»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.
tyil KotlinFan: on my machines 00:12
00:12 Zoffix joined
Zoffix KotlinFan: I use it for onelines all the time. Our dev bots are all in Perl 6. Also contemplating rewriting our large internal web app at work in Perl 6, though not sure when I'll get a chance to do so. Gonna use Cro. I hear it's popular: mi.cro.services 00:13
Hotkeys I use it in codegolf and my advent of code solutions :)
Zoffix Oh, cool. Edument has a course "Perl 6 for Programmers": edument.se/en/coursedetail/Perl-6-...rogrammers 00:15
KotlinFan but Perl 5 still exists 00:16
Zoffix KotlinFan: yeah, it's a different language. Sorta like C and C++
KotlinFan you know Kotlin language : kotlinlang.org/ ? 00:18
Zoffix First time hearing of it
Slayerk KotlinFan is an obvious troll 00:19
wow didn't realise it's been more than 1 hour since his last post 00:20
Zoffix What makes trolls obvious?
KotlinFan just comparing the hyped Kotlin language with Pearl 6 00:21
Slayerk It's like that former US Supreme Court Judge once said
"I can't describe it, but when I see it I know it"
Zoffix KotlinFan: looks nice. Too bad parens on function calls appear to be mandatory. I hate parens :D
Hotkeys Zoffix++ 00:22
00:23 zacts joined
Zoffix This is neat: try.kotlinlang.org/#/Examples/Dele...roperty.kt 00:23
We sorta have it in the form of a Proxy.
oorrrr..
eco: Proxee 00:24
buggable Zoffix, Proxee 'A more usable Proxy with bells': github.com/zoffixznet/perl6-Proxee
Zoffix ^_^
KotlinFan thanks Zoffix...I will have a look
Zoffix KotlinFan: yeah, flipping through those examples I see a bunch of similarities with what Perl 6 offers. So if you're a Kotlin Fan, you might enjoy Perl 6 too :) 00:25
KotlinFan ok nice...I will really have a look on Pearl 6 00:26
Zoffix \o/
00:28 sena_kun left
Zoffix [Coke]: this ever happen to you? Ran spelling test twice without making any changes, first time pass, second time fail: gist.github.com/zoffixznet/0883e28...35e15203c1 And another run didn't show one misspelled word on first run; only did on others 00:43
00:47 KotlinFan left
Zoffix m: sub z($i) { await |$i }([(start {sleep 2;}, start { sleep 4; }), (start {sleep 2;}, start { sleep 3; })]); say now - INIT now 00:51
evalable6 0.004645
Zoffix ah, nm 00:52
Thought maybe there's a race somewhere in the Procs that I see missing spell words
00:54 aborazmeh joined, aborazmeh left, aborazmeh joined
Slayerk Night all 00:55
00:55 Slayerk left
Zoffix Or maybe that is it. Why doesn't this wait 4 seconds? 00:56
m: sub z($i) { await |$i }; [(start {sleep 2;}, start { sleep 4; }), (start {sleep 2;}, start { sleep 3; })].map: { z $_ }; say now - INIT now
evalable6 0.0046257
Zoffix Night
Slayer
m: sub z($i) { await |$i }; [(start {sleep 2;}, start { sleep 4; }), (start {sleep 2;}, start { sleep 3; })]».<>.flat.map: { z $_ }; say now - INIT now 00:57
evalable6 4.0170170
Zoffix This waits. Somethings offcolour with the await sub
00:58 troys is now known as troys_
mspo off color remarks you say 01:01
01:01 astj joined
mspo to pieces you say 01:02
to shreds you say
01:04 Actualeyes left
Geth doc: b71dffa07b | (Zoffix Znet)++ | xt/aspell.t
Fix incorrect awaition of results

Fixes flops on spelling tests
01:04
Zoffix [Coke]: nm, it was just a bug in the test: github.com/perl6/doc/commit/b71dffa07b
01:05 astj left, astj joined 01:09 cdg joined
Zoffix Ah, it's actually an eagerness bug(?) in one of await() candidates 01:11
m: multi sub z(Iterable:D $i) { $i.eager.map: { z $_ } }; multi sub z(*@z) { say "there" }; z ([<a b c>], [<a b c>]),
evalable6
Zoffix oh wait, nm
ZofBot: never you mind!
ZofBot Zoffix, I recently also found that not HLLizing yourself, makes code faster
Zoffix Oh wait, yeaj 01:12
m: multi sub z(Iterable:D $i) { say "here"; $i.map: { z $_ } }; multi sub z(*@z) { say "there" }; z ([<a b c>], [<a b c>]),
evalable6 here
here
here
Zoffix m: multi sub z(Iterable:D $i) { say "here"; eager $i.map: { z $_ } }; multi sub z(*@z) { say "there" }; z ([<a b c>], [<a b c>]),
evalable6 here
here
there
there
there
here
there
there
there
Zoffix a bug with .map sinkage prolly actually
01:13 cdg left 01:14 pilne joined 01:19 ecocode left
Zoffix nm, no bug (well, one in &await(Iterable:D)` 01:19
A trap really:
m: <foo bar ber>.map: *.say
evalable6 foo
bar
ber
Zoffix m: <foo bar ber>.map: *.comb.map: *.say
evalable6
Zoffix Filed D##1718 01:22
Filed D#1718
synopsebot D#1718 [open]: github.com/perl6/doc/issues/1718 [LHF] Traps candidate? Sinkage of maps
01:22 Zoffix left 01:26 ecocode joined, comborico1611 left 01:38 lookatme joined
lookatme o/ 01:39
01:45 someuser left, Sgeo_ joined, someuser joined 01:47 Sgeo left 02:08 manchicken left 02:10 Cabanossi left 02:12 Cabanossi joined 02:15 Actualeyes joined 02:17 adu joined 02:23 pilne left
tbrowder hi, lookatme! 02:34
lookatme hi 02:35
adu hi
Hotkeys hi 02:36
adu is anyone going to say anything else? 02:37
02:37 itaipu left
tyil hi 02:38
adu lol
tyil the advent post is late today :(
02:41 vichib joined, Cabanossi left 02:42 Cabanossi joined 02:46 ilbot3 left
lookatme lol 02:49
02:50 Zoffix joined
Zoffix tyil: perl6advent.wordpress.com/2017/12/...hnologies/ 02:52
.tell Ven your post was still listed as a draft, 3hr past midnight. Based on irclog.perlgeek.de/perl6/2017-12-19#i_15600534 I assumed that was a mistake and published the post. I hope it was ready 😓 02:53
yoleaux Zoffix: I'll pass your message to Ven.
