/ 09:12
lookatme m: multi trait_mod:<is>(Any:D $s,:$thing){say "\"$s\" has $thing."};my $a is thing<foo>; 09:13
evalable6 (exit code 1) 04===SORRY!04=== Error while compiling /tmp/3pRvfDT05d
Can't u…
lookatme, Full output: gist.github.com/4fcd4324585205f523...d6b9aabae7
lookatme m: multi trait_mod:<is>(Any:D $s,:$thing){say "$s has $thing."};my $a is thing<foo>;
evalable6 (exit code 1) 04===SORRY!04=== Error while compiling /tmp/qPnSp0V8NH
Can't u…
evalable6 lookatme, Full output: gist.github.com/84d3e5fe0758d8aa73...cb232300ce 09:13
Zoffix ZzZombo: you declared a subroutine. Unlike operators, you can't install arbitrarily-named traits and are restricted to `is`, `does`, `hides`, `will`, `of`, `returns`, and `handles` as primary trait keyword, but you can modify the name that follows it by using named attributes
lookatme m: multi trait_mod:<is>(Variable $s,:$thing){say "$s has $thing."};my $a is thing<foo>;
evalable6 Variable<94018092016616> has foo.
Zoffix m: multi infix:<has> (Any $s, Pair $thing) {say "\"$s.perl()\" has $thing."}; my $a has :thing<foo>; 09:16
evalable6 "Any" has thing␉foo.
Zoffix m: multi prefix:<thing> (\s) {s}; multi infix:<has> (Any $s, $thing) {say "\"$s.perl()\" has $thing."}; my $a has thing<foo>; 09:17
evalable6 "Any" has foo.
Zoffix :)
ZzZombo Why the choice of trait names limited? 09:30
Zoffix ZzZombo: Presumably because they get special treatment. There's more limitation: some of them can only be used with typenames. 09:32
lizmat because you would need a slang for it
m: multi trait_mod:<foo>(Variable $s,:$thing){say "$s has $thing."};my $a foo thing<foo>
evalable6 (exit code 1) 04===SORRY!04=== Error while compiling /tmp/HijVx_oDQM
Two ter…
lizmat, Full output: gist.github.com/25617c97851d4547bb...3c6b5a4017
buggable New CPAN upload: P5sleep-0.0.1.tar.gz by ELIZABETH cpan.metacpan.org/authors/id/E/EL/...0.1.tar.gz 09:35
lizmat sleep in P5 returns number of seconds actually slept 09:40
sleep in P6 returns Nil 09:41
Zoffix lizmat: looking at your code, you need to .Int the seconds too 09:42
El_Che where do the p5 modules appear? Not on metacpan it seems
Zoffix $ perl -wlE 'say sleep 0.6'
0
lizmat Zoffix: am assuming Time::HiRes is installed, I guess 09:43
Zoffix El_Che: metacpan is for Perl 5 modules. These are Perl 6 modules.
lizmat hmmm...not even with Time::Hires
El_Che they don't appear on modules.perl6.org/, hence my question 09:44
Zoffix $ perl -MTime::HiRes=sleep -wlE 'say sleep .6'
0.600061
Need to also import the routine
lizmat ah, yes
Zoffix eco: from:cpan author:ELIZABETH
buggable Zoffix, Found 34 results: Sub::Name, List::Util, Scalar::Util, Sub::Util, Tie::StdArray. See modules.perl6.org/s/from%3Acpan%20...AELIZABETH 09:45
New CPAN upload: P5built-ins-0.0.5.tar.gz by ELIZABETH cpan.metacpan.org/authors/id/E/EL/...0.5.tar.gz
Zoffix El_Che: ^ seem to show up. Note that it'd take some time for P5sleep to show up, since it's a new module
El_Che thx
buggable New CPAN upload: P5sleep-0.0.2.tar.gz by ELIZABETH cpan.metacpan.org/authors/id/E/EL/...0.2.tar.gz 09:55
stmuk does the "start virtualbox windows for the first time in months and install 100GB of updates" dance 09:59
moritz with 5 reboots!
lizmat afk for a few hours&
El_Che stmuk: please be patient
TEttinger stmuk: there's an offline installer, I heartily recommend it 10:04
download.wsusoffline.net/
El_Che TEttinger: is it called ubuntu
TEttinger I use it every time I have a windows machine that's years behind on updates 10:05
stmuk the highlight is the unchecking of the "always trust Oracle" box for VB itself
TEttinger it handles the reboots when necessary and keeps upating
El_Che I had a windows 7 laying around missing about 3 months op patches
it took literally more than 24h non stop
TEttinger yep 10:06
stmuk TEttinger: I've got "defer updates" checked which actually stops new features and allows security updates
TEttinger windows 10?
