»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.
babydrop committable6: 2016.11: say permutations(3000).elems 00:00
committable6 babydrop, ¦«2016.11:»: No build for this commit
babydrop orly
committable6: 2016.10: say permutations(3000).elems
committable6 babydrop, ¦«2016.10:»: No build for this commit
babydrop committable6: 2016.11 say permutations(3000).elems
committable6 babydrop, ¦«2016.11»: 1
babydrop committable6: 2016.11 say +permutations(3000)
committable6 babydrop, ¦«2016.11»: Cowardly refusing to permutate more than 20 elements, tried 3000␤ in block <unit> at /tmp/gRUULhX7Nt line 1␤␤Actually thrown at:␤ in block <unit> at /tmp/gRUULhX7Nt line 1␤ «exit code = 1»
babydrop ^ that's cause the .elems form ends up with a List of one item, with Failure in it.
So TimToady's suggestion was is it may make a sense for methods that may return lists to throw instead of failing. 00:01
But now I'm seeing that would affect a ton of methods, making our failure much less useful, so perhaps we need to think of an alternate approach to resolve the original issue
s/our failure/our failure mechanism stuff/; 00:02
00:02 pyrimidine joined
AlexDaniel babydrop: as for the policy, I don't think anything is needed. lizmat++ was quick enough to fix all previous regressions that were caused by optimizations and stuff 00:05
babydrop
.oO( policy? )
00:06
m: multi foo(42) { 42}; for ^100_000 { my $x = foo 42; }; say now - INIT now
camelia rakudo-moar 4fd6e9: OUTPUT«1.6167577␤»
00:06 Actualeyes left
AlexDaniel
.oO( the one you mentioned? Nevermind )
00:06
babydrop m: multi foo(42) { 42}; multi foo([Failure:D]) { "got it" }; for ^100_000 { my $x = foo 42; }; say now - INIT now
camelia rakudo-moar 4fd6e9: OUTPUT«1.59374864␤»
babydrop huh
m: multi foo(42) { 42}; for ^200_000 { my $x = foo 42; }; say now - INIT now
camelia rakudo-moar 4fd6e9: OUTPUT«2.76862588␤»
babydrop m: multi foo(42) { 42}; multi foo([Failure:D]) { "got it" }; for ^200_000 { my $x = foo 42; }; say now - INIT now
camelia rakudo-moar 4fd6e9: OUTPUT«3.06208905␤» 00:07
babydrop m: say 3.06/2.76
camelia rakudo-moar 4fd6e9: OUTPUT«1.108696␤»
00:07 pyrimidine left
AlexDaniel babydrop: the failure thingy is still a bit over my head, so no ticket from me :P 00:07
babydrop Well, I guess that may be one way to do it, if we want to slow everything down by 10% :o
00:07 pierre_ joined
samcv babydrop, i'm back 00:08
babydrop AlexDaniel: I'll rely to TimToady's omniscience ;)
samcv: right. So you wanted to make rakudo.org better?
samcv yes :)
00:09 SS_ joined
AlexDaniel :| 00:10
m: say (5,6,7).pick(-1)
camelia rakudo-moar 4fd6e9: OUTPUT«MVMArray: Index out of bounds␤ in block <unit> at <tmp> line 1␤␤»
AlexDaniel j: say (5,6,7).pick(-1)
camelia rakudo-jvm 8ca367: OUTPUT«VMArray: Index out of bounds␤ in method pull-one at gen/jvm/CORE.setting line 18366␤ in method push-exactly at gen/jvm/CORE.setting line 2685␤ in method push-at-least at gen/jvm/CORE.setting line 2704␤ in method reify-at-least at gen/jvm/CORE.se…»
babydrop um, totally forgot my passowrd...
00:10 lucasb joined
AlexDaniel commit: 2015.12,HEAD say (5,6,7).pick(-1) 00:11
committable6 AlexDaniel, ¦«2015.12,HEAD»: MVMArray: Index out of bounds␤ in block <unit> at /tmp/s0KdTK5nQ3 line 1␤ «exit code = 1»
AlexDaniel ah, okay!
babydrop Apparently we have 53 new comments on rakudo.org 00:12
samcv: hm, I don't see an option to add a new user, so maybe I don't have sufficient perms 00:13
samcv hmm
babydrop timotimo: moritz: we have a volunteer to make rakudo.org better—samcv—would it be possible to hook them up with a user account for wordpress? 00:14
lucasb post comments? all I see is "Comments Off" O.o
babydrop lucasb: I see "53 comments in need of moderations" 00:15
Maybe it's old ones, before comments were turned off.
lucasb ah
spam maybe 00:16
00:17 secwang joined, secwang left
AlexDaniel by the way, are there any plans for uint/int overhaul? 00:19
00:19 secwang_ joined
AlexDaniel because I see this problem like everywhere 00:19
m: say (4,5,6).head(-99999999999999999)
camelia rakudo-moar 4fd6e9: OUTPUT«()␤»
AlexDaniel m: say (4,5,6).head(-999999999999999999)
camelia rakudo-moar 4fd6e9: OUTPUT«()␤»
AlexDaniel m: say (4,5,6).head(-9999999999999999999)
camelia rakudo-moar 4fd6e9: OUTPUT«(4 5 6)␤»
babydrop Well volunteered!
lucasb m: say 'let it roll baby roll,'.roll(-1) 00:20
camelia rakudo-moar 4fd6e9: OUTPUT«(let it roll baby roll, let it roll baby roll, let it roll baby roll, let it roll baby roll, let it roll baby roll, let it roll baby roll, let it roll baby roll, let it roll baby roll, let it roll baby roll, let it roll baby roll, let it roll baby roll, le…»
babydrop lucasb: you can fix that one :)
It's easy.
lucasb AlexDaniel++ I enjoy your bug reports. keep them coming 00:21
AlexDaniel thanks! 00:22
00:26 BenGoldberg joined, SS_ left
AlexDaniel wait, what… 00:30
bisect: old=2016.09 say (4,5,6).head(-9999999999999999999) 00:31
bisectable6 AlexDaniel, Bisecting by output (old=2016.09 new=4fd6e94) because on both starting points the exit code is 0
AlexDaniel, bisect log: gist.github.com/aa711737cdac63fa6d...c7aed86079
AlexDaniel, (2016-10-12) github.com/rakudo/rakudo/commit/69...d25f0a0810
AlexDaniel bisect: old=2016.09 say (4,5,6).tail(-9999999999999999999)
bisectable6 AlexDaniel, Bisecting by exit code (old=2016.09 new=4fd6e94). Old exit code: 0
AlexDaniel, bisect log: gist.github.com/a16e3b9e00de523784...9f3cfbc915
AlexDaniel, (2016-10-22) github.com/rakudo/rakudo/commit/a3...381615961c
00:34 BenGoldberg left, BenGoldberg joined, secwang_ left 00:39 canopus left, aborazmeh joined, aborazmeh left, aborazmeh joined 00:42 BenGoldberg left, BenGoldberg joined
babydrop committable6: 2015.07 sub f(::T) { say (T,) }; f(10) 00:44
committable6 babydrop, ¦«2015.07»: Method 'gist' not found for invocant of class 'T'␤ in sub f at /tmp/NnA1MD9qU9:1␤ in block <unit> at /tmp/NnA1MD9qU9:1␤ «exit code = 1»
babydrop fgzvsx d nbret
gah
There's nothing more infurriating than having copy-paste issues
szx vfdg 00:45
'<>GKM:RADEN
AlexDaniel: join #zofbot for a sec, I wanna show something and the copy-paste is annoying me
00:46 canopus joined 00:48 BenGoldberg left 00:52 pyrimidine joined, labster left 00:53 BenGoldberg joined
AlexDaniel #zofbot is a weird place. People are hugely outnumbered by bots there, yet bots are abused constantly 00:54
babydrop hah 00:55
00:57 pyrimidine left
babydrop m: put class {}.^methods».name.grep: /say/ 00:57
camelia rakudo-moar 4fd6e9: OUTPUT«␤»
babydrop `say` is a method provided provided by Mu: github.com/rakudo/rakudo/blob/4fd6...Mu.pm#L470 00:58
How come it doesn't show up in output?
m: put class {}.^methods(:all)».name.grep: /say/ 00:59
camelia rakudo-moar 4fd6e9: OUTPUT«say␤»
babydrop Forgot about that
New Advent Post: "Day 7 — Set In Your Ways: Perl 6’s Setty and Baggy Types": perl6advent.wordpress.com/2016/12/...ggy-types/ 01:02
.tell Technaton the blog post I mentioned a day ago, is up: perl6advent.wordpress.com/2016/12/...ggy-types/
yoleaux2 babydrop: I'll pass your message to Technaton.
yoleaux babydrop: I'll pass your message to Technaton.
babydrop 0.o
yoleaux: why did you have to come back!
01:03 yoleaux2 left 01:06 labster joined
AlexDaniel babydrop: “I’ll be using fancy-pants Unicode versions of operators and symbols in this post” yet no unicode quotes :E 01:11
01:11 skids left
lucasb I rakudobugged. thanks for the help AlexDaniel and babydrop! 01:11
01:17 BenGoldberg left 01:20 BenGoldberg joined
babydrop AlexDaniel: I don't like them. 01:20
AlexDaniel babydrop: who?
bots?
ah
no
quotes
01:33 ssotka left 01:34 lucasb left 01:40 Actualeyes joined, lucasb joined
lucasb m: a.:: 01:42
camelia rakudo-moar 4fd6e9: OUTPUT«===SORRY!===␤MVMArray: Can't pop from an empty array␤»
01:42 yqt left
lucasb m: a.::(1) 01:42
camelia rakudo-moar 4fd6e9: OUTPUT«===SORRY!===␤cannot stringify this␤»
01:45 cdg left
AlexDaniel cannot stringify what? :) 01:50
lucasb can't touch this
AlexDaniel lucasb: yea, both LTAs, not sure if there's anything else about it 01:51
the biggest problem there is that there's no line number 01:52
so good luck searching for the cause in your code, heh!
lucasb ok, I don't know if there's anything to do. I just don't like to see internal errors leaking
AlexDaniel lucasb: RT #129306 similar resolved ticket 01:56
01:59 lucasb left 02:01 pyrimidine joined
samcv naice. regarding the Atom/github syntax highlighter 02:01
started the process of reforming the code so I in the future can just generate it for all the quoting constructs 02:02
that's my future plan. so then it can support every quoting construct
:)
02:02 djbkd joined
samcv and also make it easy to make a change in only one spot and have it affect all of them 02:03
02:03 pyrimidine left 02:06 djbkd left 02:11 kalkin- left, kalkin-_ joined 02:13 kurahaupo_ joined 02:14 kurahaupo__ left 02:28 Herby_ joined
Herby_ o/ 02:28
Just wanted to say thanks to those who are writing for the 'Perl 6 Advent Calendar'. I'm enjoying the posts 02:30
02:31 effbiai joined 02:35 FROGGS left 02:36 itcharlie1 joined 02:38 labster left
AlexDaniel bisect: say ‘hi’.indent(9999999999999999999) 02:40
bisectable6 AlexDaniel, Bisecting by exit code (old=2015.12 new=4fd6e94). Old exit code: 1
AlexDaniel, bisect log: gist.github.com/e26ea2e21999e3c9cb...040eeffdcd
AlexDaniel, (2016-07-08) github.com/rakudo/rakudo/commit/58...4545fb17e5
AlexDaniel MasterDuke: ↑, not that the old behavior was any better, but I'm still mentioning it because perhaps you want to be aware of this :)
MasterDuke interesting, i've never touched indent(). i guess it uses x() under the hood 02:41
AlexDaniel committable6: 2016.06,HEA Dsay ‘f’ x 9999999999999999999 02:44
committable6 AlexDaniel, ¦«2016.06»: ===SORRY!=== Error while compiling /tmp/6XSmEtAG6t␤Undeclared name:␤ Dsay used at line 1␤ «exit code = 1»␤¦«HEA»: Cannot find this revision
AlexDaniel committable6: 2016.06,HEAD say ‘f’ x 9999999999999999999 02:45
committable6 AlexDaniel, ¦«2016.06»: repeat count (-8446744073709551617) cannot be negative␤ in block <unit> at /tmp/WttTnuZY7q line 1␤ «exit code = 1»␤¦«HEAD»:
AlexDaniel ya
02:45 aindilis joined
AlexDaniel committable6: 2016.06,HEAD say ‘f’ x -10 02:45
committable6 AlexDaniel, ¦«2016.06,HEAD»:
AlexDaniel MasterDuke: arguably the current behavior is better because the repeat count can actually be negative 02:47
hahaha
02:48 mr-foobar left, FROGGS joined 02:49 mr-foobar joined 02:50 BenGoldberg left 02:51 BenGoldberg joined 02:54 Ben_Goldberg joined, BenGoldberg left, Ben_Goldberg is now known as BenGoldberg 02:57 Ben_Goldberg joined, BenGoldberg left, Ben_Goldberg is now known as BenGoldberg 02:59 labster joined 03:00 Ben_Goldberg joined, BenGoldberg left, Ben_Goldberg is now known as BenGoldberg 03:02 Actualeyes left 03:04 pyrimidine joined 03:05 Sgeo joined 03:06 kurahaupo__ joined, kurahaupo_ left 03:11 Sgeo left
Herby_ Anyone here working puzzles from Advent of Code this year? 03:15
adventofcode.com/
03:18 mrsolo joined, noganex_ joined 03:19 cfa joined 03:21 noganex left
cfa m: S/d// given 'abc' 03:22
camelia ( no output )
cfa m: say S/d// given 'abc'
camelia rakudo-moar 4fd6e9: OUTPUT«abc␤»
cfa m: say S:g/d// given 'abc' 03:23
camelia rakudo-moar 4fd6e9: OUTPUT«()␤»
cfa bug?