02:54 aborazmeh left 02:57 ilbot3 joined, ChanServ sets mode: +v ilbot3 03:00 wamba left 03:02 adu left
[Coke] Zoffix: nope, haven't that particular spelling failure mode. :( 03:06
Zoffix++
I am working on the code for my post now. 03:07
lookatme why day-20-advancements-in-macrotechnologies is under 2017/12/19 03:08
[Coke] Style comments welcome: gist.github.com/coke/2ceeca4f48080...565bbae6fd (incomplete) 03:10
Zoffix FWIW all the warnings from compilation makes it hard to find where proper errors are at: temp.perl6.party/snapshot2.png
03:11 Cabanossi left
Zoffix lookatme: fixed perl6advent.wordpress.com/2017/12/...hnologies/ 03:11
It might be a glitch with WP 'cause I'm still in timezone that's in 19th. Published date showed 23:00 of 19th 03:12
03:12 Cabanossi joined
[Coke] Cube::Three.new().R.U2.Rʼ.Uʼ.R.Uʼ.Rʼ.Lʼ.U2.L.U.Lʼ.U.L leaves the cube in the correct config, yay. 03:13
Geth doc: f49cadb3d5 | (Zoffix Znet)++ | 3 files
Fix typos
lookatme oh
mspo WP has gmtime and the blog's local time
I don't think it talks about *your* tz at all? 03:14
Zoffix IME it's UI uses GMT in some places and local TZ in other places.
*its
releasable6: release 03:17
releasable6 Zoffix, Next release will happen when it's ready. No blockers. 0 out of 133 commits logged
Zoffix, Details: gist.github.com/53f53659173238574a...bac11ad6ce
Zoffix m: "/tmp/foo".IO.spurt: "abc"; with "/tmp/foo".IO.open { .seek: 3, SeekFromBeginning; .read(3).say; .read(3).say; .eof.say }
evalable6 Buf[uint8]:0x<>
Buf[uint8]:0x<>
True
Zoffix m: "/tmp/foo".IO.spurt: "abc"; with "/tmp/foo".IO.open { .seek: 4, SeekFromBeginning; .read(3).say; .read(3).say; .eof.say }
evalable6 Buf[uint8]:0x<>
Buf[uint8]:0x<>
False
Zoffix fun :)
See the fix, but gonna keep it in my head. In syncfile.c -> mvm_eof the check is statbuf.st_size == seek_pos but should be statbuf.st_size <= seek_pos, for cases where we seeked past end 03:19
*keep until past release I meant
Ad after that's fixed the check for zero-size chunks can go away here: github.com/rakudo/rakudo/commit/dc...4aa83dR240 03:20
wc
03:20 Zoffix left
Morfent so rakudo 2016.10 installs fine on openbsd ignoring the difficulty with getting java installed 03:27
but it has the same problem with dns resolution as freebsd has 03:28
nqp itself installs fine, it's just the jdk that's awkward to install 03:30
03:36 astronavt joined
Morfent i'm thinking there might be a difference between gnu and bsd's implementations of getdomainname that might be screwing things up 03:44
03:45 Rawriful left
Morfent that's more a moarvm thing though 03:45
03:57 SCHAPiE left 04:04 SCHAPiE joined 04:09 piojo left 04:15 Actualeyes left 04:28 Herby_ joined
Herby_ o/ 04:28
04:31 agwind joined
Herby_ \o 04:33
Geth doc: da5a52a6ba | (Zoffix Znet)++ | 2 files
Fix nbsp xtest issues
04:39
04:41 Cabanossi left 04:42 Cabanossi joined 04:50 troys_ is now known as troys 04:55 Cabanossi left, astronavt left 04:56 Cabanossi joined 05:02 agwind left 05:03 agwind joined 05:25 agwind left 05:27 aindilis left 05:28 aindilis joined 05:30 troys is now known as troys_ 05:37 khw left 05:41 Herby_ left 05:46 AlexDaniel left 05:52 troys_ is now known as troys 06:08 Cabanossi left 06:09 cdg joined 06:11 Cabanossi joined 06:14 cdg left 06:27 geospeck joined 06:30 geospeck left 07:05 darutoko joined 07:09 Cabanossi left 07:11 Cabanossi joined
nine m: say <what> 07:12
evalable6 what
nine moritz: was probably some network issue? 07:13
moritz nine: possibly. I'm restarting camelia now, and hope she re-joins 07:15
07:18 piojo joined 07:19 lowbro joined, lowbro left, lowbro joined
piojo I found something wrong with the way perl6 executes subprocesses on windows. 07:19
Almost every command I've tried is failing. For example, this gives no output: perl6 -e "shell 'python -c \"print(4)\"'"
And simple stuff like shell('perl -e "use v5.18;"') gives a bad exit code. 07:21
(Actually, that one fails with a different error--"Can't find string terminator '"' anywhere before EOF at -e line 1." 07:22
07:22 astj left, wamba joined
piojo ). I think rakudo must be trying to do Bash shell quoting in CMD 07:22
nine piojo: do those examples work when you put them into scripts or execute them from the REPL? 07:23
07:23 astj joined
piojo nine: I believe they also fail in scripts, because I noticed this due to Perl5::Inline having a related build failure 07:23
And these commands work when typed on the command line without being invoked by rakudo. 07:24
07:24 astj left, astj joined
piojo But the "Can't find string terminator" error is exactly what happens if I try to run a command with 'an argument' instead of "an argument" 07:25
so I'm pretty sure illegal quoting is the problem, or at least part of it. I think this worked a few months ago 07:27
any idea what logic handles shell quoting?
