Perl 6 - the future is here, just unevenly distributed

IRC log for #perl6, 2016-08-04

Perl 6 | Reference Documentation | Rakudo

| Channels | #perl6 index | Today | | Search | Google Search | Plain-Text | summary

All times shown according to UTC.

Time Nick Message
00:06 BenGoldberg joined #perl6
00:09 Actualeyes joined #perl6
00:12 pierre joined #perl6
00:17 pierre joined #perl6
01:00 cyphase joined #perl6
01:00 khw joined #perl6
01:19 cyphase joined #perl6
01:26 pierre joined #perl6
01:46 kalkin-_ joined #perl6
01:46 pierre joined #perl6
01:47 ilbot3 joined #perl6
01:47 Topic for #perl6 is now »ö« Welcome to Perl 6! | https://perl6.org/ | evalbot usage: 'p6: say 3;' or rakudo:,  or /msg camelia p6: ... | irclog: http://irc.perl6.org or http://colabti.org/irclogger/irclogger_logs/perl6 | UTF-8 is our friend!
01:48 freezerburnv joined #perl6
01:49 holyghost joined #perl6
01:57 pierre joined #perl6
02:02 vcv Any modules or tricks for pretty printing .gist output?
02:04 Leamas joined #perl6
02:09 gfldex vcv: there are 2 modules in the eco system that dump data
02:19 holyghost joined #perl6
02:19 Zoffix joined #perl6
02:21 Zoffix m: class Foo { has $.bar }; say "The code is in {.file} on line {.line}" given Foo.^can('bar')[0];
02:21 camelia rakudo-moar 4693a5: OUTPUT«concatenate requires a concrete string, but got null␤  in block <unit> at <tmp> line 1␤␤»
02:21 Zoffix Is there a way to get where the attribute is defined at? The line and number?
02:22 vcv Thanks
02:30 noganex joined #perl6
02:30 dalek doc: 25ce0e5 | (Aleks-Daniel Jakimenko-Aleksejev)++ | / (3 files):
02:30 dalek doc: Minor fixes
02:30 dalek doc: review: https://github.com/perl6/doc/commit/25ce0e539a
02:32 wamba joined #perl6
02:32 dalek modules.perl6.org: cd7572b | (Zoffix Znet)++ | README.md:
02:32 dalek modules.perl6.org: s/browers/browsers/
02:32 dalek modules.perl6.org: review: https://github.com/perl6/modules.perl6.org/commit/cd7572b696
02:32 dalek perl6.org: fa1f145 | (Zoffix Znet)++ | README.md:
02:32 dalek perl6.org: s/browers/browsers/
02:32 dalek perl6.org: review: https://github.com/perl6/perl6.org/commit/fa1f1459ad
02:34 * gfldex .oO( Did you mean "brewers"? )
02:35 AlexDaniel m: brewerse
02:35 camelia rakudo-moar 4693a5: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Undeclared routine:␤    brewerse used at line 1. Did you mean 'reverse'?␤␤»
02:36 AlexDaniel by the way, this is a good one: 🤦
02:45 Zoffix Well, that's interesting....
02:45 Zoffix m: class Foo { has $.bar; method bar('foo') {}  }; say "The code is in {.file} on line {.line}" given Foo.^can('bar')[0];
02:45 camelia rakudo-moar 4693a5: OUTPUT«The code is in <tmp> on line 1␤»
02:46 Zoffix oh :S locally it gives me 'The code is in gen/moar/m-CORE.setting on line 587'
02:47 Zoffix Ah
02:47 Zoffix m: class Foo { has $.bar; multi method bar('foo') {}  }; say "The code is in {.file} on line {.line}" given Foo.^can('bar')[0];
02:47 camelia rakudo-moar 4693a5: OUTPUT«The code is in gen/moar/m-CORE.setting on line 587␤»
02:50 Zoffix m: class Foo { multi method bar('foo') {}  }; say "The code is in {.file} on line {.line}" given Foo.^can('bar')[0];
02:50 camelia rakudo-moar 4693a5: OUTPUT«The code is in gen/moar/m-CORE.setting on line 587␤»
02:50 Zoffix 😿
02:51 Zoffix m: class Foo { proto method bar(|) {}; multi method bar('foo') {}  }; say "The code is in {.file} on line {.line}" given Foo.^can('bar')[0];
02:51 camelia rakudo-moar 4693a5: OUTPUT«The code is in <tmp> on line 1␤»
02:53 Zoffix m: class Foo { ␤␤multi method bar('foo') {}␤␤␤ multi method bar('meow') {} }; say .^can('bar')[0].cando(\($_, 'meow'))[0].line given Foo
02:53 camelia rakudo-moar 4693a5: OUTPUT«6␤»
02:54 Zoffix m: class Foo { has $.bar;␤␤multi method bar('foo') {}␤␤␤ multi method bar('meow') {} }; say .^can('bar')[0].cando(\($_)) given Foo
02:54 camelia rakudo-moar 4693a5: OUTPUT«()␤»
02:55 Zoffix m: class Foo { has $.bar }; say Foo.^can('bar')[0];
02:55 camelia rakudo-moar 4693a5: OUTPUT«bar␤»
02:55 Zoffix :/
02:55 Zoffix m: class Foo { ␤␤␤␤has $.bar }; say Foo.^can('bar')[0].cando(\(Foo))[0].line;
02:55 camelia rakudo-moar 4693a5: OUTPUT«1␤»
02:56 Zoffix Something is really stinks in this code path.
02:56 Zoffix m: class Foo { has $.bar = 42;␤␤multi method bar('foo') {}␤␤␤ multi method bar('meow') {} }; Foo.new.bar
02:56 camelia rakudo-moar 4693a5: OUTPUT«Cannot resolve caller bar(Foo.new(bar => 42): ); none of these signatures match:␤    (Foo $: Str $ where { ... }, *%_)␤    (Foo $: Str $ where { ... }, *%_)␤  in block <unit> at <tmp> line 6␤␤»
02:57 Zoffix :/
02:58 kid51 joined #perl6
03:01 Zoffix I guess it sees I provided a `bar` method so it doesn't auto-generate an accessor.
