»ö« Welcome to Perl 6! | perl6.org/ | evalbot usage: 'p6: say 3;' or rakudo:, or /msg camelia p6: ... | irclog: irc.perl6.org or colabti.org/irclogger/irclogger_logs/perl6 | UTF-8 is our friend!
Set by moritz on 22 December 2015.
00:00 mempko joined 00:03 n9l left 00:04 konsolebox joined 00:11 perlbot left 00:12 pierrot left, pierrot joined 00:13 perlbot joined 00:17 Nanjizal left 00:26 rindolf left 00:34 FonzieGonzalez joined 00:46 yoleaux joined, ChanServ sets mode: +v yoleaux 01:00 cdg_ joined 01:04 cdg left 01:05 cdg_ left 01:06 markong left 01:07 astj joined 01:14 FonzieGonzalez left 01:15 itaipu joined 01:19 comborico1611 joined, DoverMo joined 01:24 aborazmeh joined, aborazmeh left, aborazmeh joined 01:25 espadrine left 01:37 comborico1611 left 01:47 troys is now known as troys_
Geth doc/master: 4 commits pushed by (Will "Coke" Coleda)++ 01:50
01:51 itaipu left 01:53 bayprogrammer joined 01:55 bayprogrammer left, bayprogrammer joined 02:06 bayprogrammer left 02:11 eliasr left 02:14 jeromelanteri joined
AlexDaniel oops, messed up my buffer and accidentally did ``git clone reddit.com/`` 02:14
I believe my computer needs to be detoxificated now 02:15
geekosaur if that actually did anything other than error out, that's ... scary 02:18
Geth doc/post-release: 52 commits pushed by (Aleks-Daniel Jakimenko-Aleksejev)++, (Luca Ferrari)++, (Will "Coke" Coleda)++, (Zoffix Znet)++, (Alex Chen)++, (JJ Merelo)++, (Jan-Olof Hendig)++, (Marcel Timmerman)++, (Itsuki Toyota)++, (Wenzel P. P. Peppmeyer)++
review: github.com/perl6/doc/compare/3f25e...5541be5df9		 02:25
DoverMo can we make perl6 linux or unix style toolchains? 02:28
like the perl5 ones 02:29
02:29 bwisti left
mspo ? 02:30
02:31 DoverMo left 02:45 travis-ci joined
travis-ci Doc build failed. Zoffix Znet 'Merge branch 'master' into post-release' 02:45
travis-ci.org/perl6/doc/builds/332109318 github.com/perl6/doc/compare/3f25e...41be5df966
02:45 travis-ci left
buggable [travis build above] ☠ Did not recognize some failures. Check results manually. 02:45
02:47 ilbot3 left 02:56 ilbot3 joined, ChanServ sets mode: +v ilbot3 03:09 yoleaux left, yoleaux joined, ChanServ sets mode: +v yoleaux 03:12 Nanjizal joined 03:18 cdg joined 03:49 ufobat_ joined, lookatme joined 03:52 ufobat left 03:57 cdg left 03:59 cdg joined 04:00 softmoth joined 04:02 Nanjizal left 04:03 shinobi-cl joined 04:04 cdg left 04:06 shinobi-cl left, shinobi-cl joined 04:07 softmoth left, softmoth joined 04:08 shinobi-cl left 04:10 shinobi-cl joined 04:11 shinobi-cl left 04:12 shinobi-cl joined 04:16 softmoth left 04:24 piojo left 04:32 Cabanossi left, Cabanossi joined 04:37 epony left 04:50 epony joined 04:54 nativecallable6 joined, ChanServ sets mode: +v nativecallable6 04:58 piojo joined 04:59 piojo1 joined, piojo left 05:00 cdg joined 05:05 cdg left 05:14 traxex left, epony left 05:19 epony joined 05:21 aborazmeh left
shinobi-cl hi all :) I got this problem: I have this array: (1,2,3,10,11,12). I want to create two from this list, grouping consecutive sets 05:32
myfunc(1,2,3,10,11,12) returns (1,2,3),(10,11,12)
the list is always made from positive integers 05:34
made of*
lookatme m: say (1,2,3,10,11,12).rotor(3); # you mean this ? 05:40
camelia ((1 2 3) (10 11 12))
shinobi-cl well... almost 05:41
m: say (1,2,3,4,10,11,12).rotor(3);
camelia ((1 2 3) (4 10 11))
05:41 Actualey` joined
shinobi-cl m: my @a = (1,2,3,10,11,12); my $low = @a.min; my $high = @a.max; my @cutter = ($low .. $high); say @a <</>> @cutter; 05:41
camelia [1 1 1 2.5 2.2 2 0.142857 0.25 0.333333 1 1 1]
shinobi-cl the idea is to identify the "gaps", in this case, when <> 1, and cut there until we find the other "1"s... or something like this 05:43
05:43 Actualeyes left
shinobi-cl of course i could do it just googling (or stackoverflowing? :) ) or taking examples from other languages, but maybe perl6 has a cooler way to do it :) 05:47
05:49 aborazmeh joined, aborazmeh left, aborazmeh joined 05:50 statisfiable6 left, statisfiable6 joined, ChanServ sets mode: +v statisfiable6 05:53 stmuk_ left 05:56 aborazmeh left 06:01 cdg joined 06:02 khw left 06:05 stmuk_ joined, geospeck joined 06:06 cdg left, geospeck left
lookatme shinobi-cl, so you want find the continuous digits sequence not equal 1 ? 06:13
shinobi-cl m: my @a = (1,2,3,10,11,12,14,16,17,18); my $i = 1; my $last = shift @a; my %sets; push %sets{$i}, $last; for @a -> $v { if ($v != $last + 1) { $i++; }; $last = $v; push %sets{$i}, $v;}; say %sets.perl;
camelia {"1" => $[1, 2, 3], "2" => $[10, 11, 12], "3" => $[14], "4" => $[16, 17, 18]}
shinobi-cl this, but more perl6-ish :) 06:14
not necessarily in a hash, it could be (1,2,3),(10,11,12),(14),(16,17,18). I dont really need the hash keys, just identify "blocks" of contiguous integers 06:15
lookatme oh I see 06:16
06:27 klapperl left, klapperl joined
lookatme not have any good idea shinobi-cl 06:30
shinobi-cl i think the rotor you mentioned might be the solution.... I'm looking into that now 06:33
06:42 epony left
shinobi-cl m: say (1,2,4,5).rotor(2 => -1); 06:42
camelia ((1 2) (2 4) (4 5))
06:49 ChoHag left 06:56 ChoHag joined 06:58 troys_ is now known as troys
lookatme shinobi-cl, but the rotor is not flexible 07:02
07:02 yangzq50 joined
shinobi-cl m: my %block; my Int $i = 1; (1,2,4,5,6,7,9,11,12,13).rotor(2 => -1).map: { push %block{$i}, $_[0]; ($_[1] == $_[0] + 1) ?? push %block{$i}, $_[1] !! $i++ }; say %block.perl; 07:03
camelia {"1" => $[1, 2, 2], "2" => $[4, 5, 5, 6, 6, 7, 7], "3" => $[9], "4" => $[11, 12, 12, 13]}
07:03 ilmari[m] left
shinobi-cl i think i'm getting close :) 07:03
lookatme I don't think its better than the usual way
Hmm, that's a better way to iterate the elements
m: my %block; my Int $i = 1; (1,2,4,5,6,7,9,11,12,13).rotor(2 => -1).map: { %block{ ($_[1] == $_[0] + 1) ?? $i !! $i++ }.push($_[0]) }; say %block.perl; 07:09
camelia {"1" => $[1, 2], "2" => $[4, 5, 6, 7], "3" => $[9], "4" => $[11, 12]}
Xliff_ Missing 13
m: my %block; my Int $i = 1; (1,2,4,5,6,7,9,11,12,13).rotor(2 => -1, :partial).map: { %block{ ($_[1] == $_[0] + 1) ?? $i !! $i++ }.push($_[0]) }; say %block.perl;
camelia Use of Nil in numeric context
{"1" => $[1, 2], "2" => $[4, 5, 6, 7], "3" => $[9], "4" => $[11, 12, 13]}
in block at <tmp> line 1
lookatme :) yeah, should add the adv 07:10
I mean the named argument
07:10 ilmari[m] joined
Xliff_ m: my %block; my Int $i = 1; (1,2,4,5,6,7,9,11,12,13).rotor(2 => -1, :partial).map: { %block{ ($_[1] == $_[0] + 1) ?? $i !! $i++ }.push($_[0]) }; 07:11
camelia Use of Nil in numeric context
in block at <tmp> line 1
07:11 yoleaux left 07:17 troys left 07:32 lowbro joined, lowbro left, lowbro joined 07:38 abraxxa joined 07:39 tyil left 07:42 tyil joined
shinobi-cl m: my %K; my Int $i = 1; (1,3,4,5,6,7,9,11,12,13).rotor(2 => -1).map: { my ($a,$b)=@_; push %K{$i}, $a; $i++ if ($a+1 != $b); LAST {push %K{$i}, $b} }; say %K.perl; 07:44
camelia {"1" => $[1], "2" => $[3, 4, 5, 6, 7], "3" => $[9], "4" => $[11, 12, 13]}
shinobi-cl m: my %K; my Int $i = 1; (1,2,4,5,6,7,9,11,12,13).rotor(2 => -1).map: { my ($a,$b)=@_; push %K{$i}, $a; $i++ if ($a+1 != $b); LAST {push %K{$i}, $b} }; say %K.perl;
camelia {"1" => $[1, 2], "2" => $[4, 5, 6, 7], "3" => $[9], "4" => $[11, 12, 13]}
shinobi-cl niiiiice, phasers are great :) 07:45
07:45 wamba joined, broquaint left
shinobi-cl thanks lookatme++ for suggesting the rotor method 07:46
lookatme YW shinobi-cl 07:47
07:47 piojo1 left
lookatme But I think it's not elegant than the normal way :) 07:48
08:07 shinobi-cl left 08:09 piojo joined 08:11 darutoko joined 08:13 kuerbis joined
kuerbis m: my @a = <a b c d>; @a.say; my $t = @a[0..2]; $t[0] = 'Z'; @a.say; 08:14
camelia [a b c d]
[Z b c d]
kuerbis Hello, is my interpretation of this code right: "my $t = @a[0..2];" does not copying, not creating a new list?