El_Che I don't know if the update mechanism is broken upstreak, or if they broke it to push people to windows 10
El_Che windowws 8 takes also that long 10:06
stmuk yeah and I've also experimented with Enterprise LTSB which I may move to 10:07
TEttinger not as much fun as enterprise LSD
El_Che defer as ha-aha-we'll-put-it-through-you-throat-anyway
stmuk Windows 10 Enterprise LTSB doesn't even have Metro or Flappy Birds
ZzZombo Huh, can I somehow override a metamodel for my class or anything for this matter? That could help me avoid to delve deep into slang creation... 10:11
AlexDani` “deep in the nqp bowels” :D 10:14
AlexDaniel squashable6: next 10:14
squashable6 AlexDaniel, ⚠🍕 Next SQUASHathon in 2 days and ≈23 hours (2018-02-03 UTC-12⌁UTC+14). See github.com/rakudo/rakudo/wiki/Mont...Squash-Day
AlexDaniel ouh wow!
totally forgot about that
sooooo let's see
AlexDaniel buggable: tag TESTNEEDED 10:16
buggable AlexDaniel, There are 31 tickets tagged with TESTNEEDED; See fail.rakudo.party/t/TESTNEEDED for details
AlexDaniel heh, roast about to reach commit 15000 10:18
squashable6: next 10:22
squashable6 AlexDaniel, ⚠🍕 Next SQUASHathon in 2 days and ≈23 hours (2018-02-03 UTC-12⌁UTC+14). See github.com/rakudo/rakudo/wiki/Mont...Squash-Day
ufobat could anyone be so kind and update the Docker image to the current rakduo star version? hub.docker.com/r/_/rakudo-star/ 11:57
moritz has no idea who can do that 11:58
stmuk its probably just bumping the version in 11:59
github.com/perl6/docker/blob/1be22...Dockerfile
MAINTAINER Rob Hoelz 12:00
AlexDaniel hoelzro: ↑ ? 12:01
moritz and then pushing the newly built image to the registry 12:02
ZzZombo m: my method(){} 12:56
evalable6 (exit code 1) 04===SORRY!04===
Type 'method' is not declared. Did you mean '…
ZzZombo, Full output: gist.github.com/cc8bbe1649c7c2bf8e...701226ffe3
ZzZombo m: my sub(){}
evalable6 (exit code 1) 04===SORRY!04===
Type 'sub' is not declared. Did you mean 'Sub…
ZzZombo, Full output: gist.github.com/85761eae8f450f7fd3...2c5acd9d11
ZzZombo m: my sub x(){}
evalable6
stmuk is still updating Win 10 64 bit and has no desire to do this with 32 bit as well 13:24
hmmm maybe appveyor offers 32 bit
DrForr m: my $p=run('perl6','-e','say 2+',:err);say $p.err.get; 13:30
evalable6 Nil
DrForr I was ratehr hoping that would capture the syntax error text... 13:33
*rather
raschipi stmuk: You should try a multi-tasking operating system. 13:34
El_Che m: my $p=run('perl6','-e','say 2+',:out);say $p.out.get;
evalable6 Nil
El_Che good try :)
stmuk raschipi: its hosted on one
jnthn DrForr: Dunno if it's something with the bot, but it works when I try it locally 13:35
DrForr Hrm. I'm doing this inside Cro, so I'm not sure if that's interfering. I'll try it outside Cro... 13:37
El_Che jnthn: does not work on my local install 13:38
jnthn Can't think how that'd make a difference
El_Che 2017.11
jnthn Just ran it on whatever I have around, which happens to be 2017.12-270-g90246e6 13:39
El_Che or maybe it does
weird output
jnthn The output will have color escapes I guess 13:40
For me those were passed straight on through
DrForr It works outside the app, so it's something in the stack I'm working with.