03:11 canopus joined #perl6
03:19 avenj joined #perl6
03:31 canopus joined #perl6
04:28 AlexDaniel m: say WHAT ‘.’.IO.absolute
04:28 camelia rakudo-moar 4693a5: OUTPUT«(Str)␤»
04:28 AlexDaniel m: say WHAT ‘.’.IO.abspath
04:28 camelia rakudo-moar 4693a5: OUTPUT«(Str)␤»
04:28 AlexDaniel something is wrong here
04:28 AlexDaniel committable: releases say WHAT ‘.’.IO.absolute
04:28 committable AlexDaniel: ¦«2015.10,2015.11,2015.12,2016.02,2016.03,2016.04,2016.05,2016.06,2016.07,HEAD»: (Str)
04:28 AlexDaniel yeah, well… yeah
04:29 AlexDaniel m: say WHAT ‘.’.IO.relative
04:29 camelia rakudo-moar 4693a5: OUTPUT«(Str)␤»
04:29 AlexDaniel m: say WHAT ‘.’.IO.parent
04:29 camelia rakudo-moar 4693a5: OUTPUT«(Path)␤»
04:39 Cabanossi joined #perl6
04:45 cibs joined #perl6
05:06 andrzejku joined #perl6
05:26 ovibos joined #perl6
05:36 ufobat joined #perl6
05:56 TEttinger joined #perl6
06:15 ufobat morning
06:18 movl joined #perl6
06:37 firstdayonthejob joined #perl6
06:37 leedo joined #perl6
06:45 holyghost_ joined #perl6
06:49 moritz good moarning
06:50 holyghost_ gm
06:51 CIAvash joined #perl6
06:56 sno joined #perl6
07:03 RabidGravy joined #perl6
07:19 darutoko joined #perl6
07:46 dakkar joined #perl6
07:48 zakharyas joined #perl6
07:58 trnh joined #perl6
08:02 jonas2 joined #perl6
08:07 sno joined #perl6
08:15 bjz joined #perl6
08:42 ufobat my perl6 terminates :/
08:42 ufobat https://gist.github.com/ufobat/5a12692130bda9d1c2a0fc17fc7c1eca
08:42 ufobat and sometimes segfaults
08:47 margeas joined #perl6
08:56 marcovorg joined #perl6
09:02 ufobat *** Error in `/home/martin/.rakudobrew/moar-2016.07.1/install/bin/moar': double free or corruption (fasttop): 0x00007f23d00c1f10 ***
09:12 maettu1 left #perl6
09:20 cyphase joined #perl6
09:26 andrewalker joined #perl6
09:30 Rawriful joined #perl6
09:39 holyghost joined #perl6
09:52 pierre_ joined #perl6
09:53 mcmillhj joined #perl6
09:56 tbrowder joined #perl6
10:05 mcmillhj joined #perl6
10:14 tbrowder /whois Zoffix
10:18 moritz I've often asked myself who I am; I wonder why an IRC server should know
10:19 mcmillhj joined #perl6
10:25 Zoffix heh
10:26 DrForr The shad^opers know.
10:28 rindolf joined #perl6
10:33 mcmillhj joined #perl6
10:38 araujo joined #perl6
10:41 kaare_ joined #perl6
10:46 mcmillhj joined #perl6
10:49 pierre_ joined #perl6
10:52 dalek doc: 851909c | (Tom Browder)++ | doc/Language/about.pod6:
10:52 dalek doc: Add onfo on this doc and its overview status
10:52 dalek doc:
10:52 dalek doc: Reference *CONTRIBUTING.md* and *S26* as the primary sources of this info.
10:52 dalek doc: review: https://github.com/perl6/doc/commit/851909c9b1
10:52 dalek doc: 36a9232 | (Tom Browder)++ | doc/Language/about.pod6:
10:52 dalek doc: fix typo
10:52 dalek doc: review: https://github.com/perl6/doc/commit/36a9232936
10:55 tbrowder moritz: I was just reading about irc commands, i guess /whois and such don't work here (at least in this entry place in my irc client). What is the most used irc client here I wonder?
10:56 Zoffix Works here. I think you typed / twice, because it's not meant to show up in the channel
10:56 tbrowder I think it's my client.
10:57 Zoffix Oh, there's actually a space before it. Type it with / as the very first character
10:57 moritz tbrowder: right, IRC commands typically have to start with a slash, no spaces before it allowed
10:57 Zoffix WHOIS is part of the IRC protocol, so I'd be very suspicious of an IRC client that doesn't support it :P
10:57 lizmat works fine in Textual  :-)
10:59 tbrowder Okay, that works, another finger fumble trying one-finger typing on an iPad with MS trying to tell me what I'm trying to say...
11:00 pierre_ joined #perl6
11:01 mcmillhj joined #perl6
11:03 kid51 joined #perl6
11:10 dalek doc: d9ae2fc | (Tom Browder)++ | doc/Language/about.pod6:
11:10 dalek doc: some tweaks
11:10 dalek doc: review: https://github.com/perl6/doc/commit/d9ae2fc70b
11:26 cooper joined #perl6
11:28 jkramer joined #perl6
11:28 jkramer Ahoy!
11:29 jkramer What's the HTTP client module of choice these days for P6? I found a few on modules.perl6.org, but all of them are apparently about to be completely rewritten or don't support HTTPS.
11:37 dalek doc: b876885 | (Tom Browder)++ | doc/Language/about.pod6:
11:37 dalek doc: change case of HTML for consistency on the page
11:37 dalek doc: review: https://github.com/perl6/doc/commit/b876885815
11:43 ZoffixMobile joined #perl6
11:43 NEveD joined #perl6
11:43 ZoffixMobile jkramer, I use HTTP::UserAgent
11:44 ZoffixMobile LWP::Simple has a nasty bug in it, rendering it useless. There's HTTP::Tinyish, but it uses 'curl' under the hood and fails if it's not installed.
11:45 ZoffixMobile And last, but not least, there's Mojo::UserAgent:from<Perl5>
11:51 jkramer I just tested Tinyish, it fails on HTTPS URLs for some reason. Weird that it uses curl, because curl obviously supports https
11:56 timotimo has anybody been using PRE and POST blocks in production code?
12:02 pierre_ joined #perl6
12:05 bjz joined #perl6
12:07 smls joined #perl6
12:18 [Coke] "moaty nine"++
12:18 smls_ joined #perl6
12:20 sena_kun joined #perl6
12:23 tbrowder ref panda and zef: both show LTA msgs when asked to (1) install a module that cannot be found or (2) install a module that is already installed (panda is the more voluble of the two).
12:25 canopus joined #perl6
12:31 timotimo %)
12:33 masak tbrowder: sounds like something that could be reported to the github issue queues, perhaps.