08:15 raschipi joined, jast joined 08:17 rindolf joined
moritz m: my @a = 0..10: say @a[0..2].^name 08:22
camelia 5===SORRY!5=== Error while compiling <tmp>
Confused
at <tmp>:1
------> 3my @a = 0..10:7⏏5 say @a[0..2].^name
expecting any of:
colon pair
moritz m: my @a = 0..10; say @a[0..2].^name
camelia List
moritz kuerbis: looks like it creates a new list to me
kuerbis thx! 08:23
08:24 cdg joined 08:28 dakkar joined
nine moritz: I finally did it! niner.name/blog/production_on_perl_6/index.html 08:28
08:29 cdg left
moritz nine: nice, thanks! 08:29
08:42 dorothyw is now known as johnnymacs
lizmat clickbaits and another Perl 6 Weekly hits the Net: p6weekly.wordpress.com/2018/01/22/...-optimism/ 08:43
eh
lizmat clickbaits p6weekly.wordpress.com/2018/01/22/...-optimism/
:-)
08:48 Guest10951 is now known as masak 09:06 sena_kun joined 09:11 yoleaux joined, ChanServ sets mode: +v yoleaux
buggable New CPAN upload: Sparrowform-0.0.4.tar.gz by MELEZHIK cpan.metacpan.org/authors/id/M/ME/...0.4.tar.gz 09:12
09:12 wamba left 09:13 lookatme left 09:18 kuerbis left
El_Che nine: are we compiling perl6 in production? I have something as well 09:18
09:21 piojo1 joined
lizmat El_Che: yes, hoping to make it a section in the P6W now :-) 09:21
buggable New CPAN upload: Sys-Hostname-0.0.2.tar.gz by ELIZABETH cpan.metacpan.org/authors/id/E/EL/...0.2.tar.gz 09:22
El_Che cool 09:23
09:24 piojo left 09:29 keyking joined 09:31 keyking left 09:33 keyking joined 09:34 bioexpress joined
bioexpress m: my Str $a = 'x'; $a .= sprintf( '%d', 20 ); say $a; 09:35
camelia Your printf-style directives specify 0 arguments, but 2 arguments were supplied
bioexpress The error message didn't help me a lot to find the error. 09:36
lizmat bioexpress: please file an issue then, we don't like LTA errors :-) 09:43
afk for a few hours& 09:44
09:48 raschipi left 09:50 lowbro left
El_Che releasable6: status 09:51
releasable6 El_Che, Next release will happen when it's ready. Blockers: github.com/rakudo/rakudo/issues?q=...%9A%A0%22. Unknown changelog format
09:51 natrys joined
bioexpress lizmat: where is the right place to file an issue? 09:52
moritz bioexpress: github.com/rakudo/rakudo/issues 09:53
09:59 wamba joined 10:00 keyking left 10:01 keyking joined
bioexpress github.com/rakudo/rakudo/issues/1437 10:04
10:08 bioexpress left 10:11 TEttinger left
melezhik I need to run a series of system commands one by one asyncronously and to be able fetch, keep track of stdout/stder from those commands 10:13
how can I do this? 10:14
jnthn Just in terms of builtins: Proc::Async. I think there's a module (something like Proc::Q) that may help also
El_Che wouldn't channels be useful?
jnthn Not obviously :) 10:15
El_Che he's using perl6, so obviously is a relative thing :)
10:15 domidumont joined
jnthn I mean, I've done lots of async process juggling in Perl 6 and nearly never reached for a Channel :) 10:15
El_Che jnthn: that said, you're an async man 10:16
:)
10:17 itaipu joined 10:20 domidumont left 10:23 domidumont joined 10:24 pmurias joined, eliasr joined
melezhik jnthn, El_Che - thanks guys, will try Proc:Q first )) 10:24
10:29 domidumont left 10:35 pmurias left
AlexDaniel El_Che: that status message didn't tell much I think :) 10:38
m: my $s = "1"; dd uniprop($s,"White_Space") 10:39
camelia Bool::True
10:39 zakharyas joined
AlexDaniel El_Che: ↑ this is not a blocker in rakudo terms, but I think samcv decided to look at it before the release 10:39
El_Che AlexDaniel: it tells me "there is some stuff left and right that need fixing first. Come again later" :)
samcv AlexDaniel: yeah i fixed it locally and should do the release tomorrow. and i fixed <:space> for real 10:40
which is pretty nice as well
El_Che samcv++ 10:41
10:41 releasable6 left
samcv AlexDaniel: thanks for bringing that to my attention by the way 10:41
El_Che releasable6 can't take it any longer
10:41 benchable6 left, bisectable6 left 10:42 benchable6 joined, bisectable6 joined, ChanServ sets mode: +v bisectable6
moritz
.oO( we're not gonna take it )
 10:42
10:42 coverable6 joined, ChanServ sets mode: +v coverable6
AlexDaniel except that everyone came back except releasable6 xD 10:43
10:43 releasable6 joined 10:44 releasable6 left, pmurias joined, releasable6 joined, ChanServ sets mode: +v releasable6
AlexDaniel releasable6: status 10:44
releasable6 AlexDaniel, Next release There is some stuff left and that needs fixing first, come again later. Blockers: github.com/rakudo/rakudo/issues?q=...%9A%A0%22. Unknown changelog format
AlexDaniel aaaaaaahhhh… that's not going to work
pmurias do we support negate not-zerowidth literals? 10:45
10:45 releasable6 left, releasable6 joined, ChanServ sets mode: +v releasable6 10:51 domidumont joined 10:57 lizmat left
AlexDaniel pmurias: probably <-[n]> <-[o]> <-[t]> 11:02
not that I know of at least. That's if I understood the question correctly
jnthn No, there's no negated literal syntax 11:03
Though if you need it it's available as <!before 'foo'> ...
