»ö« Welcome to Perl 6! | perl6.org/ | evalbot usage: 'p6: say 3;' or rakudo:, or /msg camelia p6: ... | irclog: irc.perl6.org or colabti.org/irclogger/irclogger_logs/perl6 | UTF-8 is our friend!
Set by moritz on 22 December 2015.
00:00 cdg joined 00:02 SHODAN joined 00:14 cdg left, troys is now known as troys_
AlexDaniel 💤 00:18
00:21 agwind joined, AndChat|688961 left 00:23 bwisti left 00:27 Rawriful joined
Geth doc: b42c58c029 | (Zoffix Znet)++ (committed using GitHub Web editor) | doc/Type/IO/Handle.pod6
Make IO::Handle.eof docs more precise

with respect to what "exhausted handle" means and the behaviour of .eof when .seeking back into content.
Spec: github.com/perl6/roast/commit/d478095cb0		 00:28
synopsebot Link: doc.perl6.org/type/IO/Handle
doc: b62849f10f | (Zoffix Znet)++ (committed using GitHub Web editor) | doc/Type/IO/Handle.pod6
Fix typo
00:32 synopsebot left, synopsebot joined, ChanServ sets mode: +v synopsebot 00:33 pilne joined 00:38 Actualeyes joined 00:40 espadrine joined 00:43 synopsebot left 00:44 synopsebot joined, ChanServ sets mode: +v synopsebot 00:46 lookatme joined
lookatme o/ 00:47
tyil \o
00:54 mikowitz left 00:55 Guest19 joined 01:03 cdg joined
pilne i've never "missed" a language as much as I miss perl6 when i choose to, or have to look elsewhere 01:04
01:06 david_xd left 01:07 cdg left 01:09 pdcawley joined 01:15 pdcawley left 01:20 pdcawley joined 01:30 Zoffix joined
Zoffix pilne: it's like we say: "Perl 6 is the last language you'll ever learn" 01:30
01:31 kerframil joined
TEttinger followed by a picture of tonya harding, signed "or else" 01:32
pilne LOL 01:33
i remember watching that coverage on a CRT TV :D
01:38 agwind left 01:41 g41j1n joined 01:50 wamba left 01:53 Praise left 02:03 noganex_ joined 02:04 piojo1 joined, olinkl_ joined, Grauwolf_ joined 02:05 parisba_ joined, jnap__ joined, BuildTheRobots_ joined 02:06 hoelzro joined 02:07 noganex left, piojo left, epony left, xi- left, Grauwolf left, afresh1 left, Zoffix left, BuildTheRobots left, parisba left, nekomune left, olinkl left, hoelzro_ left, jnap_ left, parisba_ is now known as parisba, jnap__ is now known as jnap_, afresh1 joined, olinkl_ is now known as olinkl, BuildTheRobots_ is now known as BuildTheRobots 02:08 epony joined, Zoffix joined, xi- joined, nekomune joined 02:09 statisfiable6 left, bisectable6 left, quotable6 left 02:10 releasable6 left, coverable6 left, benchable6 left, squashable6 left 02:11 greppable6 left, greppable6 joined, jast left 02:17 agwind joined 02:18 jast joined
riceandbeans how stark of a difference is there going from 5 to 6? 02:23
Hotkeys riceandbeans: here's a quick rundown docs.perl6.org/language/5to6-nutshell 02:29
b2gills riceandbeans: I have a small test file that tries to do a bunch of things in such a way that it works for both. In that file the workarounds far outweigh the code
Hotkeys I probably should've read context 02:30
riceandbeans How about performance of 6 vs 5? 02:31
ugh the whitespace thing is going to take getting used to 02:33
b2gills There are some examples where Perl 6 can beat Perl 5. Writing Perl 6 is nicer. 02:34
Hotkeys There's a reason "Perl 6 is the last language you'll ever learn" :P 02:35
It may not beat other languages performance-wise (yet) but it's ever improving and has come a long way since christmas 2015
and writing p6 is very enjoyable
riceandbeans well, the only one it needs to compete against is perl 5 02:37
that's all I'm comparing it to
02:37 g41j1n left
Geth doc: c57a6e0033 | (Brad Gilbert)++ (committed using GitHub Web editor) | doc/Type/atomicint.pod6
Remove extra - in atomic-[inc|dec]-fetch
 02:43
synopsebot Link: doc.perl6.org/type/atomicint
02:44 cdg joined 02:45 ilbot3 left 02:48 cdg left 02:53 Rawriful left 02:56 ilbot3 joined
lookatme :) 03:00
03:04 agwind left 03:12 agwind joined 03:20 eliasr left, agwind left 03:25 flatwhite joined, kerframil left, flatwhite is now known as kerframil 03:27 Mikasi joined 03:34 pilne left