nine I know Inline::Perl5's installation worked on Windows at one time at least.
piojo qx fails the same way 07:30
07:30 troys left 07:33 domidumont joined 07:35 cdg joined 07:40 domidumont left, cdg left, domidumont joined, espadrine left 07:49 Hotkeys left 07:55 ZzZombo left 07:57 ZzZombo joined 08:11 cdg joined 08:12 brakmic joined 08:16 cdg left 08:18 nightfrog left 08:19 sena_kun joined 08:22 wamba left 08:24 rindolf joined, Cabanossi left 08:26 Cabanossi joined 08:40 itaipu joined
lookatme piojo, the string terminator is CMD's problem, I think 08:44
piojo lookatme: that's taking "not my problem" to a whole new level :P
"not a bug, just use bash" 08:45
lookatme Is Perl6 has a shortcut start Perl6 REPL like Python ?? 08:46
jast just run 'perl6'
lookatme :) No, user don't want open a cmd, just want click it
jast then make a shortcut that runs perl6 08:47
lookatme This is Windows logical :)
08:54 pdcawley joined 09:04 domidumont1 joined 09:07 domidumont left 09:11 riceandbeans joined
riceandbeans what's the difference between rakudo and perl6? 09:11
I thought rakudo was just a jvm implementation of it and now I'm reading you can have a backend that is C based and that one was written specifically for rakudo perl6 so I'm wondering what differs 09:12
lookatme Perl6 is a language, like Python
rakudo like CPython 09:13
or other Python
riceandbeans OK, but then where do you fit in MoarVM
09:13 lookatme left
riceandbeans :( 09:13
09:15 pdcawley left 09:23 Cabanossi left 09:26 Cabanossi joined 09:28 dakkar joined
masak just found the concept en.wikipedia.org/wiki/Muntzing which is similar in spirit to bug golfing 09:40
09:42 domidumont joined 09:44 domidumont2 joined, pmurias joined 09:45 domidumont1 left
pmurias riceandbeans: rakudo can target both JVM and MoarVM 09:45
09:46 eliasr joined
El_Che pmurias: add js to the moarvm mix :) 09:46
riceandbeans When was non jvm support added, but still how does rakudo differ from whatever perl6 guideline
I don't remember the term used, but I remember reading a perl6 has to do this and that and anything that does this and that is a perl6
El_Che spec tests? 09:47
test suite?
09:47 domidumont left
riceandbeans maybe I'm not making sense 09:47
pmurias riceandbeans: Perl 6 is the language, Rakudo is the implementation 09:48
riceandbeans how does the rakudo implementation of perl6 differ from any others
09:48 Ven` joined
pmurias riceandbeans: other implemantations are as far as I know mostly not actively developed 09:48
riceandbeans: Perl6 is Python, Rakudo is CPython and Jython 09:49
Ven` o/
riceandbeans right right it's just weird because of the varied backends and it being the only one
Ven` Zoffix: very sorry about that. Since I had set a "release at", I thought "publish" would instead publish it immediately... Oops.
pmurias riceandbeans: it makes sense to reuse the frontend 09:50
riceandbeans why not just call it perl6?