(And if we ever got such a syntax, it's probably just compile into that too) 11:04
11:04 mahafyi joined 11:05 yoleaux left 11:06 yoleaux joined, ChanServ sets mode: +v yoleaux 11:08 cdg joined 11:09 keyking left
samcv pmurias: i fixed the <:space> text btw. you (i think it may be you) fudged it for js. well it's not puhed it. i will push tomorrow 11:10
11:12 cdg left 11:27 lizmat joined 11:36 faraco joined
moritz 11:45 < pmurias> do we support negate not-zerowidth literals? 11:52
11:53 yangzq50 left
moritz the question is always what people actually *want* when they want to negate a regex, or a literal 11:53
*any* string that isn't the literal? any string that doesn't start with the literal?
any string that doesn't contain the literal?
any string where no character matches the literal (but a zero-width string would match that too)? 11:54
jnthn True :) 11:55
I guess all of those are more clearly written with things we already do have :)
jkramer Is the order of pairs in hashes/bags/... still randomized like in P5? 11:56
moritz potentially, yes 11:57
12:06 clarkema joined
faraco docs.perl6.org/type/Hash#Looping_o...and_values 12:06
lizmat and the weekly is on HN again: news.ycombinator.com/item?id=16211864 12:10
and of course none of the comments go about the actual post 12:12
12:13 domidumont left, lizmat left, lizmat joined
stmuk_ lobste.rs/ may be more friendly 12:14
El_Che lizmat: I am kind of puzzled by the people that moved to python. The last years I have mostly seen people moving away from it
(maybe it's just the devops world)
tadzik El_Che: this lobsters thing looks nice :) 12:15
piojo1 m: say (bag 'a').pick(1); 12:17
camelia P6opaque: get_boxed_ref could not unbox for the representation '20' of type Scalar
in block <unit> at <tmp> line 1
piojo1 m: say (bag 'a').pick;
camelia a
piojo1 am I doing something crazy, or is Rakudo?
lizmat piojo1: looks like a bug 12:18
piojo1 lizmat: thanks. I'll report it.
12:19 itaipu left
lizmat m: dd (bag "a").pick(1) # also seems related to .gist 12:21
camelia ("a",).Seq
lizmat m: (bag "a").pick(1).gist 12:22
camelia P6opaque: get_boxed_ref could not unbox for the representation '20' of type Scalar
in block <unit> at <tmp> line 1
12:22 domidumont joined 12:23 markong joined
El_Che m: say (bag 'a', 'b').pick(3) 12:24
camelia (b a)
El_Che m: say (bag 'a', 'b').pick(2)
camelia P6opaque: get_boxed_ref could not unbox for the representation '20' of type Scalar
in block <unit> at <tmp> line 1
El_Che one off error?
masak whoa :)
timotimo moar should be able to give you the representation name instead of the id here 12:25
lizmat m: (bag <a b c d e>).pick(1).gist
camelia P6opaque: get_boxed_ref could not unbox for the representation '20' of type Scalar
in block <unit> at <tmp> line 1
12:25 Nanjizal joined
lizmat it's Seq.gist complaining I think 12:25
m: (bag <a b c d e>).pick(1).cache[0] 12:26
camelia P6opaque: get_boxed_ref could not unbox for the representation '20' of type Scalar
in block <unit> at <tmp> line 1
12:27 itaipu joined
timotimo um, that works locally 12:28
lizmat representation 20, what is that ? 12:30
timotimo bisectable6: (bag <a b c d e>).pick(1).cache[0]
bisectable6 timotimo, Bisecting by exit code (old=2015.12 new=8bd9fd3). Old exit code: 0
12:30 cdg joined
bisectable6 timotimo, bisect log: gist.github.com/227613b2654c80d429...32fe34b16c 12:30
timotimo, (2017-07-22) github.com/rakudo/rakudo/commit/75...05f8bd8592
timotimo 76:#define MVM_REPR_ID_P6bigint 20 12:31
that's reprid 20 12:32
12:33 faraco left
timotimo ok, with all the right versions i get that error, too 12:34
so maybe bumping nqp and moar would help
12:35 cdg left
timotimo so, the explosion happens in Baggy.pm in the pull-one method. since it's trying to get a bigint from something, and the only bigint op in there is nqp::sub_I($!todo, 1, Int), i'd assume that $!todo wasn't initialized properly? 12:36
lizmat hmmm looks like there's a decont missing somewhere ?
yeah, investigating that
12:36 Nanjizal left, reportable6 left
timotimo hm, you need to decont that, huh? 12:36
maybe
12:37 reportable6 joined, ChanServ sets mode: +v reportable6
timotimo doesn't explain why it doesn't explode with newer nqp&moar :) 12:41
12:42 Nanjizal joined 12:49 player joined
piojo1 Is there any reason Mix doesn't support .pick, or is it that nobody had time to implement it yet? 12:50
player Hi. What's the analogue of break <label> in Perl 6? 'last' seems to exit from the innermost loop
piojo1 player: there's a label example here: docs.perl6.org/language/control#LABELs
timotimo i think we do have last + label
pmurias El_Che: what are they moving towards? 12:51
piojo1 m: say (a => 0.97, b => 0.03).Mix.pick;
camelia .pick is not supported on a Mix
in block <unit> at <tmp> line 1
timotimo piojo1: it's because mix supports non-integral weights, and pick is supposed to remove an item it picked from the mix
piojo1 timotimo: I think that's .grab, not .pick 12:52
timotimo no, grab is pick but also modifies the original object
m: <a a a b c>.pick(10).say
camelia (a a b c a)
timotimo you see how i ask for 10, but only get 5?
piojo1 timotimo: ahh, thanks. I should have been using .roll for what I wanted 12:53
timotimo the error message on the mix should probably point that out 12:54
Geth doc: fdbf475805 | (Luca Ferrari)++ | doc/Language/operators.pod6
Add infix minmax documentation to operators.

Tested on Rakudo version 2017.11 built on MoarVM version 2017.11
synopsebot Link: doc.perl6.org/language/operators
Geth doc: 6736c6511a | (Luca Ferrari)++ | doc/Language/operators.pod6
Improve minmax: example with same values.
doc: d1b0571701 | (Luca Ferrari)++ | doc/Language/operators.pod6
minmax documentation: list and hash.

Close #1734
12:55 pmurias left 12:56 pmurias joined
El_Che pmurias: go, js 12:57
pmurias: even java (spring) 12:58
12:59 stmuk joined 13:00 scimon joined
scimon (Turns out my attempt to not read this was failing dismally. So I figured I might as well log in again). Hi. 13:00
13:00 stmuk_ left
lizmat scimon o/ 13:01
El_Che pmurias: I am under the impression that Python got a firm foot in science (although at a course I followed they stressed that you'll end up writing a lot of C) and related fields. While it seemed to be losing web developement en devop stuff (microservices and the like)
scimon: welcome back
scimon I'm seeing a lot of places still using python for web dev but node is getting a large mind share. 13:02
(Frankly I think that's weird because it's a horrible language for a lot of back end stuff. Async DB interaction is just a pain IMO)
El_Che scimon: it's still there (as is Ruby), buy my very-personal-impression is a trend 13:03
scimon: we have big python people at $work contemplating leaving python and move to php7 for web stuff
scimon: grain of salt of course, anecdotical at best
scimon I think a plus point at the moment is the rise of devops and containers. Where you have teams being able to pick their own tools that they have to support things can change. 13:04
php7?
Urgggh.
El_Che it's also the impression I get at FOSDEM. Ruby is bleeding badly, Python seem to be over the top
scimon: php7 got a lot of stuff right (according to them)
scimon Ah well. I am possibly guilty of reacting to PHP that way many people react to Perl. 13:05
3 years doing PHP development is enough for me.
El_Che it seems to have becomed fast 13:06
if so, cheers for the php devs
scimon Well done them.
I might even take a look.
lizmat github.com/rakudo/rakudo/commit/7361fe2b6d # piojo1
El_Che I had a look last week on how to containerized their projects
piojo1 thanks lizmat! I confess I don't understand the cause, so I can't offer an opinion on the "magic" :) 13:12
13:13 Zoffix joined
Zoffix Looking at the fix and the crash, the magic is that add_I doesn't auto-decont 13:14
m: my $x = 42; use nqp; say nqp::add_I(2, $x, Int)
camelia P6opaque: get_boxed_ref could not unbox for the representation '20' of type Scalar
in block <unit> at <tmp> line 1
Zoffix m: my $x := 42; use nqp; say nqp::add_I(2, $x, Int)
camelia 44
jnthn None of the _I things to 13:15
*do
lizmat so I wonder why we didn't spot this earlier 13:16
anyhows, fixed now
jnthn Is there a test added to cover it also?
