»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.
00:13 ugjka left 00:20 Bigger left 00:24 Actualeyes joined 00:29 wamba left 00:33 pierre_ joined 00:36 espadrine left 00:39 labster left 00:53 rurban1 left 00:58 aborazmeh joined, aborazmeh left, aborazmeh joined 01:04 AlexDaniel joined 01:07 xinming left 01:08 xinming joined
jnthn So I finally took the time to write up a full answer on the mutator question. I'm glad, because it comes up every so often, and now I've got my answer to point to: 6guts.wordpress.com/2016/11/25/per...ood-thing/ :) 01:10
Also my browser doesn't seem to think spell checking is something it wants to do, and it's 2am, so something will have sneaked past me. 01:11
But I'm going to sleep now. :)
01:12 xinming left
viki jnthn: night \o :) 01:12
01:13 xinming joined
jnthn 'night o/ 01:17
01:20 xinming left 01:21 xinming joined
viki jnthn++ great post :) 01:24
jnthn: the main thing I've learned is that if I repeat a bad idea often enough, you'll write great blogs on the topic ;) 01:25
01:25 xinming left 01:26 xinming joined
samcv is there something i can use that is faster than JSON Tiny, (tried JSON fast but it will fail sometimes saving my hashes) 01:27
for saving data from hash tables that hold strings
doesn't have to be JSON at all
01:30 Eddward joined
viki So .perl then? :) 01:32
m: my %h = :42foo; :43bar; my $data = %h.perl; my %new-hash = $data.EVAL; say %new-hash 01:33
camelia rakudo-moar 64343d: OUTPUT«WARNINGS for <tmp>:␤Useless use of ":43bar" in sink context (line 1)␤{foo => 42}␤»
viki m: my %h = :42foo, :43bar; my $data = %h.perl; my %new-hash = $data.EVAL; say %new-hash
camelia rakudo-moar 64343d: OUTPUT«{bar => 43, foo => 42}␤»
Eddward Is there a short hand for assigning the same value to multiple keys in a hash? 01:37
MasterDuke m: my @a = <a b>; my %h; %h{@a} = 2 xx @a.elems; dd %h 01:39
camelia rakudo-moar 64343d: OUTPUT«Hash %h = {:a(2), :b(2)}␤»
Eddward rakudo: my %p ; %p<a b c> = 1 xx *; %p.perl.say 01:40
camelia rakudo-moar 64343d: OUTPUT«{:a(1), :b(1), :c(1)}␤»
Eddward That works, thanks! 01:41
MasterDuke np
01:41 ocbtec left
viki m: my %p = <a b c>.Bag; dd %p 01:43
camelia rakudo-moar 64343d: OUTPUT«Hash %p = {:a(1), :b(1), :c(1)}␤»
MasterDuke i really like xx *, hadn't thought of that 01:44
viki Yeah, neat
01:45 labster joined
Eddward I'm tweaking blogs.perl.org/users/ken_youens-cla...eader.html 01:45
my %pause is default(.05); 01:46
%pause<! ? .> = .3 xx *;
%pause<, ;> = .1 xx *;
:
sleep %pause{$_};
viki Eddward: but what about the default? 01:47
Eddward my %pause is default(.05);
viki doh
I need to go sleep :D
Eddward I've been testing with camelia. I don't know if it actually works. 01:48
Is there a debian repository for current * releases? 01:49
viki No idea 01:50
MasterDuke not sure about a repo, but i believe El_Che makes packages
viki That's just for regular Rakudo. github.com/nxadm/rakudo-pkg/releases
Eddward cool
viki huggable: deb :is: CentOS and Debian Rakudo packages: github.com/nxadm/rakudo-pkg/releases 01:51
huggable viki, Added deb as CentOS and Debian Rakudo packages: github.com/nxadm/rakudo-pkg/releases
dalek c: dc154aa | MasterDuke17++ | doc/Type/Hash.pod6:
Add examples using slices
 01:57
synopsebot6 Link: doc.perl6.org/type/Hash
MasterDuke Eddward++ 01:58
02:08 eisen74 left 02:21 pierre_ left 02:25 kalkin- left, kalkin-_ joined 02:27 pierre_ joined 02:32 pierre_ left 02:33 labster left, grondilu left 02:37 rpburkholder joined 02:39 setty1_ joined 02:41 setty1 left 02:48 ugjka joined 02:52 eisen74 joined 02:56 eisen74 left 03:01 pierre_ joined 03:02 BenGoldberg joined 03:06 xtreak joined 03:19 AlexDaniel left
perlawhirl anybody awake 03:21
geekosaur not very 03:23
kalkin-_ half
03:23 kalkin-_ is now known as kalkin-
Eddward turkey coma here 03:24
MasterDuke barely
perlawhirl how does 'perl6 --doc=Markdown' or perl6 --doc=HTML' work? does it just look for any module called 'Pod::To::Foo'
or are HTML and Markdown special cased? 03:26
MasterDuke perlawhirl: github.com/rakudo/rakudo/blob/nom/...s.nqp#L853 03:28
perlawhirl ahh, thanks
03:28 noganex_ joined
perlawhirl MasterDuke++ 03:28
03:29 canopus left 03:31 noganex left 03:35 canopus joined 03:36 imcsk8_ joined 03:39 imcsk8 left 03:41 labster joined
perlawhirl can I access a modules $=pod from a script? 03:41
ie, from a separate file
geekosaur pretty sure you'd have to actually load the module
perlawhirl so say i have 'Foo.pm'. in another file, say 'foo.p6' I can 'use Foo;' but can I then access Foo.pm's $=pod from foo.p6? 03:42
PerlJam perlawhirl: I think you're Foo.pm would need a name to attach it to. (i.e. module Foo or package Foo or class Foo etc.) Then you'd get at it with Foo.WHY 03:48
s/you're/your/ 03:49
ugexe run('perl6', '--doc', $path-to-code) 03:50
geekosaur it also looks like $=pod should have structure including module names, so it should be somewhere inside $=pod?