12:34 awwaiid I cant' remember and google is failing -- is there a magical var that has the current for loop count? like for @something -> $n { ... var has 1, 2, 3, ... }
12:35 smls awwaiid: Not exactly, but you can use  $++
12:35 yoleaux 2 Aug 2016 17:11Z <TimToady> smls: you can autogen the function directly: my &prod = &prefix:<[*]>; say prod 1..5
12:35 yoleaux 2 Aug 2016 17:31Z <TimToady> smls: and the reason we don't have a .product to go with .sum is that .sum can be heavily optimized for huge ranges like sum 1 .. 10¹⁰⁰⁰ whereas heavy use of factorials would probably just cache the sequence from [\*]
12:35 ufobat https://gist.github.com/ufobat/68979f84a4830b55fe9ff6878436d782 :-)
12:36 awwaiid smls: $++ not in docs.perl6.org index :(
12:36 perlpilot m: my @a = "a".."z";  for @a.kv -> $i, $v { say "$i: $v" }
12:36 camelia rakudo-moar 1488b3: OUTPUT«0: a␤1: b␤2: c␤3: d␤4: e␤5: f␤6: g␤7: h␤8: i␤9: j␤10: k␤11: l␤12: m␤13: n␤14: o␤15: p␤16: q␤17: r␤18: s␤19: t␤20: u␤21: v␤22: w␤23: x␤24: y␤25: z␤»
12:36 smls awwaiid: $ is an implicitly declared state variable. You can use it like any other state variable, e.g. increment it with ++ :)
12:36 perlpilot Using .kv is more better though  ;)
12:37 awwaiid smls: implicit to what scope?
12:37 awwaiid perlpilot: ah .kv; I kept trying to just do @stuff -> $a, $n but got htem two at a time. forgot about .kv
12:38 freezerburnv joined #perl6
12:39 smls awwaiid: Each instance of $ is its "own" scope
12:39 smls m: for ^5 { say $++; say $++ }
12:39 camelia rakudo-moar 1488b3: OUTPUT«0␤0␤1␤1␤2␤2␤3␤3␤4␤4␤»
12:39 perlpilot awwaiid: you can also do:  my @a = "a".."z";  for @a.pairs -> $p { say "{$p.key}: {$p.value}" }
12:39 awwaiid m: for <a b c> -> $a { say "$a - {$++} {$++}" ; for @stuff -> $b { say "inner $b - {$++}" } } # fascinating
12:39 camelia rakudo-moar 1488b3: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Variable '@stuff' is not declared␤at <tmp>:1␤------> 3c> -> $a { say "$a - {$++} {$++}" ; for 7⏏5@stuff -> $b { say "inner $b - {$++}" } ␤»
12:39 itaipu joined #perl6
12:39 awwaiid m: my @stuff = <a b c>; for @stuff -> $a { say "$a - {$++} {$++}" ; for @stuff -> $b { say "inner $b - {$++}" } } # fascinating
12:39 camelia rakudo-moar 1488b3: OUTPUT«a - 0 0␤inner a - 0␤inner b - 1␤inner c - 2␤b - 1 1␤inner a - 3␤inner b - 4␤inner c - 5␤c - 2 2␤inner a - 6␤inner b - 7␤inner c - 8␤»
12:39 pierre_ joined #perl6
12:40 awwaiid so implicit to most local block. gotchya
12:40 vcv Having a brain fart here... is there a way to shorten this? @a = sort *<foo>, @a;
12:40 jnthn @a .= sort *<foo>
12:41 smls jnthn: Isn;t there a colon missing after 'sort'
12:41 smls or parens
12:42 kent\n joined #perl6
12:42 kent\n joined #perl6
12:42 jnthn ah, yes
12:42 jnthn @a .= sort(*<foo>)
12:43 vcv ah there it is.  i had tried without the parens. thanks!
12:44 lichtkind__ joined #perl6
12:47 wamba joined #perl6
12:51 zakharyas joined #perl6
12:56 sufrostico joined #perl6
12:58 AlexDaniel joined #perl6
13:05 skids joined #perl6
13:06 ufobat i am disapointed :( my perl6 segfaults so often :(
13:07 jnthn Have you golfed the problem and reported it?
13:07 ufobat no :-( i dont know how, its a lot of supplies and stuff
13:08 jnthn Can you at least write up instructions for how it can be reproduced and file those? :)
13:08 mcmillhj joined #perl6
13:08 jnthn That's a starting point for others to be able to golf/solve it
13:10 ufobat there are a couple of github repos involved, all with changes ... *sigh*
13:10 ufobat i am playing around with p6w
13:12 ufobat jnthn,  i will try :-)
13:12 jnthn ufobat: OK, thanks.
13:18 santhosh_k joined #perl6
13:19 ufobat i am afraid that i will add quite a useless bug report :/
13:20 santhosh_k left #perl6
13:20 jkramer What's the name of that P5 module that took a list of words and return "word1, word2, word3 and word4"?
13:20 jkramer And what's the easiest way to do that in P6? :)
13:22 jkramer Found it, Text::Conjunct :)
13:24 * moritz would have expected Lingua::EN::Enumeration or something like that
13:29 * nine wouldn't be surprised if both existed
13:30 mst and at least one more in a really bizarre namespace ... that will eventually turn out to be better
13:31 nine As long as there's no ::ButMaintained version of it...
13:31 * [Coke] tries out 'task' again based on awwaiid's suggestion
13:32 [Coke] ::Butt::Maintained
13:32 smls jkramer: This seems to work... https://gist.github.com/smls/98e28d5c23783d9e1ad86f25b6bb1e15
13:34 smls too bad a signature like  ($conj, *@words, *$word)  is not allowed in Perl 6...
13:35 unmatched} jkramer: the easiest way would be to use the Perl 6 module for it :) http://modules.perl6.org/dist/Lingua::Conjunction
13:36 moritz [Coke]: unmaintained butts can quickly lead to the proverbial pain in the butt :-)
13:37 jkramer smls: Nice, guess I'll use that then. I was playing with the Z operator to find the most awesome shortest way to do it, but can't come up with something clever that actually works :)
13:38 jkramer unmatched}: That's even better cleverer :)
13:40 ptolemarch joined #perl6
13:40 AlexDaniel m: say WHAT ‘.’.IO.absolute; say WHAT ‘.’.IO.abspath
13:40 camelia rakudo-moar 1488b3: OUTPUT«(Str)␤(Str)␤»
13:40 AlexDaniel can somebody comment on this ↑ ?
13:41 AlexDaniel docs claim that .absolute should return a Path object
13:41 AlexDaniel which is a bit confusing because .abs*path* is possibly a better candidate for that behavior
13:41 unmatched} bisect: say WHAT ‘.’.IO.abspath
13:41 bisectable unmatched}: On both starting points (good=2015.12 bad=1488b39) the exit code is 0 and the output is identical as well
13:41 bisectable unmatched}: Output on both points: (Str)
13:42 AlexDaniel but none of these is actually returning a Path, so what's the actual problem?
13:42 unmatched} I'd expect IO too, especially since it stringifies to a Str anyway
13:42 AlexDaniel is it a documentation bug? Or is it a rakudo bug? And if that's the problem with rakudo, what would be the right expected behavior?
13:43 unmatched} Unless I'm misremembering, I think there was a conversation on this topic, and the argument was that internally the first result would be a Str, so that's returned and if the user wants an IO, they'd call .IO. Returning an .IO would require turning the .Str to .IO and if the user wanted a Str, there'd be this unwanted shuffling between Str -> IO -> Str.