El_Che well, the error is printed by the perl6 cmd, not the $p.out.get
ah 13:41
err
wrong copy paste
sorry
jnthn DrForr: Maybe try using Proc::Async directly (Proc is just a wapper around that), to rule out any oddities with that 13:42
raschipi Perl6 emits color codes without checking if those are appropriate? Should at least ask if isatty(3). 13:43
jnthn raschipi: Yeah, probably should. Feel free to submit at patch, if time/interest. :)
DrForr jnthn: I tried that as well, the example was too long to paste to the bot, but I'll try it again.
jnthn s/at/a 13:45
DrForr I've also tried it asynchronously, though I'll admit that it's probably me in that case. 13:47
At least I know it's not something I'm doing that's *egregiously* stupid. 13:48
AlexDaniel m: my $p=run('/home/alex/.rakudobrew/bin/perl6','-e','say 2+',:out);say $p.out.get; 14:02
evalable6 04===SORRY!04=== Error while compiling -e
Missing required term after infix
at -e:1
------> 03say 2+08⏏04<EOL>
expecting any of:
prefix
term
Nil
AlexDaniel ‘perl6’ is not in PATH
because, well… which perl6 do you want? :)
the one that the bots runs with, or the one that you're running the code with?
so HEAD or whatever the bot has?
DrForr digs around in evalbot source. 14:07
AlexDaniel evalable6: source 14:10
evalable6 AlexDaniel, github.com/perl6/whateverable
AlexDaniel well, let's add the current one to the path
lizmat DrForr: re Spreadsheet::(Read|Write)Excel: modules.perl6.org/dist/P5pack:cpan:ELIZABETH 14:11
DrForr lizmat: Nod. I didn't say I'd be the one to write it though :) 14:13
lizmat ah, ok :-) 14:14
just wanted to let you know that at least you don't need to look at pack/unpack (hopefully) 14:15
AlexDaniel actually, thanks everyone 14:24
this is probably a nice addition
buggable New CPAN upload: Terminal-Table-0.0.7.tar.gz by ALOREN cpan.metacpan.org/authors/id/A/AL/...0.7.tar.gz 14:25
AlexDaniel because now you can bisect by running perl6 with different flags and stuff
let's see if it works…
araraloren :) Happy to see it was working 14:26
AlexDaniel evalable6test: run ‘perl6’, ‘-e’, ‘say $*PERL.compiler’
evalable6: run ‘perl6’, ‘-e’, ‘say $*PERL.compiler’
evalable6 rakudo (2018.01.60.g.2.c.36.ab.2.ef)
AlexDaniel yay?
tbrowder lizmat: if you have in mind p5 modules to be p6’d, maybe you could get people to commit to converting them in a list something like mos5 wantd for p5 to p6 (in some kind of order of importance as you see it). 14:26
AlexDaniel c: 2017.12 run ‘perl6’, ‘-e’, ‘say $*PERL.compiler’ 14:26
committable6 AlexDaniel, ¦2017.12: «rakudo (2017.11.137.gc.84.ed.2942)␤»
AlexDaniel hm this one is a bit weird 14:27
araraloren but.. seems like we not support ASCIIDOC ?
AlexDaniel c: 2017.12^ run ‘perl6’, ‘-e’, ‘say $*PERL.compiler’
committable6 AlexDaniel, ¦2017.12^: «rakudo (2017.11.136.g.0700.cdbf.0)␤»
tbrowder uh, “most wanted for p5 to p6...”
araraloren I am not using markdown
AlexDaniel IIRC there's a bug in whateverable that some tagged releases are built before the tag is pulled, so they end up with weird versions. But everything is technically correct anyway 14:28
m: my $p=run('perl6','-e','say 2+',:err);say $p.err.get;
evalable6 04===SORRY!04=== Error while compiling -e
AlexDaniel m: my $p=run('perl6','-e','say 2+',:err);say $p.err.slurp; 14:29
evalable6 04===SORRY!04=== Error while compiling -e
Missing required term after infix
at -e:1
------> 03say 2+08⏏04<EOL>
expecting any of:
prefix
term
AlexDaniel DrForr: ↑
does anybody want to try bisecting REPL now? :)
AlexDaniel you'd just need to feed the stdin into it correctly, I think 14:29
araraloren :( So our module page not support ASCIIDOC ?