Zoffix riceandbeans: very stark, if you're aiming at native Perl 6, rather than just speaking with a heavy Perl 5 accent. They're different languages. Performance-wise: rakudo (perl6 compiler) is currently slower in many areas, but in the past year+ it caught quite a lot. I'd say if viewed as a landscape, rakudo's is a very uneven one--writing something differently can sometimes give 10x boost--while perl5's is 03:35
smooth. Specific area where perl5 is a lot faster is regexes: in rakudo they both seen little optimization so far and are more elaborate than perl5's (you get proper Match objects for all Captures for example). BUT, whenever I write Perl 5, I often have to write stuff while mubling "this is so much easier to write in perl 6" :) *Programmer's time* is a lot more optimized in Perl6 imo
riceandbeans: it's like the brochure says (page 11): "Consice" is the word :) github.com/perl6/marketing/blob/ma...ochure.pdf 03:36
... if spelt right :P
My personal pain-point currently is lack of polished, top-notch, perl6 modules (Mojo::Pg, Mojolicious), not perf 03:38
stressing on *polished* 03:40
\o
03:41 Zoffix left 04:00 astj left 04:01 astj joined 04:04 astj left, astj joined 04:06 mahafyi joined, Cabanossi left 04:07 Cabanossi joined 04:10 astj left 04:11 astj joined 04:15 mahafyi left
riceandbeans :( 04:23
I do perl5 every night at home working with mojo::pg and a full mojo app 04:24
how is moarvm vs the jvm performance?
I mean, after the five years of waiting for a jvm to start
random, I saw a funny youtube of larry wall and he referenced companies really close to me, so I got curious and I was like....where does he work? I didn't figure it out in my short search before I lost interest, BUT his personal page mentions his church, which I can almost walk to 04:26
04:30 troys_ is now known as troys 04:50 cpup left 04:56 cpup joined 05:12 perigrin_ joined, kshannon_ joined 05:13 perigrin left, kshannon left 05:14 cdg joined 05:17 poohman joined 05:18 poohman left 05:19 poohman joined, poohman left, cdg left 05:33 aindilis left, traxex left 05:38 traxex joined 05:54 espadrine left 05:56 khw left 06:02 cpup left 06:10 SCHAPiE left 06:19 cpup joined 06:20 shadowpaste0 joined, [Coke]_ joined, [Coke]_ left, [Coke]_ joined 06:21 BooK_ joined 06:22 ccntrq- joined 06:26 hoelzro left, robinsmidsrod left, shadowpaste left, buggable left, simcop2387 left, [Coke] left, leedo left, ZofBot left, plicease left, kent\n left, peteretep left, BooK left, jcallen left, gdonald left, NEveD left, ccntrq left, shadowpaste0 is now known as shadowpaste 06:27 troys is now known as troys_ 06:28 greppable6 left, greppable6 joined 06:29 darutoko joined 06:31 Guest19 left, hoelzro joined, robinsmidsrod joined, buggable joined, simcop2387 joined, leedo joined, ZofBot joined, plicease joined, kent\n joined, peteretep joined, jcallen joined, gdonald joined, NEveD joined, leguin.freenode.net sets mode: +vv buggable ZofBot 06:35 greppable6 left, greppable6 joined
piojo1 How do you indent and brace multi-line method parameter lists? (things can get really long with types, where, default values, and argument unpacking) 06:48
06:48 hoelzro left, robinsmidsrod left, buggable left, simcop2387 left, leedo left, ZofBot left, plicease left, kent\n left, peteretep left, jcallen left, gdonald left, NEveD left, SCHAPiE joined
piojo1 is there any commonly preferred style? 06:48
06:50 domidumont joined, yoleaux left 06:51 preaction left
piojo1 (I mean signatures, not invocations) 06:51
06:53 preaction joined 06:54 domidumont left 07:01 Mikasi left 07:06 konsolebox left, lancew left 07:08 lancew joined, konsolebox joined 07:09 domidumont joined 07:13 domidumont left 07:14 domidumont joined 07:15 konsolebox left 07:16 troys_ is now known as troys, konsolebox joined
moritz piojo1: perlpunks.de/paste/show/5a3b607e.727d.9a 07:19
07:20 pdcawley left