Maybe I'm just dumb, I think it's confusing 09:51
Like with Ruby when I had to figure out what happened when they had MRI and merged YARV with it 09:52
Ven` riceandbeans: because when you say "I'm using Ruby" you can't really know if it's MRI, MRB, JRuby, Rubinius, ... 09:53
El_Che riceandbeans: the idea is to leave to door open for alternative implementations 09:54
riceandbeans what was the last attempt at one? 09:55
masak Ven++ # perl6advent.wordpress.com/2017/12/...hnologies/ 09:56
current status about infix:<ff> -- it sits in a branch, and it's "one failing test from being merged" 09:57
but behind that test is an unexpected host of conceptual challenges that I'm currently grappling with
El_Che riceandbeans: en.wikipedia.org/wiki/Perl_6#Histo...mentations 09:58
masak I have a solution, but I feel a bit like the pirate chief who enters a deal with higher powers without fully understanding the consequences
Ven`: in order to address lizmat++'s (and possibly others') confusion in the backlog, maybe add `# 007 code` at the top of the top code snippet? 09:59
10:00 parv joined
Ven` Fair enough :-). 10:03
10:09 scimon joined
masak oh, I meant as a *comment* *in* the code snippet 10:09
sorry, backseat editing is Hard™
Ven` I know you did, but I think it's fine that way, isn't it? 10:10
El_Che masak: you're doing fine :)
masak on the hypothesis that people skim sometimes, and the eye seeks out the first non-para thing
at least mine do
by the way, I realized far too late in 007's trajectory that the examples/ directory shouldn't be filled with fluff like fizzbuzz.007 :P 10:11
moritz masak: git rm or git mv can help :-)
masak it's one of those growth effects. the fizzbuzz example was impressive in 2016, when we were still fleshing out the basic language. less so in late 2017, when we're racing towards full macro capability. 10:12
moritz: aye; the plan is to phase out the old examples gradually as we land real macro examples
riceandbeans well, let's see if rakudo compiles for me
El_Che git mv examples fluff; mkdir example
masak hah 10:13
we're pretty good at testing the examples -- and those tests are very usefull because they exercise the language
I've decided that the old examples get to survive as such tests
but yes, maybe renamed as "fluff" or similar
10:33 domidumont joined 10:36 domidumont2 left, Cabanossi left 10:54 Actualeyes joined
piojo masak: IMO we don't test examples in modules nearly enough 11:01
I mean external files. Not necessarily modules. 11:02
11:05 Cabanoss- joined 11:09 Cabanoss- is now known as Cabanossi 11:10 cdg joined 11:14 cdg left 11:15 Zoffix joined, Hotkeys joined, araraloren joined
Zoffix nine: camelia is still not back. Note that when you ran `m: say <what>` it was the backup evalable6 bot that responded, not camelia 11:15
sjn .u  11:19
yoleaux No characters found
sjn huh
Zoffix piojo: "that's taking "not my problem" to a whole new level :P" it's more of taking a hard problem that doesn't seem to have a single solution: blogs.msdn.microsoft.com/twistylit...wrong-way/
piojo: it even affects run() with $*EXECUTABLE on windows, because it's a .bat file that uses cmd.exe and so needs cmd.exe's special quoting: RT#132258 11:20
synopsebot RT#132258 [new]: rt.perl.org/Ticket/Display.html?id=132258 [SECURITY][WINDOWS] `run "perl6" ...` can be made to execute shell commands
11:21 domidumont left
piojo Zoffix: thanks, that will be an interesting read. BTW, I just bisected it and reported it: github.com/rakudo/rakudo/issues/1325 11:21
11:22 Cabanossi left
piojo Zoffix: I use cygwin, and I completely bypass perl6.bat, because frankly it's rubbish 11:22
(It might not be its fault--I shouldn't judge)
11:23 Cabanossi joined
Zoffix There was also this R#1306 11:23
synopsebot R#1306 [closed]: github.com/rakudo/rakudo/issues/1306 [RFC] [RFC] add Proc/Proc::Async arg quoting modifiers
piojo but a cygwin rakudo wrapper can handle any type of arguments, whereas the included perl6.bat can't
11:24 Cabanoss- joined
nine Zoffix: looks like packages from the outside don't make it back to virbr0, so the VM never gets them 11:27
Zoffix piojo: so you're using Cygwin with the R#1325? May be worth mentioning. Also, mention the Rakudo version you're using
synopsebot R#1325 [open]: github.com/rakudo/rakudo/issues/1325 Shell quoting is wrong on Windows
riceandbeans my rakudo-test failed :\ 11:28
piojo Zoffix: I reproduced using CMD, for cleanest results
11:28 Cabanossi left, Cabanoss- is now known as Cabanossi, wamba joined
Zoffix riceandbeans: what OS are you using? Also, what are you building? 11:31
piojo: but you'd need proper quoting for CMD.exe, wouldn't you?
piojo: I get the same sort of issues when I put qx/perl -e "say 42"/ in a script and run it with perl 5 11:33
piojo Zoffix: I meant my comment about the "perl6.bat" argument handling in relation to one of the other links you mentioned. that seems like a windows batch programming bug, not a perl issue at all
Zoffix: It works for me. It should be: print qx/perl -E "say 42"/ 11:35
Zoffix Ah 11:36
Yeah, works for me too 11:37
riceandbeans Zoffix: DragonFly BSD 11:43
Zoffix: just followed the steps on building 11:44
piojo Zoffix: the least wrong solution is to let the user decide how to quote, and pass the string verbatim to CMD 11:47
i.e., apply no additional substitution or logic.
I think we can do better, but that's the easy correct solution. Unless I'm missing a detail of how CMD is invoked. 11:48
11:49 cpage_ joined, nightfrog joined
Zoffix riceandbeans: I don't think we have any volunteers currently actively ensuring builds work on BSDs, but you could report your issue, along with full output of failures, so someone could take a look: github.com/rakudo/rakudo/issues/new 11:50
El_Che it builds on osx, so other bsd flavours should work, no? 11:56
riceandbeans it's not quite like linux where the 'flavours' are related
with linux it's the same kernel, just different versions and different userland tools on top 11:57
El_Che darwin was based on freebsd, I thought
riceandbeans with the BSDs the kernels are different in each one, the userlands are different in each, and nothing is inherently compatible from one to the other, they share common ancestry
Zoffix AlexDaniel`: "<AlexDaniel> fwiw branches are cheap so this shouldn't be stopping anyone from committing (not to master tho)". On personal experience, it ain't that cheap. It's branches on 4 different repos + you either can't ver bump or have to suffer through merge conflicts. Currently, I can't be arsed to go through this. My 2 cents. 11:58
11:58 philomath_ joined
lizmat Ven`: looks like some <ff> are missing from your blogpost ? 11:58
riceandbeans darwin took parts from various BSDs at various versions over time, a fair amount from freebsd, but it is its own kernel 11:59
lizmat Ven` : shouldn't "macro infix:(lhs, rhs)" be "macro infix:<ff>(lhs, rhs)" ???