13:44 moritz then the docs should be updated
13:44 AlexDaniel there are some tests that do 「.absolute.IO」, which *maybe* means that indeed it should be a Str… couldn't find any tetst for .abspath :(
13:44 AlexDaniel tests*
13:44 moritz I dimly recall that the return type changed at some point
13:44 tbrowder masak: WILCO
13:45 AlexDaniel committable: releases say WHAT ‘.’.IO.abspath
13:45 committable AlexDaniel: ¦«2015.10,2015.11,2015.12,2016.02,2016.03,2016.04,2016.05,2016.06,2016.07,HEAD»: (Str)
13:45 AlexDaniel committable: releases say WHAT ‘.’.IO.absolute
13:45 committable AlexDaniel: ¦«2015.10,2015.11,2015.12,2016.02,2016.03,2016.04,2016.05,2016.06,2016.07,HEAD»: (Str)
13:48 AlexDaniel unmatched}: interestingly, the way I stumbled upon this is because I had somethisg like this in the code: ‘…’.IO.abspath.IO
13:48 AlexDaniel this .IO….IO looked weird, so I started looking for alternatives
13:51 AlexDaniel after all, it is a bit weird to have two methods that return exactly the same thing
13:51 unmatched} Yeah, that's LTA and the returned type can be argued both ways.
13:52 unmatched} I actually see I've tacked on a spurious .IO call on a return value of a .child due to this issue.
13:52 unmatched} (in one of my recent codes)
13:53 unmatched} benchable: for ^10000 { my $x = "foo".IO }
13:53 benchable unmatched}: ¦«for»:Cannot find this revision
13:53 unmatched} benchable: HEAD for ^10000 { my $x = "foo".IO }
13:53 benchable unmatched}: ¦«HEAD»:0.2336
13:54 unmatched} benchable: HEAD for ^10000 { my $x = "foo".IO.Str }
13:54 benchable unmatched}: ¦«HEAD»:0.2350
13:54 smls unmatched}: "internally the first result would be a Str" -- wouldn't that be true for all IO functions?
13:54 tbrowder mazak: issue was already there for panda, new issue filed for zef
13:54 unmatched} If there were voting, I'd vote for all such methods to return an IO::Path rather than a plain string
13:54 smls Seems like a dubious reason to avoid returning an IO in this one case.
13:54 smls yeah, what unmatched} said.
13:55 perlpilot unmatched}: I would vote the same
13:55 AlexDaniel m: ‘test’.IO.abspath.say
13:55 camelia rakudo-moar 1488b3: OUTPUT«/home/camelia/test␤»
13:55 AlexDaniel m: ‘-test’.IO.abspath.say
13:55 camelia rakudo-moar 1488b3: OUTPUT«␤»
13:56 AlexDaniel what the f
13:56 AlexDaniel m: ‘-test’.IO.absolute.say
13:56 camelia rakudo-moar 1488b3: OUTPUT«␤»
13:56 AlexDaniel so .absolute actually calls .abspath
13:57 atta joined #perl6
13:57 perlpilot AlexDaniel: huh ... there's an explicit check for $!path.starts-with('-')
13:58 smls Why does it not like filenames that starty with dashes?
13:58 AlexDaniel perhaps someone wanted to handle ‘-’ specially
13:58 AlexDaniel and wrote starts-with instead?
13:59 AlexDaniel OK rakudobugging this, but don't know what to do with .absolute vs .abspath thing
13:59 smls Does it have to do with $*PROGRAM  returning "-e".IO in case of oneliners?
14:00 * lizmat refrains from commentary
14:01 * unmatched} wonders how come
14:01 AlexDaniel you can still use ‘./-foo’ to make it work, but come on…
14:03 smls AlexDaniel: Yeah, I hope it's a bug and not a "feature", because this is the kind of thing that security exploits are made of.
14:05 perlpilot It's a "feature" that's been there since 2014 when the code read: $!path.substr(0,1) eq '-'  instead of using starts-with()
14:06 perlpilot Still not sure what that's about other than smls's guess.
14:06 AlexDaniel perlpilot: the first commit seems to be 5870906, but that's still in 2014, yeah
14:07 AlexDaniel also, that's when .absolute was changed to return Str
14:07 nine One day I'm gonna have a close look at the newio branch and figure out, what exactly we missed out on. And will probably light a candle.
14:08 smls $  mkdir test;  cd test;  touch -- -a;  perl6 -e 'dd dir[0].absolute'
14:08 smls ""
14:08 perlpilot smls: aye, that's an unfortunate side-effect
14:14 smls perlpilot: I had hoped that maybe at least dir() is aware of it and produces "./-e".IO, but no.
14:14 freezerburnv joined #perl6
14:15 smls If this isn't fixed, I forsee it'll become an attack vector against server applications etc. written in Perl 6... :(
14:19 AlexDaniel #128840
14:19 synopsebot6 Link:  https://rt.perl.org/rt3//Public/Bug/Display.html?id=128840
14:20 AlexDaniel by the way, does text/html bug reports work OK for everyone?
14:21 ufobat jnthn, https://rt.perl.org/Public/Bug/Display.html?id=128839
14:21 AlexDaniel ufobat++
14:23 hcit joined #perl6
14:23 unmatched} AlexDaniel: looks fine on my box
14:30 AlexDaniel by the way, maybe it's not too bad because ‘’.IO is not valid
14:30 ugexe .absolute will always give the absolute path while .Str will give the path that the path was originally created with. And .abspath is not documented
14:30 AlexDaniel ugexe: https://docs.perl6.org/type/IO$COLON$COLONPath#method_abspath
14:32 ugexe ah i meant speculations. see https://github.com/rakudo/rakudo/pull/666
14:35 margeas joined #perl6
14:36 rindolf joined #perl6
14:37 smls what kind of URL is "type/IO$COLON$COLONPath" o.O
14:37 AlexDaniel https://rt.perl.org/Ticket/Display.html?id=126262
14:38 AlexDaniel ok, given that there's a pull request #666 and also a bug report, not going to submit anything.
14:39 unmatched} smls: https://docs.perl6.org/type/IO::Path#method_abspath also works :)
14:39 unmatched} smls: as for the reason for the URL being so colonically-inclined, it's a side effect of making files not contain any special characters.
14:40 unmatched} (so they don't conflict on the server side, since they're all just static files on disk)
14:40 [Coke] : is technically reserved or somehow not allowed.
14:40 lizmat AlexDaniel: https://github.com/rakudo/rakudo/commit/02fdcf9a1f
14:41 unmatched} : can't be used in filenames on Windows and I think linux uses them as PATH separators too
14:41 AlexDaniel lizmat: that was quick. Thanks!