piojo1 moritz: thanks. I think I'll follow your lead and break the old perl tradition of same-line praces 07:21
07:23 troys left 07:24 NEveD joined, hoelzro joined, peteretep joined, gdonald joined, robinsmidsrod joined, leedo joined, jcallen joined 07:25 buggable joined, ZofBot joined, simcop2387 joined, plicease joined
piojo1 For consistency, do you use classical next-line braces when the argument list isn't long? pastebin.com/c8LjkPPT 07:26
07:26 kent\n joined
moritz piojo1: no, I use cuddled braces there 07:27
piojo1 moritz: thanks 07:28
I see your longer style is a hybrid of cuddled and next-line braces :)
07:29 cognominal joined
moritz piojo1: not really; in the multi-line example, the { is still cuddled with the ) on the same line 07:29
07:32 konsolebox left 07:34 lancew left
piojo1 moritz: yes, but I see the line-break as being part of the next-line style. Though I suspect your perspective is coming from a different place (list formatting) than mine (function formatting) 07:34
07:34 wamba joined, konsolebox joined 07:35 lowbro joined, lowbro left, lowbro joined 07:36 cognominal left 07:38 domidumont left 07:39 domidumont joined
Geth Pod-To-HTML: fluca1978++ created pull request #29:
Use of a debug environment variable.		 07:48
07:59 traxex left, abraxxa joined 08:03 aborazmeh joined, aborazmeh left, aborazmeh joined 08:04 traxex joined 08:15 aborazmeh left 08:20 Grauwolf_ left, aborazmeh joined, aborazmeh left, aborazmeh joined, Grauwolf joined 08:30 wamba left 08:33 kerframil left 08:36 dakkar joined 08:41 aborazmeh left 08:43 konsolebox left 08:44 bioexpress joined 08:45 konsolebox joined, pmurias joined 08:48 eliasr joined 08:49 yoleaux joined, ChanServ sets mode: +v yoleaux 08:51 scimon joined
scimon Morning all. 08:51
bioexpress Hello, how could I rewrite this code gist.github.com/anonymous/1eec3377...ef847b05d9 to get a better error message? 08:56
scimon One sec. 09:00
moritz bioexpress: subset PositiveInt of Int where * > 0 09:02
09:02 parv joined
moritz bioexpress: and then use PositiveInt as a type constraint 09:02
I think you can also do things like
m: subset ABC of Int where { $_ > 0 or fail "must be positive" }; my ABC $x = -3;
evalable6 (exit code 1) must be positive
in block <unit> at /tmp/0TWYzjcHRr line 1
scimon Both of those are great answers ++moritz 09:04
09:05 konsolebox left 09:09 rindolf joined 09:12 konsolebox joined 09:15 cdg joined 09:19 cdg left 09:24 wamba joined 09:45 konsolebox left 09:46 konsolebox joined
Ulti scimon++ for todays adventing 09:49
scimon Cheers 09:50
09:55 vichib left 09:56 Actualeyes left 09:57 Actualeyes joined 10:00 traxex left, traxex joined 10:02 traxex left 10:06 traxex joined, jeromelanteri joined 10:09 sena_kun joined
bioexpress moritz: thx! 10:14
10:21 parv left
Geth perl6.org: packtpartner++ created pull request #89:
Update index.html		 10:22
10:25 xinming_ joined
Geth perl6.org: 43387f37c7 | Packt++ (committed by Moritz Lenz) | source/resources/index.html
Update index.html
 10:27
10:29 xinming left 10:51 cdg joined 10:55 bioexpress left, cdg left 10:57 abraxxa left 10:58 abraxxa joined 11:16 kerframil joined 11:48 kaiwkx joined, kaiwkx left 11:50 void1 joined, void1 left, void1 joined, kaiwkx joined 11:53 kaiwkx left 11:56 eliasr left 12:19 strangemk2 joined 12:21 strangemk2 left 12:22 void1 left 12:26 void1 joined, void1 left, void1 joined 12:28 geospeck joined 12:33 xcm left 12:34 xcm joined 12:35 mahafyi joined 12:39 lowbro left 12:40 void1 left
AlexDaniel “looks like Camelia lost its $*IN” hmmm 12:40
evalable6: say slurp
evalable6 ♥🦋 ꒛㎲₊⼦🂴⧿⌟ⓜ≹℻ 😦⦀🌵 🖰㌲⎢➸ 🐍💔 🗭…
AlexDaniel, Full output: gist.github.com/e3280956c818adc6e3...6481481775
mahafyi what makes a module 'polished'? with reference to what Zoffix mentioned earlier today? I asked the ther day about what makes a line of cod elegant? is it being as concise as possible, as performant as possible, should it be OO programming?