ugexe i guess you probably want the structured $=pod 03:51
geekosaur but I don't know offhand how that works
perlawhirl ugexe: yeah. it's just for some minor customer formatting of a module. i could always just add a cmdline option to my modul that spits out $=foo.perl, and eval it into my formatter 03:52
but i might be overthinking it
PerlJam: yep... my module is a class, so Foo.WHY works for me 03:53
thakns
erm thanks
hmm, maybe not. thanks anyways people... I can do what I want by writing a Pod::To::Foo module, but was just trying to be lazy :D 03:55
04:00 kyclark_ joined
ZzZombo I don't understand where am I supposed to look for trait docs, the syntax and whatnot? 04:02
PerlJam ZzZombo: docs.perl6.org/ search for "trait" perhaps 04:05
MasterDuke docs.perl6.org/type/Signature#Para..._Modifiers
PerlJam MasterDuke++ 04:06
perlawhirl Unfortunately, i don't think all the traits are documented, at least not in the one plce 04:07
some are documented under Routine
docs.perl6.org/type/Routine
PerlJam There don't seem to be docs on how to make your own traits either 04:08
perlawhirl and also some under Variable (because, they are variable traits) 04:09
docs.perl6.org/type/Variable
PerlJam Don't forget Attribute too 04:10
perlawhirl Perljam: what do you want to do with a custom trait? could a Role serve your purpose?
04:10 pierre_ left
perlawhirl PerlJam: yeah, maybe Language needs a new 'Traits' page 04:11
PerlJam I don't want to do anything with them at the moment, but making traits is a reasonable design alternative sometimes.
04:12 kyclark_ left 04:16 khw left 04:20 pierre_ joined
PerlJam Earlier today I translated some Perl 5 code that uses PDL to do some computations into Perl 6. Mostly it looks the same except that I used @foo instead of piddle $foo and there are many operators encased in «». And I got to eliminate a routine because rotor() exists :) 04:21
If I would have had a routine that could fit a non-linear curve, I could have finished the translation 04:22
perlawhirl rotor is awesome... doing the same thing in languages that do not have a similar routing is typically very ugly 04:23
04:24 pierre_ left
PerlJam yep. 04:27
Actual line in the code: my @segs = @p.rotor($n => -($n / 2)); # 50% overlap 04:28
:-)
perlawhirl nice. 04:29
04:38 pierre_ joined 04:52 kyclark_ joined 04:53 kyclark_ left 04:59 Eddward left
gfldex PerlJam: the ENODOC has been noted 05:03
05:10 Cabanossi left
TEttinger what does rotor do? I have seen it here a few times as part of code reductions 05:10
the name isn't at all familiar
05:11 Cabanossi joined
PerlJam TEttinger: it lets you do sliding windows on a list with windows of different sizes (if you want) and different size gaps (if you want) between the windows. 05:12
so you turn a single list of values into multiple lists of values, each list being your "window" into the original list
TEttinger oh yeah, that's handy. clojure has it, java oh hell no 05:13
PerlJam m: my @a = ^10; dd @a.rotor(3 => 2);
camelia rakudo-moar 64343d: OUTPUT«((0, 1, 2), (5, 6, 7)).Seq␤»
PerlJam m: my @a = ^10; dd @a.rotor(3 => 2, :partial);
camelia rakudo-moar 64343d: OUTPUT«((0, 1, 2), (5, 6, 7)).Seq␤»
PerlJam m: my @a = ^10; dd @a.rotor(4 => 2, :partial); 05:14
camelia rakudo-moar 64343d: OUTPUT«((0, 1, 2, 3), (6, 7, 8, 9)).Seq␤»
TEttinger so the 2 is the gap?
PerlJam yep
and a negative gap allows for overlap
m: my @a = ^10; dd @a.rotor(4 => -2);
camelia rakudo-moar 64343d: OUTPUT«((0, 1, 2, 3), (2, 3, 4, 5), (4, 5, 6, 7), (6, 7, 8, 9)).Seq␤»
PerlJam and I'm clearly tired as I can't think to demo the :partial :) 05:16
m: my @a = ^10; dd @a.rotor(3 => 1);
camelia rakudo-moar 64343d: OUTPUT«((0, 1, 2), (4, 5, 6)).Seq␤»
PerlJam m: my @a = ^10; dd @a.rotor(3 => 1, :partial);
camelia rakudo-moar 64343d: OUTPUT«((0, 1, 2), (4, 5, 6), (8, 9)).Seq␤»
PerlJam there
it allows for windows that aren't "full" or a "partial window" 05:17
TEttinger clojure does something similar, it just expresses gap differently. (partition 3 5 (range 10)) ; returns ((0 1 2) (5 6 7)) 05:18
(partition 4 2 (range 10)) ; returns ((0 1 2 3) (2 3 4 5) (4 5 6 7) (6 7 8 9))
it is, agreed, really useful
partition-all is the :partial thing 05:19
m: my @a = ^10; dd @a.rotor(4 => -4); 05:20
PerlJam can you give multiple different window/gap sizes with partition?