14:42 [Coke] (er, my comment was for URLs). but filenames, also, yes, can be problematic.
14:42 lizmat afk&
14:45 cdg joined #perl6
14:56 ugexe the one thing i would have liked most from newio was IO::Huh
15:08 benchable joined #perl6
15:10 BrassLantern joined #perl6
15:10 BrassLantern joined #perl6
15:15 RabidGravy joined #perl6
15:18 jkramer https://docs.perl6.org/language/operators#prefix_temp - can anyone explain what is true? > "temporizes" the variable passed as the argument, which means it is reset to its old value on scope exit. (This is similar to the local operator in Perl 5, except that temp does not reset the value).
15:18 jkramer Does it reset or not? :)
15:18 jkramer And if not, what's the point of it?
15:19 unmatched} jkramer: my $x = "foo"; { temp $x; say $x; $x = "bar"; say $x }; say $x; in P6 will print "foo bar foo", but when using local in P5 it'll print "[uninitialized variable warning] bar foo"
15:19 jnthn m: my $a = 42; { temp $a; say $a; $a++; say $a; }; say $a
15:19 camelia rakudo-moar 02fdcf: OUTPUT«42␤43␤42␤»
15:19 jnthn And then what unmatched} said :)
15:20 unmatched} It's reset *to old value* when scope is left, but unlike P5, it's NOT reset to an undefined value inside the scope
15:20 cdg joined #perl6
15:20 jkramer Ah, I get it. Thanks!
15:21 gfldex the reference to P5 doesn't belong there
15:21 jast it could probably be clarified apart from that
15:21 unmatched} jkramer: feel free to improve it by submitting a PR:
15:21 unmatched} jkramer: https://github.com/perl6/doc/blob/master/doc/Language/operators.pod6#L1112
15:21 unmatched} gfldex: agreed
15:22 unmatched} All P5 references should be confined to 5to6 guides IMO
15:22 jast actually I guess it's clear enough if you simply remove the remark about P5
15:23 jkramer jast: Agreed, that was the only confusing part about it really :)
15:23 jast minor clarification about exactly what old value it is reset to might be good (value at the time the variable is temporized, presumably)
15:23 AlexDaniel unmatched}: I agree
15:24 jast m: my $a = 42; { $a++; temp $a; $a++; say $a; }; say $a
15:24 camelia rakudo-moar 02fdcf: OUTPUT«44␤43␤»
15:24 AlexDaniel hm, how many references to other languages are there?
15:25 AlexDaniel possibly none
15:26 acrussell joined #perl6
15:31 nightfrog joined #perl6
15:39 mspo joined #perl6
15:42 Xliff .tell FROGGS all tests in 07dtd now pass. Thanks!
15:42 yoleaux Xliff: I'll pass your message to FROGGS.
15:45 NEveD joined #perl6
15:50 cdg joined #perl6
15:51 unmatched} star: say .parts.join.split("g")[*-1] // .Str given $*PERL.compiler.version
15:51 camelia star-m 2016.04: OUTPUT«20164␤»
15:52 unmatched} Oh... never mind. I was about to ask a question, but then realized where my mistake is at :)
15:52 gregf_ m: say $*PERL.compiler.version
15:52 camelia rakudo-moar 02fdcf: OUTPUT«v2016.07.1.126.g.02.fdcf.9␤»
15:53 gregf_ m: say .split("-").map(&tclc) given "foo-bar-baz"
15:53 camelia rakudo-moar 02fdcf: OUTPUT«(Foo Bar Baz)␤»
15:56 unmatched} m: say .Str ~~ /"g"/ ?? .parts.join.split("g")[*-1] !! .Str given $*PERL.compiler.version
15:56 camelia rakudo-moar 02fdcf: OUTPUT«2fdcf9␤»
15:56 unmatched} star: say .Str ~~ /"g"/ ?? .parts.join.split("g")[*-1] !! .Str given $*PERL.compiler.version
15:56 camelia star-m 2016.04: OUTPUT«2016.04␤»
15:56 unmatched} (that's what I was going for, if anyone was curious; the commit or tag for the currently running Rakudo build)
15:57 nightfrog joined #perl6
16:15 AlexDaniel is there any way to tweak the priority of a multi?
16:15 psch what does "priority" mean there?
16:16 hahainternet here's a stupid question, i'm designing a litte storage class (actually in Go but that's besides the point) and i'd like a way to key items that isn't a string, but is as close to a string as possible. ie a ruby symbol
16:16 unmatched} Move it in the source
16:16 hahainternet that way it cannot trivially be mistaken for a string
16:17 hahainternet i see on the perl6 from ruby page that colon pair syntax is a 'workaround' for this behaviour
16:17 hahainternet but is there anything nicer?
16:18 unmatched} m: multi foo ($ where * < 10) { say "meow" }; multi foo ($ where * < 100) { say "moo" }; foo 9
16:18 camelia rakudo-moar 02fdcf: OUTPUT«meow␤»
16:18 unmatched} m: multi foo ($ where * < 100) { say "moo" }; multi foo ($ where * < 10) { say "meow" }; foo 9
16:18 camelia rakudo-moar 02fdcf: OUTPUT«moo␤»
16:19 unmatched} hahainternet: "actually in Go"....
16:19 hahainternet heh that's a very nice compromise, and i assume lexical declarations will overtake globals?
16:19 * unmatched} presses the "Flamewar" button
16:19 hahainternet unmatched}: i write it for my day job
16:19 unmatched} Traitor! :P
16:19 hahainternet but just in case anyone asked for code samples
16:19 hahainternet go has no symbols, nor any succinct way of creating a non string stringy thing
16:20 hahainternet so, whenever i come across a scenario like this i turn to perl6 and usually find it solved elegantly and succinctly
16:20 unmatched} hahainternet: not sure what you're asking my multis above were an example related to AlexDaniel's question.
16:21 hahainternet unmatched}: i know, see my question above your replies :)
16:21 psch hahainternet: fwiw, the thing we use keys most for in perl6 is hashes, which coerce their keys to Str
16:21 psch hahainternet: having non-Str keys is what takes extra work
16:21 hahainternet psch: oh i didn't even realise they were coerced
16:22 unmatched} hahainternet: sure, but wouldn't the folks in #go-lang know more about how to solve it? The Perl 6 way is to parametarize the hash to not Str its keys, but I'm unsure how that can be translated into Go
16:22 psch m: sub f { }; my %h; %h{&f} = "foo"; say %h.keys
16:22 camelia rakudo-moar 02fdcf: OUTPUT«Sub object coerced to string (please use .gist or .perl to do that)  in block <unit> at <tmp> line 1␤(f)␤»
16:22 hahainternet unmatched}: oh i already know the answer in Go
16:22 psch hahainternet: see the warning :)
16:22 hahainternet psch: so is there an easy non string keyed hash declaration?