12:40 Praise joined, Praise left, Praise joined 12:41 kerframil left
mahafyi since i am working towards making a module , though its going to be spectacularly slow i suspect 12:41
AlexDaniel mahafyi: for me it means well-thought-out. Also it should be properly maintaned. Concise, fast, OO — these really don't matter IMO. 12:43
12:44 reportable6 left, unicodable6 left, greppable6 left, reportable6 joined, squashable6 joined, unicodable6 joined, ChanServ sets mode: +v unicodable6, greppable6 joined, benchable6 joined, ChanServ sets mode: +v benchable6
Hotkeys I'd say conciseness is good as long as you aren't hindering readability 12:44
scimon I'd prefer a documented and tested module that's maybe not as fast as it could be over a lightning fast one I have to trawl through the code of in order to work out how to use it. 12:52
mahafyi well then i will skip attempt to use OO for now at least... i think i can get the functional programming part of it 12:56
12:56 vichib joined 13:00 cpage_ joined
scimon One of the strengths of the language is you can mix and match. 13:04
Including functional with types :)
13:11 Zoffix joined
Zoffix mahafyi: by "polished" I meant well-designed, well-tested, well-documented, and well-battle-tested. Any time I reach for Mojo::Pg, I can easily look up in the docs what I need and solve my task at hand with a single method call chain of objects that fit one another; all without any bugs or issues. By constrast, with Perl 6's DBIish, I have to read the source code to find most of the methods, which are 13:15
inconsistently named, awkward to use, and last time I used that module, it ate up all the RAM on my server.
jast anything that is inspired by DBI would be awkward to use 13:16
anyway, "polished" is a woolly term to use, but for libraries what I care about most is that they're easy to use, common cases don't need too much useless boilerplate (ideally none at all), but special cases are still possible without too much monkeying around 13:17
13:18 cpage_ left
mahafyi Zoffix : :) i will try my hand , though no deadline in sight, all i know is i will work daily at it. 13:18
jast one way to make a not-so-polished library is to design it up front to be able to do "absolutely anything" but to achieve anything, the library's user has to set up a few dozen context objects and strategies and adapters and callback chains 13:19
Zoffix Totally.
13:22 Zoffix left
scimon Good points 13:30
13:37 cpage_ joined 13:41 jeromelanteri left
mahafyi and we should try to reuse modules wherever possible? 13:44
13:45 cpage left 13:48 cpage joined
scimon Generally reinventing the wheel is a bad plan. 13:48
(So if a module exists to do something it makes sense to try and use it)
jast it's not always easy to decide 13:49
generally if a module covers the thing you need fairly well, it tends to be a good idea to use it
13:49 lizmat left
jast but it depends on how well and according to which criteria the module is being maintained, too 13:49
for example, if the maintainers of a module like to move fast and break things, you'll quickly end up being stuck with an old version forever, or you'll end up maintaining your own fork 13:50
Geth doc: 77bbde9d9e | (Luca Ferrari)++ | doc/Language/terms.pod6
Fix 'too early' ambigous statement in constant term explaination.

As proposed in Pull Request <github.com/perl6/doc/pull/1686> and related discussion, it is probably better to rephrase the 'too early' statement. Since the original pull request author gfldex did not comment out the review for almost three weeks, I'm committing this with the only aim to help readers better understanding the paragraph.