camelia rakudo-moar 64343d: OUTPUT«(timeout)»
PerlJam m: my @a = ^10; dd @a.rotor(1 => 2, 3 => 2);
camelia rakudo-moar 64343d: OUTPUT«((0,), (3, 4, 5), (8,)).Seq␤»
PerlJam m: my @a = ^20; dd @a.rotor(1 => 2, 3 => 2); 05:21
camelia rakudo-moar 64343d: OUTPUT«((0,), (3, 4, 5), (8,), (11, 12, 13), (16,)).Seq␤»
TEttinger in one call? not that I'm aware of. it's possible but would be tricky to implement
perlawhirl @rotar can also do things like this
TEttinger why the closing comma after (0,)
perlawhirl m: say (^12).rotor(3,2,1)
camelia rakudo-moar 64343d: OUTPUT«((0 1 2) (3 4) (5) (6 7 8) (9 10) (11))␤»
PerlJam TEttinger: because it's a one element list. 05:22
TEttinger ohhh
that makes sense
I do feel like there's some deeply good ideas in perl 6 that don't really have much chance to get discovered because the language itself is so massive and people may not try to learn the whole (still evolving) thing 05:24
it's interesting. as I've learned more programming languages I've really started to return to ones that aren't "maximalist" designs, since I have a chance to actually understand the subtleties of a simpler language where I absolutely would not with Perl6 for years 05:27
PerlJam That's why it's important the people write books and blogs and such. Different people coming from different points of view will highlight different aspects of the language.
Perl 5 has the same problem even if it's on the order of an aleph-0 language and Perl 6 is an aleph-1 :-) 05:28
TEttinger better than a nun-NaN
I don't know what an aleph-n is 05:29
is it like a Kardashev civilization type? 05:30
(very low numbers currently expected, very high numbers are only found in science fiction)
PerlJam en.wikipedia.org/wiki/Aleph_number 05:32
similar I guess
05:34 pierre_ left 05:35 pierre_ joined
perlawhirl I've never fully accepted the argument that a language is too big. I don't know all of perl 5 but I use it to great effect. 05:36
There's a lot of stuff in your kitched... but you don't need all of it to make an omelette
Perl 6 is big, but most people will use a fraction of, and everyone's fraction, tho it overlaps, is slightly different 05:37
PerlJam perlawhirl: exactly
There are bits of Perl I've *never* used in the 24-ish years I've been programming in it. There are bits I've tried to use, but they turned out to not be really useful (reset is an example of the former and study (when it did something) an example of the latter) 05:40
in a sense those things I've never used aren't really part of the language for me, though they may be for someone else 05:41
perlawhirl agreed. the language feature you've never used is integral to someone else 05:42
well... not integral to _someone_ :D ... i mean... is an integral feature _for_ them. my english isn't working today. 05:43
05:45 pierre_ left 05:52 pierre_ joined 05:54 rpburkholder left 05:57 pierre_ left 06:00 rurban joined, rurban left 06:01 dayangkun joined 06:03 skids left 06:30 zacts left, xtt joined 06:35 domidumont joined 06:36 rpburkholder joined, perlawhirl left 06:37 xtt2 joined 06:41 pierre_ joined, eisen74 joined, xtt left
gfldex m: &trait_mod:<is>.WHAT.say 06:42
camelia rakudo-moar 64343d: OUTPUT«(Sub)␤»
06:43 BenGoldberg left 06:48 zacts joined 06:53 RabidGravy joined 06:55 domidumont left 07:00 aborazmeh left 07:04 dayangkun left 07:06 rurban joined
dalek c: f06eaad | gfldex++ | doc/Type/Sub.pod6:
doc trait_mod (yes, we totally forgot about that)
 07:10
synopsebot6 Link: doc.perl6.org/type/Sub
gfldex PerlJam++ # for spotting the ENODOC 07:11
07:12 seatek joined 07:14 bjz joined
gfldex m: multi sub trait_mod:<is> (Sub $s, :$foo) is foo { say 'is foo called' }; sub bar {}; &trait_mod:<is>(&bar, :foo); 07:21
camelia rakudo-moar 64343d: OUTPUT«is foo called␤is foo called␤»
gfldex Perl 6 is fun :)
07:22 RabidGravy left
dalek c: f383d01 | gfldex++ | doc/Type/Sub.pod6:
show how to use traits at runtime
 07:23
synopsebot6 Link: doc.perl6.org/type/Sub
07:30 pierre_ left
dalek c: f6140d8 | gfldex++ | doc/Language/ (2 files):
link to trait_mod
 07:31
07:31 pierre_ joined 07:32 xiaomiao left
ZzZombo gfldex++ 07:32
thanks for that 07:33
gfldex ZzZombo: it's not complete. Traits are rather involved. 07:36
ZzZombo I know. And BTW, how do you pass more arguments? like "sub x is XYZ (param,:param2(val))"? 07:38
gfldex i don't think you can without changing the grammar 07:39
m: class A {}; multi sub trait_mod:<is>(Variable, :$foo){}; my $a is :foo(1, A => 1, 3) = A.new;
camelia rakudo-moar 64343d: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Invalid name␤at <tmp>:1␤------> 3it_mod:<is>(Variable, :$foo){}; my $a is7⏏5 :foo(1, A => 1, 3) = A.new;␤»
ZzZombo what do you mean then by "A single named argument defines the secondary name and may carry arguments when the trait is called"?