Zoffix jkramer: as far as the language is concerned: yes hashes/maps/setties/baggies/mixies are randomized, but .say/.print/.perl output them in sorted order.
lizmat jnthn: not yet afaik
Zoffix jkramer: there's still a ticket for the latter part RT#131244 13:17
synopsebot RT#131244 [new]: rt.perl.org/Ticket/Display.html?id=131244 Baggy/Setty .Str/.gist/.perl needs to guarantee order, like Map/Hash do
buggable New CPAN upload: Sparrowform-0.0.5.tar.gz by MELEZHIK cpan.metacpan.org/authors/id/M/ME/...0.5.tar.gz 13:22
piojo1 are there any tests for mix/bag/set? I couldn't find any when grepping around rakudo/t 13:27
13:28 broquaint joined
lizmat ls t/spec/*/bag* 13:28
ls t/spec/*/mix*
ls t/spec/*/set*
Zoffix uses `tree -f | grep -i SOME_TERM` 13:29
piojo1 lizmat: in which repository? 13:30
Zoffix and if that fails `grep -FIRn SOME_TERM`
piojo1: github.com/perl6/roast/
piojo1 Zoffix: ahh, thanks
moritz ack -i some_term
Zoffix piojo1: if you run `make spectest` in rakudo's checkout, it'll automatically clone roast into t/spec
piojo1 I just discovered ag. It's a little faster than ack :)
Zoffix piojo1: and after you write the test, you can check that it runs with `t/fudgeandrun t/spec/blah/blah/test.t` 13:31
piojo1 thanks! 13:32
13:33 mempko left
Zoffix (there's also ZScript where you can run `z t blah/blah/test.t`) 13:33
huggable: zscript 13:34
huggable Zoffix, Helper script for Rakudo Perl 6 core development: github.com/zoffixznet/z
El_Che lizmat: the fosdem thing, is that tomorrow (when irc is faster than searching your mailbox)?
lizmat yes
El_Che ok, making arrangements, thx 13:36
timotimo help, p6weekly keeps putting my twitter handle in messages it pre-makes when you hit the "share via twitter" button 13:37
lizmat ?? 13:40
timotimo twitter.com/ApopheozRu/status/9557...3444775936 13:43
Zoffix lol 13:44
moritz timotimo: maybe that's configured somewhere in that wordpress instance? 13:45
if so, maybe lizmat can help
lizmat eh... I have no idea :) 13:46
timotimo i already disconnected my accounts from the "sharing" page in the wordpress site settings
Zoffix timotimo: based on this, if this is the plugin you use, you'd just remove the username here: cdn-images-1.medium.com/max/1600/0...7Iw2fx.jpg ?
timotimo i don't have that entry in my settings drowdpon 13:47
aha!
there's a "sharing" entry there where i put the twitter username at some point 13:48
so should i put in a different username?
Zoffix @perl6org ? 13:49
timotimo that was my first thought, too
oh, wow. the views are *way* up today 13:50
3k views by 2.5k visitors
El_Che impressive
moritz no objections against using @perl6org there 13:51
timotimo the stats for 2018-01-22 through 2018-01-28 are currently at 3.3k views, 2.6k visitors. the periods before that were around 1k views, 500 visitors. the last peak was 2017-08-21 through 2017-08-27 with 5.5k views by 4.2k visitors 13:53
El_Che reddit?
timotimo 2k referred by hacker news, 76 by twitter, 65 from search engines, 50 from hckrnews.com and 30 from reddit 13:54
for the current post
El_Che maybe hn's popularity is worth investing time replying there, but those kind of discussions look tiring 13:55
lizmat yeah, I guess I'm going to have me a login after all 13:56
El_Che be careful not to get burnt out
Zoffix Heh :) I briefly looked at the comments; looks like the majority are by the marketing and branding experts telling core devs to give up :)
El_Che lizmat: I got a reddit login since yesterday :) 13:57
lizmat yeah, the irony wasn't lost on me :-)
13:58 aborazmeh joined, aborazmeh left, aborazmeh joined 13:59 stmuk_ joined 14:00 yoleaux left, yoleaux joined, ChanServ sets mode: +v yoleaux, telex left 14:01 telex joined 14:02 stmuk left, cdg joined
jnthn Heh, so first it was "Perl 6 is vaporware", then it was "Perl 6 still isn't released yet", and now it's "can't you just give up?" Dammit, even naysayers ain't what they used to be. :P 14:03
Zoffix :) 14:04
14:07 itaipu left, cdg left, jeromelanteri left 14:10 cdg joined
moritz now they are "time to recognize that the cost is sunk" 14:11
lizmat Hacker News telling me: You're posting too fast. Please slow down. Thanks 14:12
El_Che you see, perl 6 may not be fast yet, bit its core devs are!
Zoffix :) 14:13
lizmat++ # responding to people
Zoffix &
14:13 Zoffix left 14:14 itaipu joined
piojo1 lizmat: I PR'd a test. I couldn't manage to run it with "fudgeandrun", so I ran it as "perl6 t/.../whatever.t". I'm not sure if that's wrong, but I'm on Windows, so I'm used to things not working. 14:14
lizmat piojo1: thanks 14:15
(shouldn't we see a notice of that PR here ?)
piojo1 that's what I thought!
maybe it needs to wait for tests to run
oh, Zoffix merged it really fast! 14:17
Thanks :)
timotimo oh, we're still putting stuff into master? 14:24
it's too late for the udp datagram class, though, i'm sure
[Coke] timotimo: on roast? yes?
yoleaux 21 Jan 2018 11:18Z <tbrowder> [Coke]: I sent the CLA via snail mail (USPS) to the address on the form.
El_Che timotimo: if you don't tell, I won't either
timotimo haha
[Coke]: there was a fix to rakudo/master on top of the roast commit 14:25
14:25 Zoffix joined, geospeck joined
Zoffix piojo1: yeah I merged; the notifications for roast/rakudo/nqp/MoarVM repos are only in #perl6-dev 14:26
-1 on merging datagram stuff right now. Let the release happen. It's already $n days late :| 14:27
timotimo right
after all, i'm not happy with the API yet
and haven't written a single test, nor even ran the spectest we already have ;)
piojo1 Time to go. Thanks for the tips about testing :) 14:29
14:29 piojo1 left
player Where are modules installed by Zef are stored? Someone suggested .perl6, but I didn't find modules in it. 14:30
timotimo you can ask zef to "locate" a module for you
Zoffix player: irclog.perlgeek.de/perl6/2018-01-20#i_15714393 14:31
14:31 Zoffix left
buggable New CPAN upload: P5caller-0.0.1.tar.gz by ELIZABETH cpan.metacpan.org/authors/id/E/EL/...0.1.tar.gz 14:32
jnthn zef locate Some::Module::Name
player I'm sorry, Zoffix, I didn't notice your post last time.
jkramer Hmm I ran into a problem here. I need to append lots of elements to an already huge array. I'd like to just $a.append($b), but .append wants a slip or whatever it's called, not an array. If I do $a.append(|$b) however, P6 dies with "Too many arguments in flattening array." 14:33
jnthn What is $b?
jkramer Do I need to push each element individually?