Thanks to gfldex for his effort.		 14:04
synopsebot Link: doc.perl6.org/language/terms
doc: ca2095be8b | (Luca Ferrari)++ | doc/Language/terms.pod6
Add link to traps sections for compile time constants.
14:10 domidumont1 joined 14:13 domidumont left 14:15 cdg joined
Geth doc: dc141b07e8 | (Luca Ferrari)++ | doc/Language/traps.pod6
Minor rewording of the constant trap section.

Use of a dash coherently between 'pre'-'compilation'. Add 'Computed' to the title. Minor rephrasing of the final paragraph.		 14:27
synopsebot Link: doc.perl6.org/language/traps
14:29 lizmat joined
scimon jast: one would hope if the authors are doing that they'd also be open to outside help. :) 14:43
14:46 david_xd joined 14:47 aindilis joined
Geth Pod-To-HTML: bb6bd8dfbd | (Luca Ferrari)++ (committed by Zoffix Znet) | lib/Pod/To/HTML.pm
Use of a debug environment variable. (#29)

  * Use of a debug environment variable.
Instead of defining a couple of methods with the same name 'Debug' to enable or disable the debugging output, borrow the same P6DOC_DEBUG environment variable. ... (17 more lines)		 14:50
14:51 lizmat_ joined
scimon So... I just tried something silly (defining a subset of a Sub that checks the arity) my attempt didn't work. But now I wonder if it's doable somehow. 14:51
Geth Pod-To-HTML: d91cc41dee | (Zoffix Znet)++ (committed using GitHub Web editor) | README.md
Document P6DOC_DEBUG env var
Pod-To-HTML: 916bd50f5e | (Zoffix Znet)++ (committed using GitHub Web editor) | README.md
Fix tpo
 14:52
Pod-To-HTML: 98cc35eb05 | (Zoffix Znet)++ (committed using GitHub Web editor) | META6.json
Bump version
scimon m: subset TwoArgs of Sub where *.arity == 2;sub caller ( TwoArgs &f, $a, $b ) { say &f($a,$b) };sub callee( $a, $b ) { $a + $b };caller( &callee, 1, 2 );
evalable6 (exit code 1) Type check failed in binding to parameter '&f'; expected Callable[Sub] but got Sub (sub callee ($a, $b) {...)
in sub caller at /tmp/MOERkS7uNS line 1
in block <unit> at /tmp/MOERkS7uNS line 1
14:52 Hotkeys left 14:53 lizmat left 14:58 Zoffix joined
Zoffix scimon: nothing silly about that. We support full signature typechecks both with subsets and even in signature itself. 15:00
m: subset IntAndStrSub where .?signature ~~ :(Int, Str); say -> Int, Str {} ~~ IntAndStrSub; say {;} ~~ IntAndStrSub
evalable6 True
False
Zoffix m: multi z(&s:(Int, Str)) { say "right sub" }; multi z(|) { say "wrong sub" }; z -> Int, Str {}; z {;}'
evalable6 (exit code 1) 04===SORRY!04=== Error while compiling /tmp/pic1Gp40_R
Strange…		 15:01
Zoffix, Full output: gist.github.com/370abc01da5192989f...0d44012ff5
Zoffix m: multi z(&s:(Int, Str)) { say "right sub" }; multi z(|) { say "wrong sub" }; z -> Int, Str {}; z {;}
evalable6 right sub
wrong sub
Zoffix should be able to use anon var too `&:(Int, Str)`, but looks like it's a parse bug there 15:02
scimon AHHHH! I knew there had to be a way to do it.
Zoffix I think that stuff is documented here: docs.perl6.org/type/Signature#Sign...e_Literals 15:03
15:03 cognominal joined
Zoffix R#1326 15:08
synopsebot R#1326 [open]: github.com/rakudo/rakudo/issues/1326 [LHF] Parse failure with signature unpacking on anon Callable param
scimon Yeah but why read stuff when you can try jamming it into the code and see what happens? :D (Generally the next thing I do it read stuff). 15:09
Zoffix I think that ticket can be fixed by just changing this line to `<subshortname> | <sigterm> | <subshortname> <sigterm>` github.com/rakudo/rakudo/blob/2024...r.nqp#L582 15:12
If anyone wants to give it a go
\o
15:13 Zoffix left, cdg left, domidumont joined, domidumont left 15:14 domidumont joined, domidumont1 left
Geth doc: 2877756577 | (Luca Ferrari)++ | doc/Language/traps.pod6
Space between 'Perl' and version ('5', '6').