07:39 wamba joined 07:40 xiaomiao joined
gfldex m: class A {}; multi sub trait_mod:<is>(Variable, :$foo){}; my $a is foo(1, A => 1, 3) = A.new; 07:40
camelia ( no output )
gfldex the single argument can take a list and you can use a subsig to destructure it
(i think) 07:41
ZzZombo I see
07:41 eisen74 left
gfldex m: class A {}; multi sub trait_mod:<is>(Variable, :$foo($first, Pait $pair, $last) ){}; my $a is foo(1, A => 1, 3) = A.new; 07:43
camelia rakudo-moar 64343d: OUTPUT«5===SORRY!5===␤Shape declaration with () is reserved;␤ please use whitespace if you meant a subsignature for unpacking,␤ or use the :() form if you meant to add signature info to the function's type␤at <tmp>:1␤------> 3multi sub trait_mod:…»
gfldex m: class A {}; multi sub trait_mod:<is>(Variable, :$foo ($first, Pait $pair, $last) ){}; my $a is foo(1, A => 1, 3) = A.new;
camelia rakudo-moar 64343d: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Invalid typename 'Pait' in parameter declaration. Did you mean 'Pair'?␤at <tmp>:1␤------> 3t_mod:<is>(Variable, :$foo ($first, Pait7⏏5 $pair, $last) ){}; my $a is foo(1, A =>␤»
gfldex m: class A {}; multi sub trait_mod:<is>(Variable, :$foo ($first, Pair $pair, $last) ){}; my $a is foo(1, A => 1, 3) = A.new;
camelia rakudo-moar 64343d: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Can't use unknown trait 'is foo' in a variable declaration.␤at <tmp>:1␤------> 3, $last) ){}; my $a is foo(1, A => 1, 3)7⏏5 = A.new;␤ expecting any of:␤ TypeObject␤ default…»
gfldex well, you can do the destructureing inside the trait ofc 07:46
psch m: multi sub trait_mod:<is>(Variable $, :$foo ($a, *@t)) { }; my $x is foo([1, 2, 3]) = 1 07:48
camelia ( no output )
psch m: sub f($a, Pair $p, $b) { }; f 1, a => 1, 2 07:49
camelia rakudo-moar 64343d: OUTPUT«Too few positionals passed; expected 3 arguments but got 2␤ in sub f at <tmp> line 1␤ in block <unit> at <tmp> line 1␤␤»
psch gfldex: ^^^ the Pair you pass is a named
m: class A {}; multi sub trait_mod:<is>(Variable, :$foo ($first, Pair $pair, $last) ){}; my $a is foo(1, 'A' => 1, 3) = A.new;
camelia rakudo-moar 64343d: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Can't use unknown trait 'is foo' in a variable declaration.␤at <tmp>:1␤------> 3$last) ){}; my $a is foo(1, 'A' => 1, 3)7⏏5 = A.new;␤ expecting any of:␤ TypeObject␤ default…»
psch m: class A {}; multi sub trait_mod:<is>(Variable, :$foo ($first, Pair $pair, $last) ){}; my $a is foo([1, 'A' => 1, 3]) = A.new;
camelia rakudo-moar 64343d: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Can't use unknown trait 'is foo' in a variable declaration.␤at <tmp>:1␤------> 3ast) ){}; my $a is foo([1, 'A' => 1, 3])7⏏5 = A.new;␤ expecting any of:␤ TypeObject␤ default…»
psch ...although that doesn't seem to be the only problem :)
we probably want more detailed error message for "can't apply this trait" i suspect 07:50
*messages
m: sub f($a, Pair $p, $b) { }; f 1, 'a' => 1, 2
camelia ( no output )
07:51 xtreak left 07:53 Ven joined 07:54 Ven left 07:55 Ven joined
psch m: multi trait_mod:<is>(Variable $a, :@foo) { }; my $x is foo[1,2,3,:1a] = 1 # vOv 07:57
camelia ( no output )
07:59 xtt joined
psch m: multi trait_mod:<is>(Variable $a, :@foo [$firstpos, *@restpos, :$firstnamed, *%restnameds]) { }; my $x is foo[1,2,3,:1a] = 1 # vOv 07:59
camelia ( no output )
psch m: multi trait_mod:<is>(Variable $a, :@foo [$firstpos, *@restpos, :$firstnamed, *%restnameds]) { }; my $x is foo[1,2,3] = 1 # vOv
camelia ( no output )
07:59 rurban left
psch so, yeah, it can work, but "Can't use unknown trait" is, in some cases, not specific enough. i'm pretty sure we should be able to figure out if we can't apply it because of an argument mismatch 08:00
08:01 xtt2 left
psch ...maybe just add the invoking argument list to X::Trait::Unknown is enough 08:05
08:08 geekosaur left 08:09 rurban joined, rurban left 08:10 geekosaur joined 08:11 dayangkun joined 08:12 pierre_ left 08:15 xtreak joined 08:18 pierre_ joined 08:21 xtt2 joined 08:22 Ven left 08:24 xtt left, mohae_ joined 08:25 zakharyas joined 08:27 mohae left
dalek c: 7588852 | gfldex++ | doc/Type/Sub.pod6:
show how to call traits with multiple arguments
 08:28
synopsebot6 Link: doc.perl6.org/type/Sub
psch gfldex++ # docs \o/ 08:29
08:29 pierre_ left 08:32 Ven joined, dayangkun left 08:33 domidumont joined 08:34 dayangkun joined 08:36 eisen74 joined, AlexDaniel joined 08:37 pierre_ joined 08:38 domidumont left 08:39 dayangkun left, domidumont joined 08:44 g4 joined, g4 left, g4 joined 08:45 bjz left 08:52 Ven left 08:57 bjz joined 09:00 labster left 09:01 labster joined 09:08 AlexDaniel left 09:12 pierre_ left 09:14 msdfsd joined, bjz left
msdfsd Im thinking about joining the perl6 train 09:14
can I make native executables from perl?