$b is an Array
14:33 yoleaux left 14:34 yoleaux joined, ChanServ sets mode: +v yoleaux
jnthn @$b 14:34
timotimo m: my $a = [1, 2, 3]; my $b = [9, 9, 9, 9, 9, 9]; $a.append($b); say $a
camelia [1 2 3 [9 9 9 9 9 9]]
timotimo m: my $a = [1, 2, 3]; my $b = [9, 9, 9, 9, 9, 9]; $a.append($b.list); say $a
camelia [1 2 3 9 9 9 9 9 9]
timotimo right
jnthn Things are generally neater if using the @ sigil for arrays :)
player Zoffix, and where precompiled modules are installed?
timotimo m: my $a = [1, 2, 3]; my $b = |[9, 9, 9, 9, 9, 9] xx 100000; $a.append($b.list); say $a
camelia [1 2 3 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 ...]
jkramer Weird, what does @ change? Doesn't it just add a type-constraint to Array or List or something? 14:36
timotimo in this case it just puts .list at the end 14:37
much like $@foo will be @foo.item
14:38 sena_kun left, Zoffix joined
Zoffix player: why? What are you trying to accomplish? 14:38
jkramer: your arrays is conted, so it gets treated as one item. See perl6advent.wordpress.com/2017/12/...oneandonly
14:39 sena_kun joined 14:40 domidumont left 14:41 domidumont joined
buggable New CPAN upload: Sparrowform-0.0.6.tar.gz by MELEZHIK cpan.metacpan.org/authors/id/M/ME/...0.6.tar.gz 14:42
14:43 sena_kun left
jkramer Zoffix: Yeah I know, I was just wondering why Array.append wouldn't accept another Array as argument :) 14:44
Zoffix player: `zef locate` gives you SHA id of the distro, with full path to the source file. The binary precomp file is in adjacent directory with the same SHA id: gist.github.com/zoffixznet/f0083de...a8717e792e
m: my @a; @a.append: [<a b c>]; dd @a
camelia Array @a = ["a", "b", "c"]
Zoffix jkramer: it does accept it.
m: my @a; @a.append: $[<a b c>]; dd @a
camelia Array @a = [["a", "b", "c"],] 14:45
Zoffix ^ there you're saying "Append one item, an Array object". Not "append items in this array"
"Eat a bag of apples" vs. "bring me a bag of apples". In the former, you're acting on the contents, in the latter you're acting on the single item: the container of other items 14:46
14:48 smash joined
Zoffix m: my @a; my $b = [<a b c>]; @a.append: $b; @a.append: @$b; @a.append: $b.cache; @a.append: $b.list; @a.append: $b<>; @a.append: $b.self; dd @a 14:49
14:49 domidumont1 joined
camelia Array @a = [["a", "b", "c"], "a", "b", "c", "a", "b", "c", "a", "b", "c", "a", "b", "c", "a", "b", "c"] 14:49
Zoffix And ^ there are various ways of remove it from the scalar container, so that its contents are worked on rather than the array itself, as a single item
jkramer Hmm I see
It does make some sense :) 14:50
14:50 domidumont left
titsuki .seen krunen 14:51
yoleaux I saw krunen 9 Jan 2016 16:24Z in #perl6: <krunen> [Coke]: (re moving Rakudo release tags) I thought I only said it was talked about at some point, but confused discussion happened. Sorry.
14:52 raschipi joined
[Coke] O_o; 14:53
Zoffix That was the Great Moved Tag rumor during 2015 Christmas. That our release was so crap we supposedly had to move the tag. 14:54
ZofBot: good times! 14:55
ZofBot Zoffix, Haven't seen it crash at all here
[Coke] Yah, i was there. :)
Zoffix :)
[Coke] I still have no idea how that got started.
14:57 geospeck left, aborazmeh left 14:59 ChoHag left 15:00 Zoffix left 15:06 ChoHag joined
moritz I had to move a tag once, because I had mixed up some numbers. But that was years before the 2015 Christmas release 15:11
jkramer Is there something nicer/shorter than this for take a range from an array without knowing if it's withing its bounds? $foo[4..^96].grep(*.defined) 15:12
buggable New CPAN upload: P5lcfirst-0.0.1.tar.gz by ELIZABETH cpan.metacpan.org/authors/id/E/EL/...0.1.tar.gz
jkramer I just want the elements in that range that exist, not a bunch of undefined values. But grepping the result seems inefficient 15:13
15:13 zakharyas left, zakharyas joined
[Coke] note that you can have an undefined value in an interior position as well. 15:14
moritz $foo.tail(-4).head(96-4) # something along those lines? 15:15
module inevitable off-by-one errors
m: say <a b c d e f>.tail(-2)
camelia ()
moritz m: say <a b c d e f>.tail(2)
camelia (e f)
moritz m: say <a b c d e f>.head(4) 15:16
camelia (a b c d)
jkramer m: <foo bar baz>.tail(-1).say
camelia ()
moritz $foo.head(96).tail(96-4)
jkramer moritz: from the tail docs: Returns an empty Seq if NUMBER <= 0
moritz jkramer: yes, I mixed them up. $foo.head(96).tail(96-4) might work
jkramer Ah right 15:17
lizmat m: my @a; @a[4] = 42; dd @a[]:v # sorta expected that to give (42,)
camelia (Any, Any, Any, Any, 42)
lizmat m: my @a; @a[4] = 42; dd @a[]:k # it does with :k
camelia (4,)
lizmat hmmm
15:18 zakharyas left
[Coke] m: my $foo = 1..10; given $foo { say $_[min(4,$_.elems)..min(96,$_.elems)]} 15:19
camelia (5 6 7 8 9 10 Nil)
[Coke] m: my $foo = 1..10; given $foo { say $_[min(4,$_.elems-1)..min(96,$_.elems-1)]}
camelia (5 6 7 8 9 10)
15:19 Nanjizal left
[Coke] m: my $foo = 1..10; given $foo.elems -1 { say $foo[min(4,$_)..min(96,$_)]} 15:20
camelia (5 6 7 8 9 10)
15:20 stmuk joined
[Coke] ^^ might be able to golf that. 15:20
(and also fix the first endpoint. :)
15:22 stmuk_ left
[Coke] m: my $foo = 1..10; given ($foo.elems-1) { say $foo[min(4,$_)..min(96,$_)]} 15:26
camelia (5 6 7 8 9 10)
[Coke] (oh, wait, that is right. :)
15:26 khisanth_ left
buggable New CPAN upload: P5ucfirst-0.0.1.tar.gz by ELIZABETH cpan.metacpan.org/authors/id/E/EL/...0.1.tar.gz 15:32
15:37 geospeck joined 15:38 geospeck left 15:39 khisanth_ joined 15:46 obfusk joined 15:49 yoleaux left 15:50 yoleaux joined, ChanServ sets mode: +v yoleaux 15:52 Ven`` joined
Ven`` .u А 15:52
yoleaux U+0410 CYRILLIC CAPITAL LETTER A [Lu] (А)
Ven`` .u а 15:54
yoleaux U+0430 CYRILLIC SMALL LETTER A [Ll] (а)
15:56 raschipi left, khw joined
jnthn m: my $vаr = 42; say $var # :D 15:57
camelia 5===SORRY!5=== Error while compiling <tmp>
Variable '$var' is not declared. Did you mean any of these?
&VAR
$vаr

at <tmp>:1
------> 3my $vаr = 42; say 7⏏5$var # :D
15:57 dakkar left
Ven`` jnthn: freenode's webchat font actually distinguishes them quite well 15:58
ilmari does perl6 have the notion of script runs?
metacpan.org/pod/release/ABIGAIL/p...cript-Runs
Ven``
.oO( Just kidding, Perl 6 is broken, we should all go home )
15:58 dakkar joined
moritz ilmari: I don't think so 15:58
jkramer m: 'lolololol' ~~ ^Inf 16:01
camelia ( no output )
jkramer m: say 'lolololol' ~~ ^Inf
camelia True
jkramer m: say ^100 ~~ ^50 16:02
camelia False
jkramer m: say 'lolololol' ~~ ^1000
camelia False
16:03 epony joined
jkramer m: say False ~~ ^Inf 16:03
camelia True
jkramer Is this a bug? Any smartmatch against ^Inf seems to be True 16:04
16:04 Zoffix joined
Zoffix m: say "lolololol" after 0 && "lolololol" before Inf 16:04
camelia True
AlexDaniel m: say ∞ ~~ ^Inf 16:05
camelia False
Zoffix m: say "z" after 0 && "z" before Inf
camelia True
16:05 wamba left
timotimo everything is less than Inf. even a dog. 16:06
Zoffix m: say "" ~~ ^Inf
camelia False
Zoffix jkramer: ^ not everything :) But yeah, every string is "before" Inf, so that'd match a lot of stuff
jkramer Shouldn't it just crash? :) 16:07
Zoffix jkramer: smartmatch never crashes 16:08
jkramer Hmm true I guess that's the point of it :)
Zoffix lizmat: BTW we already have like 1010 modules. I just looked at the build log and 12 have JSON errors in their META files so they're likely not on the site (PRs sent) 16:09
16:09 st_elmo joined
jkramer From the smartmatch docs: Any Range in stringy range X.min le $_ le X.max 16:10
But:
m: say (^Inf).min le 'lololol' and 'lololol' le (^Inf).max
camelia WARNINGS for <tmp>:
True
Useless use of "le" in expression "'lololol' le (^Inf).max" in sink context (line 1)
jkramer What
Zoffix m: 'lolololol' ~~ 1010101011..Inf
camelia ( no output )
Zoffix m: say 'lolololol' ~~ 1010101011..Inf
camelia True
jkramer That doesn't happen locally o_O
Zoffix jkramer: precedence of `and` is lower than arguments to `say`
jkramer m: say ((^Inf).min le 'lololol' and 'lololol' le (^Inf).max)
camelia False
jkramer Ah
Yeah I didn't use say in the repl :)
Anyway the smartmatch definition from the docs seems to disagree with the implementation 16:11
Zoffix jkramer: on which page is it on?