 15:16
doc: 4a19bcbb4c | (Luca Ferrari)++ | doc/Language/traps.pod6
Add C<> blocks in titles and paragraphs.

For a more coherent appearence and behavior, use C<> when referring to types in titles and paragraphs.
synopsebot Link: doc.perl6.org/language/traps
doc: 474dca2866 | (Luca Ferrari)++ | doc/Language/traps.pod6
Little rewording on the BUILDALL paragraph.

Promoted the paragraph as a 'main' one, removing parentheses; linked to object creation for a better explaination of BUILD, BUILDALL, callsame.
Minor typos fixed.
mahafyi :) i am able to answer and control a telephony channel using Asterisk PBX AGI (Astersisk Gateway Interface) now. Instead of the perl5 RETURN, I simply used SAY and it worked ok. the hard part is structuring the whole thing into a module now, i have to get started with subroutines in earnest next i suppose. 15:17
scimon That sounds pretty impressive :) 15:19
15:20 jargan joined 15:21 jast left, jargan is now known as jast 15:23 tabv left, tabv_ joined, tabv_ is now known as tabv 15:24 zakharyas joined
mahafyi as i said, if a code executes without errors, that itself is good for me, lol. 15:24
perl6/asterisk/agi/t1.p6,123: @*ARGS.perl is an extract from logs, it is supposed to give all the arguments passed to script , i have : say "VERBOSE @*ARGS.perl 1"; 15:25
where VERBOSE is the asterisk agi command (exectues ok) , @*ARGS.perl is the arguments array elements printed and the last 1 is simply the verbosity level in asterisk (can be safely ignored for our immediate purpose) 15:26
15:26 cdg joined 15:28 domidumont1 joined, lizmat_ left
mahafyi ok never mind , i got it 15:31
15:31 domidumont left, zakharyas left
mahafyi now the script erros out : from asterisk debug i see - AGI Rx << VERBOSE["123"] 1 . it should be VERBOSE 123 1 , so thats going to be easy enough, but it looks like i have to tear away to do some work (for which i am paid) 15:33
15:33 zakharyas joined 15:36 kaare_ joined 15:37 david_xd left 15:40 cognominal left
mahafyi right now, i am simply writing say 'this command + params' and say 'that command + parms' ; the whole script is only that, get args, get env setting from asterisk config files, inherit all system variables for use . i have to figure out how this can be a module, how a user can use this beneficially instead of writing everything all over. 15:41
15:44 troys joined 15:48 lowbro joined, lowbro left, lowbro joined 15:49 geospeck left 15:55 geospeck joined 15:57 david_xd joined 15:59 geospeck left, geospeck joined 16:00 geospeck left 16:02 pmurias left 16:05 [Coke]_ is now known as [Coke] 16:06 cdg left, cdg joined 16:07 geospeck joined 16:08 geospeck left 16:09 wamba left 16:11 cdg left 16:15 cognominal joined, kent\n left 16:16 wamba joined, kent\n joined 16:17 itaipu joined 16:19 cognominal left 16:20 cognominal joined
Geth doc/master: 8 commits pushed by (Luca Ferrari)++ 16:21
16:39 silug joined 16:41 troys is now known as troys_ 16:46 troys_ is now known as troys 16:47 domidumont1 left 16:52 lowbro left
[Coke] m: enum sides <Back Front>; my %a := :{Back=>10, Front=>20}; say %a.perl; say %a{Back}; 16:53
evalable6 :{:Back(10), :Front(20)}
(Any)
[Coke] How to make that print "10" ?
AlexDaniel m: enum sides <Back Front>; my %a := :{Back=>10, Front=>20}; say %a.perl; say %a{~Back}; 16:56
evalable6 :{:Back(10), :Front(20)}
10
AlexDaniel hmmm
[Coke] ah, the hash init isn't doing what I thought, perhaps.