gfldex no and there is no way to make bundles yet either but that is a planned feature
psch well, r-j bundles are kinda sorta possible, but experimental at best 09:15
i'm not sure how well it works with included modules, in any case
msdfsd well 09:17
here's to holding for perl7
09:17 msdfsd left, xtreak left
psch ...wow 09:17
09:18 xtreak joined
gfldex Perl 09:18
Perl 6 is meant to last for 100 years. So that could be a rather long wait for you.
psch they've already left... vOv 09:19
DrForr Another satisfied customer.
psch and i doubt they're reading the logs -- research doesn't seem to be a thing, considering we have like at least half a dozen tickets concerning that "binary deploy" thing
timotimo the problem clearly is that now that we've had the stable release of perl6, we can't add anything until perl7 09:20
that's how software works, isn't it?
psch exactly, everything is 1.0 forever
i mean, Perl 5 has been the same for 25 years too vOv 09:21
timotimo famous examples include windows version 2000.0, and google chrome version 54 million 2840.100
09:21 pierre_ joined
DrForr Why take the time to think when you can just Google an opinion and believe whatever the first link is? 09:24
psch honestly, i interpreted "joining the train" as "helping develop" at first anyway 09:25
09:26 pierre_ left
gfldex i interpret "joining the train" as an empty phrase used by one who has nothing to say 09:26
psch right, that's probably more sensible. along the lines of "i want to seem interested but only if you deliver literally everything i demand, and i fully know some demands aren't particularly reasonable" or something..? 09:28
09:28 andrzejku joined
gfldex my guess would be that the swiftly leaving fellow was send by his boss to check out this new Perl 6-thing he read about in a magazin 09:30
09:30 pierre_ joined 09:31 rindolf joined
DrForr Can't please 'em all. 09:33
El_Che DrForr: if someone could, it would be you :) 09:39
good day
09:41 labster left
DrForr Wow. Thanks. 09:41
09:41 BenGoldberg joined
DrForr Incidentally Perl6::Parser now passes 94% of its tests. 09:42
timotimo that's good news! 09:46
09:47 BenGoldberg left
seatek i heard that perl7 will have trains... 09:48
timotimo i like trains. 09:49
DrForr Take a look at the test suite to see how much it already handles, I think it currently handles files up to 40-50 lines of somewhat dense code.
timotimo that's ... not very much :(
seatek that's why we need trains
09:50 chaon joined
DrForr Oh, it's not limited in size in the way you're implying. 09:50
That's just because RosettaCode samples top out at 50-60 lines, it's perfectly capable of thousands of lines. 09:51
timotimo OK :)
rosettacode.org/wiki/Snake#Perl_6 09:52
^- quite simple code, though.
09:53 zakharyas1 joined 09:56 Ven joined, bjz joined 09:59 wamba left 10:04 chaon left
DrForr github.com/drforr/perl6-Perl6-Pars...e/master/t 10:04
Most of the existing issues are (1) comments and (2) contextualizer variables such as $[1]. 10:05
timotimo mhm 10:06
DrForr Contextualizer is actually trivial, I was going to take care of it last night but got distracted. 10:07
10:13 pierre_ left 10:14 andrzejku left, chee is now known as javascriptisnom 10:17 javascriptisnom is now known as chee 10:20 zakharyas1 left 10:21 pierre_ joined 10:26 perlawhirl joined, pierre_ left 10:32 xtreak left, Ven left 10:52 ugjka left 10:57 eisen74 left, pierre_ joined, pierre_ left 11:00 AlexDaniel joined 11:10 ugjka joined 11:13 gregf_ joined
viki is watching Toronto Perl Mongers slowly imploding :( 11:13
One of the emails from organizers gist.github.com/zoffixznet/c2ac5c6...2c00a01060 11:14
11:27 xtt2 left 11:28 rpburkholder left
TEttinger it is toronto and it is probably snowing since it's roughly wintertime. this is a time of year when I'd expect a little less commute tolerance from people not close-by 11:35
I can't recall the last time I've seen snow
viki heh 11:36
Snow wouldn't affect my commute :) And I don't think it's snowing proper yet
TEttinger is viki zoffix?
viki Who is Zoffix?
TEttinger gist is a bit of a giveaway 11:37
viki :)
TEttinger I wonder if there's some emoji IRC and Unicode could work together to support, that is the "Zoffix Synonym Indicator Emoji" 11:38
DrForr Who is Zoffix Galt? 11:39
TEttinger the lizard wizard vizier 11:40
11:44 wamba joined 11:46 Ven joined 11:52 bjz_ joined 11:53 bjz left
viki .ask MasterDuke can you look at this ticket? IIRC we decided this was a DITHWIDT and wasn't worth a fix, right? rt.perl.org/Ticket/Display.html?id=128551 11:53
yoleaux viki: I'll pass your message to MasterDuke.
viki m: say %(:4̈a) 11:55
camelia rakudo-moar 64343d: OUTPUT«{a => 0}␤»
psch m: say <4̈>.perl 11:56
camelia rakudo-moar 64343d: OUTPUT«"4̈"␤»
psch m: say val("4̈").perl
camelia rakudo-moar 64343d: OUTPUT«"4̈"␤»
viki Oh, wait, that was for fancy-pants Unicode up in there.