(or am i missing something regarding :g?)
AlexDaniel cfa: well… 03:25
bisect: say S:g/d// given 'abc'
bisectable6 AlexDaniel, Bisecting by output (old=2015.12 new=4fd6e94) because on both starting points the exit code is 0
AlexDaniel, bisect log: gist.github.com/28dcb535797f4cd527...a9ad94386b
AlexDaniel, (2016-10-23) github.com/rakudo/rakudo/commit/b7...e8f5c77f41
AlexDaniel cfa: so yes, this is a bug 03:26
cfa AlexDaniel: thanks
also, hello (sorry, i entered rather abruptly)
AlexDaniel cfa: and turns out there was no test for it so we did not catch its appearance
cfa: will you submit a bug report?
cfa sure, where can i find information on filing one?
AlexDaniel huggable: bug 03:27
huggable AlexDaniel, nothing found
AlexDaniel huggable: rakudobug
huggable AlexDaniel, [email@hidden.address] or use perl6 query on rt.perl.org ; see github.com/rakudo/rakudo/#reporting-bugs
cfa AlexDaniel: thanks, i'll do so within the hour
AlexDaniel cfa: basically you just have to write an email to [email@hidden.address]
cfa nods
AlexDaniel cfa: please also note in your report that bisectable says that b7201a8f223 is the cause 03:28
committable6: b7201a8f223^,b7201a8f223 say S:g/d// given 'abc'
committable6 AlexDaniel, ¦«b7201a8f223^»: abc␤¦«b7201a8»: ()
cfa will do
AlexDaniel babydrop: that will be a 6th [REGRESSION] ticket today :) 03:29
cfa: and thanks for coming here, this is actually important 03:30
… though it would've been better if you asked this question one month ago :) 03:31
committable6: 2016.10,2016.11,HEAD say S:g/d// given 'abc' 03:32
committable6 AlexDaniel, ¦«2016.10»: abc␤¦«2016.11,HEAD»: ()
03:35 aborazmeh left 03:42 ChoHag left, stanley joined 03:43 ChoHag joined, FROGGS left, FROGGS joined
cfa AlexDaniel: :) 03:45
tailgate m: my @directions = <N E S W>; say @directions.grep-index('E'); 03:47
camelia rakudo-moar 4fd6e9: OUTPUT«No such method 'grep-index' for invocant of type 'Array'␤ in block <unit> at <tmp> line 1␤␤»
tailgate m: my List @directions = <N E S W>; say @directions.grep-index('E');
camelia rakudo-moar 4fd6e9: OUTPUT«Type check failed in assignment to @directions; expected List but got Str ("N")␤ in block <unit> at <tmp> line 1␤␤»
AlexDaniel tailgate: what's grep-index?
m: my @directions = <N E S W>; say @directions.grep(‘E’, :k); 03:48
camelia rakudo-moar 4fd6e9: OUTPUT«(1)␤»
tailgate doh
AlexDaniel ↑ that's what you want, I guess?
geekosaur the @ already says list; the type is that of elements
(well, it says Positional, but that gets you List by default...) 03:49
03:49 Herby_ left, rindolf left
tailgate yeah 03:50
cfa AlexDaniel: submitted 03:53
AlexDaniel cfa++ 03:56
04:02 labster left 04:03 Sgeo joined, Ben_Goldberg joined 04:04 BenGoldberg left, Ben_Goldberg is now known as BenGoldberg
AlexDaniel m: say <a b c>[*-18446744073709551618] 04:06
camelia rakudo-moar 4fd6e9: OUTPUT«b␤»
AlexDaniel :-|
wait, what 04:08
m: my @a = <a b c>; @a[*-18446744073709551616]:delete; say @a
camelia rakudo-moar 4fd6e9: OUTPUT«[a b c]␤»
AlexDaniel m: my @a = <a b c>; @a[*-18446744073709551617]:delete; say @a
camelia rakudo-moar 4fd6e9: OUTPUT«[a b]␤»
AlexDaniel m: my @a = <a b c>; @a[*-18446744073709551618]:delete; say @a
camelia rakudo-moar 4fd6e9: OUTPUT«[a (Any) c]␤»
04:13 Ben_Goldberg joined, BenGoldberg left, Ben_Goldberg is now known as BenGoldberg 04:14 BenGoldberg left 04:18 itcharlie1 left
ugexe how do I submit my post for the advent? 04:24
04:25 AlexDaniel left
samcv sweet :) i have the generation of all the q[] qq[] and Q[] constructs fully automated 04:30
i have q<< >>, q (( )), q[[ ]], q<< >>, q/ /, q{ }, q ' ', q" " 04:31
what more to add now? now i only have to add them to the list and it'll generate it for me. very happy with this 04:32
no more copy pasting
04:33 xtreak joined 04:35 aindilis left 04:36 Actualeyes joined 04:49 pierre_ left 04:56 pierre_ joined 05:01 pierre_ left
samcv Wow. ok i finally got using any delimiter you want (same char) like q%this%. 05:13
this is coming together really well
05:14 yoleaux left 05:15 parisba_ joined, bisectable6 left, committable6 left, PotatoGim_ joined, benchable6 left, parisba left 05:16 _28_ria left, unicodable6 left, evalable6 left, parisba_ is now known as parisba, zostay joined, _28_ria joined, mephinet joined 05:17 BinGOs joined, PotatoGim_ is now known as PotatoGim 05:20 clkao joined, TeamBlast joined 05:23 dj_goku joined, mithaldu_ joined 05:24 djbkd joined, BuildTheRobots joined 05:25 resol joined 05:29 xinming left, xinming joined 05:31 dylanwh joined 05:32 Sgeo left 05:43 resol left 05:45 FROGGS left, asdasadasdasd joined
asdasadasdasd hi 05:46
05:47 brillenfux joined
asdasadasdasd whats the use of sub usage (;$)? 05:47
samcv ? 05:49
asdasadasdasd, explain? 05:51
05:55 Cabanossi left, telex left, Cabanossi joined 06:05 Tonik joined 06:10 khw left 06:16 dontbe^2 left, dontbe^2 joined 06:17 pierre_ joined 06:18 rumble joined, djbkd left 06:19 rumble is now known as grumble 06:29 dontbe^2 left 06:30 pyrimidine left, perigrin joined, pyrimidine joined 06:31 zostay left, zostay joined 06:32 telex joined, TEttinger left 06:33 TEttinger joined 06:34 xxpor joined 06:35 domidumont joined 06:37 dontbe^2 joined 06:43 Sgeo joined 06:45 domidumont left 06:52 RabidGravy joined
samcv m: sub circumfix:<⟅ ⟆> ( Mu $a ) is raw { bag($a) }; say ⟅'test'⟆ 06:54
camelia rakudo-moar 4fd6e9: OUTPUT«bag(test)␤»
samcv u: ⟅
u: bag delimiter
07:00 bjz joined 07:04 pierre_ left, ufobat joined 07:05 dj_goku_ joined, dj_goku_ left, dj_goku_ joined, Khisanth joined 07:07 pierre_ joined
moritz samcv: re rakudo.org, please /msg me your email address 07:09
samcv awesome
messaged
07:09 dj_goku left, CIAvash joined
moritz invitation sent 07:11
07:11 seatek joined 07:12 ponbiki_ is now known as ponbiki 07:15 domidumont joined
seatek m: sub t(*%h) { say 'hi' }; my $a = 'key1'; my $b = 'key2'; t($a => 1, $b => 2); 07:17
camelia rakudo-moar 4fd6e9: OUTPUT«Too many positionals passed; expected 0 arguments but got 2␤ in sub t at <tmp> line 1␤ in block <unit> at <tmp> line 1␤␤»
seatek m: sub t(*%h) { say 'hi' }; my $a = 'key1'; my $b = 'key2'; t(a => 1, b => 2);
camelia rakudo-moar 4fd6e9: OUTPUT«hi␤»
seatek i had no idea that would happen :)
07:17 kurahaupo__ left
seatek is there a way to pass a variable key name in that form? i may have gone brain dead too. 07:18
m: sub t(*%h) { say 'hi' }; my $a = 'key1'; my $b = 'key2'; t({$a => 1, $b => 2}); 07:20
camelia rakudo-moar 4fd6e9: OUTPUT«Too many positionals passed; expected 0 arguments but got 1␤ in sub t at <tmp> line 1␤ in block <unit> at <tmp> line 1␤␤»
seatek m: sub t(*%h) { say 'hi' }; my $a = 'key1'; my $b = 'key2'; t(($a => 1, $b => 2));
camelia rakudo-moar 4fd6e9: OUTPUT«Too many positionals passed; expected 0 arguments but got 1␤ in sub t at <tmp> line 1␤ in block <unit> at <tmp> line 1␤␤»
seatek m: sub t(*%h) { say 'hi' }; my $a = 'key1'; my $b = 'key2'; t(({$a => 1, $b => 2}));
camelia rakudo-moar 4fd6e9: OUTPUT«Too many positionals passed; expected 0 arguments but got 1␤ in sub t at <tmp> line 1␤ in block <unit> at <tmp> line 1␤␤»
seatek :)
m: sub t(*%h) { say 'hi' }; my $a = 'key1'; my $b = 'key2'; t('a' => 1, 'b' => 2); 07:21
camelia rakudo-moar 4fd6e9: OUTPUT«Too many positionals passed; expected 0 arguments but got 2␤ in sub t at <tmp> line 1␤ in block <unit> at <tmp> line 1␤␤»
07:21 domidumont left
seatek m: sub t(*%h) { say 'hi' }; my $a = 'key1'; my $b = 'key2'; t(a => 1, b => 2); 07:21
camelia rakudo-moar 4fd6e9: OUTPUT«hi␤»
07:22 domidumont joined, yoleaux joined, ChanServ sets mode: +v yoleaux
ufobat m: sub t(*%h) { say dd %h }; my $a = "key1"; my $b = "key2"; t(|{"a" => 1, "b" => 2}); 07:23
camelia rakudo-moar 4fd6e9: OUTPUT«{:a(1), :b(2)}␤Nil␤»
seatek oh, do the thingy! hadn't tried that ;) 07:24
yes that works :) thanks ufobat++ ! :) 07:25
ufobat wow, the first time i could help someone!
seatek ah, the unfurling pipe solves every problem ;) 07:26
well glad it was me! :)
07:26 labster joined 07:28 wamba joined 07:34 _28_ria left 07:37 bjz left 07:38 cibs left 07:39 cibs joined 07:45 darutoko joined 08:01 dontbe^2 left 08:02 dontbe^2 joined, pyrimidine left, pyrimidine joined 08:03 notbenh left, notbenh joined 08:05 jonas joined 08:06 xtreak left 08:07 xtreak joined 08:11 abraxxa joined 08:15 bjz joined 08:19 ponbiki left, dontbe^2 left 08:23 pierre_ left 08:27 ponbiki joined, ponbiki is now known as Guest97524 08:29 pierre_ joined 08:30 seatek left 08:32 dontbe^2 joined, bjz left 08:33 cognominal left 08:34 cognominal joined, zakharyas joined 08:36 zakharyas left 08:38 zakharyas joined 08:46 wamba left 08:48 cognominal left 08:51 bjz joined 08:53 cognominal joined 08:55 xtreak left 08:59 bjz left, g4 joined, g4 left, g4 joined 09:08 pmurias joined
pmurias hi 09:08
09:08 dakkar joined
samcv hmm it seems that not all the same delimiters that are ok for q[] work for #`[ multi line comments. hmm 09:11
m: #`“ comment ” say 'hi'
camelia rakudo-moar 4fd6e9: OUTPUT«hi␤»
09:11 pyrimidine left
samcv hmm or maybe 09:11
maybe i messed up 09:12
yeah they seem to work. never mind
good that saves me trouble
09:13 bjz joined, pyrimidine joined 09:21 Actualeyes left
dalek : 6aa6800 | ab5tract++ | misc/perl6advent-2016/schedule:
Add myself to the calendar

I took the last slot due to tuits and theme (snow on Christmas Eve, no matter where you live). The post I displaced explicitly told me to remove itself from the listing.