AlexDaniel` Zoffix: ok point taken. Sorry for the delay :)
riceandbeans osx binaries aren't abi compatible with freebsd
Ven` lizmat: fixed 12:00
parv riceandbeans, 1 test has been failing consistently on FreeBSD 11. I had filed a bug report on Github.
riceandbeans anyway, I'm kind of passing out it's like 4am here I'll file an issue later
Ven` seems like the MD converter kept that as <ff>, and wp removed it...
riceandbeans parv: I'll look intoi t more when I'm more awake
parv ok. for ref: rt.perl.org/Public/Bug/Display.html?id=131237
ok filed not on Github but on RT 12:01
riceandbeans well, before I lay down parv , t/05-messages/02-errors.t ................... Failed 1/1 subtests
12:01 cpage_ left
parv riceandbeans, that's different than mine (t/04-nativecall/02-simple-args.t line 62) 12:02
Zoffix AlexDaniel`: it might be fairly easy if the dev just keeps master HEAD of all rakudo/nqp/MoarVM like timotimo does and builds stuff separately but I don't know how to do this :o
piojo Zoffix: I'm now thinking we shouldn't quote/escape/modify shell() args at all, because it's convoluted. First we would need to parse it into individual arguments, then re-package them as a properly formatted string. But that goes against the whole point of what shell() claims to do.
shell() is supposed to let the shell do all the parsing. 12:03
Zoffix piojo: would be good to have comments on the ticket so others could read them too :)
piojo good point, thanks
parv rest, riceandbeans. I am going to too 12:04
Zoffix & 12:06
12:06 Zoffix left, parv left 12:11 mahafyi joined
AlexDaniel` sjn: use u: instead of .u :) 12:11
u: 
unicodable6 AlexDaniel`, U+E0FF <Private Use-E0FF> [Co] ()
12:20 philomath_ left
stmuk parv: I thought I'd seen a full PASS on FreeBSD since Apr but will check 12:31
12:33 ZzZombo left 12:34 ZzZombo joined 12:44 squashable6 left, reportable6 left, unicodable6 left, greppable6 left, benchable6 left, squashable6 joined, reportable6 joined, ChanServ sets mode: +v reportable6, greppable6 joined, unicodable6 joined, ChanServ sets mode: +v unicodable6, benchable6 joined
perlawhi1l hrm 12:45
hi perlers
can I do sets of lists?
m: say Set((0, 1), (1, 1), (1, 2)) (|) Set((1, 1), (2, 1), (2, 2))
evalable6 set(0 1 2)
perlawhi1l setops (eg. ∪ ) seem to work fine with typed Hashes (ie. %set{List})... can I Type my sets? 12:46
scratch that... typed Hashes ain't working. So... how can i do List Sets? 12:50
I guess... I can... stringify my Lists for the keys :/ 12:51
moritz lists are not value types (they are lazy), so they make terrible hash or set keys 12:53
12:54 cpage_ joined 12:56 philomath_ joined, domidumont joined
perlawhi1l hmm, yeah... i just found some relevant discussion 12:57
irclog.perlgeek.de/perl6/2016-09-23#i_13260617
seems theres also the issue of how List's are compared
so... looks like stringifying my keys is the best option so far. perhaps we need a new 'FrozenList' type, lol 12:58
or is that ImmutableArray (an oxymoron in Perl 6 :D) 12:59
moritz we might need a tupel type 13:00
we used to have Parcel that kinda, somewhat fit that niche
but it's been removed during the Great List Refactoring
perlawhi1l as long as we as a community agree to saying it "tupple" and not "toople", i'm down 13:01
actually yeah! resurrecting Parcel would be a cool thing
moritz it's tupper ware for memory! 13:03
lizmat moritz: I have a patch for making Lists that consist of decontainerized values be a value type 13:06
perlawhi1l oooooh
lizmat but it has a few issues and it will be definitely post 2017.12
perlawhi1l lizmat++
Ven` lizmat++
perlawhi1l that's cool... nice to know it's in the works
Ven` having tooples sounds cool :D.
perlawhi1l damnit Ven! 13:07
I bet you say "jif", too
lizmat fwiw, I do think Lists and Arrays are a bit too conflated atm
stmuk parv: roast has PASS but you are right t/04-nativecall/02-simple-args.t from moar tests has been failing on FreeBSD obviousily for some time
Ven` perlawhi1l: I actually alternate between both pronunciations :P 13:08
13:08 pmurias left
Ven` just to confuse people 13:08
13:08 pmurias joined
lizmat
.oO( you say tomato )
13:08
Ven` it's a-boot time. 13:09
perlawhi1l Parcel has such a perlish vibe to it... though, adding another Positional container isn't going to help the conflation issues lizmat mentioned 13:10
13:12 Zoffix joined
Zoffix "adding another Positional container isn't going to help the conflation issues".... So put it out into ecosystem as a module. 13:12
m: say Set($(0, 1), $(1, 1), $(1, 2)) (|) Set($(1, 1), $(2, 1), $(2, 2)) 13:13
evalable6 set((0 1) (1 1) (1 1) (1 2) (2 1) (2 2))
Zoffix perlawhi1l: ^ they were just being all flattened in a slurpy
perlawhi1l Zoffix++ 13:14
Zoffix m: my $l = (1, 1); say Bag($(0, 1), $l, $l, $(1, 1), $(2, 1), $(2, 2)){$l} 13:15
evalable6 2
Zoffix There's this problem, but it should be resolved by the patch lizmat++ mentioned 13:16
(would then give `3` as result there)
13:17 cpage_ left
lizmat hmmm... seems I misplaced the patch... 13:27
Ah, I realise what the problem was: from the .WHICH of an object itself you cannot see whether something is a value type or not
m: my $a = 42; say $a.WHICH # disregards container here 13:28
evalable6 Int|42
Zoffix Ah
lizmat m: my $a = (1,2,3); say $a[0].WHICH # no container 13:29
evalable6 Int|1
Zoffix m: use nqp; -> \obj { my \z = nqp::istype(obj, Iterable) ?? obj !! obj.list; dd z }( :foo({:ber(80)}).Pair )
evalable6 (:foo({:ber(80)}),)
lizmat m: my $a = [1,2,3]; say $a[0].WHICH # has container, but still disregarded
evalable6 Int|1
13:29 Hotkeys left
Zoffix m: use nqp; -> \obj { my \z = nqp::istype(obj, Iterable) ?? obj !! obj.list; dd z }( (:foo({:ber(80)}),) ) 13:29
evalable6 (:foo({:ber(80)}),)
Zoffix Same result, right?