Juerd [Coke]: Well, if anything goes, just remove the : before {}
[Coke]: Why is that there in the first place? :) 16:57
[Coke] Juerd: because I want an object hash.
I didn't want the stringified versions.
Juerd I don't know what that is
AlexDaniel ( docs.perl6.org/type/Hash#Non-strin...ject_hash) )
[Coke] a hash with objects as keys instead of strings
Juerd Thanks
[Coke] (making it work with strings only for now) 16:58
AlexDaniel aaaaaaaah
m: enum sides <Back Front>; my %a := :{(Back)=>10, (Front)=>20}; say %a.perl; say %a{Back};
evalable6 :{sides::Back => 10, sides::Front => 20}
10
Juerd I was just about to say: they're strings, not objects :) 16:59
AlexDaniel that's just the pair notation
AlexDaniel facepalms
17:00 abraxxa left
Juerd TIL object hashes exist and you need parens to actually use it :) 17:00
[Coke] ah, AlexDaniel++
*facepalm* and I did this the right way someone else using Pair.new explicitly. :| 17:01
timotimo well, the pair notation will autoquote the LHS if it looks identifiery
[Coke] *somewher
scimon So... as usual when I get shown something like this I try to jam something crazy in. 17:10
m: my %h{Any};%h{any(1,2)} = "1 or 2";say %h{$_} for 1..3 17:11
evalable6 1 or 2
1 or 2
(Any)
scimon Junctions as hash keys. Because why not?
AlexDaniel scimon: :| 17:12
scimon But.... it doesn't do what I thought it did :(
17:13 geospeck joined, Zoffix joined
Zoffix scimon: because Junctions aren't Any, they don't fit when you assign to the key. You need to type-constraint on Mu 17:14
And then you'd hit RT#125614 , for which there's a fix in a branch, but the fix is blocked by RT#128905 17:16
synopsebot RT#125614 [open]: rt.perl.org/Ticket/Display.html?id=125614 [BUG] Can't store whole 1&2 junctions in Mu-keyed hash in Rakudo
RT#128905 [open]: rt.perl.org/Ticket/Display.html?id=128905 [BUG] Some core Routines do not do Callable role
Zoffix s/because//; 17:17
scimon Well.... by the time I think of a good use case for then it might be working :)
Zoffix scimon: what did you think it did?
Cause your code assins the string value to keys 1 and 2 and then correctly looks those up 17:18
scimon Yeah... I realised this after I tried adding a none(1,2) junction as a Key as well.
(I kinda liked the idea of a none junction as a default key) 17:19
Zoffix Ah yeah, the autothreading doesn't care about the type of junction IIUC
17:20 cognominal left
Zoffix m: my %h{Mu}; my $j := none 1, 2; %h.AT-KEY($j) = "1 or 2"; say %h.AT-KEY: $_ for 1, 2, $j 17:20
evalable6 (Any)
(Any)
1 or 2
scimon I shall continue poking about at stuff. 17:22
Zoffix scimon: good :) Report all the explosions if you find any: github.com/rakudo/rakudo/issues/new
scimon Will do
I was interested in the WHICH results for sets being consistent allowing for === matching of them. 17:25
17:26 zakharyas1 joined
Zoffix m: enum sides <Back Front>; (my %a{sides}){Back, Front} = 10, 20; say %a.perl; say %a{Back}; 17:30
evalable6 (my Any %{sides} = sides::Back => 10, sides::Front => 20)
10
17:31 scimon left, Zoffix left 17:32 khw joined 17:39 mahafyi left, dakkar left 17:40 zakharyas left 17:41 zakharyas1 left 17:54 rindolf left 17:56 rindolf joined 18:03 wamba left 18:09 nativecallable6 left, coverable6 joined, nativecallable6 joined 18:11 domidumont joined 18:17 sjoshi joined 18:18 sjoshi left 18:19 sjoshi joined, sjoshi left 18:21 raiph joined, sjoshi joined 18:23 domidumont left, cognominal joined 18:24 darutoko left 18:25 domidumont joined 18:28 cognominal left, eliasr joined, cognominal joined 18:33 troys is now known as troys_ 18:36 sjoshi left 18:37 sjoshi joined 18:39 sjoshi left 18:41 geospeck