.tell MasterDuke never mind, I now recall what you and TimToady were fixing. This seems to be a different bug that silently fails to do the right thing 11:57
yoleaux viki: I'll pass your message to MasterDuke.
MasterDuke viki: yeah, similar, but not exactly the same. the other ticket (can't remember the #) was about using 'No ' characters instead of 'Nd'. TimToady ruled against supporting (or specifically checking for) 'No' 11:59
yoleaux 11:53Z <viki> MasterDuke: can you look at this ticket? IIRC we decided this was a DITHWIDT and wasn't worth a fix, right? rt.perl.org/Ticket/Display.html?id=128551
11:57Z <viki> MasterDuke: never mind, I now recall what you and TimToady were fixing. This seems to be a different bug that silently fails to do the right thing
11:59 wamba left
psch m: say :١a # this one i guess? 12:00
camelia rakudo-moar 64343d: OUTPUT«Unexpected named argument 'a' passed␤ in block <unit> at <tmp> line 1␤␤»
psch m: say (:١a) # this one i guess?
camelia rakudo-moar 64343d: OUTPUT«a => 1␤»
psch or support for it, probably
viki That's not a No :)
psch m: say uniprop("١") 12:01
camelia rakudo-moar 64343d: OUTPUT«Nd␤»
viki It's the outlier crop of numbers-that-aren't-digits
psch ahh, 'we decide to allow No instead Nd' is what MasterDuke was saying TimToady was saying..`?
+of
MasterDuke psch: the opposite
viki m: :⅒42
camelia rakudo-moar 64343d: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Bogus statement␤at <tmp>:1␤------> 3:7⏏5⅒42␤ expecting any of:␤ colon pair␤»
viki m: :⅒a 12:02
camelia rakudo-moar 64343d: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Bogus statement␤at <tmp>:1␤------> 3:7⏏5⅒a␤ expecting any of:␤ colon pair␤»
psch m: say uniprop "⅒"
camelia rakudo-moar 64343d: OUTPUT«No␤»
psch okay i'm bad at reading right now apparently vOv
MasterDuke 'Nd' is allowed. 'No' is not allowed, and TimToady also thought we shouldn't specifically check for them and give them their own error message 12:03
heh, 'No' makes things hard to parse. "No No is not allowed"
12:03 AlexDaniel left
MasterDuke s/things/sentences 12:04
psch always these persky words with their meanings *shakes fist*
DrForr inb4 Buffalo buffalo buffalo ... :) 12:06
12:07 Ven left, TEttinger left 12:08 rindolf left, Ven joined 12:09 Ven left, rindolf joined 12:12 Ven joined 12:15 Ven left 12:19 user9 left 12:20 kd_christoph joined, kd_christoph left 12:21 user9 joined, Ven joined
timotimo how do i figure out where in memory my executable is mapped? and especially how big the mapping is and where exactly it starts (and ends) 12:24
arnsholt Isn't that one of those things that modern OSes go out of theur way to make hard to figure out? 12:28
DrForr mutters something about Rowhammer. 12:29
timotimo well, address layout randomization, sure
but i'd like to just concatenate some data to my binary and get at that data later on
or some other simple way to put data into an executable that hopefully doesn't involve running a full compiler
arnsholt I think there's a separate section in the executable format for that 12:31
Googling seems to indicate that in ELF at least it's the rodata section
I'm curious what you need this for though
timotimo *cough cough* 12:32
arnsholt Sekkrit projects? =)
timotimo i don't want to get people's hopes up
anyway, there's lots and lots of stuff that comes after .rodata in the binary i'm looking at right now 12:33
the last thing i can see is .symtab and .strtab (which seem to sit at the same offset)
12:34 rpburkholder joined
arnsholt At any rate, ELF stuff is where you're gonna be wanting to look 12:34
timotimo on linux at least
arnsholt Yeah 12:35
OS X and Windows have different executable formats
perlawhirl hmm... who feels like helping me bikeshed a module name
I extended Pod::To::Markdown and used a little creative pod to create fenced code blocks 12:36
gist.github.com/0racle/e092c62fadd...579118c8b8
I have a module, but just questioning what a good name for it is... Pod::To::Markdown::SOMETHING
I'm already using this module to render the README.md for my Net::Netmask module... it's really nice to be able to have it automatically generated with code blocks 12:37
timotimo i don't want to have to write a custom ld script :\ 12:39
12:42 rpburkholder left
arnsholt timotimo: For this kind of stuff, I'm afraid it's hard to get out of that 12:48
And then you get to implement bindings to the right C functions/libraries to read the data back in =) 12:49
timotimo get what out of what?