m: dd gather { deepmap {.take}, (:foo{:80ber},) } 13:30
evalable6 (:foo({:ber(80)}),).Seq
Zoffix m: dd gather { deepmap {.take}, (:foo{:80ber}.Pair) }
evalable6 (:ber(80),).Seq
Zoffix Yet, different result here. It goes through here, so I'd thought based on first code above the result for the second code would be the same :S github.com/rakudo/rakudo/blob/4fca...1447-L1449 13:31
13:32 ruschlem_ joined
Zoffix Ooh, wait. No it doesn't go through there :P 13:32
The .Pair version goes through here: github.com/rakudo/rakudo/blob/4fca...ps.pm#L652
lizmat well, the coremap / deepmap / whatever map is still a bit of a hacky area of Rakudo, imo
Zoffix Gonna try to add a `Pair` candidate, 'cause currently `gather { @meows».take }` breaks up Pairs into values, which is kinda sucky 13:33
m: dd gather { %(:42foo, :bar{:_{:_{:40meows}}})».take } 13:34
evalable6 (40, 42).Seq
Zoffix m: dd gather { (:42foo, :bar{:_{:_{:40meows}}})».take }
evalable6 (:foo(42), :bar({:_(${:_(${:meows(40)})})})).Seq
lizmat yeah, that difference feels wrong 13:35
a hash should be an iterable of Pairs
Zoffix mhm
13:35 itaipu left, philomath_ left
lizmat not of values 13:35
m: my %h = a => 42, b => 666; dd $_ for %h 13:36
evalable6 :a(42)
:b(666)
lizmat m: dd $_ for a => 42, b => 66
evalable6 :a(42)
:b(66)
Zoffix AlexDaniel`: actually, it's 6 repos, not 5 :) nqp + rakudo + MoarVM + roast + docs :) 13:37
AlexDaniel` :) 13:38
Zoffix I mean 5 repos not 4. 13:39
ZofBot: counting is hard. Let's go shopping.
ZofBot Zoffix, But a strong man does need an occasional holiday, and I made up my mind that I would get it
13:39 wamba left
Zoffix You're having way too many holidays, robot. 13:39
pmurias What do robots celebrate? Judgment Day? 13:41
Zoffix Holy Robanukah 13:42
theinfosphere.org/Robanukah
moritz Independence Day! :-9 13:43
13:47 poohman joined
Zoffix Man, &nodemap is document, but it has precisely 1 test for it in the spec: github.com/perl6/roast/blob/master...per.t#L425 13:48
*documented
13:48 itaipu joined
Zoffix R#370 13:50
synopsebot R#370 [closed]: github.com/rakudo/rakudo/pull/370 Main single positional
Zoffix S#370
synopsebot S#370 [open]: github.com/perl6/roast/issues/370 `nodemap` is virtually untested
Geth doc: c625923d77 | (Zoffix Znet)++ (committed using GitHub Web editor) | doc/Type/Any.pod6
Make `does` more palatable on a plural

Also make other linkable things in paragraph actual links readers could follow
13:54
synopsebot Link: doc.perl6.org/type/Any
13:54 |oLa| joined 13:57 |oLa| left 13:59 itaipu left 14:04 itaipu joined 14:06 pdcawley joined, MasterDuke left, wamba joined 14:09 Rawriful joined 14:12 Rawriful left 14:13 ruschle75 joined, ruschle75 left, ruschle57 joined, ruschle57 left 14:14 ruschlem_ left, ruschle20 joined, ruschle20 left 14:15 ruschlem_ joined, ruschlem_ left, ruschlem_ joined, ruschlem_ left 14:16 rindolf left, ruschlem_ joined, ruschlem_ left 14:21 pdcawley left 14:32 poohman left 14:34 domidumont1 joined 14:36 cdc joined 14:37 Thrush joined, domidumont left
Thrush You know how: $string = $string.flip; can be shortened to: $string .= flip; ? Well, what would be the equivalent of a chain, like this: $string = $string.lc.flip.tc.flip; ? 14:42
Zoffix m: my $string = 'meows'; $string .= lc .= flip .= tc .= flip; say $string 14:43
evalable6 meowS
14:44 wamba left
Thrush Thanks, Zoffix. Is there a way for me to look at that so it seems more logical? I just can't see how replacing all the '.' with '.=' would work. 14:45
Zoffix: Or rather, "should" work.