left 18:43 cognominal left 18:48 SHODAN left 18:49 SHODAN joined 18:51 domidumont left 18:53 setty1 joined 18:56 wamba joined 18:58 domidumont joined 19:00 domidumont left 19:01 domidumont joined 19:02 natrys joined 19:11 geospeck joined 19:16 geospeck left, domidumont left 19:18 poohman joined 19:27 vichib left 19:42 lizmat joined 19:44 poohman left 19:59 geospeck joined 20:00 cognominal joined 20:21 traxex_ joined 20:22 traxex left 20:23 committable6 left, bloatable6 left, bisectable6 joined, ChanServ sets mode: +v bisectable6, committable6 joined, ChanServ sets mode: +v committable6, bloatable6 joined, quotable6 joined, ChanServ sets mode: +v quotable6, statisfiable6 joined 20:26 Hotkeys joined 20:32 cdg joined 20:41 brrt joined 20:57 brrt left 20:58 evalable6 left, releasable6 joined, evalable6 joined 21:08 lizmat left 21:19 zakharyas joined 21:21 zakharyas left 21:22 zakharyas joined 21:37 itaipu left 21:38 itaipu joined 21:47 zakharyas left, zakharyas joined 21:50 zakharyas1 joined 21:54 zakharyas left 22:03 cdg left 22:05 cdg joined 22:09 troys_ is now known as troys 22:10 itaipu left 22:11 cdg left 22:15 geospeck left 22:29 wamba left 22:33 setty1 left 22:35 Zoffix joined 22:44 sena_kun left 22:45 natrys left 22:48 zakharyas1 left
Zoffix Happy 2nd pre-Birthday, Perl 6! :) 22:56
Who wants some celebration cake? temp.perl6.party/cake.jpg
ZofBot: here, have a slice, buddy: 🍰 22:57
ZofBot Zoffix, and I think so far it's even eluded jnthn++
jnthn The cake. It eluded me. /o\ 22:59
22:59 someuser_ joined
rindolf Zoffix: hi 23:01
geekosaur ...it gets content from irc?
(I remember sending the IRC message that came from) 23:02
23:02 someuser left
AlexDaniel /o/ 23:03
jnthn I'm curious what the context was :)
But anything that eluded me in the past sure as heck is going to at the moment.
geekosaur home/allbery/.config/hexchat/logs/ZNC [ttuttle ds] - FreeNode/#perl6-dev.log-2017 Dec 11 23:35:33 <geekosaur>the signal values thing is an open bug 23:04
(about $*SIGNAL (I think?) mismatches)
oh, second line lost because of slashes, whoops
Zoffix jnthn: it was how to fix differences between static Signal enums and dynamic Kernel.signal lookups. They have different values (at least on some OSes, like Windows): irclog.perlgeek.de/perl6-dev/2017-...i_15563701 23:05
geekosaur yes, that
Zoffix geekosaur: the brain is primed with a random book from Project Gutenberg and then it's also fed from IRC :)
And when the bot starts giving too much nonsense, I reset the brain and load another book. 23:06
geekosaur also I need to correct that path at some point
23:06 cdg joined 23:07 lizmat joined
geekosaur yeh, I'd seen the book matches but not IRC content before 23:08
23:09 kerframil joined
jnthn Zoffix: ah, that... 23:11
jnthn kinda hoped somebody else might nail that one
Zoffix :D
jnthn Will get to it eventually though, if nobody else does 23:12
23:12 cdg left
jnthn In the meantime, I should probably get to bed and continue trying to sleep away this flu thing... 23:12
o/ 23:13
Zoffix \o
timotimo Zoffix: that cake has coconut shavings as its outer layer? 23:23
Zoffix yeah
timotimo what's on the inside? 23:26
Zoffix Should be vanilla cake/cream. (I'm yet to cut it)
timotimo that does sound kinda nice 23:27
Zoffix :)
timotimo workrave tells me to stop keyboardwork for the day :\
AlexDaniel disabled it temporarily
timotimo happy be-earlied birthday to perl6 :P
Zoffix \o/
23:49 cdg joined 23:50 cpage__ joined 23:52 cpage_ left, cpage__ is now known as cpage_

Powered by Cro and the Raku® Programming Language.

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

Thank you!