arnsholt "Get out of" intended to mean "escape" in that case
Or was it the second line you were wondering about? 12:53
timotimo i didn't understand the second line, either ;)
i might end up with just fopen on /proc/self/exe :\ 12:54
also, we could totally have ABSOL P6 based on bulbapedia.bulbagarden.net/wiki/Abs...%C3%A9mon)
(Almost But Still Only Like Perl6)
arnsholt Well, once you've stashed data somewhere in the executable, you presumably want to get it back out. There's probably a C library for that
timotimo if i had a section that goes at the exact end of my binary, i could just grab the address of that section and go nuts 12:55
13:14 cygx joined
cygx timotimo: there are sevral ways to embed binary data into executables, some more hacky than others 13:15
but what do you mean by 'doesn't involve running a full compiler'
timotimo cygx: ideally, i'd have the program prepared to Deal With It, and then just use cat myprogram extradata > enhancedprogram 13:17
if you want to help me even more, how do i tell ld that i want it to make a .so file usable that i manually mmapped (or rather: that's Just Already There)
cygx well, with gcc, there's stuff like balau82.wordpress.com/2012/02/19/l...-with-gcc/ 13:18
personally, I've also just hot-patched a char[] of sufficient size that held a marker string 13:19
timotimo okay, but that's hardly a step up from turning my binary blob into a const char * ...
cygx then, you can append data to the executable and open the file from disk
timotimo i feared as much.
cygx on windows, the proper way to do it would involve resources 13:20
timotimo i don't know anything about that 13:21
but that also involves a compiler?
well ... linker
cygx yes
timotimo then i'll go with fopen on /proc/self/exe
13:22 araujo left
timotimo do you have a tip for the .so thing? 13:22
i mean i could mmap it to a file descriptor, and tell dlopen to open /proc/self/fd/myfiledescriptor
13:22 araujo joined
viki m: gist.github.com/zoffixznet/c90ebeb...b4ceffce3f 13:23
camelia rakudo-moar 64343d: OUTPUT«True␤<anon|176537216>.new␤<anon|176537216>.new␤<anon|176537216>.new␤<anon|176537216>.new␤<anon|176537216>.new␤<anon|176537216>.new␤<anon|176537216>.new␤<anon|176537216>.new␤<anon|176537216>.new␤<anon|176537216>.new␤<anon|176537216>.ne…»
dalek c: 86e56be | ZzZombo++ | doc/Type/Sub.pod6:
Update Sub.pod6
 13:24
synopsebot6 Link: doc.perl6.org/type/Sub
viki Any idea why my push-all impl is broken? Line 50 is called like 200 times, when it should be just 4 times and the .elems gives 0, when it should give 5
moritz ZzZombo: could you please write a bit better commit messages? "Update Sub.pod6" doesn't tell me what you changed, or why 13:25
ZzZombo for such small changes?
psch "Fix typo" is still more meaningful than "Update $file" 13:26
moritz ZzZombo: sure. "Fix typo" isn't more to write either
ZzZombo I didn't write it
moritz then please do in future :-)
timotimo that's what github suggests
psch ah, web commit?
cygx timotimo: I don't think here's an official way to do such a thing
timotimo i think you're right
13:27 Ven left
cygx it's possible, but you'll likely have to get your hands rather sirty, so to speak 13:27
*dirty
first google result: github.com/m1m1x/memdlopen
timotimo fantastic! only 500 lines of code! 13:30
that opens /proc/self/maps %) 13:31
i'll go with mmap + /proc/self/fd/*, then :) 13:32
this will be the worst proof of concept ever conceived, i'm sure
13:34 pierre_ joined
viki Any ideas? 13:37
Man, this 5-minute chunk of code is now taking 40minutes :(
What I don't get even more is if there's a bug that makes it go 200 iterations, but does it stop there and not go forever? :/ 13:38
dudz evening
dudz tips non existing hat
DrForr Afternoon. How go things?
dudz good thank you very much. 13:39
waiting for my hair to dry and for my body to commence shutdown procedure then i can get some zed's 13:40
DrForr Good luck with that.
13:40 khw joined
dudz what's all the entire gang upto this fine <timezone> ? 13:40
viki Trying to make a fast flat.iterator.push-all 13:41
And failing miserably
13:41 chacalmaneiro joined
DrForr At work, finished with some blogs.perl.org stuff and moving on to other work. 13:41
chacalmaneiro test
viki chacalmaneiro: test successful
psch m: use Test; ok 1
camelia rakudo-moar 64343d: OUTPUT«ok 1 - ␤»
dudz push is not array related right? 13:43
psch m: say Array.^can('push')
camelia rakudo-moar 64343d: OUTPUT«(Method+{<anon|54735920>}.new push Method+{<anon|54735920>}.new)␤»
DrForr The first rule of Tautology club shall be "The initial rule of Tautology club shall be the rule that is observed when joining the club."
psch dudz: it is, but push-all isn't push, it's part of the Iterator/Iterable interface
13:44 wamba joined, Ven joined
dudz an iterator doesn't have to be a number does it? 13:44
viki dudz: an iterator knows how to produce a sequence of numbers. 13:45
13:45 chacalmaneiro left
viki dudz: and you can ask it for, say, one value and it'll produce it. The .push-all asks for all values and asks for them to be pushed into a target. 13:45
ZzZombo m: say Str.^lookup('Int')
camelia rakudo-moar 64343d: OUTPUT«Int␤»
ZzZombo m: say Str.^lookup('Int').perl
viki docs.perl6.org/type/Iterator
camelia rakudo-moar 64343d: OUTPUT«method Int (Str:D $: *%_) { #`(Method|50741328) ... }␤»
dudz checking 13:46
timotimo wow 13:47
src/core/bytecodedump.c:425:1: internal compiler error: in get_dynamic_type, at ipa-polymorphic-call.c:1665
dudz arh, iterate as in, to proceed through right, not just counting
*question mark placed in the right place ^ 13:48
jnthn timotimo: Uh, wow...