09:22
c: 71eb971 | titsuki++ | doc/Language/syntax.pod6:
Delete >>.say
09:30
c: 7b3c1d0 | titsuki++ | doc/Language/syntax.pod6:
Merge pull request #1037 from titsuki/fix-hyper-say

Delete >>.say
09:30 n0xff joined
09:30 bjz left
timotimo yes 09:31
good idea
09:32 xtreak joined 09:41 labster left 09:44 bjz joined 09:49 TEttinger left 09:50 seatek joined 09:55 rindolf joined, aindilis joined
pmurias TimToady: should 2**9999999999999999999999 return Inf or raise an X::Numeric::Overflow exception? 10:00
10:07 TEttinger joined
DrForr X::Numeric::You'veGottaBeKiddingMe 10:11
jnthn X::Numeric::FFS::Man::There'sLessAtomsInTheUniverse 10:12
DrForr There's only 10**86 or so, IIRC.
jnthn m: say 10**86 < 2**9999999999999999999999 10:13
camelia rakudo-moar 4fd6e9: OUTPUT«Numeric overflow␤ in block <unit> at <tmp> line 1␤␤Actually thrown at:␤ in block <unit> at <tmp> line 1␤␤»
jnthn oh heh :)
m: say 10**86 < 2**999999999999999999999
camelia rakudo-moar 4fd6e9: OUTPUT«Numeric overflow␤ in block <unit> at <tmp> line 1␤␤Actually thrown at:␤ in block <unit> at <tmp> line 1␤␤»
jnthn m: say 10**86 < 2**9999999999999999999
camelia rakudo-moar 4fd6e9: OUTPUT«Numeric overflow␤ in block <unit> at <tmp> line 1␤␤Actually thrown at:␤ in block <unit> at <tmp> line 1␤␤»
jnthn m: say 10**86 < 2**999999999999
camelia rakudo-moar 4fd6e9: OUTPUT«Numeric overflow␤ in block <unit> at <tmp> line 1␤␤Actually thrown at:␤ in block <unit> at <tmp> line 1␤␤»
jnthn Wow :)
m: say 10**86 < 2**9999999999
camelia rakudo-moar 4fd6e9: OUTPUT«Numeric overflow␤ in block <unit> at <tmp> line 1␤␤Actually thrown at:␤ in block <unit> at <tmp> line 1␤␤»
jnthn m: say 10**86 < 2**999999
camelia rakudo-moar 4fd6e9: OUTPUT«True␤»
jnthn Yeah, OK, now I see how big that number is :D
10:14 TEttinger left
samcv i don't think it should return Inf.. that just sounds like more trouble 10:18
10:23 sena_kun joined 10:28 AlexDaniel joined
AlexDaniel pmurias: well, TimToady was the one who added X::Numeric::Overflow for this case 10:29
10:29 zakharyas left 10:34 zakharyas joined 10:36 ChoHag left
pmurias AlexDaniel: yes, but the implementation depends on a bug 10:36
AlexDaniel which bug? 10:37
pmurias pow_I is broken 10:39
nqp::pow_I
on all backends (js, moar, jvm) 10:40
in different ways :(
j: say((-1) * 999999999)
camelia rakudo-jvm 8ca367: OUTPUT«-999999999␤»
pmurias j: say((2) * 9999999999999999999998) 10:41
camelia rakudo-jvm 8ca367: OUTPUT«19999999999999999999996␤»
AlexDaniel m: say((-1) * 999999999)
camelia rakudo-moar 4fd6e9: OUTPUT«-999999999␤»
AlexDaniel m: say((-1) ** 999999999) 10:42
camelia rakudo-moar 4fd6e9: OUTPUT«-1␤»
AlexDaniel … right …
pmurias j: say((2) ** 99999999999998) 10:43
camelia rakudo-jvm 8ca367: OUTPUT«Inf␤»
pmurias j: say((2) ** 99999999999997)
camelia rakudo-jvm 8ca367: OUTPUT«-Inf␤»
AlexDaniel by the way, *able6 bots are down because they cannot connect for some reason
pmurias AlexDaniel: ^^ # why is this -Inf? 10:44
AlexDaniel pmurias: no, this is definitely broken
10:45 pierre_ left
AlexDaniel the whole point of the fix was to make sure it does not return shit like this :) 10:45
rt.perl.org/Ticket/Display.html?id=125811 10:46
10:47 pierre_ joined
AlexDaniel pmurias: wait, actually, what is the problem exactly? 10:48
(-1) ** stuff seems to be working fine
and everything else too 10:49
on moarvm at least
not on JVM, but isn't it a problem of jvm? :)
*backend
pmurias m: say((-1) ** 99999999999999999999999999999999999999998); 10:50
camelia rakudo-moar 4fd6e9: OUTPUT«Numeric overflow␤ in block <unit> at <tmp> line 1␤␤Actually thrown at:␤ in block <unit> at <tmp> line 1␤␤»
10:51 pierre_ left
masak pmurias: I think most operations on doubles that overflow return Inf rather than throwing an exception. 10:51
AlexDaniel well, are you saying that this is LTA?
masak no, I'm saying that there's IEEE 754 precedent -- and it makes sense to follow it 10:52
AlexDaniel that was the question to pmurias actually
seatek my first reaction was to return unicode crying cats, but that sounds better probably. 10:53
AlexDaniel (if (-1) ** somethinghuge should throw instead of returning -1 or 1)
masak that one feels to me like a NaN-returning operation 10:54
but my intuition might be off here
DrForr I wouldn't expect there to be any special casing for odd/even integers if the actual number would ordinarily be outside integer range. 10:55
masak I'm probably basing that on "the corresponding sequence doesn't converge", which makes NaN reasonable
DrForr: even in the presence of bigints? is it difficult to find out if a huge bigint is odd or even?
DrForr Well, my reasoning is just that if you're using (-1)**$n, then you're going to be doing something else with $n that will carry *that* into NaN territory. 10:57
10:57 yoleaux left
AlexDaniel m: say((-1) ** 4294967295) 10:58
camelia rakudo-moar 4fd6e9: OUTPUT«-1␤»
AlexDaniel m: say((-1) ** 4294967296)
camelia rakudo-moar 4fd6e9: OUTPUT«Numeric overflow␤ in block <unit> at <tmp> line 1␤␤Actually thrown at:␤ in block <unit> at <tmp> line 1␤␤»
AlexDaniel m: say((1) ** 9999999999999999999999999999999999999999999999999999999999999999999)
camelia rakudo-moar 4fd6e9: OUTPUT«1␤»
11:00 yoleaux joined, ChanServ sets mode: +v yoleaux
AlexDaniel uummmmmmmmmmm 11:01
dalek c: 905377f | Altai-man++ | doc/Type/ (3 files):
Minor fixes: missing whitespace, redundant brackets, redundant `is export`.
11:02 xtreak left
AlexDaniel m: say (42 ** 999).chars 11:02
camelia rakudo-moar 4fd6e9: OUTPUT«1622␤»
AlexDaniel m: say (42 ** 999.0).chars
camelia rakudo-moar 4fd6e9: OUTPUT«3␤»
babydrop asdasadasdasd: that looks like a Perl 5 prototype. Try asking in #perl
AlexDaniel m: say 42 ** 999.0
camelia rakudo-moar 4fd6e9: OUTPUT«Inf␤»
AlexDaniel this is probably expected, but it did surprise me a bit 11:03
DrForr Or put differently, tacking (-1)**$n onto a term isn't likely to matter because the rest of the term based on $n will probably already be overflowing.
11:13 peteretep joined 11:16 ChoHag joined, bisectable6 joined, ChanServ sets mode: +v bisectable6, evalable6 joined, unicodable6 joined 11:17 benchable6 joined, committable6 joined, ChanServ sets mode: +v committable6 11:18 lucasb joined 11:19 cognominal left
arnsholt DrForr: Well, that'd depend on whether the rest of the term is sub-linear in $n or linear or whatever =) 11:19
11:19 committable6_ joined 11:21 benchable6 left, committable6 left
babydrop Man, I wish brian would at least pop into this channel and just drop his code in for 2 seconds before writing his "tips". www.learningperl6.com/2016/12/05/q...-off-base/ 11:23
That sprintf concoction is awful.
And in the comments they uglify it even more to workaround a Failure.
In 2 seconds in this channel, someone would say "hey, how about you write that as @*ARGS[0].parse-base(8).base(2) instead" 11:24
11:24 xtreak joined, cognominal joined
babydrop GRUMBLE! 11:24
</uncoffeenated-rant>
arnsholt parse-base is pretty new 11:25
I remember the discussion of it, a month ago maybe?
11:25 zostay left
babydrop star: "22".parse-base(8).base(2).say 11:25
camelia star-m 2016.10: OUTPUT«No such method 'parse-base' for invocant of type 'Str'␤ in block <unit> at <tmp> line 1␤␤»
AlexDaniel babydrop: I agree
11:25 zostay joined
babydrop committable6_: 2016.11 "22".parse-base(8).base(2).say 11:25
committable6_ babydrop, ¦«2016.11»: 10010
babydrop arnsholt: it's in latest star at least 11:26
The problem is many people view brian as a knowledgeable authority and eat up all his tips as top-shelf goodies
AlexDaniel and it is kinda scary because he is writing a book… 11:27
babydrop At least the book will be reviewed
AlexDaniel 🤷
El_Che If the tone is friendly, I am pretty sure he'll appreciate the input 11:28
babydrop u: 🤷
unicodable6 babydrop, U+1F937 SHRUG [So] (🤷)
El_Che (not "pretty sure", but I hope)
babydrop El_Che: on what? the tip is published already
dalek c/consistent-return-types: 40a97e3 | Altai-man++ | doc/Type/ (2 files):
Changes need to resolve a merge conflict.
c/consistent-return-types: ae816b9 | Altai-man++ | t/return-type.t:
Temporary test file deletion
El_Che babydrop: I mean that the tip can be edited as it's not set in stone 11:29
babydrop Well, I left the .parse-base version in the comments. 11:35
Tho not in "the tone is friendly" :} 11:36
lucasb I don't see, the comments must be moderated 11:37
11:38 travis-ci joined
travis-ci Doc build failed. Altai-man 'Temporary test file deletion' 11:38
travis-ci.org/perl6/doc/builds/181930636 github.com/perl6/doc/compare/88c2f...816b9600c7
11:38 travis-ci left
lucasb book author isolated from collective community knowledge and best practices? 11:39
babydrop lucasb: just for his "tips". The private book repo has several of core hackers added to oversee stuff
lucasb ah, alright. I feels happy with this fact 11:40
babydrop Yeah, the comment says "awaiting moderation". I just wrote The much simpler way to write that is as: @*ARGS[0].parse-base(8).base(2) Instead of that ugly sprintf concoction. .parse-base also handles negative numbers, unlike the :16() radix literal syntax.
CIAvash another example is www.learningperl6.com/2016/11/20/t...-creation/ it may give the idea that it is the way to do it. unfortunately, I replied on twitter instead of commenting. fortunately TimToady mentioned the better way of doing it 11:41
but not everyone is going to read the comments
11:43 stanley left 11:46 stanley joined
pmurias masak: (-1) ** $big-number being NaN seems like the worst option possible 11:48
lucasb BTW, babydrop++ for making syntax-highlighted post 11:51
11:51 xtreak left
babydrop pmurias: why NaN? I see masak mentioning Inf 11:52
oh nm 11:53
11:54 xtreak joined, xtreak left
babydrop the IEEE says overflows give the Inf with sign of the limit, but I don't how we can figure out a sign here, so maybe it is specialcased to something else 11:54
m: say (-125)**(1/3) 11:55
camelia rakudo-moar b2ac4e: OUTPUT«NaN␤»
babydrop rather see this DTRT than NaN :D
m: say (my num $ = -1) ** (my num $ = 999)
camelia rakudo-moar b2ac4e: OUTPUT«This type cannot unbox to a native number: P6opaque, Int␤ in block <unit> at <tmp> line 1␤␤»
babydrop m: say (my num $ = -1e0) ** (my num $ = 999e0)
camelia rakudo-moar b2ac4e: OUTPUT«-1␤»
babydrop m: say (my num $ = -1e0) ** (my num $ = 1.4e308) 11:56
camelia rakudo-moar b2ac4e: OUTPUT«1␤»
babydrop hm, right 'cause we don't have enough precision for the last digits, it'll always be an even number so we can assume a 1 on super large numbers 11:57
oh wait
m: say (my num $ = -1e0) ** ((my num $ = 1.4e308) + (my num $ = 1))
camelia rakudo-moar b2ac4e: OUTPUT«This type cannot unbox to a native number: P6opaque, Int␤ in block <unit> at <tmp> line 1␤␤»
babydrop m: say (my num $ = -1e0) ** ((my num $ = 1.4e308) + (my num $ = 1e0))
camelia rakudo-moar b2ac4e: OUTPUT«1␤»
babydrop Well, the power is calcualated first and it ain't got enough precision still... so... it sorta-kinda makes sense? :D
pmurias babydrop: isn't it -Inf if the base is negative and the exponent is odd?
babydrop: we are talking about bigints here 11:58
broquaint m: sub($x) { "got $x" }((1,2)) 11:59
camelia rakudo-moar b2ac4e: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Variable '$x' is not declared␤at <tmp>:1␤------> 3sub(7⏏5$x) { "got $x" }((1,2))␤»
babydrop pmurias: we'd should check IEEE 754, it goes pretty well into defining powers
broquaint Can anonymous subs have signatures?
babydrop pmurias: it may list -1 ** Inf as a spoecial case 12:00
DrForr m: return sub($x){$x++}
camelia rakudo-moar b2ac4e: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Variable '$x' is not declared␤at <tmp>:1␤------> 3return sub(7⏏5$x){$x++}␤»
babydrop broquaint: put space after `sub`
broquaint m: sub ($x) { "got $x" }((1,2))
camelia ( no output )
babydrop broquaint: otherwise it thinks you're trying to call a function named sub()
broquaint babydrop: Thanks!