jkramer design.perl6.org/S03.html#Smart_matching
Zoffix jkramer: that's not documentation. That's historical, archived speculations 16:12
jkramer Oh
timotimo look at the big red bar it has at the top
Zoffix jkramer: docs.perl6.org/ is the documentation and for smartmatch, you'd mostly look at the docs for the "ACCEPTS" method for the RHS's type (tho I see that seems to be missing for Range type)
jkramer An up-to-date table like on the other site would be nice 16:14
Zoffix imagines it'd be huge 16:16
16:16 cdg left
Zoffix jkramer: actually, almost-table already exists: docs.perl6.org/routine/ACCEPTS.html :) 16:17
jkramer Zoffix: Yeah but as you said, it's incomplete :) 16:18
Geth doc: 0797ea6d4c | (Zoffix Znet)++ (committed using GitHub Web editor) | doc/Type/List.pod6
Remove rogue pod element
synopsebot Link: doc.perl6.org/type/List
Geth doc: 0048029c68 | (Zoffix Znet)++ (committed using GitHub Web editor) | doc/Type/Numeric.pod6
More precise desc. for Numeric.ACCEPTS
 16:22
synopsebot Link: doc.perl6.org/type/Numeric
16:23 yoleaux left, yoleaux joined, ChanServ sets mode: +v yoleaux
player Zoffix, Actually I want to build new instance of compiler and make sure that installed modules don't class with modules of another instance 16:26
Zoffix player: rakudobrew does precisely that; lets you have multiple versions of the compiler with own sets of modules (but don't know if you can have multiple instances of same version): github.com/tadzik/rakudobrew 16:28
player: zef also has `--install-to` option you can use to tell it to install module to specific location (it might be possible to stick that into zef's config file) 16:29
--install-to=inst#/some/path # yes `inst#` bit is part of it 16:30
m: say 42 ~~ (*); say (*) ~~ 42
camelia True
{ ... }
Zoffix m: say 42 ~~ (*); say ((*)) ~~ 42
camelia True
False
16:30 troys joined, clarkema left
Zoffix How come only one set of parens is needed on RHS to avoid curry, but on LHS two sets are needed? 16:30
Geth doc: 7babb3dbcc | (Zoffix Znet)++ (committed using GitHub Web editor) | doc/Type/Whatever.pod6
Expand Whatever.ACCEPTS

  - Document :U candidate
  - Include examples		 16:32
synopsebot Link: doc.perl6.org/type/Whatever
ilmari moritz: I think it might have been Karl Williamson who came up with the idea (and then implemented it for perl 5). feel free to steal it for perl 6 ;) 16:33
jkramer m: my $ = 'oops typo' 16:34
camelia ( no output )
jkramer m_ my $ = 'lol'; say $
Zoffix jkramer: $ is an anon state var 16:35
m: say $++ while ++$ < 3
camelia 0
1
jkramer Huh 16:36
Zoffix no idea where they're documented...
moritz ilmari: I'm thinking about a 6ish way to design it. If we had <scriptrun>, we could combine it with other regexes easily as <scriptrun> & <other regex>
because we already have the & operator 16:37
ilmari moritz: and that requries the two regexes to match the same part of the string?
moritz ilmari: correct 16:38
ilmari that is a very neat way of combining them (but I don't want to think about backtracking)
moritz it's matched with the NFA, at least if <other regex> is declarative 16:39
Zoffix jkramer: found it: docs.perl6.org/syntax/state#The_$_Variable 16:40
jkramer Zoffix: Thanks!
Geth doc: e1a6ab5f73 | (Zoffix Znet)++ (committed using GitHub Web editor) | doc/Language/variables.pod6
Add more index terms for anon state vars

Closes #1742		 16:43
synopsebot Link: doc.perl6.org/language/variables
player Zoffix, where precompiled modules are installed by default? If I have only one instance of compiler?
Zoffix player: if you run zef locate SomeModule it'll tell you
$ zef locate Test 16:44
===> From Distribution: CORE:ver<6.c>:auth<perl>:api<>
Test => /home/zoffix/rakudo/install/share/perl6/sources/C712FE6969F786C9380D643DF17E85D06868219E
I have rakudo build in ~/rakudo and ^that's where it installs them. Yours might be different
Zoffix & 16:46
16:46 Zoffix left 16:48 mempko joined 16:55 abraxxa left 16:58 wamba joined 17:00 domidumont1 left 17:11 lolo78 joined, lolo78 left 17:17 Ven`` left, cdg joined 17:20 scimon left 17:22 cdg left 17:24 mempko left 17:26 clarkema joined 17:27 domidumont joined
stmuk twitter.com/matz_translated <=- there is auto translated perl6 comment, unfortunately (or maybe fortunately) it's unclear what it means 17:29
jkramer Can I put operators in a variable as some kind of curried function? Similar to `let x = (*)` in Haskell for example?
17:29 troys is now known as troys_
geekosaur m: my &op = &[+]; say op(2, 3) 17:31
camelia 5
jkramer geekosaur: Sweet, thanks! 17:32
DrForr There's a more official way to do that as well; I think it's mentioned as *like* curry in the docs, but not quite the same. 17:33
17:33 telex left 17:35 yoleaux left, yoleaux joined, ChanServ sets mode: +v yoleaux
timotimo &[+] is syntactic sugar for &infix:<+>, which is the proper, full name of the operator 17:36
17:36 telex joined 17:40 gregf_ left 17:43 evalable6 left, evalable6 joined 17:44 dakkar left
jkramer m: say (0, 1, * + * ...^ * > 100) 17:45
camelia (0 1 1 2 3 5 8 13 21 34 55 89)
jkramer What is this, some kind of haskell-like list comprehension?
Is it documented somewhere? :)
timotimo it's the sequence operator, and it's beautiful
docs.perl6.org/language/operators#...e_operator - for some reason putting "..." in the search doesn't find it 17:46
jkramer Yup that's exactly what I tried. :) Thanks :)
17:46 natrys left
DrForr Egads. I'm trying to install react + a grid layout, and *these* people give us grief over installation?... 17:48
17:48 Zoffix joined
Zoffix The like curry thing is "assuming" method 17:49
m: my &add-five = &[+].assuming: 5; say add-five 42
camelia 47
Zoffix weird; search index is listed as `X<|...,operators>X<|...^,operators>X<|…,operators>X<|…^,operators>`, yet only unicode forms turn up 17:52
DrForr .assuming, yes, that's what I had in mind.