Zoffix m: my $string = "meows"; $_ = .lc.flip.tc.flip with $string; say $string # I'd write it like that tho 14:46
evalable6 meowS
Zoffix Thrush: picture the "." and ".=" as operators between a thing on the left and a method name on the right. If you write it as $string .= lc.flip.tc.flip you're saying "call .lc on $string and put it into $string, then on the result call the rest of the methods and discard that result" 14:47
14:47 khw joined
Zoffix $string .= lc.flip.tc.flip; is like ($string .= lc).flip.tc.flip 14:47
14:48 geospeck joined 14:49 samcv joined
geekosaur this belongs in traps. or we should reconsider precedence somehow 14:51
Zoffix Or alternatively: "." means "call method and return result"; ".=" means "call method and assign result back into original thing"; Then, $string .= lc.flip.tc.flip => "call lc and save result into original, then call flip and return result, call lc and return result and call flip and return result" 14:52
14:53 grumble left
Thrush Zoffix: So what if I were to use: $string .= lc.flip.=tc.flip ? How would the mix of '.' and '.=' play out? 14:55
14:56 jast left
Thrush m: my $string = "meows"; $string .= lc.flip.=tc.flip; say $string; 14:57
evalable6 (exit code 1) Cannot modify an immutable Str (swoem)
in block <unit> at /tmp/HWWj9VqldU line 1
Zoffix m: say '$string .= lc.flip.=tc.flip'.trans: [<. .=>] => ['call method and return result', 'call method and assign the value to LHS']
evalable6 $string call method and assign the value to LHS lccall method and return resultflipcall method and assign the value to LHStccall method and return resultflip
Zoffix m: say '$string .= lc.flip.=tc.flip'.trans: [<. .=>] => [' (call method and return result) ', ' (call method and assign the value to LHS) ']
evalable6 $string (call method and assign the value to LHS) lc (call method and return result) flip (call method and assign the value to LHS) tc (call method and return result) flip
Zoffix m: say '$string .= lc.flip.=tc.flip'.trans: [<. .=>] => [' (call method and return result) ', ' (call method and assign the value to LHS and return result) ']
evalable6 $string (call method and assign the value to LHS and return result) lc (call method and return result) flip (call method and assign the value to LHS and return result) tc (call method and return result) flip 14:58
Zoffix Thrush: ^ that'd happen :) And the error you saw above is because one of the assignments tries to assign the value into an immutable value returned by one of the methods
14:58 jast joined
Zoffix & 14:58
15:00 rindolf joined
Thrush Zoffix: Okay, thank you. You've been a great help. 15:01
15:02 piojo_ joined 15:06 grumble joined 15:11 Thrush left, wamba joined 15:22 brakmic left
Zoffix \o/ 15:22
15:26 grumble left 15:28 pmurias left 15:29 pmurias joined
Geth doc: 72e179f57a | (Elizabeth Mattijsen)++ | 2 files
Document .skip(WhateverCode)
15:31
[Coke] realizes that there's no way he's going to finish this entire module by friday and instead concentrates on getting the blog post to a satisfying conclusion with what's available. 15:33
Zoffix perks up upon hearing that Friday is that close... 15:34
15:34 comborico1611 joined
lizmat [Coke]: would it help if we would switch slots ? 15:34
aka 22 <-> 24
15:35 grumble joined
[Coke] lizmat: ... crap, that might give me enough time. :|\ 15:35
lizmat so is that a yes or a no ?
15:35 brakmic joined
lizmat :-) 15:35
[Coke] Eh. I think it'll still be fine as is. if you give me more time, I will expand to fill it, and that's not necessarily good; I think we'll end up with a decent post. 15:36
I appreciate your kind offer, however.
lizmat ok, well, let me know if you can't make the deadline anyway... I intend to finish mine tonight or early tomorrow
[Coke] Roger. 15:37
15:38 Zoffix left 15:46 wamba left 15:47 Ven` left, gregf_ left, dmaestro left 15:49 geospeck left, grumble left 15:52 grumble joined 15:54 Houl joined 15:57 wamba joined 15:58 troys joined 16:01 domidumont1 left 16:02 Ven` joined, gregf_ joined, dmaestro joined, brakmic left 16:05 piojo_ left 16:06 araraloren left, david_xd joined 16:15 piojo_ joined
scimon Mines all ready to go :) Looking forward to it. 16:19
moritz \o/ 16:21
16:25 itaipu left
El_Che boom 16:30
16:30 pmurias left 16:32 xinming joined 16:35 piojo_ left 16:45 khisanth_ left, lowbro left 16:46 itaipu joined 16:48 pmurias joined 16:51 poohman joined 16:58 khisanth_ joined 17:02 david_xd left
lizmat notes that Lock::Async is undocumented :-( 17:03
17:03 itaipu left 17:11 philomath_ joined 17:17 philomath_ left 17:24 zakharyas joined 17:32 scimon left 17:34 dakkar left 17:39 david_xd joined 17:47 wamba left 17:55 poohman left 17:59 itaipu joined 18:05 ExtraCrispy joined 18:08 coverable6 left, nativecallable6 left 18:09 coverable6 joined, nativecallable6 joined, ChanServ sets mode: +v nativecallable6 18:11 Morfent left 18:20 darutoko left
Geth doc: 3afac8fcc2 | (Tom Browder)++ (committed using GitHub Web editor) | doc/Language/pod.pod6
mark formats that are NYI
18:23
synopsebot Link: doc.perl6.org/language/pod
18:23 daxim left
lizmat tyil.nl/articles/funding-yourself-...loper.html # tyil++ 18:23
*and* at #2 at news.ycombinator.com 18:24
18:25 mahafyi left 18:36 itaipu left
[Coke] wonders if we have any developers using patreon or some other system. 18:36
perl in general, not just sixers.