DrForr Shouldn't do this while I'm looking at perl5 code :)
babydrop IEEE 754: www.csee.umbc.edu/~tsimo1/CMSC455/I...4-2008.pdf 12:02
search for pow to find the powers stuff
babydrop runs to catch the bus
pmurias babydrop: it's not -1 ** Inf it's -1 ** (number we arbitrarily decided is too big)
12:02 Skarsnik joined
broquaint m: sub ($x) { say "got $x $x.WHAT" }((1,2)) 12:02
camelia rakudo-moar b2ac4e: OUTPUT«got 1 2 1 2.WHAT␤»
babydrop pmurias: and "too big" is Inf with the sign of the limit 12:03
broquaint I'm failing to grok string interpolation :/
babydrop m: sub ($x) { say "got $x $x.WHAT()" }((1,2))
camelia rakudo-moar b2ac4e: OUTPUT«Use of uninitialized value of type List in string context.␤Methods .^name, .perl, .gist, or .say can be used to stringify it to something meaningful.␤ in sub at <tmp> line 1␤got 1 2 ␤»
babydrop m: sub ($x) { say "got $x $x.^name()" }((1,2))
camelia rakudo-moar b2ac4e: OUTPUT«got 1 2 List␤»
babydrop broquaint: you need () on methods
broquaint babydrop: Thanks again :)
babydrop and .WHAT returns a type object which is undefined and doesn't stringify well
broquaint Good to know :) 12:06
babydrop m: say (-1)**Inf 12:14
camelia rakudo-moar b2ac4e: OUTPUT«1␤»
12:14 bjz_ joined
babydrop m: say (-1)**-Inf 12:14
camelia rakudo-moar b2ac4e: OUTPUT«1␤»
babydrop oh right
12:15 bjz left
babydrop pmurias, so I guess the super large power would always result in 1 12:15
pmurias babydrop: no, "too big" is a big integer bigger than 2**32 12:16
12:17 CIAvash left
babydrop I missed the start of this discussion, but how come we are arbitrarily deciding thw number is too big? Doesn't even P5 let you have numbers as big as your ram or something? 12:17
pmurias m: say( (-1) ** 17179869184)
camelia rakudo-moar b2ac4e: OUTPUT«Numeric overflow␤ in block <unit> at <tmp> line 1␤␤Actually thrown at:␤ in block <unit> at <tmp> line 1␤␤»
pmurias babydrop: P5 doesn't have builtin bigints 12:18
babydrop ok
pmurias babydrop: we make that decision to prevent things because 2 ** 17179869184 is too big a number to store naively in ram 12:19
12:20 CIAvash joined
pmurias babydrop: you might actually fit in that one but if the exponent gets a bit bigger it won't fit for sure 12:21
babydrop pmurias, yet it behaves like I described: treating uber large numbers as Inf i.imgur.com/Bw3oPvh.png
12:22 skids joined
babydrop the first one that still gave -1 seems to be even bigger than what ours dies with 12:25
pmurias babydrop: P5 turns big integers into doubles
babydrop ah
ok
ggoebel slightly off topic question... a while back someone posted a url to a web based "learn to program" site which taught a programming (javascript I think) by having you learn enough code to escape increasingly difficult ascii art prisons. Does anyone remember what that site was? I think one of my teenagers might really like that approach. 12:26
pmurias babydrop: I'll just fix (-1)**$big-number to return the proper answer 12:27
babydrop pmurias, which one is it? 12:28
12:30 kurahaupo__ joined
moritz 1 if $big-number is even, -1 if it's odd 12:34
babydrop so even if the number is too big to handle we still can figure out its oddness? weird 12:37
or should I say that's... odd :D 12:39
12:41 cibs left
DrForr Well, we'd need the string representation, figure out its base and work from that; I ... won't make the other comment(s) that I ordinarily would. 12:42
babydrop What comments?
ggoebel found it... alexnisnevich.github.io/untrusted/ 12:43
12:43 cibs joined 12:44 canopus left 12:45 mawkish joined
DrForr returns to watching www.youtube.com/watch?v=_wfR4hRE9zk while trying to get some work done. 12:47
12:47 AlexDaniel left 12:49 canopus joined, cognominal left 12:50 cognominal joined
El_Che DrForr: it's safer than doing some glass blowing while trying to look at your code 13:01
DrForr With my code? Not by much. :) 13:02
13:05 pierre_ joined
ggoebel safer than a tesla lightning gun... youtu.be/_fTC_Ud_k3U 13:05
skip to 5m30s
13:10 pierre_ left, lizmat joined, AlexDaniel joined 13:11 lizmat_ joined
lucasb can the output of 'perl -h' be de-indented 4 spaces? 13:12
github.com/rakudo/rakudo/blob/nom/...er.nqp#L66
just strip the 4 space indentation from that string 13:13
or use a heredoc, if the literal code layout matters
babydrop lucasb: would you submit a PR? 13:14
lucasb idk if heredoc was not used in the first place because of some optimization or limitation of using heredocs in NQP
babydrop: which version do you prefer? stripped indetation with quotes or heredocs?
ilmari
.oO( indented heredocs? )
13:15
13:15 lizmat left
lucasb about .indent($negative-number) 13:16
babydrop lucasb: I don't think heredocs work in nqp (or I've just tried wrong). I'd be OK with the message just moved in source code flush to the left side 13:17
nqp: " x".indent(-1)
camelia nqp-moarvm: OUTPUT«Cannot find method 'indent': no method cache and no .^find_method␤ at <tmp>:1 (<ephemeral file>:<mainline>)␤ from gen/moar/stage2/NQPHLL.nqp:1536 (/home/camelia/rakudo-m-inst-2/share/nqp/lib/NQPHLL.moarvm:eval)␤ from gen/moar/stage2/NQPHLL.nqp:1739 (/ho…»
Skarsnik hello, here some result of my perf test: gist.github.com/Skarsnik/e2d531f10...646ad1d89f not sure how useful is that x)
lucasb m: say " foo\n bar".indent(-2)
camelia rakudo-moar b2ac4e: OUTPUT«foo␤bar␤»
[Coke] ggoebel++ # was probably me. I've lost a few hours to that site. :)
lucasb m: say " foo\n\n bar".indent(-2)
camelia rakudo-moar b2ac4e: OUTPUT«foo␤␤bar␤»
babydrop lucasb: src/Perl6/Compiler.nqp is nqp code, tho, isn't it? 13:18
lucasb babydrop: yes, nqp. I'll just strip the indentation then
babydrop lucasb: I guess I should've asked first... how come we're de-denting it?
lucasb babydrop: sorry, didn't understand. was a joke? :) 13:19
13:19 lukaramu joined
babydrop lucasb: lol, no. I mean what's the reason for this change? To make prettier or was there another reason? 13:20
13:20 lizmat_ left
lucasb well :) I pay attention to the output of "program --help" messages 13:20
babydrop heh 13:21
lucasb: OK. Well, submit a PR :)
lucasb when I saw the 4 spaces indentation... it looked strange
but yes, just a stylistic, cosmetic change
babydrop Sounds good 13:22
lucasb back to negative de-indent...
m: say " foo\n \n bar".indent(-2)
camelia rakudo-moar b2ac4e: OUTPUT«Asked to remove 2 spaces, but the shortest indent is 1 spaces␤ in block <unit> at <tmp> line 1␤foo␤␤bar␤»
13:22 AlexDaniel left
lucasb " foo\n\n bar".indent(-2) # empty line, no warnings 13:23
" foo\n \n bar".indent(-2) # line with white space, warnings
babydrop Skarsnik: yeah, lizmat++ did lots of magicks around 0.6 and we saw a huge drop: twitter.com/zoffix/status/796810512986238978
lucasb: that makes sense to me. 13:24
(the warning)
lucasb ok, for a second I thought about being more permissive with blank lines made just of whitespace characters
babydrop is getting distracted again with IRC /o\ 13:25
lucasb but, it's ok, I accept the warning
Skarsnik babydrop, I think it's still very slow for what I am testing ^^
13:26 AlexDaniel joined
Skarsnik 1.2 sec to create 6000 objects is not that great x) (for the gumbo test) 13:26
babydrop Skarsnik: yup. Would be nice to have more hands doing core hacking.
Skarsnik and I will probably use a libxml2 binding for gptrixie because 20 sec is annoying when debugging stuff x) 13:27
pmurias news.perlfoundation.org/2016/12/gra...erl-6.html - brain d foy's grant proposal 13:31
moritz uhm, what? 13:32
the kickstarter I can understand
but kickstarter *and* tpf grant?
DrForr Uh... 13:33
lucasb With all this much funding, I hope the book will be freely downloadable from the internet
DrForr falls out of his chair laughing.
AlexDaniel O_o
13:34 pyrimidine left
moritz that's just not good style 13:34
13:34 cognominal left, nicq20 joined
babydrop that's bullshit 13:35
moritz first doing the kickstarter, saying "hey, I need some money so that I can write a Perl 6 book". So people fund it. And then, "oh, btw, I need money to write a Perl 6 book"
13:35 wamba joined
moritz will sleep over it for a day before commenting on the blog 13:36
lucasb will both this book and the "Think Perl 6" one be published by oreilly then?
AlexDaniel can we instead give this money to gfldex for his work on the docs?…
babydrop Indeed.
13:36 cognominal joined
babydrop And based on brian's "tips" most of the work for the book will be done by, ahem, "self-selected backers" 13:36
lucasb (well, oreilly has released some free books in the past)
AlexDaniel sorry, but this is fucked up 13:37
moritz or me, to motivate by book efforts some more? :-)
babydrop Yeah
Well, unlike the moritz's more rational approach, I will comment right now with the emotions driving me pen!
AlexDaniel babydrop: actually there is some text trying to explain the situation 13:38
moritz there is 13:39
but it's still fucked up
babydrop It's the same bullshit he peddled when people asked him wtf he was kickstarting a book when O'Reilly is publisher
moritz the kickstarter was there to raise the money to write the book
the kickstarter is funded
babydrop Yes.
moritz why can't that be enough to write the book?
if it's not, then that's brian's error, and not something the TPF should square for him 13:40
AlexDaniel
.oO( more money = better book? )
u: facepalm
u: face palm
babydrop And why was the proposal made the day after kickstarter got funded? He waited for it, to ensure people were on the hook for 37 grand before being pissed off
unicodable6 AlexDaniel, Found nothing!
AlexDaniel, U+1F926 FACE PALM [So] (🤦)
AlexDaniel how many grants were rejected? 13:41
moritz a few 13:42
lucasb moritz: the hyperlinks in the book mailing lists came all changed by click tracking, no?
moritz lucasb: yes, I'm afraid; the mailchimp free tier doesn't allow me to disable that :/
lucasb moritz: ok
pmurias babydrop: the TPF didn't always publish grant proposal instantly in the past 13:43
babydrop pmurias: noted 13:45
seatek i agree that would be a bad allocation of money. much better spent paying for work on core stuff
lucasb moritz: your blog posts and consequently the book you intend to write will be all written in POD5 (or Pod6) ? 13:46
13:46 sufrostico left 13:47 sufrostico joined 13:49 prammer joined
perlpilot er ... what? 13:49
I think I would rather TPF give $$$ to moritz or gfldex or some *other* voice writing Perl 6 docs/books/tutorials/etc rather than give a big chunk to bdf who already has a kickstarted funded. 13:50
moritz lucasb: nope, Markdown so far
13:50 AlexDaniel left
mspo I thought asciidoc was the go-to for books 13:51
lucasb moritz: ah, ok
seatek i agree. or why not spend 1/5th that amount and get someone to come up with a manageable installer that's easy for people, to do as a local user or system-wide. That's like a fundamental operational thing that should be there for people so they can actually deploy stuff
moritz seatek: want to write one? apply for a grant!
I'd support *that*
making Rakudo reloctable
13:51 cdg joined
seatek i would feel guilty taking money for something liket hat 13:52
moritz mspo: it might be, but LeanPub supports markdown + git + results in Dropbox + selling through the website, and I didn't want to take ownership of any toolchain myself
13:52 AlexDaniel joined
arnsholt is picky enough that for a book it'd be RST, plus custom plugins to do TikZ figures 13:53
mspo moritz: I think oreilly and no-starch are both asciidoc based
babydrop what's the word for the typical way to write code?
mspo moritz: markdown is a little.. lacking?
arnsholt (Or just writing LaTeX directly, but that's less useful for producing e-versions)
mspo babydrop: canonical 13:54
babydrop mspo: thanks!
arnsholt Idiomatic perhaps?
mspo idiomatic is also good
arnsholt Canonical sounds a bit overly "one true way" to me
mspo normative
arnsholt That's the word! =D 13:55
mspo arnsholt: I have exactly the same reaction to "idiomatic" :)
DrForr Back when I was originally working with ORA they were accepting LaTeX but that had to be converted to PDF or Word through their markup chain.
babydrop mspo: oh idiomatic! yeah
arnsholt TBH, I prefer writing LaTeX over markdown and friends 13:56
mspo golang loves "idiomatic"
sena_kun is there a list of characters/escape sequences that can be used to wrap a string literal? 13:59
seatek for the book grant -- he says books sales aren't what they used to be. that means people aren't buying books as much. so why should tpf pay for books to be written about p6? he kinda gives the best argument against funding that grant himself 14:00
DrForr sena_kun: Sure, there's a section called 'quoting' in the docs.