16:22 psch m: sub f { }; my %h; %h{&f} = "foo"; # just to prove its not the enumeration
16:22 camelia rakudo-moar 02fdcf: OUTPUT«Sub object coerced to string (please use .gist or .perl to do that)  in block <unit> at <tmp> line 1␤»
16:22 psch m: my %h{Any}; sub f { }; %h{&f} = "foo";
16:22 camelia rakudo-moar 02fdcf: ( no output )
16:23 psch hahainternet: FSDO "easy"
16:23 unmatched} m: my %hash{Any}; %hash{&say} = 'tis a say sub'; say %hash.keys; say %hash{&say}
16:23 camelia rakudo-moar 02fdcf: OUTPUT«(sub say (| is raw) { #`(Sub|71440392) ... })␤tis a say sub␤»
16:23 hahainternet haha no i get Any, but the question is more about what's the stringiest but non string type in perl6
16:23 psch hahainternet: the RHS-y way is :{}
16:23 psch m: say Stringy
16:23 camelia rakudo-moar 02fdcf: OUTPUT«(Stringy)␤»
16:23 hahainternet i guess that wasn't clear from my previous question sorry
16:23 psch ...probably?
16:24 hahainternet i'd imagine that matches a string though
16:24 NEveD joined #perl6
16:24 psch m: say 5 ~~ Stringy
16:24 camelia rakudo-moar 02fdcf: OUTPUT«False␤»
16:24 psch m: say "5" ~~ Stringy
16:24 camelia rakudo-moar 02fdcf: OUTPUT«True␤»
16:24 psch well, Str are Stringy..?
16:24 hahainternet i'm looking for something that is not
16:24 psch m: say Stringy.^mro
16:24 camelia rakudo-moar 02fdcf: OUTPUT«Method 'mro' not found for invocant of class 'Perl6::Metamodel::ParametricRoleGroupHOW'␤  in block <unit> at <tmp> line 1␤␤»
16:24 hahainternet a symbol and a string in ruby are distinct
16:24 hahainternet i mean obviously this is doable a billion different ways with a few characters
16:25 hahainternet but that's no fun
16:25 psch afaik symbols in perl6 are Str keys that point to (some type of) container in a Stash
16:25 psch m: say CORE::.WHAT
16:25 camelia rakudo-moar 02fdcf: OUTPUT«(PseudoStash)␤»
16:25 psch m: say CORE::<&say>.key
16:25 camelia rakudo-moar 02fdcf: OUTPUT«Method 'key' not found for invocant of class 'Sub'␤  in block <unit> at <tmp> line 1␤␤»
16:25 psch m: say CORE::.keys
16:25 camelia rakudo-moar 02fdcf: OUTPUT«(SIGSEGV WhateverCode Slip &callframe Pair SIGHUP &flat &RETURN-LIST utf8 &infix:<∖> CurrentThreadScheduler PromiseStatus StringyEnumeration Distribution &splice &postcircumfix:<{ }> &callsame &GATHER Backtrace PF_INET6 &sinh &infix:<lt> &print &asech NF…»
16:26 psch m: say CORE::.keys[0].WHAT
16:26 camelia rakudo-moar 02fdcf: OUTPUT«(Str)␤»
16:26 MilkmanDan joined #perl6
16:26 zacts joined #perl6
16:26 psch so what i guess the difference between symbols and strings is in ruby is solved by parser state in perl6
16:27 hahainternet in pseudocode: my %h{Symbol-like-type}; %hash{:symbol-syntax}; %hash<whatever> # <-- compile time error (ideally)
16:27 psch well, plus some workarounds, like ."foo"()
16:28 hahainternet also wow i went from %h to %hash, i'm tired
16:28 psch maybe enums fit that symbol-like-type..?
16:29 hahainternet yeah an enum would be fine, just requires declaration ahead of time i think
16:29 hahainternet enums-on-demand i guess
16:29 psch you can also type a hash to a specific type and use subtypes as key i suppose
16:29 hahainternet or the pair hack on the ruby to perl6 wiki page, but that's just a hack
16:29 psch although that's probably a slightly different direction that the original intent
16:30 hahainternet psch: oh sure, but my question is really all about succinctness
16:30 hahainternet how to do it in the fewest characters
16:30 hahainternet strikes me as a particularly nice feature in ruby, i initially hated it
16:30 hahainternet but now see its appeal
16:30 hahainternet course i bet ruby doesn't catch it at compile time but i may be wrong
16:33 psch m: my %h{Pair}; %h{:foo} = 1; say %h.perl; %h<foo> # well, the syntax works :P
16:33 camelia rakudo-moar 02fdcf: OUTPUT«(my Any %{Pair} = (:foo) => 1)␤Type check failed in binding to key; expected Pair but got Str ("foo")␤  in block <unit> at <tmp> line 1␤␤»
16:33 psch it's not compile time though, but that seems to be a thing we general don't do
16:34 harmil_wk oh hey, should I be rakudobugging that file/line issue on Int.^methods from yesterday? I can't recall if that was something folks wanted.
16:35 unmatched} harmil_wk: I'd say no. I'm currently writing a blog post on the topic and will release a module that makes stuff like that easier
16:35 unmatched} Probably will be done today.
16:35 unmatched} harmil_wk: it'd be:   .sourcery.say for Int.^methods;
16:36 unmatched} And it'll give you correct file/location of actual Rakudo source files along with a link to GitHub
16:36 AlexDaniel m: class Foo { multi method foo ($ where * < 10) { say "meow" } }; class Bar is Foo { multi method foo ($ where * < 100) { say "moo" } }; Bar.new.foo(5)
16:36 camelia rakudo-moar 02fdcf: OUTPUT«moo␤»
16:37 AlexDaniel unmatched}: how can I move it here?
16:37 harmil_wk Hmm... is post-processing really the right approach? Then, to find out where my stuff is defined I look at MyClass.^methods, but for core stuff (well, some core stuff) I need a module?
16:38 unmatched} harmil_wk: not exactly. For core stuff you *do* get the correct location where the stuff is defined, it's just not too useful if you want to *develop* rakudo and I see no issues with using a module as a helper feature for core development rather than returning false information for core subs
16:38 unmatched} AlexDaniel: no idea. There's actually a ticket from Ovid on the topic.
16:38 * unmatched} tries to find it
16:38 unmatched} buggable: rt
16:39 buggable unmatched}, TOTAL: 1364, UNTAGGED: 615, BUG: 417, LTA: 94, JVM: 62, NYI: 32, SEGV: 28, RFC: 26, UNI: 25, PERF: 19, POD: 14, CONC: 11, @LARRY: 10, TODO: 9, PRECOMP: 8, GLR: 6, BUILD: 5, STAR: 4, WEIRD: 3, MOARVM: 2, OSX: 2, CPP: 1, SPEC: 1, LIST: 1, LHF: 1, SPESH: 1, DOCS: 1, NATIVECALL: 1   Details: http://bug.perl6.party/1470328748.html
16:39 AlexDaniel unmatched}: by the way, this question is related to IRC::Client. I can't get “help” command to work in bisectable because bisectable treats it like a source to execute
16:40 AlexDaniel unmatched}: of course I can change the regex to ignore [help|source|url]‘?’? but that's a bit LTA
16:41 unmatched} AlexDaniel: there, placing it first in the source should work.