Zoffix Filed as D#1743 17:54
synopsebot D#1743 [open]: github.com/perl6/doc/issues/1743 Glitch with search and `...` op
17:58 troys_ is now known as troys 17:59 Nanjizal joined 18:06 clarkema left 18:09 Zoffix left, cdg joined, setty1 joined
jkramer When an object has an attribute that holds a Sub, how do I call it? $foo.attribute($arg) calls the getter apparently, not the Sub 18:14
18:14 preaction left 18:15 cdg left
jkramer Ah got it. $foo.attribute.(...) 18:16
18:17 mempko joined
timotimo yeah, another . or ().($args) 18:26
buggable New CPAN upload: Inline-Perl5-0.30.tar.gz by NINE cpan.metacpan.org/authors/id/N/NI/....30.tar.gz 18:32
player When 2018.01 will be released? 18:34
El_Che soon 18:35
any time
player Isn't there fixed schedule or plan? 18:36
El_Che yes, it should be out, but the core devs are doing some last minute fixing
AlexDaniel well, the minute is extended enough to make sure the release is great :) 18:42
player: if you're looking for rakudo star, then it usually comes out a bit later than the rakudo release 18:43
18:44 darutoko left 18:45 yoleaux left, yoleaux joined, ChanServ sets mode: +v yoleaux 18:50 Zoffix joined
Zoffix Maybe we should try out a new system? A week before release MoarVM/nqp/rakudo/roast get `pre-release` branch created and tests get run on it + users get invited to test it out. All the devs continue working on master and nothing but critical bug fixes get into `pre-release` branch. 18:51
I'm using post-release branch and it's slightly annoying because Issues don't auto-close and you need to pull regularly to avoid conflicts. I also see some devs still commit fixes to master. And really, a small last minute fix could bring in a huge bug with it. 18:52
El_Che Zoffix: I could love RC tars, that way I could use a branch of rakudo-pkg as a canary 18:53
[Coke] FWIW, I'd be happy with something more closely resembling gitflow, yes. (maybe have the branch be release-2018.05 if we don't want to reuse the branch name.)
El_Che as it tests several Linux distro's (cfr the 2017.12 bug)
[Coke] and we could lock down who can commit to the release branch via github.
(down to the release manager by default)
player AlexDaniel, I mean compiler releases that is released every month
AlexDaniel player: we will release rakudo this month, yes :) 18:54
Zoffix Boom: we reached a milestone. 1002 modules in ecosystem: modules.perl6.org/search/?q= 18:55
lizmat whee!
AlexDaniel Zoffix: something like this, yeah maybe. There's a little issue is that releases will be branched out
which is not necessarily a problem, just something to keep in mind 18:56
gitflow is an overkill though
18:56 cdg joined 19:00 zakharyas joined
Zoffix It'd just be handy for users across 4 repos to not care (i.e. not worry about pushing something too soon) when release happens. Especially since two of those repos are in general "perl6" group with lots of users having push access. 19:01
+ PRs won't be forgotten about ("oh can't merge now because too close to release. I'll merge later"... *forgotten*) 19:02
And release manager would have full control over which bugs are considered critical and are allowed to be cherry-picked after the large part of testing happened already 19:04
AlexDaniel stmuk: insert obligatory complaint about the lack of lts branches here :) 19:08
Zoffix :)
AlexDaniel stmuk: but actually pinging you so that maybe you have some ideas on the matter 19:09
19:09 mempko left 19:10 troys is now known as troys_ 19:12 Kaiepi joined
stmuk blinks 19:13
19:16 stmuk_ joined
stmuk_ I prefer the term MTS (Medium Term Support) anyway 19:17
promises less :)
19:18 stmuk left
El_Che stmuk_: don't promise, it's implied :) 19:19
19:21 zakharyas left
AlexDaniel
.oO( “short term support” because let's be realistic about the alternative )
 19:21
19:21 zakharyas joined, |oLa| joined 19:22 comborico1611 joined
stmuk_ Best Of Effort MTS for the Enterprise 19:24
19:29 ChoHag left
comborico1611 Feels like a Friday, no? 19:33
Zoffix Nope :) 19:34
19:35 mempko joined 19:36 yoleaux left, quotable6 left, yoleaux joined, ChanServ sets mode: +v yoleaux, quotable6 joined 19:39 Nanjizal left
comborico1611 I read some crap about Microsoft today. 19:40
Well, it's actually some bs slick marketing by Microsoft. 19:41
19:47 clarkema joined, Zoffix left 19:48 itaipu left
player I want to execute few external commands sequentially even if some of them fails, however straightforward(run 'false'; run 'message') doesn't print 'message' 19:51
run 'false'; run 'echo', 'msg' 19:52
19:54 ChoHag joined 19:55 Kaiepi left
geekosaur capture and use the result of "run" instead of letting it be sunk 19:55
19:55 Kaiepi joined
geekosaur if the command fails, default sink will trip the Failure. if you capture and test the result then it will not throw 19:56
19:56 itaipu joined 19:58 Kaiepi left 20:01 cdg left 20:02 Kaiepi joined 20:03 Nanjizal joined 20:04 clarkema_ joined, clarkema_ left 20:05 clarkema left 20:06 mahafyi left, itaipu left 20:09 TeamBlast left 20:10 TeamBlast joined 20:11 Zoffix joined
Zoffix FWIW, there's no Failure in this case. Only the Proc object and Proc with non-zero .exitcode throws when sunk. 20:11
player: just do: |run 'false'; |run «echo msg»;
(`|` in this context is a slip operator that will just call .Slip and wrap your Proc into a Slip (which is a List) and List.sink is a no-op and Proc won't get sunk and won't explode) 20:13
20:19 Ven`` joined 20:21 zakharyas left, yoleaux left, yoleaux joined, ChanServ sets mode: +v yoleaux
player Thanks, Zoffix. 20:24
It's better than using $ = run ... 20:25
geekosaur does that ensure sequentiality though? 20:26
geekosaur does not recall
Zoffix Good point 20:28
player: my bad don't use that. 20:29
20:29 konsolebox left
mst if it's a slip followed by a semicolon, surely the semicolon ensure sequentiality? 20:30
Zoffix player: generally, I'd imagine the normal way to do it would be to just write `so run «...»` but per R#1406 that warns currently annoyingly
synopsebot R#1406 [open]: github.com/rakudo/rakudo/issues/1406 Warning about non-useless use of `so`
Zoffix mst: a Proc is a Proc::Async behind the scenes that fakes being sync. And if you don't capture its output, the syncing points are .so/.Bool/.sink/.exitcode methods 20:31
Zoffix tries to verify
mst ah. so maybe |run('false').so ? 20:32
Zoffix No need for | in that case. The mere .so is enough. parentheses are gross tho :P
This runs sequentially tho: |run «perl6 -e "sleep 1; say q|meow|"»; |run «perl6 -e "say q|foo|"»; sleep 3 20:33
s: Proc, 'so', \()
SourceBaby Zoffix, Sauce is at github.com/rakudo/rakudo/blob/7361...Mu.pm#L106
Zoffix s: Proc.new, 'so', \()
SourceBaby Zoffix, Sauce is at github.com/rakudo/rakudo/blob/7361...Mu.pm#L106
Zoffix :(
Ahhh 20:34
player: nevermind, it's safe to use. If you don't capture anything it waits for end.
"self!wait-for-finish unless $!out || $!err || $!in;"
though you could use run('...').so to be really safe :} 20:35
20:35 konsolebox joined 20:37 comborico1611 left
Geth doc: 38ed8fbc7f | (Zoffix Znet)++ (committed using GitHub Web editor) | doc/Type/IO.pod6
Expand `run` section

  - Show how nice «...» is with `run`
  - Include info that sunk procs explode and how to deal with it		 20:37
synopsebot Link: doc.perl6.org/type/IO
Zoffix e: sub prefix:<⛵> { my $ = @_; Nil }; ⛵run 'false' 20:42
evalable6
Zoffix e: sub prefix:<⛵> { my $ = @_; Nil }; ⛵run 'false'; ⛵run «echo "Tis good"»
evalable6 Tis good
Zoffix git commit; git push; (to generate some publicity for "ZOMG! They got Unicode operatorz!" :)) 20:43
e: sub prefix:<⛵> (|) {}; ⛵run 'false'; ⛵run «echo "Tis good"» # overthinking it :) 20:44
evalable6 Tis good
20:45 comborico1611 joined
comborico1611 Anyone ever heard of Vagrant server? 20:46
Zoffix I heard the name, but no idea what that is. 20:47
DrForr I've used it before, but quite a while back. Basically the same idea as Docker, IIRC. 20:48
comborico1611 The author of this PHP book is using it. I'm new to the scene, if you catch my jive. I just don't want to be uber-hacked. 20:49
pwned-hacked, we would call it.
Not to be confused with phone hacked.