[Coke] note that the timing on the grant proposal is out of bdf's control. there is a gap between submittal and publication.
sena_kun DrForr, ah, thanks. 14:01
14:01 pierre_ joined
DrForr [Coke]: Yes, having written at least one myself there's a process that's been gone through here. 14:01
14:02 prammer left
moritz arnsholt: I like latex for scientific stuff, but I find it painful when using many code examples 14:03
also, if the toolchain is under my control, I'm under constant compulsion to tweak the result
when it's not, there's only so much I can do, and I spend the rest of the time actually writing stuff :-)
14:04 synopsebot6 joined 14:05 prammer joined
[Coke] DrForr: yes, I was addressing others in the backlog who were concerned with the timing. 14:06
14:06 aindilis left
DrForr Ah. 14:07
[Coke] (thanks for confirming)
nicq20 So, I remember a small discussion here about using a ':' in place of '{ }' on certain things. Is that mentioned anywhere in the docs? 14:09
Can't seem to find the discussion though... 14:10
timotimo : doesn't replace { }, it replaces ( )
nicq20 timotimo: That explains why I'm having some trouble understanding it.... 14:12
[Coke] Note that I've seen very few grant requests come in for perl 6 projects through the normal grant request process. (most have been tied to the one-time Hague grant, or to the new core maintenance fund)
nicq20 timotimo: Is it mentioned in the docs though? 14:13
timotimo i don't know 14:14
moritz m: say "foo\x[ab]" 14:15
camelia rakudo-moar b2ac4e: OUTPUT«foo«␤»
moritz m: say "foo\x[bar]"
camelia rakudo-moar b2ac4e: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Unrecognized backslash sequence: '\x'␤at <tmp>:1␤------> 3say "foo\x[ba7⏏5r]"␤ expecting any of:␤ argument list␤ double quotes␤ hex character␤ term␤»
nicq20 So if ':' replaces '( )', why can I do something like `do: say "hi";`? 14:17
seatek nicq20 it's like for function/method calls -- $these.things: 4, 5, 9; 14:18
and other stuff. it's kinda mashed up around all over ;)
nicq20 Huh, ok. TIMTOWTDI? 14:19
arnsholt moritz: -Oproductivity is definitely an excellent idea! =)
moritz nicq20: : doesn't generally replace (), it's just for methods calls 14:20
nicq20 So if it's just for method calls why does it work for other things? 14:21
seatek well, why does a '-' work for more than one thing?
pmurias .seen TheLemonMan
yoleaux I saw TheLemonMan 6 Sep 2016 14:07Z in #perl6: <TheLemonMan> ugexe, zef's working perfectly so far :)
nicq20 seatek: Because it's overloaded I guess? 14:22
14:23 pyrimidine joined
nicq20 seatek: Oh! nm, I see what you mean. 14:23
seatek colons can be used for stuff like $this<key>:exists ??
or for my::class::here 14:24
just meant to make stuff look easy to read where it makes sense, for you and the compiler
sometimes $this.thing: here, there; looks better, and sometimes $this.thing(here, there) looks better or is more readable 14:25
nicq20 seatek: Hmm. Ok, that makes sense. I might try to write some docs for it then. Any ideas as to where in the code to start looking for it? 14:26
seatek no, i'm a USER ;)
14:27 pyrimidine left
moritz nicq20: src/Perl6/Grammar.pm 14:28
14:28 pyrimidine joined
moritz possibly postfixish is a good starting point 14:28
nicq20 moritz: Thank you!
seatek i love being a user. it's so relaxing 14:29
someone bring me a mai tai
nicq20 seatek: ikr
moritz nicq20: acutally, token methodop
seatek :) 14:30
babydrop nicq20: run some code with --target=parse, it'll list tokens in src/Perl6/Grammar.nqp 14:31
seatek does rakudo have to worry about building/installing on the BSDs? 14:32
babydrop what do you mean by "have to worry"? :)
[Coke] Still 2-3 gaps in the advent calendar. First gap is 2-3 days out!
seatek hehe i was going to leave that up for interpretation ;)
moritz babydrop: ah, good idea. In this case it seems to confirm my suspicion that methodop parses the method call
babydrop seatek: we support as many platforms as there are volunteers willing to make it work on
moritz seatek: if I remember correctly, rakudo works on FreeBSD 14:33
seatek i don't know anything about their hierarchies
14:33 pyrimidine left
seatek nor very little about the windowses 14:33
[Coke] It would be great if people who are booked for days 12-24 could move their post sooner so the gaps are later, giving us more time to fill them.
(or maybe just 12-23)
moritz [Coke]++ # ringing the advent calendar drum 14:34
seatek is most of the current install base linux-y?
14:34 cognominal left
moritz seatek: yes. And some Windows, some MacOS 14:35
babydrop [Coke]: I have an already-written post ready to go for any spot. I just don't list it on the schedule to make empty spots visible.
So if there's a gap, I can fill it.
14:35 sufrosti1o left
dalek : e8d1490 | (Will Coleda)++ | misc/perl6advent-2016/schedule:
remove published note on day 1
14:35
seatek moritz: i tried recently on windows and it was very problemmatic to get a good environment that included git as well, which seems necessary and is separate
14:36 sufrostico left
moritz seatek: I know; windows for developers is still a bit PITA 14:36
seatek: that's why I'm glad there rakudo .msi's for download
14:37 sufrostico joined 14:38 huggable joined, sufrosti1o joined
seatek moritz: yeah. i'll try playing around with some linux-based install choices if i get the time, just to see what's involved. i have a feeling with all the insane compile cacheing that it's not super simple 14:38
and with that kind of cacheing, permissions are certainly an issue. 14:39
moritz seatek: one of the real blockers for a versatile installer is that rakudo isn't relocatable
that is, you need to know where it's going to be installed *by the time you compile rakudo*
14:40 lizmat joined
seatek this might be a little fun. i'll need to tackle it here for myself anyway 14:40
eventually
might be a good excuse to finally poke at some bsd too 14:42
14:42 cognominal joined 14:45 pyrimidine joined 14:46 AlexDaniel left, prammer left, CIAvash left 14:47 CIAvash joined 14:48 prammer joined, Actualeyes joined, asdasadasdasd left 14:49 abraxxa left 14:50 pyrimidine left, _ramix_ joined
babydrop m: say 43.pick(-1) 14:52
camelia rakudo-moar b2ac4e: OUTPUT«MVMArray: Index out of bounds␤ in block <unit> at <tmp> line 1␤␤»
babydrop m: say 43.roll(-1)
camelia rakudo-moar b2ac4e: OUTPUT«(43 43 43 43 43 43 43 43 43 43 43 43 43 43 43 43 43 43 43 43 43 43 43 43 43 43 43 43 43 43 43 43 43 43 43 43 43 43 43 43 43 43 43 43 43 43 43 43 43 43 43 43 43 43 43 43 43 43 43 43 43 43 43 43 43 43 43 43 43 43 43 43 43 43 43 43 43 43 43 43 43 43 43 43 43 …»
babydrop Throw or treat as zero?
moritz throw
babydrop OK.
moritz in fact, I'd do it as a where-constraint on the argument 14:53
babydrop tis slow :(
m: for ^100_000 { sub (Int) {}(42) }; say now - INIT now; 14:54
camelia rakudo-moar b2ac4e: OUTPUT«0.11964007␤»
babydrop m: for ^100_000 { sub (UInt) {}(42) }; say now - INIT now;
camelia rakudo-moar b2ac4e: OUTPUT«0.5588977␤»
babydrop m: for ^100_000 { sub (Int $x) {?($x > 1)}(42) }; say now - INIT now; 14:55
camelia rakudo-moar b2ac4e: OUTPUT«0.12155607␤»
babydrop >= 0 I mean
jnthn You'd need to include an actual throw to really compare though
timotimo and invoke the throw, too, not just go via the good path :)
jnthn Well, not invoke it on the good case :)
babydrop jnthn: but I'm comparing the common good case
_ramix_ Hi. I have the modules.perl6.org/dist/FileSystem::Capacity module in the ecosystem. I would like to add a vbs script in the /bin directory to be able to run it from the same module with run or shell. It's possible? How could I do it? Thanks!
timotimo :D
babydrop It becomes 5 times slower, just to exclude the uncommon bad case 14:56
moritz :(
jnthn babydrop: For sure, but there's a tad more to it than that.
timotimo we could teach the binding lowerer about unsigned ints - might want to introduce a new primitive for that? 14:57
jnthn Firstly, a bit big of the cost difference is probably inlining related
timotimo hm, what were they called again
jnthn *big bit
moritz _ramix_: what's a vbs script?
timotimo visual basic script
moritz why not just a Perl 6 script?
you can also call that from the shell
brrt so it's for windows, it'd need to be added to $PATH
moritz and there's tooling for that
_ramix_ visual basic script to acces to filesystem object in win32 14:58
jnthn There where costs that, but the pick is likely too big for that in the first place
[Coke] "don't mince words, bones, what do you REALLY think?"
jnthn Not saying `where` is cheap, just that compared in the real scheme of things it may be a few percent.
brrt ultimately, this is about our module installers. and I don't know if they can do what you ask
timotimo don't mince bones?
moritz don't mend bones 14:59
babydrop jnthn: OK, I'll stick a UInt on it then. If you ignore any performance differences, it's a much cleaner solution
moritz or so :-)
15:00 AlexDaniel joined
lucasb I just read the comment on the grants post 15:01
I'm not sure how harsh it was, but I'm glad we have someone passionate about p6 :) 15:03
15:04 kurahaupo__ left 15:05 cognominal left
sena_kun Why does gist.github.com/Altai-man/78ead16e...c30a14fa1b throws an exception now even if the output was clear(i.e. nothing was written to stderr afaik)? 15:05
15:05 sufrostico left
moritz sena_kun: what exception? 15:05
15:06 sufrosti1o left
sena_kun moritz, not an exception, rather something was tapped from stderr. 15:06
15:06 skids left, pyrimidine joined
babydrop sena_kun: output shows it's just an empty string 15:06
Maybe that's a quirk of ls?
15:07 sufrostico joined
nicq20 moritz: So, I was wrong (as usual). When calling `do: 1+1`, 'do:' is seen as a label and never actually called. 15:07
moritz sena_kun: so what exactly is tapped from stderr?
sena_kun babydrop, it can be, but I tried other commands previously(firstly, it was call to perl6 binary)...
moritz, an empty string.
pmurias what does 'say 2**10000000' in rakudo-moar currently do on OS X? 15:08
15:08 sufrosti1o joined
babydrop sena_kun: oh, yeah, just tried with perl6 -e 'say "hi"'. That's a bug, would you please report it? 15:09
sena_kun babydrop, I'll try, but I might ask some help about it. (:
15:10 brillenfux left, khw joined
babydrop "Proc::Asyn sends empty trying to stderr tap on program exit, even though the program did not send anything' 15:10
15:10 cdg_ joined
babydrop sena_kun: oh, and stdout tap too 15:11
same thing.
Maybe it's that slurp-rest: :close thing we added awhile back?
15:11 pierre_ left
sena_kun babydrop, yes, I'll create a ticket now, just some time to register is needed... 15:11
15:12 cognominal joined
babydrop on RT? The New Ticket button is disabled. Just email to [email@hidden.address] 15:12
15:13 cdg left
sena_kun Ah, okay. 15:13
lucasb funny
15:13 sufrostico left, sufrosti1o left
lucasb slurp-rest was added exactly because people didn't like slurp(:close) 15:14
babydrop weird
lucasb and later, slurp-rest got its own :close parameter 15:15
15:15 _ramix_ left, sufrostico joined
lucasb > await Proc::Async.new('foo').start 15:15
===SORRY!===
no such file or directory
15:15 sufrosti1o joined
babydrop :( 15:16
lucasb would be interesting to output program name or command line that was unsuccessful
babydrop Also bug
lucasb: there actually is logic for that :(
I recall bikeshed with the @.command attribute
lucasb and also... I thought "===SORRY!===" was for compile-time errors? how does it shows at runtime?
ugexe that was for Proc
babydrop ugexe: but Proc::Async uses Proc udner the hood 15:17
And I recall propagating that command stuff for Proc::Assync
lucasb: would you bugreport that, please? Just email to [email@hidden.address]
dalek /titsuki-advent: 6e0deb2 | titsuki++ | misc/perl6advent-2016/schedule:
Add mysef as a 11th writer

I have already blogged the Japanese version of this post in Qiita for an unofficial Perl 6 advent calender. Where Qiita is a blog service mainly for Japanese engineers:
  qiita.com/advent-calendar/2016/perl6
  qiita.com/titsuki/items/1ad338c6cf14a59be3e8
So I'll write the English version for this advent calendar.
babydrop wooohoo \o/ titsuki++ 15:18
sena_kun titsuki++
nicq20 pmurias: FYI that `say 2**10000000` is still running.
timotimo cool
15:18 kurahaupo__ joined
sena_kun babydrop, code quoting in the mail is just ">"? 15:19
babydrop sena_kun: just put it as is. Makes it easier to copy-paste 15:20
sena_kun babydrop, thanks.
15:20 cibs left 15:21 cibs joined 15:22 nicq20 left, ufobat left
sena_kun Reported. 15:25
babydrop sena_kun++
15:26 AlexDaniel left
lucasb it's not just stderr. it also sends a empty line to stdout in the end 15:26
15:27 CIAvash left
sena_kun lucasb, I mentioned that in the title and the description too. (: 15:27
Also, is there a chance for me to try to patch it? 15:28
lucasb sena_kun: oh, right. now I noticed. thanks
15:29 AlexDaniel joined
lucasb my $p = Proc::Async.new('bash', '-c', 'echo out; echo err 1>&2'); 15:30
$p.stdout.tap(-> $x { say "out: $x" });
$p.stderr.tap(-> $x { say "err: $x" });
^^ just sharing another way to show the behaviour
babydrop sena_kun: sure! The code's in github.com/rakudo/rakudo/blob/nom/...c/Async.pm
dalek : 6e0deb2 | titsuki++ | misc/perl6advent-2016/schedule:
Add mysef as a 11th writer

I have already blogged the Japanese version of this post in Qiita for an unofficial Perl 6 advent calender. Where Qiita is a blog service mainly for Japanese engineers:
  qiita.com/advent-calendar/2016/perl6
  qiita.com/titsuki/items/1ad338c6cf14a59be3e8
So I'll write the English version for this advent calendar.
15:31
: b42b969 | (Will Coleda)++ | misc/perl6advent-2016/schedule:
Merge pull request #23 from perl6/titsuki-advent

Add mysef as a 11th writer
babydrop sena_kun: you can also try bugging AlexDaniel to run that code on bisectable to figure out what commit introduced that behaviour
sena_kun babydrop, thanks! I'll try to dig in then. 15:32
ping AlexDaniel
15:32 CIAvash joined
[Coke] we now have 1-2 gaps left! 15:33
babydrop sena_kun: this may be a candidate: github.com/rakudo/rakudo/commit/e4...8b78321c8f
So you could try building a version prior to that commit (or just change that Supplier::Preserving back to Supplier and see if the issue goes away). And if it does, we'd need to figure out how to address that commit and the current issue 15:34
moritz titsuki++
15:34 prammer left
moritz [Coke]: how is your post coming along? 15:34
timotimo it could be we're giving the empty string from inside moarvm 15:35
[Coke] moritz: AHAHAHAHAHAHAHAAHAHAHAH
I'm sure I'll finish something in time. :)
timotimo that sounds ... bad?
babydrop [Coke] has a secret plan :}
15:35 skids joined
[Coke] I have 3 paragraphs written towards a fresh impl. of a perl 6 module re-impl of a brand new p5 module. 15:35
babydrop $ ./perl6 -e '42.pick: -1' 15:36
Coercion to UInt out of range. Is: -1, should be in 0..Inf
in block <unit> at -e line 1
Is that good 'nuf?