16:41 harmil_wk unmatched}: Well, I don't know what your module does, so I can't really speak in an informed way. It just feels like when as user wants to know, "hey, how exactly is this defined," it's logical to expect that they get that answer the same way for Int as for MyClass.
16:41 unmatched} In buggable I did a whole new plugin for such stuff
16:41 AlexDaniel unmatched}: well, not really, because “help” is in the base class, and bisectable itself is a subclass
16:42 unmatched} harmil_wk: but that's my point, they *do* get that answer right now. The `.line` and `.number` aren't bogus for Int.^methods, they point to the actual, existing, and correct location in the setting file.
16:42 AlexDaniel maybe that's not how it should be, I'm not sure now. But it worked great until I stumbled upon this
16:43 unmatched} That's really a Perl 6 limitation and I think an area that should be more defined and maybe more featureful
16:43 AlexDaniel hm
16:43 unmatched} (I can't find the ticket on the ticket page, but searching for tickets from Ovid will give it)
16:44 unmatched} m: "{&say.file}:{&say.line}".say;
16:44 camelia rakudo-moar 02fdcf: OUTPUT«gen/moar/m-CORE.setting:29424␤»
16:44 cooper joined #perl6
16:45 unmatched} harmil_wk: ^ so here, if you open m-CORE.setting file on line 29424 you'll find the definition for sub `say`, but if we instead return src/core/io_operators.pm:17, that location does not refer to any actual files or line numbers on user's disc. It refers to the location in Rakudo's source and the user would have to obtain that source *at the correct commit* to view the definition. I can't see the rationale for
16:45 unmatched} doing so.
16:47 unmatched} AlexDaniel: here: https://rt.perl.org/Ticket/Display.html?id=123577
16:48 unmatched} Though I don't think what he argues for is possible. We can't know which where {} is more specific in cases like * < 100 vs * < 10 given 9
16:48 unmatched} So the simplest solution is like now, to go by source position. This creates an edge case where we don't know what the source position is, so...
16:49 harmil_wk unmatched}: isn't that a bit like saying, "it points to the correct post-C-preprocessor pile of unreadable mash that your program got turned into" when working in C? I'll grant that m-CORE.setting is much more readable than most post-C-preprocessor droppings, but still...
16:49 * unmatched} refrains from proposing any possibly-stupid ideas here
16:49 ilmari harmil_wk: that's why the c preprocessor inserts #line directives, so that error messages refer to the input files
16:49 AlexDaniel unmatched}: thanks
16:49 pierre_ joined #perl6
16:50 jnthn Maybe for CORE.setting we simply should not show line numbers unless some I'm-a-Rakudo-developer environment flag is set. Then people can't be confused by them. :P
16:50 unmatched} harmil_wk: I don't care about C. m-CORE.setting is nothing more than @all-source-files.join("#line 1 $_.filename"), so the readability argument is a poor one.
16:51 unmatched} I rather have the location in a 40,000-line file than just an essentially random filename I'd have trouble locating.
16:51 unmatched} And that argument assumes there isn't a CoreHackers::Sourcery that can give you all the nice details, which there will be tonight :)
16:51 harmil_wk In fact, settings already has directives like "#line 1 src/core/core_prologue.pm" so really the work is done on the generation side, we just need $?FILE and $?LINE to play ball.
16:53 AlexDaniel yeah, I also don't think that it is possible. But 123577++ anyway because it raises some very good questions
16:53 unmatched} harmil_wk: play ball how? What will you be doing exactly with those $?FILE and $?LINE?
16:53 wamba joined #perl6
16:53 mcmillhj joined #perl6
16:53 harmil_wk unmatched}: let me see if I can dredge up the perl6-language list discussion about this. I think it will be illustrative. We've been around this particular mulberry bush a few times in the past.
16:55 unmatched} harmil_wk: those $?FILE and $?LINE don't refer to any specific entitiy. You need the commit number or the tag for the specific Rakudo build that generated those $?FILE and $?LINE and even when you do have that commit number, you have to obtain the source code at that specific commit, for the information contained in those variables to be of any use.
16:55 unmatched} By comparison, the current .line and .file do refer to the actual source code
16:56 [Coke] unmatched}: where is the code for buggable?
16:56 unmatched} buggable: source
16:56 buggable unmatched}, See: https://github.com/zoffixznet/perl6-buggable
16:56 harmil_wk unmatched}: I'm not sure why you're talking about source control, here.  I thought we were talking about files, not version control metadata. My copy of rakudo might be in CVS source control...
16:56 * harmil_wk rushes to hide his CVS checked out copy of rakudo under some books...
16:56 unmatched} :)
16:57 unmatched} harmil_wk: OK. Then what file would $?FILE refer to?
16:57 unmatched} For example, for &.say
16:58 unmatched} My module gives me src/core/io_operators.pm, line 20, at commit c843682 and if I combine those three bits of info, I get the actual entity and indeed that's the definition of sub say() for the current compiler I'm running on: https://github.com/rakudo/rakudo/blob/c843682/src/core/io_operators.pm#L20
16:59 Actualeyes left #perl6
16:59 unmatched} If I run {say "{.file}:{.line}" given &say}, it gives me gen/moar/m-CORE.setting:28036 and indeed, if I open that file, I get the location of where the sub say() is defined.
16:59 harmil_wk I'm having trouble finding the conversation. I have a vague memory it was in the mid 2000s, but I think the idea was that $?FILE and $?LINE should be settable lexically so that you can describe where the Perl 6 you're currently parsing was generated from. Then each subsequent line should increase $?LINE. This is almost exactly what the CPP does, but it uses comments.
17:00 harmil_wk As far as things like commits... I think that's a great place for a module to come into play
17:00 unmatched} harmil_wk: sure, that's probably doable, but I my point is that were it done, the information contained in $?FILE $?LINE would be somewhat useless.
17:00 cooper joined #perl6
17:00 cooper joined #perl6
17:01 harmil_wk unmatched}: If it seemed like I was saying that your module wasn't worth writing, I'm absolutely not. I think it's great. You want more metadata, and that's wonderful.
17:01 unmatched} And that's why I'm talking about commits. You need to know which commit the setting was compiled from to make sense of what such $?FILE and $?LINE are talking about
17:01 unmatched} harmil_wk: no, that's not how I interpreted it.