lizmat tbrowder: yes, I will but after I get back from my PR&R 14:30
unless someone beats me to it, of course :-)
tbrowder: it most definitely should not depend on me :-)
DrForr AlexDaniel: Thanks. 14:32
Oh, probably linefeed issues, the output has linebreak... 14:33
*linebreaks
Geth whateverable: 2e06be9a04 | (Aleks-Daniel Jakimenko-Aleksejev)++ | lib/Whateverable.pm6
Tweak PATH to make perl6 available

Phixes #279
AlexDaniel DrForr: ↑ and that's how it looks in code :)
tbrowder well, putting a list on the site should be reasonably easy, the hard part is getting thoughts on the importance of modules to be converted. several folks have said p6 is not a good language for beginners because of lack of advanced features--those folks should be making entries on the "most wanted p5 to p6" list. 14:34
AlexDaniel oh, that's exactly what I needed by the way…
daxim pmurias, redd.it/7ty4xn 14:35
Geth doc: 8fdecffaa7 | (Zoffix Znet)++ | doc/Language/quoting.pod6
Clarify :b quoter tweak

It doesn't imply `q`, since `q` has only rudimentary backslashing and is silent on unrecognized sequences, while in `:b` they throw.
Rakudo impl: github.com/rakudo/rakudo/commit/deffe54b8d Spec: github.com/perl6/roast/commit/284a97d31d		 14:37
synopsebot Link: doc.perl6.org/language/quoting
AlexDaniel 6c: run :ENV(%(MVM_COVERAGE_LOG => ‘/anoethusntoahu’), <perl6 -e>, ‘say ‘hello’’
committable6 AlexDaniel, gist.github.com/f3c10a2818550e8e2b...0681d60797 14:38
AlexDaniel 6c: run :ENV(%(MVM_COVERAGE_LOG => ‘/anoethusntoahu’)), <perl6 -e>, ‘say ‘hello’’
committable6 AlexDaniel, ¦6c (27 commits): «Unexpected named argument 'ENV' passed␤ in block <unit> at /tmp/I3myGKQ3o0 line 1␤␤ «exit code = 1»» 14:38
AlexDaniel 6c: run :env(%(MVM_COVERAGE_LOG => ‘/anoethusntoahu’)), <perl6 -e>, ‘say ‘hello’’
committable6 AlexDaniel, gist.github.com/b3aca17b4e1bae843b...65a7788d89
DrForr Grump. I think it was actually an amalgam of three issues... 14:40
El_Che s/^G/T/ 14:41
DrForr Yep, the slurp + two other things I noticed fixed it. 14:42
AlexDaniel++
lizmat afk& 14:51
DrForr It's not quite ready for prime time - there are two more features I need to add, but this should give you an idea of what I've been doing. Gimme a few ere... 14:52
*here
DrForr github.com/drforr/perl6-App-Learn-Perl6 14:55
scimon So I just did a little test, writing a simple REST API (that simulates rolling a bunch of dice, not external dependencies) in Mojo::Lite, Cro and express. All in basically the simplest run types so no WSGI NGINX or other stuff. 15:07
AlexDaniel squashable6: next
squashable6 AlexDaniel, ⚠🍕 Next SQUASHathon in 2 days and ≈18 hours (2018-02-03 UTC-12⌁UTC+14). See github.com/rakudo/rakudo/wiki/Mont...Squash-Day
scimon Mojo and express clocked in at 2 seconds for 100 requqests to roll 1000 1000 sided dice (because generating lots of random numbers is fun). 15:08
Cro took 3 seconds for the same.
(This is sequential requests, using a for loop and curl).
So... that's not too shabby.
ZzZombo m: role R {};sub x() does R {} 15:11
evalable6 (exit code 1) 04===SORRY!04=== Error while compiling /tmp/7YTrCaj5OV
Cannot …
ZzZombo, Full output: gist.github.com/860367b59513d9035d...5c14ffca4c
ZzZombo m: role R {};sub x does R () {}
evalable6 (exit code 1) 04===SORRY!04=== Error while compiling /tmp/d_ZQeaDzqg
Missing…
ZzZombo, Full output: gist.github.com/d6e792870b2250a407...b40240a1c1
ZzZombo Why? 15:12
jnthn What are you expecting to happen? 15:13
ZzZombo apply the role to the sub?