15:36 n0xff left
babydrop It basically can take an Inf or Whatever too, so the conversion into an Int is deeper in the guts 15:36
15:36 wamba left
babydrop I made it convert to UInt instead. 15:36
I just imagine AlexDaniel creating a new ticket saying the message is still LTA :P 15:37
jnthn babydrop, sena_kun: I suspect the empty string comes from github.com/rakudo/rakudo/blob/nom/...der.pm#L36 15:40
babydrop oh cool 15:41
jnthn Also, that XXX comment is no longer true 'cus I fixed that nasty bug :)
sena_kun good.
jnthn But the emit will happily spit out an empty string there
The one on line 29 actually does check 15:42
tbh I'd kinda expect code to be robust enough to cope, but I'm happy to tweak this not to emit empty strings :)
babydrop m: multi foo($) { nextwith 42; say "meow" }; multi foo (Int) { say "ha!" }; foo 'x' 15:43
camelia ( no output )
babydrop How come no output?
m: multi foo($) { callwith 42; say "meow" }; multi foo (Int) { say "ha!" }; foo 'x'
camelia rakudo-moar 5bacb0: OUTPUT«meow␤»
babydrop m: multi foo($) { nextwith 42; say "meow" }; multi foo (Int) { say "ha!" }; foo 42
camelia rakudo-moar 5bacb0: OUTPUT«ha!␤»
babydrop weird 15:44
sena_kun jnthn, just give me some time and wait 'til I will return here with (no?) result. (:
jnthn babydrop: Because nextwith iterates a pre-determiend candidate list
(Determined at the point of dispatch)
Use samewith to redispatch
15:45 prammer joined
babydrop Oh. Thanks. jnthn++ 15:45
m: multi foo($) { samewith 42; say "meow" }; multi foo (Int) { say "ha!" }; foo 'x' 15:46
camelia rakudo-moar 5bacb0: OUTPUT«ha!␤meow␤»
15:46 zakharyas left
babydrop hmm 15:46
m: multi foo (Int) { say "ha!" }; multi foo($) { samewith 42; say "meow" }; foo 'x'
camelia rakudo-moar 5bacb0: OUTPUT«ha!␤meow␤»
babydrop m: multi foo (Int) { say "ha!" }; multi foo($) { nextwith 42; say "meow" }; foo 'x'
camelia ( no output )
babydrop I was hoping to redispatch without returning 15:47
so the output would be just "ha"
15:47 ufobat joined
babydrop oh well 15:48
Well, I don't entirely get it, I guess. 15:49
nextwith just moves further in the candidates list? So it'll only try any candidates that haven't been tried for that call already? 15:50
moritz I think so, yes
babydrop OK
jnthn Conceptually, it works like this: when you do a multi dispatch, instead of just calculating the best candidate, we also calculate all possible candidates that would have matched. nextwith takes you do the next one in that list. 15:51
*to 15:52
babydrop Makes sense. Thanks.
pmurias jnthn: I made a pull request to fix pow_I on moar 15:55
jnthn: rakudo also needs to be fixed and I'll start working on that too 15:56
babydrop \o/
jnthn pmurias: Thanks 15:58
babydrop pmurias: if you've got tuits, pow_In also needs some lovin' and there are three tickets affected by it. I found one fix (github.com/MoarVM/MoarVM/pull/442) but I don't understand WHY it fixes it, so I closed it in fear it'd be breaking other things.
Well, I understand why it fixes it, but I don't understand what that code is for and what my change may be breaking.
jnthn pmurias: Will look shortly 15:59
babydrop Well, you were right. The pick with UInt candidate is not slow and is even about 4% faster. But the error message sucks: gist.github.com/zoffixznet/32b9f1a...3ea09a9a8f 16:06
Because it doesn't tell me how to fix it. 16:07
moritz babydrop: does the error message improve if you add an Int candidate with a where clause?
oh, that might give an ambiguous dispatch
babydrop Well, UInt *is* an Int with a where clause. It's a subset. And the candidates in the error actually list it as `Int where { ... }` candidate which makes the error that much more confusing 16:08
And `Inf` candidate is listed as `Num $ where { ... }` :/
16:11 sena_kun left
dalek osystem: a2ed138 | (Shoichi Kaji)++ | META.list:
Add WaitGroup to ecosystem

See github.com/skaji/perl6-WaitGroup
16:13
jnthn pmurias: The patch seems legit :) 16:14
Will let the OSX travis tests have at it before merging :) 16:15
16:15 g4 left
dalek : 21eea09 | (Will Coleda)++ | misc/perl6advent-2016/schedule:
move brrt++ up to the 9th.
16:18
16:18 seatek left
[Coke] brrt++ 16:18
so now we have 1-2 gaps, at 5 days out and maybe 3 days out. 16:19
Hopefully the 12th is far enough out that someone could consider writing something from scratch in that time.
jnthn I'll see how soon I get my one ready 16:26
(No promises! :))
(Other than to have it at latest by when I already signed up for)
16:26 mawkish left 16:27 AlexDani` joined
babydrop AlexDani`: you around? 16:28
AlexDani` yes
came here because somebody wanted to bug me :)
babydrop lol
AlexDani`: what's your opinion on this error: gist.github.com/zoffixznet/32b9f1a...3ea09a9a8f
AlexDani` babydrop: hmmm, why is it an error? 16:29
16:29 telex left, mawkish joined
AlexDani` should it be an error? 16:29
babydrop AlexDani`: -1 is not a valid thing to give to .pick
Well, what should it do? 16:30
16:30 yqt joined, telex joined
babydrop AlexDani`: another alternative impl lent me this error: "Coercion to UInt out of range. Is: -1, should be in 0..Inf" 16:32
AlexDani` m: say (42,50,60).roll(0) 16:33
camelia rakudo-moar 5bacb0: OUTPUT«()␤»
AlexDani` m: say (42,50,60).roll(-1) 16:34
camelia rakudo-moar 5bacb0: OUTPUT«(60 50 50 60 42 42 50 60 50 60 60 42 60 42 60 50 50 42 42 42 42 60 50 50 42 60 42 42 60 60 50 60 60 50 50 50 60 42 60 42 60 50 42 50 60 60 60 42 42 60 42 42 60 50 60 60 42 42 42 42 42 60 50 50 50 42 42 50 50 50 60 50 42 50 60 50 50 50 42 50 50 42 50 42 42 …»
AlexDani` ah right
geekosaur mm. "rakudobrew triple" (even after purging the existing one) dies after installing nqp because nqp-m has a relative path for ../install/bin/moar that isn't correct for the path rakudo's Configure.pl is run from
AlexDani` babydrop: if I recall correctly, some things actually produce empty output when you give them negative values
m: say 42 xx -10
camelia rakudo-moar 5bacb0: OUTPUT«()␤»
AlexDani` like this, I guess?
geekosaur suspects this is not rakudobrew's fault; any relative path like that is questionable at best
AlexDani` babydrop: so I don't really know if it should throw or not 16:35
babydrop Yeah, that's kind of inconsistent :/
AlexDani` babydrop: if you want it to throw, then “none of these signatures match” is definitely LTA
but perhaps a good step forward anyway
babydrop Well, my original gut reaction was to treat it as zero. Then someone in the channel suggested the throwage 16:36
AlexDani` babydrop: ask TimToady?
babydrop m: [][-1] # I mean, this throws
camelia rakudo-moar 5bacb0: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Unsupported use of a negative -1 subscript to index from the end; in Perl 6 please use a function such as *-1␤at <tmp>:1␤------> 3[][-1]7⏏5 # I mean, this throws␤»
AlexDani` it throws because people expect it to index from the end
babydrop m: ''.match: /^/, :nth(-1); # I mean, this throws 16:37
camelia rakudo-moar 5bacb0: OUTPUT«Attempt to retrieve before :1st match -- :nth(-1)␤ in block <unit> at <tmp> line 1␤␤»
16:37 domidumont left
babydrop Kinda like we need to plan this stuff out more. Some sort of grand "Perl 6's Awesome Failure Mode Plan" 16:38
That deliniates all of such cases and tells what we do.
And deliniates the awesomification level all errors need to achieve.
AlexDani` so what code should I run on bisectable? 16:39
babydrop AlexDani`: oh. 16:40
AlexDani`: I think we found it actually :)
AlexDani`: but it was this ticket: rt.perl.org/Ticket/Display.html?id...et-history
AlexDani` ah
wow, that's an interesting ticket 16:41
16:45 CIAvash left 16:46 captain-adequate joined, mawkish left 16:47 mawkish joined
captain-adequate Congratulations on reaching your funding goal moritz 16:47
Really looking forward to seeing the perl6 book :) 16:48
Or was it Brian that's writting that. 16:49
16:50 wamba joined
AlexDani` 16:50
captain-adequate: yea, it was not moritz…
captain-adequate Oops... ha ha...
Brian D Foy
16:51 mawkish left
DrForr Dors anyone mind if I move my posting to yhe 10th? 16:51
16:51 mawkish joined
moritz captain-adequate: but I wouldn't say no to a few k USD :-) 16:51
babydrop DrForr: as opposed to... which date? Tomorrow?
AlexDani`: you may enjoy answering this question: www.reddit.com/r/perl6/comments/5g...s/dawh3oj/
(how to type unicode ops)
AlexDani` “Do you have a browser tab open” hahahahaha 16:52
“are you memorizing the character codes” :|
babydrop Yeah, I laughed too. But I guess if you've never heard of other methods that'd be your ownly guess
DrForr Yes, I'm scheduled tomorrow but din't have as nuch time as I hopwd tonight.
babydrop I do have some codes memorized, actually.
captain-adequate Ha ha... Well thanks for the news letter moritz.
babydrop Like − and ♥ 16:53
DrForr: no worries, I schedule my secret post.
moritz captain-adequate: you're welcome 16:54
DrForr Excelebt, I'll be resdy to toll gor the 10th. I'lk move aroubd when I get honr.
(on my phone, if it isn't obvious from tyops.) 16:55
16:58 pyrimidine left, adrusi joined 17:00 mawkish left 17:01 mawkish joined
[Coke] I'll move it now. 17:01
17:01 Skarsnik_ joined 17:02 Skarsnik__ joined
babydrop [Coke]: it's "How To Make Use And Abuse Perl 6 Subsets" 17:03
[Coke]: it's "How To Make, Use, And Abuse Perl 6 Subsets"
dalek : 4722580 | (Will Coleda)++ | misc/perl6advent-2016/schedule:
move DrForr down.

Also, surprise Zoffix++ post to fill the gap.
17:04
[Coke] babydrop++ Zoffix++
17:04 Skarsnik left, pierre_ joined 17:06 Skarsnik_ left 17:08 ssotka joined
broquaint m: sub ext-s { ($($^x, $^y), $^z).flat } ext-s(1,2,()); 17:08
camelia rakudo-moar 5bacb0: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Strange text after block (missing semicolon or comma?)␤at <tmp>:1␤------> 3sub ext-s { ($($^x, $^y), $^z).flat }7⏏5 ext-s(1,2,());␤ expecting any of:␤ infix␤ infix stopper␤…»
17:09 pierre_ left
broquaint m: sub ext-s { ($($^x, $^y), $^z).flat }; ext-s(1,2,()) 17:09
camelia ( no output )
broquaint m: sub ext-s { ($($^x, $^y), $^z).flat }; say ext-s(1,2,())
camelia rakudo-moar 5bacb0: OUTPUT«((1 2) ())␤»
broquaint m: my $r = ($(1,2), ()).flat; say $r 17:10
camelia rakudo-moar 5bacb0: OUTPUT«((1 2))␤»
broquaint Why is the empty list flattened in the latter case but not the former?
lucasb it is itemized 17:11
m: say ($( ), $( )).flat
camelia rakudo-moar 5bacb0: OUTPUT«(() ())␤»
lucasb m: say ($( ), $( )).flat.perl
camelia rakudo-moar 5bacb0: OUTPUT«($(), $()).Seq␤»
babydrop m: sub ext-s { gather deepmap *.take, ($($^x, $^y), $^z) }; say ext-s(1,2,()) 17:12
camelia rakudo-moar 5bacb0: OUTPUT«(1 2)␤»
broquaint Where's the itemization occurring (given the expressions should be identical)?