Thanks, DrForr. I just needed some reassuranace. 20:50
20:50 Zoffix left
DrForr Most of the cracking these days is supposed to be defended against as 'docker image stop; docker image delete ; docker image create... 20:51
El_Che whut? 20:52
DrForr Or not. Feel free to ignore me ; I'm staring at React code. 20:54
El_Che comborico1611: vagrant was the shit before docker arrived
now it's more of an abstraction for vm's
AlexDaniel fwiw « » is not nice with run, it's basically a jump from the shell hell to, erm, exactly the same but just on perl 6 level
comborico1611 El_Che: Yeah, I'm using it in a vm. 20:55
I don't actually see the VM, though. I guess it's there. Haha.
El_Che comborico1611: our spectre and meltdown friends should make it easier nowadays to steal info from outside the vm
comborico1611 I was expecting a graphical VM. It's Virtual Box.
El_Che although it's hard
AlexDaniel demonstrating it in the docs without any precautions is just harmful, and even more so if the example has just two arguments 20:56
El_Che comborico1611: what I would do is recreating the environment with virtualbox or docker myself
self and you learn also the lower layers
full-stack sticker for your laptop and stuff :)
21:00 espadrine joined 21:01 player left, yoleaux left, yoleaux joined, ChanServ sets mode: +v yoleaux
Geth doc: a8b908539d | (Aleks-Daniel Jakimenko-Aleksejev)++ | doc/Type/IO.pod6
We'd rather not advice «» to those who may not know enough

  «» is basically a jump from the shell hell to exactly the same but
on perl 6 level. Those who want to do it may do it, but if we really want to document this, then it needs to come with a warning and a blurb on what this may lead to.
Also, in that particular example there are just two arguments so using «» is not really justified.
Half-reverts 38ed8fbc7f1005db4cea683bb8dad3324da196ac.		 21:05
synopsebot Link: doc.perl6.org/type/IO
21:08 Zoffix joined
Zoffix AlexDaniel: what's "shell hell"? 21:08
AlexDaniel «rm $file» 21:09
Zoffix AlexDaniel: that doesn't tell me anything
AlexDaniel so you forget to put quotes and then things go bad 21:10
21:11 Zoffix left
AlexDaniel Zoffix: like? if the file has spaces, then it's split into two arguments 21:11
and so you delete wrong stuff
I think you already understand that, not exactly sure what kind of justification you're expecting from me 21:12
El_Che $file="" #init; ...; «rm -rf $file/»
21:12 rindolf left
AlexDaniel the essential issue in shells is that the “default” behavior is doing something that you don't want most of the time 21:12
and then perl 6 came along and brought that “awesome” behavior in «» 21:13
well it's better because you don't have to use «» of course, and so I personally don't
21:13 st_elmo left
AlexDaniel except in code golf maybe :) 21:14
21:14 st_elmo joined
AlexDaniel I'm not against documenting it, just not in that form 21:15
like “You can also use «» but be aware that «» splits stuff by whitespace which may result in problems that you typically have in shells, …” and then some examples
well, something like that
and I care about it that much because almost in any shell script out there you'll find issues like this 21:18
jnthn AlexDaniel: I guess you're aware of this, but the <<...>> form are called shell quotes :) 21:19
AlexDaniel and I recall reporting at least one serious security issue in an open-source project
yea, am aware
.oO( but post-modern shells don't do that! )
 21:20
jnthn I very rarely find myself using them, tbh
One of those "nice to have now and then" things 21:21
AlexDaniel IIRC fish shell has a different understanding of things, and does not need quotes while operating properly
jnthn But not a form I'd emphasize or default to, especially in the docs
21:22 ChoHag left
tyil should we add Perl 6 to this list: en.wikipedia.org/wiki/Comparison_o..._languages 21:25
gfldex tyil: more propaganda is always better :) 21:26
AlexDaniel by the way, fwiw 21:30
it's not enough to put quotes in run «rm "$file"» to make it actually work
well, on perl 6 yes, enough. But for rm itself you need -- also 21:31
21:31 stmuk joined
AlexDaniel level 21:31
tyil gfldex: on which repo would an issue for this be most suited? 21:32
AlexDaniel (so run «rm -- "$file"», which yea is arguably easier to read than ‘rm’, ‘--’, $file) 21:33
21:33 stmuk_ left
moritz run <rm -->, $file 21:34
AlexDaniel I guess we have a winner :) 21:35
moritz: I never thought about that.
moritz++
21:35 yoleaux left 21:36 yoleaux joined, ChanServ sets mode: +v yoleaux
AlexDaniel but maybe it doesn't work like that? 21:37
need | in front of <> I think
run |<rm -->, $file
gfldex tyil: github.com/perl6/marketing 21:38
AlexDaniel gfldex: we shouldn't view wikipedia as a marketing platform
21:38 mempko left
AlexDaniel even though it factually is :) 21:38
tyil I was thinking of mu
since I couldnt find anything else that had a fitting description 21:39
gfldex AlexDaniel: that we are to humble to marketing might the the reason why we are shit at it
21:40 yoleaux left, yoleaux joined, ChanServ sets mode: +v yoleaux
AlexDaniel runs to bed quickly so that he stops telling people what they should not do 21:40
I think it's a doc issue 21:41
we can doc the relevant parts, and maybe that'll be usable as a source for wikipedia purposes 21:42
21:44 ChoHag joined 21:45 preaction joined 21:49 trnh joined 21:51 cdg joined 21:52 mempko joined
tyil that sounds pretty valid 21:53
21:54 Ven`` left 21:56 Ven`` joined, troys_ is now known as troys 21:58 Nanjizal left
DrForr Well, that was rather unfun but I've got React and the react-grid-layout module playing with Cro. 22:01
[Coke] DrForr: that still sounds funner than $dayjob. :)
DrForr I have *got* to fix irssi's nick higlight color, I can't make out yellow-on-white. 22:03
22:04 cdg_ joined
[Coke] is just black & white here. :) 22:04
22:05 mempko left 22:07 cdg left 22:08 Ven`` left 22:11 Ven`` joined, lostinfog joined, lostinfog left, comborico1611 left
DrForr Incidentally grid layouts don't appear to nest, unless I've done something horribly wrong, which *is* remotely possible. 22:20
22:20 setty1 left 22:22 squashable6 left, unicodable6 left, committable6 left, bloatable6 left, squashable6 joined, trnh left, unicodable6 joined, committable6 joined, ChanServ sets mode: +v committable6, bloatable6 joined
El_Che DrForr: it's not irssi, but your terminal colour layout 22:26
DrForr Oooh, in fact it *really* doesn't want you nesting grids, as my most recent CLI invocation 'killall -9 firefox' can attest to. 22:28
That was fun. G'night. 22:29
22:31 trnh joined 22:34 st_elmo left 22:37 geraud joined 22:40 notostraca joined 22:45 stmuk_ joined, ChoHag left 22:46 notostraca is now known as TEttinger 22:47 stmuk left 22:53 Ven`` left 22:57 ChoHag joined 22:59 Ven`` joined 23:05 itaipu joined 23:08 domidumont left, Ven`` left 23:09 domidumont joined, greppable6 left, greppable6 joined, ChanServ sets mode: +v greppable6 23:11 Ven`` joined
timotimo DrForr: if it's css grids you can definitely nest them 23:16
23:16 mempko joined
timotimo css grids being the new awesome sauce 23:16
23:18 comborico1611 joined
buggable New CPAN upload: Net-IP-Lite-Perl6-1.0.0.tar.gz by TBROWDER cpan.metacpan.org/authors/id/T/TB/...0.0.tar.gz 23:22
New CPAN upload: Net-IP-Lite-Perl6-1.0.0.zip by TBROWDER cpan.metacpan.org/authors/id/T/TB/...-1.0.0.zip
New CPAN upload: Geo-Ellipsoid-Perl6-1.0.0.tar.gz by TBROWDER cpan.metacpan.org/authors/id/T/TB/...0.0.tar.gz
23:23 Ven`` left, Ven` joined 23:24 pmurias left
MasterDuke when does it come back around to doing everything with HTML tables again? 23:34
23:35 traxex joined 23:46 ChoHag left, sftp left 23:47 ChoHag joined, yoleaux left 23:48 yoleaux joined, ChanServ sets mode: +v yoleaux, sftp joined 23:49 comborico1611 left

Powered by Cro and the Raku® Programming Language.

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

Thank you!