DrForr What could the role give the subroutine that would be relevant? 15:14
jnthn But a sub isn't a type, and the does trait modifier is defined as doing compile-time role composition into a type
ZzZombo m: role R {};sub x(){};&x does R;say &x ~~ R
evalable6 True
DrForr Attributes work with clases, not subroutines. Same with methods.
ZzZombo how is sub not a type? 15:15
ZzZombo m: role R {};sub x(){};&x does R;say &x ~~ R,&x.WHAT 15:15
evalable6 True(Sub+{R})
jnthn The way to do this is to write a routine trait that receives the routine as a parameter and then does the application of the role
ZzZombo m: role R {};sub x(){};&x.WHAT.say 15:16
evalable6 (Sub)
jnthn A sub is an *instance* of the Sub class
Contrast it with class C does R { } where you're declaring a type C
ZzZombo DrForr, in my case I just apply a role to tag a method as some kind, like Delphi attributes, but without any other fields (docwiki.embarcadero.com/RADStudio/T...s_(RTTI)). 15:18
szabgab hi there! 15:19
yoleaux 18 Aug 2017 12:42Z <Zoffix> szabgab: RE smoking Rakudo itself. There's no real setup yet. There's Toaster ( rakudo.party/post/Perl-6-Release-Q...em-Toaster ) that's currently being run on just Debian. Eventually that'll expand to more OSes. Of course, anyone can do `make stresstest` in a Rakudo checkout to stresstest it on their own systems.
szabgab It seem yoleaux just knows what I wanted to ask :)
I just looked at smoke.perl6.org/report and it returns 500
SmokeMachine m: multi trait_mod:<does>(&func, Mu:U $bla) {&func does $bla}; role R {}; sub ble does R {}; say &ble.WHAT 15:20
evalable6 (Sub+{R})
szabgab is that site supposed to work?
DrForr ZzZombo: Yeah, I was jus thinking about your potential use case and I was probably being unnecesarily harsh. 15:20
SmokeMachine ZzZombo: ^^ 15:21
jnthn We could in theory make it work, it just feels slightly off to me
ZzZombo SmokeMachine, I'll take a look. Thank you. 15:22
jnthn It's not so much effort to write a multi trait_mod:<is>(Routine $r, ...) { ... }
MasterDuke szabgab: i don't think that site has been updated at all recently 15:24
szabgab It is linked under "smoke testing" on perl6.org/community/ so if that site is not in use then maybe it should not be linked either. 15:26
stmuk testers.perl6.org/ also seems dead 15:43
Geth whateverable: f3b3e352d7 | (Aleks-Daniel Jakimenko-Aleksejev)++ | lib/Whateverable.pm6
Fix Bisectable and others not getting the right PATH

Fixes an oops in 2e06be9a049854d6cb5a5913a28b09122afc5e7d
Phixes #280		 16:45
jkramer m: (1 xx 5).combinations(1..3).say 17:37
evalable6 ((1) (1) (1) (1) (1) (1 1) (1 1) (1 1) (1 1) (1 1) (1 1) (1 1) (1 1) (1 1) (1 1) (1 1 1) (1 1 1) (1 1 1) (1 1 1) (1 1 1) (1 1 1) (1 1 1) (1 1 1) (1 1 1) (1 1 1))
jkramer Is there a smarter way to get a list of combinations without the duplications, instead of first generating lots of duplicates and then using .unique? 17:38
With smarter I mean faster, when using a really long list of input values :)
That possibly are all the same
hoelzro stmuk, AlexDaniel: it's a matter of bumping the version there and updating the library file in the images repo 17:44
I can do it later today
I totally missed the announcement that R* just came out =S
stmuk ++hoelzro 17:50
ufobat btw, the term "sigilless variable" is in the glossary but does not have any text to it: docs.perl6.org/language/glossary#S...s_Variable 19:32
bazzaar \o perl6 21:51
hi, I'm trying out Cro and IO.spurt of a http request header/body response to file(s), would be grateful if someone could look at gist.github.com/bazzaar/040af6a440...file1-txt, the result is at the bottom of the gistfile 21:55
I'm using Rakudo Star 2018.01, but I got the same result with 2017.10 this am 21:58
geekosaur you will probably need to either extract the response content yourself, or explicitly request the response be converted to Str 22:07
a response is not just the content! it's also a response code and various headers telling you/the application how to interpret the response content
geekosaur I don;t know Cro offhand, so I don't knwo if it has e.g. Cro::HTTP::Response.Str 22:08
DrForr Wish this hadn't come up while I wasn't buried in OSCON stuff. 22:09
(just put a Cro application up on GH, hence my interest.)