lucasb assigning it to $^z itemizes 17:13
broquaint Aha, thanks! :)
lucasb m: say { $^a }.(()).perl 17:14
camelia rakudo-moar 5bacb0: OUTPUT«$()␤»
lucasb m: say { @^a }.(()).perl
camelia rakudo-moar 5bacb0: OUTPUT«()␤»
17:16 prammer left, prammer joined
broquaint Super, I'm now dangerously close to a functioning port of microKanren.scm :) 17:17
babydrop \o/
17:20 AlexDani` left, jdv79_ joined, domidumont joined 17:21 Skarsnik_ joined 17:22 jdv79_ left 17:25 Skarsnik__ left, pyrimidine joined 17:28 nicq20 joined 17:30 pyrimidine left 17:31 dakkar left, pmurias left 17:34 cognominal left 17:35 cognominal joined 17:38 Guest97524 is now known as ponbiki 17:40 AlexDani` joined 17:47 zakharyas joined 17:48 pyrimidine joined, CIAvash joined 17:50 cognominal left, mawkish left, Skarsnik__ joined 17:51 pyrimidine left, mawkish joined, pyrimidine joined 17:54 Skarsnik_ left 17:55 CIAvash left 18:00 Skarsnik_ joined, mohae joined 18:01 pmurias joined, CIAvash joined 18:02 Skarsnik joined 18:03 Skarsnik__ left, mohae_ left 18:04 yqt left 18:05 Skarsnik_ left 18:06 cognominal joined 18:11 Actualeyes left 18:14 cgfbee joined 18:15 r3m joined 18:20 nicq20 left 18:21 AlexDani` left 18:22 AlexDani` joined 18:23 skids left 18:27 mawkish left, mawkish joined 18:29 AlexDani` left
DrForr [Coke]: Thanks, just got in and was going to do the edit :) 18:31
18:35 TheDir joined
pmurias how do I print out 2 ** (-1020) without it being rounded down to zero? 18:38
18:38 zakharyas left
pyrimidine re: bdf's TPF grant proposal, I agree that, in light of the Kickstarter being funded, that someone else should get the benefit of that proposal 18:39
18:39 mr_ron joined
babydrop m: printf "%.2200f", 2 ** (-1020) 18:40
camelia rakudo-moar f367e4: OUTPUT«0.00000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000…»
babydrop pmurias: I *think* that's the way. But I think we may have some method for that
pyrimidine saying that, I also think [Coke] is right regarding the timing. It is worth pointing out the (non-funded) help that he is getting, but I don't think it's helpful to completely slag on his proposal. The initial comment comes across that way 18:41
IMO
babydrop pyrimidine: I wouldn't be so pissed of if Brian was competent in Perl 6. 18:42
m: say (2 ** (-1020)).base-repeating: 10 18:43
camelia rakudo-moar f367e4: OUTPUT«No such method 'base-repeating' for invocant of type 'Num'␤ in block <unit> at <tmp> line 1␤␤»
babydrop mmkay
psch babydrop++
babydrop m: say (2 ** (-1020)).FatRat.base-repeating: 10
camelia rakudo-moar f367e4: OUTPUT«(0 )␤»
mr_ron I recently came across an IRC discussion of unicode number subscripts and superscripts and related gist irclog.perlgeek.de/perl6/2016-04-14#i_12340102 and gist.github.com/AlexDaniel/c89bd2786f9b63f31e4c 18:44
m: my $x₁ # error expected ...
camelia rakudo-moar f367e4: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Bogus postfix␤at <tmp>:1␤------> 3my $x7⏏5₁ # error expected ...␤ expecting any of:␤ constraint␤ infix␤ infix stopper␤ postfix␤ statement end␤ …»
psch u: ₁ 18:45
unicodable6 psch, U+2081 SUBSCRIPT ONE [No] (₁)
psch, U+0020 SPACE [Zs] ( )
psch mr_ron: afaik, No doesn't fit into <ident>
babydrop m: say ₁ + ₁ # :D
camelia rakudo-moar f367e4: OUTPUT«2␤»
18:45 ufobat left
psch mr_ron: we have some hackery in Perl6::Grammar to parse them as literal numbers, but... 18:46
u: arabic one
unicodable6 psch, U+0661 ARABIC-INDIC DIGIT ONE [Nd] (١)
psch, U+06F1 EXTENDED ARABIC-INDIC DIGIT ONE [Nd] (۱)
psch, U+08EA ARABIC TONE ONE DOT ABOVE [Mn] (◌࣪)
mr_ron psch: yes irc discussion link mentions that I think
psch m: my $x١
camelia ( no output )
unicodable6 psch, gist.github.com/2c1b824480c9ea0b05...e37395c97c
psch m: my $x١ = 5; say $x١
camelia rakudo-moar f367e4: OUTPUT«5␤»
psch mr_ron: well, i'm wrong
mr_ron: it's not really about No, it's about some cases arbitrarily working and others not
DrForr babydrop: FWIW he explained his reasoning in the original Kickstarter, but I'm not happy about the code quality getting slung around either. The lack of consultation is giving me flashbacks to Matt's Script Archive and in all the bad ways. 18:47
psch mr_ron: if you're willing and able to work out a consistent rule and convince @LARRY that it's better than what we have you got at least me behind you
babydrop mr_ron: and me too. I'd love to have subscripts on terms
psch mr_ron: as far as i'm concerned, we should really stick to unicode categories with those things, cause we pride ourself on, well, sticking to unicode categories
mr_ron Wondering about documentation like docs.perl6.org/language/syntax#Identifiers not advertising unicode identifiers ... 18:48
18:48 AlexDani` joined
DrForr mr_ron: Well volunteered? 18:48
psch ++mr_ron
babydrop maybe the rule can be something like "you can stick a No at the end" (I'm just guessing here, I've not seen the <ident> or what issues a No in it may cause)
psch babydrop: well, ² is No too, isn't it 18:49
and we kinda want that as &infix:<**>.assuming(*, 2)
babydrop m: '²'.uniprop.say
camelia rakudo-moar f367e4: OUTPUT«No␤»
psch so that's hard
babydrop oh yeah :(
mr_ron Also math unicode identifiers are mentioned on rosetta code rosettacode.org/wiki/Unicode_varia...mes#Perl_6 but seem to me to be missing from github.com/perl6/roast/blob/master.../unicode.t ??
18:50 cognominal left
babydrop m: my \term<2²> { 42 }; say 2² 18:50
camelia rakudo-moar f367e4: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Term definition requires an initializer␤at <tmp>:1␤------> 3my \term7⏏5<2²> { 42 }; say 2²␤»
babydrop m: sub term<2²> { 42 }; say 2²
camelia rakudo-moar f367e4: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Missing block␤at <tmp>:1␤------> 3sub term7⏏5<2²> { 42 }; say 2²␤ expecting any of:␤ new name to be defined␤»
babydrop m: sub term:<2²> { 42 }; say 2²
camelia rakudo-moar f367e4: OUTPUT«42␤»
psch yeah, that's an annoying bit of work there 18:51
i'm not sure "check if the lexical exists otherwise assume superscript means pow" is a good idea
18:51 firstdayonthejob joined
psch ...actually i'm pretty sure it's a bad idea 18:51
babydrop yeah
psch so we cannot exclusively rely on unicode category there
mr_ron Any dispute with trying to add math identifier tests to S02-lexical-conventions/unicode.t ? 18:52
18:52 cognominal joined
DrForr Doesn't Unicode TR27 have some things to say about this? 18:52
psch mr_ron: if it's used on rosettacode i wouldn't object 18:53
18:53 AlexDani` left
psch mr_ron: considering TimToady++ did at least a lot of those 18:53
DrForr (math identifiers, super and subscripts, that is?)
18:54 domidumont left
mr_ron DrForr: Draft Unicode Technical Report #27 - Unicode Consortium ?? 18:55
18:55 dudz joined
DrForr One of those; it talks about math operations and such. 18:55
19:00 FROGGS joined
babydrop DrForr: TBH, I'm still not following bdf's "reasoning". He says O'Rly is already commited, but then writes "I've shifted O'Reilly risk tolerance by taking on some of it myself". OK, so you launched a Kickstarter for $37K. Considering he plans to finish the book by August, his "risk" is he'd make "only" $4,000 per month, assuming NOT A SINGLE PERSON BUYS THE BOOK EVER. Hell, I'd love some of that sort of risk. 19:01
But wait! Presumably at the same time he started a Kickstarter, he submitted a TPF grant, for another $10,000 of that juicy "risk". And to my knowledge didn't mention it on his Kickstarter (did he?). So what happened there? If he needed 50 grand of "risk", why not just kickstart for 50 grand?
psch honestly, i most just don't see how bdf is involved with perl6 19:02
that's my primary concern
babydrop ditto
psch i mean, i remember him being around occasionally about a year or one and a half years back or so
and, well, considering i'm only closing in on three years of spotty attendance myself i probably don't have that much of a point to stand on anyway 19:03
but if you're writing a "learn the essentials but also significant detail" book about a language that (kind of) recently released its 1.0 spec you probably should pay attention to what's likely to change in the future..? 19:04
19:04 CIAvash left
psch and, well, we're pretty explicit about "be on irc" wrt the pulse of langdev 19:04
19:05 CIAvash joined
DrForr I repeatedly mention IRC at the very *start* of my talks on the cummuity slides. 19:09
*community # evil keyboard. 19:10
19:10 mr_ron left, mawkish left 19:11 mawkish joined
DrForr FWIW I'm all for the book, and I want to see it out in people's hands. But the lack of community interaction worries me. As it does most everyone here, so I'm not going to flog that dead horse. 19:12
On the other hand, being distracted by an IRC window constantly open when you're trying to write is annoying too. 19:14
perlpilot DrForr: were you going to be co-author on the book or something else? 19:15
19:16 darutoko left
DrForr I don't think that was ever his intention; part my reading too much into it and part miscommuication. 19:17
psch .oO( the evil keyboard strikes again )
perlpilot okay.
19:19 domidumont joined
perlpilot So ... I find I'm much less enamored of the book then. 19:19
19:20 cognominal left
DrForr I'm just glad that the kickstarter text and now this proposal came out before I was more involved. I don't need the drama I can easily see ensuing. 19:20
babydrop perlpilot: there's plenty of fish in the sea: perl6book.com/
19:20 cognominal joined
[Coke] btw, bdf will be speaking at a NYC perl meetup tomorrow. 19:21
perlpilot babydrop: oh sure ... It's just that there's a certain gravitas associated with "Learning Perl" and "Programming Perl" and O'Reilly publications...
psch honestly, this whole situations seems to strongly involve a rock and a hard place 19:22
perlpilot babydrop: and what I don't want to see happen is a publication that #perl6 thinks sucks.
pyrimidine babydrop, I guess my point is, bdf took the initiative to write up and get a Kickstarter rollnig (then funded), and also had the initiative to write up a TPF grant. That's a lot of work 19:23
[Coke] hurls www.meetup.com/The-New-York-Perl-M...235779597/
psch i don't see a way without dangerous -- if not straight-out damaging -- amount of drama happening
babydrop pyrimidine: and my point bdf is asking for a whole lot of money for work he's not competent to perform well. 19:24
pyrimidine: and being secretive about it
pyrimidine I do think it would have helped having an open coversation w/ the community here on how a 'Learning Perl6' would look.
exactly
babydrop screw it.
psch (it would definitely include a nbsp between the 'l' and the '6')
babydrop blocks out all of LP6 drama and focuses on code from now on
pyrimidine babydrop: apologies for the noise. I do agree with you on 90% of it. The tone just was off to me 19:25
pyrimidine now needs to work out why my IRC client drops/reconnects so often 19:28
19:29 RabidGravy left 19:30 bjz joined 19:31 bjz_ left 19:34 zakharyas joined 19:35 RabidGravy joined 19:38 mls joined 19:39 CIAvash left 19:42 Charlie is now known as cw
jdv79 just fyi. its very irritating that there is no easy way to nav the advent entries. 19:44
i scroll to the bottom of the front page and it auto lengthens and i cant find tbe calendar selector thing 19:45
on my phone
19:46 domidumont left
[Coke] we are ever in search of a wordpress theme that works excellently everywhere. 19:46
babydrop jdv79: I don't see an option to display the calendar on mobile version as well... Sorry 19:48
jdv79 is it too much to ask for useable nav controls?
babydrop It *is* there on desktop version
19:48 maybekoo2 joined
babydrop jdv79: we don't have a lot of control over this specific installation as we're using the free tier account on wordpress.com 19:48
jdv79 maybe its time to do it on hack in p5 or p6 19:49
psch jdv79: i thought advent 2017 is the one that we host ourselves?
babydrop jdv79: well volunteered! :D
jdv79 anyway. yay. i gave up:(
oh, really? 17 huh. cool! 19:50
19:50 cognominal left, ssotka left
babydrop Granted. It's perfectly possible there *is* a way to make it sane and I just don'tknow it :) 19:51
It's not like I know anything about Wordpress
You can also visit reddit.com/r/perl6 all the posts are listed there
psch so, for 2017 we obviously need Pod::To::HTTPS
babydrop eww pod
:) 19:52
psch oh cmon, Pod6 is pretty good
babydrop Markdown kinda spoiled me, I guess
psch i mean, it still has an overabundance of =, but still
babydrop I'm more annoyed by overabundance of < >... or rather of «<>|» stuff
vs. [``]() in Markdownb 19:53
I guess editor snippets is a way to go there
psch hm, maybe we need different doc preprocessors too :P
...actually i don't know 19:54
19:54 AlexDaniel left
psch i mean, i've written some Pod6, but not a lot 19:54
i wasn't terribly bothered by it, not even by the fact that i had to look it up by example in the synopses
babydrop Markdown::To::Pod :)
psch heh 19:55
use Doc::Markdown; BEGIN %*LANG<DOC> = Doc::Markdown # was more my line of thought :P
...actually, i'm not sure in how far we have Pod6 in the language braid 19:57
moritz %?LANG<DOC>, if anything
psch that'd be a nice project for something like GSOC i guess
m: say %?LANG
camelia rakudo-moar 2d0198: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Variable '%?LANG' is not declared␤at <tmp>:1␤------> 3say 7⏏5%?LANG␤»
psch m: say %*LANG
camelia rakudo-moar 2d0198: OUTPUT«Dynamic variable %*LANG not found␤ in block <unit> at <tmp> line 1␤␤Actually thrown at:␤ in block <unit> at <tmp> line 1␤␤»
moritz setting dynamic variables at compile time doesn't seem to make too uch sense to me 19:58
since they are inherently tied to the run time
psch moritz: right. i'll concede to any critique on the semantics there
babydrop m: BEGIN say %*LANG
TimToady %*LANG should be considered deprecated
camelia rakudo-moar 2d0198: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤An exception occurred while evaluating a BEGIN␤at <tmp>:1␤Exception details:␤ No such method 'WHICH' for invocant of type 'Perl6::Grammar'␤ in code at <tmp> line 1␤␤»
babydrop m: BEGIN say %*LANG.^name 19:59
camelia rakudo-moar 2d0198: OUTPUT«BOOTHash␤»
TimToady the current language braid should be a property of the current cursor, not a dynvar
psch TimToady: so where *do* we get to it? 20:00
TimToady via some method on the current cursor
20:01 cognominal joined
TimToady once we get there, it will be possible to save a "current language" for reference in a separate semantic pass on the AST 20:03
as it is currently, the dynvar goes out of scope before the next pass
pragmas likewise should come through the cursor, not through a dynvar 20:04
the cursor is the goto place for all things that define the current language, so using dynvars in parallel is a smell 20:05
20:05 vike joined
psch ah, okay 20:05
so any assignment to anything belonging to the current cursor (i don't think $¢ currently works in the mainline..?) translates to a BEGIN..? 20:07
or am i mixing semantics there? afaiu, BEGIN is the time where we can add new bareword, which additions to the language braid might want to do? 20:08
20:20 mawkish left 20:21 mohae_ joined 20:23 sena_kun joined, mohae left 20:25 mawkish joined 20:26 pierre_ joined 20:31 dudz left, dudz joined, pierre_ left 20:34 labster joined
dalek osystem: 7396929 | RabidGravy++ | META.list:
Add Manifesto

  github.com/jonathanstowe/Manifesto
20:38
RabidGravy boom 20:39
moritz bang! 20:44
20:46 mawkish left
moritz m: my %h is default([]); %h<a>.push: 1; %h<b>.push: 2; say %h 20:53
camelia rakudo-moar 2d0198: OUTPUT«{}␤»
moritz m: my %h; %h<a>.push: 1; %h<b>.push: 2; say %h 20:54
camelia rakudo-moar 2d0198: OUTPUT«{a => [1], b => [2]}␤»
moritz does this look like a bug to anybody else?