17:03 unmatched} As I've said, even without the module, I feel that changing Rakudo to give original source locations rather than setting location for core subs would provide less information than we currently do.
17:04 unmatched} And would introduce the inconsistency for .file and .line's behaviour
17:04 * unmatched} leaves to do $work
17:05 hankache joined #perl6
17:05 hankache hola #perl6
17:06 AlexDaniel o/
17:10 PapaChub joined #perl6
17:13 spider-mario joined #perl6
17:23 mohae joined #perl6
17:27 dsp-_ joined #perl6
17:33 setty1 joined #perl6
17:41 Fleurety joined #perl6
17:43 dalek doc: a92017a | nsvedberg++ | doc/Language/nativecall.pod6:
17:43 dalek doc: Change 'mayor' to 'major' in my_version pointers example
17:43 dalek doc: review: https://github.com/perl6/doc/commit/a92017a597
17:43 dalek doc: 6d12e31 | (Aleks-Daniel Jakimenko-Aleksejev)++ | doc/Language/nativecall.pod6:
17:43 dalek doc: Merge pull request #795 from nsvedberg/nativecall-typo-fix
17:43 dalek doc:
17:43 dalek doc: Change 'mayor' to 'major' in my_version pointers example
17:43 dalek doc: review: https://github.com/perl6/doc/commit/6d12e31158
17:44 unmatched} harmil_wk: one thing I'm thinking is for .file/.line on core stuff to return an object that stringifies to settings.file/line, but has properties that refer to the original file/line (including the commit). That's just thoughts on the interface; I've no idea what sort of implementation/performance details we're looking at
17:44 unmatched} m: &say.file.^name.say
17:44 camelia rakudo-moar 02fdcf: OUTPUT«Str␤»
17:44 unmatched} Another contender for returning IO::Path rather than plain Str ^
17:44 unmatched} m: use Test; &is.file.say
17:44 camelia rakudo-moar 02fdcf: OUTPUT«perl#sources/C712FE6969F786C9380D643DF17E85D06868219E (Test)␤»
17:45 unmatched} hm.... or maybe not
17:46 cdg joined #perl6
17:46 labster joined #perl6
17:47 unmatched} (keeping in mind the location in original sources is different from offset from `#line 1 src/core/foo.pm6` line, due to backend-specific clauses and stripped comments)
17:47 [Coke] don't we have the rakudo tag in the compiler version?
17:50 unmatched} Yeah
17:50 unmatched} m: my $where = .Str ~~ /g/ ?? .parts.join.split("g")[*-1] !! .Str given $*PERL.compiler.version;
17:50 camelia rakudo-moar 02fdcf: ( no output )
17:50 unmatched} m: say .Str ~~ /g/ ?? .parts.join.split("g")[*-1] !! .Str given $*PERL.compiler.version;
17:50 camelia rakudo-moar 02fdcf: OUTPUT«2fdcf9␤»
17:50 unmatched} star: say .Str ~~ /g/ ?? .parts.join.split("g")[*-1] !! .Str given $*PERL.compiler.version;
17:50 camelia star-m 2016.04: OUTPUT«2016.04␤»
18:00 firstdayonthejob joined #perl6
18:14 lostinfog joined #perl6
18:16 khw joined #perl6
18:16 webstrand joined #perl6
18:28 FROGGS joined #perl6
18:29 harmil_wk unmatched}: sounds reasonable
18:34 AndyBotwin joined #perl6
18:34 acrussell joined #perl6
18:35 grondilu m: say $*PROGRAM
18:35 camelia rakudo-moar 02fdcf: OUTPUT«"<tmp>".IO␤»
18:38 avenj joined #perl6
18:43 moritz note that camelia fudges the name of tempfiles in output
18:52 shlomif joined #perl6
18:52 sno joined #perl6
19:09 sno joined #perl6
19:11 dalek doc: 82a3be5 | (Jan-Olof Hendig)++ | doc/Type/Setty.pod6:
19:11 dalek doc: Added docs for Setty.Bag
19:11 dalek doc: review: https://github.com/perl6/doc/commit/82a3be5d10
19:18 sjoshi joined #perl6
19:23 stv joined #perl6
19:25 stv left #perl6
19:32 * unmatched} downloads the sound of a pin dropping
19:37 dalek doc: f50b20a | (Jan-Olof Hendig)++ | doc/Type/Setty.pod6:
19:37 dalek doc: Added docs for Setty.BagHash
19:37 dalek doc: review: https://github.com/perl6/doc/commit/f50b20a902
19:49 dalek doc: 40d8a1e | coke++ | doc/Language/about.pod6:
19:49 dalek doc: use nbsp
19:49 dalek doc: review: https://github.com/perl6/doc/commit/40d8a1e78c
20:18 AngeloMichael joined #perl6
20:20 zakharyas joined #perl6
20:24 skids joined #perl6
20:27 xfix joined #perl6
20:29 rindolf joined #perl6
20:29 khw joined #perl6
20:39 espadrine joined #perl6
20:48 Zoffix left #perl6
20:57 rindolf joined #perl6
20:57 Khisanth joined #perl6
21:02 mcmillhj joined #perl6
21:04 _28_ria joined #perl6
21:07 vittooz joined #perl6
21:08 RabidGravy ha, DARPA project called CAMELIA http://www.defenseone.com/technology/2016/08/if-theres-malware-your-toaster-us-military-wants-find-it/130488/
21:08 hankache ha!
21:12 rindolf joined #perl6
21:13 sufrostico joined #perl6
21:17 Xliff m: $*PERL.compiler.version.say
21:17 camelia rakudo-moar 02fdcf: OUTPUT«v2016.07.1.126.g.02.fdcf.9␤»
21:19 Xliff star: $*PERL.compiler.version.say
21:19 camelia star-m 2016.04: OUTPUT«v2016.04␤»
21:21 vittooz joined #perl6
21:25 margeas joined #perl6
21:40 itaipu joined #perl6
21:44 margeas joined #perl6
21:46 tbrowder left #perl6
21:47 NEveD joined #perl6
21:49 margeas joined #perl6
21:58 PapaChub joined #perl6
22:01 Sgeo__ joined #perl6
22:10 cdg joined #perl6
22:19 TEttinger joined #perl6
22:27 Zoffix joined #perl6
22:30 freezerburnv joined #perl6
22:31 skids joined #perl6
22:52 labster joined #perl6
23:13 kurahaupo joined #perl6
23:23 sufrostico joined #perl6
23:36 pierre_ joined #perl6
23:47 freezerburnv joined #perl6
23:50 BenGoldberg joined #perl6
23:57 pierre_ joined #perl6

| Channels | #perl6 index | Today | | Search | Google Search | Plain-Text | summary

Perl 6 | Reference Documentation | Rakudo