bazzaar geekosaur: thanks for your words of wisdom, that's given me a good lead, I will go away and investigate further. 22:10
timotimo bazzaar: there's a support channel for cro on freenode, too, btw! it'? #cro 22:47
it's #cro
bazzaar: irclog.perlgeek.de/cro/2018-01-24 - does this help perhaps? 22:50
bazzaar timotimo: thanks for the #cro channel link, I'd forgotten about it. In this case I was sure if it was an IO related thing
timotimo depending on what you want to do, you could put the code earlier in the pipeline where the body hasn't been parsed yet 22:51
or feed it into a RequestSerializer, like the one the http client uses
bazzaar timotimo: that's brilliant advice, thanks so much :-) 22:52
bazzaar *wasn't (previous previous) 22:54
El_Che DrForr: you planning to go delirium or something on friday? Someone on twitter asks 22:55
(I am not sure I can make it myself)
DrForr Aroo? Yeah, I'll be there. 22:59
I'm still in the throes of writing here, so I haven't checked recently.
jnthn bazzaar: What exactly are you after, out of curiosity? 23:04
DrForr I didn't see a mention or DM, what message is it?
jnthn bazzaar: To write out the exact set of bytes that come back over the network without any decoding applied at all?
Anyway, $resp.Str for the first spurt will get you the response line and headers 23:05
await $resp.body-blob a few lines up will get you the bytes making up the response body, except that decoding chunked encoding will have been handled for you 23:06
El_Che DrForr: someone sent me a private msg asking if we've planned something
bazzaar jnthn: to be honest, I'm just aiming to do some text retrieval from a web search page, and I thought I'd have a go using cro, and started off using the example in the http request example 23:07
DrForr Oh. Have them DM me or whatever as well, but as I recall last time, it really doesn't open for our purposes until around 4pm. 23:08
bazzaar jnthn: thanks for the tip re. $resp.Str and $resp.body-blob 23:10
jnthn bazzaar: oh, then what I said will do it, and if you know the body is text then $resp.body-text will get you a Str instead of a Blob :) 23:10
jnthn .body will look at the set of available body parsers and Do The Right Thing 23:11
The default set includes a JSON parser, so if you get a JSON response then .body will already be decoded
bazzaar jnthn: thanks again, and thanks for Cro :-) 23:12
lizmat waves from FRA 23:16
jnthn bazzaar: Welcome :) 23:17
lizmat: Hmm, the lounge there has beer on tap and sausages? :)
El_Che lizmat will rock the boat!
lizmat well, actually from a new hotel just outside of the airport 23:18
and yes, the bar has beer on tao and sausages 23:19
it *is* germany after all :-_
)
jnthn
.oO( Is the typing due to the beer on tap? :-) )
 23:20
timotimo lizmat: have a safe trip!
lizmat well, some rum and cider already passed the revue 23:21
:-)
timotimo: thanks
El_Che lizmat: take care! 23:32
and if a ghost ask for a present of your mother tell it you love reggae
buggable New CPAN upload: P5sleep-0.0.3.tar.gz by ELIZABETH cpan.metacpan.org/authors/id/E/EL/...0.3.tar.gz 23:35
comborico1611 So I read some tutorials and still dont' know how to customize a major mode in Emacs . . . 23:47
All I want to do is turn off electric-indent-mode from off html-mode.
lizmat
.oO( what a dreadlock holiday )
 23:52
sleep&
ilmari comborico1611: (add-hook 'html-mode (lambda () (electric-indent-local-mode 0))) 23:56
html-mode-hook, even
comborico1611 Thank you so much. Trying it now. 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!