babydrop yeah
moritz I was curious whether it created two distinct arrays, but didn't expect to see %h empty 20:55
moritz rakudos submitbug
20:55 pyrimidine left
masak aye, bug 20:56
20:56 El_Che joined 20:58 FROGGS left
masak not sure which of your two hypothesized behaviors I would expect, even though the "two distinct arrays" feels both nicer and more magical 20:58
moritz I leave that open for the purpose of this bug report 21:00
the question is really: does the 'is default' trait clone the default value before assigning it 21:02
the right solution would be to capture the default as a thunk
and evaluate the thunk each time it's needed 21:03
just like we handle defaults in signatures
but I'm afraid our current trait system doesn't make that accessible
21:05 Skarsnik_ joined
jnthn m: gist.github.com/jnthn/ad63b599d601...5e1fd4d86f 21:05
21:05 cognominal left
camelia rakudo-moar 2d0198: OUTPUT«(0 1 2 3 4 5 6 7 8 9)␤» 21:05
21:06 pyrimidine joined
jnthn RabidGravy: Guess you've got more sophisticated uses for Manifesto, but ^^ :) 21:06
moritz jnthn: sleep sort ftw
RabidGravy :) 21:08
21:08 Skarsnik left
RabidGravy yeah 21:08
I couldn't think of an example that didn't have a dependency 21:09
jnthn But it strikes me the module code might golf down some to a supply block
21:11 pyrimidine left, cognominal joined 21:15 sena_kun left 21:20 Tonik left 21:21 donaldh joined
RabidGravy that thought did occur to me but I couldn't work out how to lose the old promises 21:24
21:26 espadrine joined 21:28 sufrosti1o left 21:29 sufrostico left 21:30 lucasb left 21:32 TEttinger joined 21:37 pyrimidine joined, AlexDaniel joined 21:40 sufrostico joined, sufrosti1o joined 21:41 rindolf left, pyrimidine left 21:44 zakharyas left, jonas left 21:45 AlexDani` joined 21:47 AlexDaniel left, AlexDani` is now known as AlexDaniel 21:50 bjz left 21:51 gsoedl joined
[Coke] ... we have a CLOSE phaser? 21:52
(and it's doc'd? :) 21:53
lizmat only works in a supply block afaik 21:55
22:08 bjz joined 22:09 pyrimidine joined 22:13 pyrimidine left 22:20 TheDir left, TheDir joined
dogbert17 is lizmat still awake? 22:20
22:22 bjz left
dogbert17 m: gist.github.com/dogbert17/41042137...c1c05055bb 22:22
camelia rakudo-moar 0428b7: OUTPUT«(signal SEGV)»
dogbert17 why does that golfattempt SEGV?
AlexDaniel bisect: gist.github.com/dogbert17/41042137...c1c05055bb
dogbert17 .oO 22:23
bisectable6 AlexDaniel, It looks like a URL, but mime type is ‘text/html; charset=utf-8’ while I was expecting something with ‘text/plain’ or ‘perl’ in it. I can only understand raw links, sorry.
AlexDaniel ah right
bisect: gist.githubusercontent.com/dogbert...tfile1.txt
bisectable6 AlexDaniel, Successfully fetched the code from the provided URL.
AlexDaniel, Bisecting by exit signal (old=2015.12 new=0428b79). Old exit signal: 0 (None)
AlexDaniel there we go
bisectable6 AlexDaniel, bisect log: gist.github.com/d020b178ffb9858385...1d02cb2046
dogbert17 cool
bisectable6 AlexDaniel, (2016-11-24) github.com/rakudo/rakudo/commit/17...e71362146d
AlexDaniel dogbert17: will you submit a bug report?
22:23 RabidGravy left
dogbert17 I guess so :-) 22:24
AlexDaniel dogbert17: perhaps you want to golf it a little bit… :)
dogbert17 I have tried
AlexDaniel lizmat: ping, ↑ (although you will probably see it anyway)
dogbert17 was hoping she was still awake
AlexDaniel dogbert17: well, we need a ticket anyway 22:25
dogbert17 indeed, I'll fix
22:25 donaldh left
AlexDaniel bisectable is so useful sometimes :O 22:26
dogbert17 indeed, AlexDaniel++
AlexDaniel MasterDuke++ also for a bunch of other whateverable features 22:28
22:29 MasterDuke joined 22:31 pmurias left 22:34 guest__ joined
dogbert17 done, RT #130294 22:34
synopsebot6 Link: rt.perl.org/rt3//Public/Bug/Displa...?id=130294
22:37 setty1 left, setty1 joined 22:40 prammer left 22:41 prammer joined
AlexDaniel dogbert17: thanks 22:41
22:44 El_Che_ joined 22:45 El_Che_ left, pyrimidine joined 22:46 lukaramu left 22:50 pyrimidine left
babydrop How can I get a list of all core types? My goal is to get all the core methods and subs and introspect their signatures, to call them with different parameters to log how they fail. 22:54
m: say keys ::CORE 22:55
camelia rakudo-moar 0428b7: OUTPUT«()␤»
babydrop I recall there was an invocation like that. Something or other
jnthn m: say SETTING::.pairs.grep(!*.value.defined) 22:58
camelia rakudo-moar 0428b7: OUTPUT«RESTRICTED-CLASS is disallowed in restricted setting␤ in sub restricted at src/RESTRICTED.setting line 1␤ in method gist at src/RESTRICTED.setting line 33␤ in block <unit> at <tmp> line 1␤␤»
22:58 guest__ left
jnthn evalable6: say SETTING::.pairs.grep(!*.value.defined) 22:59
evalable6 jnthn, rakudo-moar 0428b79: OUTPUT«(exit code 1) RESTRICTED-CLASS is disallowed in restricted setting␤ in sub restricted at src/RESTRICTED.setting line 1␤ in method gist at src/RESTRICTED.setting line 33␤ in block <unit> at /tmp/ywtw9hRrfp line 1␤»
jnthn oh come on, stop it with this security stuff :P
22:59 mr_ron joined
babydrop I get No such method 'WHICH' for invocant of type 'Perl6::Metamodel::ContainerDescriptor' 22:59
but a bunch of stuff if I reverse the conditional.
Thanks, I think I can figure out the rest :)
AlexDaniel yea. Not yet, not yet 23:02
23:07 prammer left
mr_ron AlexDaniel: I just read your comment on my issue about unicode identifiers. I admit to being a relic of the ASCII and even EBCDIC era but think the Unicode vs Texas page is too much digging for someone who just wants to know about using Unicode for identifiers and numbers. 23:08
github.com/perl6/doc/issues/1039
23:08 pyrimidine joined
mr_ron I guess I am thinking in terms of advertising a feature rather than providing spec. 23:09
AlexDaniel hmmm
mr_ron: perhaps you can suggest something specific? 23:10
or maybe even submit a pull request
mr_ron I want to know about the use of subscripts and the delta symbol for an identifier. It would seem to be a syntax issue and not about a texas/ascii alternative to Unicode. 23:11
babydrop mr_ron: delta is not a symbol, it's a greek letter, and letters are allowed in identifiers. 23:12
AlexDaniel “Any codepoint that has the Ll (Letter, lowercase), Lu (Letter, uppercase), Lt (Letter, titlecase), Lm (Letter, modifier), or the Lo (Letter, other) property can be used just like any other alphabetic character from the ASCII range.”
which means that Ll, Lu, Lt, Lm, Lo are allowed in identifiers
everything else… not really
mr_ron Yes - that is in Unicode vs Texas - but I am not looking for an alternative ASCII coding 23:13
AlexDaniel u: ₆
unicodable6 AlexDaniel, U+2086 SUBSCRIPT SIX [No] (₆)
babydrop mr_ron: basically uncode stuff has "properties" so valid stuff in identifiers goes by those properties. For example, is an uppercase letter, so it has Lu property and Lu properties are allowed
23:13 pyrimidine left
babydrop *delta is an uppercase letter 23:13
AlexDaniel so ₀₋₉ are not going to work in identifiers
which is sad, I know!
babydrop m: '−'.uniprop.say 23:14
camelia rakudo-moar 0428b7: OUTPUT«Sm␤»
AlexDaniel
.oO( but it is even sadder that they don't work as array subscripts )
babydrop heh
23:14 Skarsnik_ left
mr_ron I stand corrected on delta but still wonder about having the info on the right page. Since it's a feature I wonder if it might not even be worth an example? 23:14
AlexDaniel mr_ron: yea, it's worth an example and worth a link to unicode_texas page :) 23:15
babydrop mr_ron: yes, that would be helpful. If you can think of a good way to document that, please submit a PR 23:16
mr_ron Thanks ... will try hopefully by sometime tomorrow.
babydrop \o/
AlexDaniel 🙌 23:19
23:20 skids joined 23:21 cognominal left 23:22 cognominal joined
tbrowder hi #perl6 23:23
AlexDaniel 🙋
dogbert17: hey 23:25
actually, it seems to be something different
lizmat++ only revealed the problem ( github.com/rakudo/rakudo/commit/17...668be71362 )
it does not segfault with MVM_SPESH_DISABLE=1 23:26
babydrop tbrowder: oh, BTW, I only today saw your email about the Advent thing, 'cause I rarely check email. 23:28
huggable: advent
huggable babydrop, github.com/perl6/mu/blob/master/mi...6/schedule
babydrop tbrowder: there's one spot you can post in, if you want. And then [Coke] won't need to write his. github.com/perl6/mu/blob/master/mi...hedule#L13 23:29
tbrowder babydrop: ok but understand it won't be "new" (but i have an idea to extend it just a bit) 23:30
23:31 telex left, maybekoo2 left
babydrop tbrowder: yes, extend it and it'll be fine, since you mention it got little exposure when you first wrote it :) Add yourself to github.com/perl6/mu/blob/master/mi...6/schedule 23:36
.tell [Coke] seems like we found a volunteer to fill up the last spot (tbrowder) so you're off the hook for your article
yoleaux babydrop: I'll pass your message to [Coke].
babydrop tbrowder: do you have access to post on perl6advent.wordpress.com ? If not, let me know your wordpress username (create one if needed) and I'll invite you. 23:37
dogbert17 AlexDaniel: an optimizer bug then? 23:38
tbrowder let me see...
23:38 revdiablo joined, pmurias joined
pmurias should we also setup travis to run nqp tests on Mac OS X? 23:39
babydrop wouldn't hurt
pmurias: you have a robomessage in -dev
23:40 Actualeyes joined, AlexDaniel left
pmurias babydrop: I deleted that branch 23:41
babydrop \o/
pmurias babydrop: if we verify that the ** works on mac os x, we should be able also to close: rt.perl.org/Public/Bug/Display.html?id=125938 23:42
23:44 Actualeyes left
babydrop .ask lizmat would you be able to test whether 2**10000000000 throws on OSX on HEAD rakudo? pmurias++ fixed RT#125938 we just need someone with OSX to verify the fix works. 23:44
yoleaux babydrop: I'll pass your message to lizmat.
synopsebot6 Link: rt.perl.org/rt3//Public/Bug/Displa...?id=125938
tbrowder babydrop: my wordpress user name is "tombrowder" and it doesn't look like i have access to perl6advent.wordpress.com (but can follow it) 23:45
babydrop tbrowder: "Invitation sent successfully" 23:46
23:46 telex joined
babydrop tbrowder: a word of warning: be sure to double check your < and > characters get rendered correctly. When you switch Visual and HTML modes in wordpress, it deletes them. I personally wrote my stuff in markdown, then placed it into a github gist, naming the file with .md extension, and then gave the URL to this script that spat out HTML code I just pasted into wordpress: 23:48
gist.github.com/zoffixznet/153bef6...46e308a53b
I s'pose any &somename; would be affected as well 23:50
23:51 cognominal left
tbrowder zoffix, er babydrop: thanks! 23:51
babydrop :) 23:52
23:52 gsoedl left 23:53 cpage_ left 23:54 pyrimidine joined 23:56 firstdayonthejob left 23:58 sufrostico left 23:59 sufrosti1o left, pyrimidine left