lizmat gabor maybe? 18:37
18:38 jast left, ExtraCrispy left
[Coke] lizmat: ah, www.patreon.com/szabgab 18:39
18:43 itaipu joined 18:44 domidumont joined, jast joined
Geth doc: 24fe425d27 | (Elizabeth Mattijsen)++ | doc/Type/Lock/Async.pod6
First stab at documenting Lock::Async
18:51
synopsebot Link: doc.perl6.org/type/Lock/Async
[Coke] lizmat: why is Kept capitalized? 18:53
lizmat m: my $p = Promise.new; $p.keep; say $p.status # that's why, although it may be a poor reason 18:54
evalable6 Kept
18:57 troys is now known as troys_ 18:58 Actualeyes left
[Coke] m: my $p = Promise.new; $p.keep; say $p.status.^name; 18:58
evalable6 PromiseStatus
19:09 wamba joined 19:13 itaipu left, cdg joined 19:18 domidumont left
timotimo [Coke]: szabgab just recently posted on his blog about perl and patreon 19:22
tyil [Coke]: yes, me :p 19:25
on the wondering if anyone uses patreon 19:26
[Coke] tyil: URL? 19:29
tyil www.tyil.nl/donate.html for the general page (exists since today), www.patreon.com/bePatron?c=1401236 for my patreon page 19:30
all feedback would be appreciated tbh 19:31
Geth doc: b5a8a43e4b | (Elizabeth Mattijsen)++ | doc/Type/Telemetry/Sampler.pod6
First stab at documenting Telemetry::Sampler
19:34
synopsebot Link: doc.perl6.org/type/Telemetry/Sampler
19:54 traxex joined
masak +1 on lower-casing "kept" when it's used normally in a sentence ;) 19:56
Geth doc: bce40f1b28 | (Elizabeth Mattijsen)++ | doc/Type/Lock/Async.pod6
s/Kept/kept/ by popular demand
20:00
synopsebot Link: doc.perl6.org/type/Lock/Async
20:22 geospeck joined
Geth doc: abbd0c9f73 | (Elizabeth Mattijsen)++ | doc/Type/atomicint.pod6
Document atomic-(inc|dec)-fetch, aka ++⚛ and --⚛
20:23
synopsebot Link: doc.perl6.org/type/atomicint
20:26 bart_ joined 20:29 AndChat|688961 joined, comborico1611 left 20:33 bart_ is now known as brrt 20:36 troys_ is now known as troys 20:40 AlexDaniel joined 20:46 Houl left 20:48 brrt left 20:51 daxim joined 20:52 david_xd left 20:55 quotable6 left, quotable6 joined, brrt joined 20:57 committable6 left, committable6 joined, ChanServ sets mode: +v committable6
AlexDaniel should “perl5” be written with nbsp? I thought it's a thing in perl 6 only 20:57
20:58 bisectable6 left, evalable6 left, bloatable6 left, statisfiable6 left, releasable6 left, bloatable6 joined, ChanServ sets mode: +v bloatable6, evalable6 joined, ChanServ sets mode: +v evalable6, bisectable6 joined, releasable6 joined, statisfiable6 joined
[Coke] both, so that you never have a dangling Perl. 21:00
21:01 armin left 21:02 armin joined
lizmat yup, that's the idea 21:06
21:07 noganex_ left 21:09 lancew joined
AlexDaniel well, I usually write it as “Perl5”… :) 21:11
don't want to offend anyone with unicode nbsp :)
21:18 troys is now known as troys_ 21:21 [particle] left 21:22 david_xd joined 21:23 geospeck left 21:27 lancew left 21:31 zacts left 21:38 [particle] joined 21:45 noganex joined 22:06 koto joined 22:07 koto left, sena_kun left 22:09 troys_ is now known as troys 22:11 zakharyas left 22:25 jast left 22:30 silug left 22:31 jast joined 22:33 zacts joined
Geth doc: 8efd70fc03 | (Elizabeth Mattijsen)++ | doc/Type/Blob.pod6
Document Blob.allocate
22:44
synopsebot Link: doc.perl6.org/type/Blob
Geth doc: a066bb9549 | (Elizabeth Mattijsen)++ | doc/Type/Buf.pod6
Document Buf.reallocate
22:55
synopsebot Link: doc.perl6.org/type/Buf
22:56 cdg left 23:04 SHODAN left 23:10 cdg joined
[Coke] lizmat++ lotsadocsa. 23:14
23:14 cdg left
lizmat ADD # Advent Driven Documentation :-) 23:14
23:22 lancew joined 23:23 cdg joined 23:25 konsolebox left 23:27 cdg left 23:29 brrt left 23:32 konsolebox joined 23:34 cdg joined 23:41 Hotkeys joined 23:42 lancew_ joined, lancew__ joined, lancew__ left 23:43 pmurias left 23:44 cdg_ joined 23:45 cdg left 23:50 Guest19 joined 23:51 Guest19 left 23:52 Guest19 joined, Guest19 left 23:53 lancew_ left
Geth doc: 19ecebc1ec | (Zoffix Znet)++ | doc/Type/Array.pod6
Revert "Revert "Document Array.gist""

This reverts commit 08c0b3656f24961a3df50e41c46fa254a29a87c9.
Method got brought back:
  github.com/rakudo/rakudo/commit/08539c43d2
  github.com/perl6/roast/commit/5a0beb5afb
23:53
synopsebot Link: doc.perl6.org/type/Array
23:53 mikowitz joined 23:56 rindolf left 23:59 cdg_ left