ZzZombo docs.perl6.org/routine/Int doesn't list Str as having Int().
Why?
viki ZzZombo: would it be listed under Str? 13:49
*wouldn't
dudz i'm due for a perl6 t-shirt once that project begins *optomistic
*oreilly perl6 book
ZzZombo Str isn't here at all.
DrForr dudz: Party on! (I'm working with him on writing it.) 13:50
viki ZzZombo: then the asnswer is: no one added it 13:51
ZzZombo Uh, I thought it was autogenerated in some way, that assembled individual pieces together?
viki ZzZombo: well, yeah, but it's taken from documents written by humans. I think in this case the method is provided by Cool 13:52
dudz yeah i thought your nickname had some marketing to it DrForr :)
viki ZzZombo: so it'd be from Language/Type/Cool.pod6
DrForr Egads, I'm a brand?
dalek c: e71be07 | ZzZombo++ | doc/Type/Metamodel/ClassHOW.pod6:
Typo
 13:53
synopsebot6 Link: doc.perl6.org/type/Metamodel/ClassHOW
dudz kinda, a well respected type, i wonder though, do you own suit's>
*?
viki blurb on doc github.com/zoffixznet/perl6.party/...oc-content
timotimo src/core/bytecodedumpsterfire.c:666 13:54
dudz i'm on my way upto lader towards a suit
DrForr Yes. I use it at OSCON, but that's about it.
dudz for now i'm a polo shirt/nice trousers that could be ironed
plus i'm a hair brusher and clean face shaver 13:55
viki m: dd IterationEnd 13:56
camelia rakudo-moar 64343d: OUTPUT«IterationEnd␤»
13:56 wamba left
timotimo why exactly does the dependencies+deserialize frame of perl6.moar "loadbytecode" ModuleLoader.moarvm 5 times in a row? :) 13:56
viki m: gist.github.com/zoffixznet/eee66dd...80ec77f56c
camelia rakudo-moar 64343d: OUTPUT«True␤[["b"], ["c"], [IterationEnd], IterationEnd]␤»
dudz when i was growing up we would call it IT, but now they added an extra C, to make it ICT, the C i am still learning that i guess.
viki :o
DrForr I'm usually pretty casual; OSCON is an exception simply because others don't really focus on it. 13:57
viki Oh, my perl6 is too old to show InterationEnd
dudz figured it out viki ?
viki dudz: I think so, I was testing for Iteration end with nqp::istype() rather than nqp::eqaddr
dudz because the latter is used in your version? 13:58
viki I was using the wrong op. 13:59
13:59 domidumont left
viki 49 minutes wasted on a dumb typo :) 13:59
dudz :D
viki Yey Programming!
dudz it's good that you found it out though 14:00
did yourself some troubleshooting and practiced some paitence, and done :)
viki I also cried a little.
dudz nice :)
14:01 Ven left 14:02 Ven joined, domidumont joined
dudz i'm gonna go hangout in a horizontal position with my eye's closed for a bit, i'll be back after :) 14:06
viki \o
dudz *night perl6 community
o/
timotimo gnite dudz 14:08
tbrowder ref: 14:10
ZzZombo what are the possible values for trait declarations? Beside 'is' and 'will'? 14:12
timotimo "will" isn't really a trait declaration. it has more to do with phasers
MasterDuke hides, does, will, of, returns, handles
14:12 zakharyas left
tbrowder ref rakudo downloadable archives: is it worth my time to try to make it able to download and install zef as part of the package, i.e., will a PR be considered that enables that? 14:13
viki tbrowder: PR to which repo?
timotimo probably rakudo/star 14:14
ZzZombo and of course you can't make your own "keywords" (what is the name for those, again?) for declaration?
viki ZzZombo: I'm sure you can with slangs
ZzZombo I fear that would be a bit far-stretched for the task at han. 14:15
Unfortunate :(
hand*
psch the keyword for declarations is declarators, no? 14:16
viki Cool, there's pi users in here.
14:17 ChoHag left
DrForr I have a Pi v3, haven't used it yet... 14:17
14:17 ChoHag joined
tbrowder viki: rakudo 14:18
i tried rakudo star but gave up...
viki tbrowder: no, that PR won't be accepted. The repo already has a means to install any module you wish 14:19
tbrowder and how is that done?
MasterDuke tbrowder: someone is working on racoon (making tools/install-dist.pl a little more robust/feature filled i think)
jnthn timotimo: Pretty sure that you can write any trait_mod:<will> that you want; doesn't have to be about phasers 14:20
tbrowder can you name the branch?
timotimo fair enough 14:21
MasterDuke tbrowder: don't know, i think nine is the one working on it 14:22
timotimo it's called raccoon
so, how would a respectable person currently write out an unsigned 64bit integer to stdout or to a file?
MasterDuke anyway, off to go hit a little ball with metal sticks 14:23
timotimo good luck!
DrForr Don't lose it!
timotimo m: say 0xff_00_11_22_33_44_55_66_77.polymod(255)
camelia rakudo-moar 64343d: OUTPUT«(221 18446762986173038694)␤»
timotimo m: say 0xff_00_11_22_33_44_55_66_77.polymod(255 xx *) 14:24
camelia rakudo-moar 64343d: OUTPUT«(221 189 181 219 25 208 73 28 8 1)␤»
timotimo m: say 0xff_00_11_22_33_44_55_66_77.polymod(255 xx *).fmt("%x")
camelia rakudo-moar 64343d: OUTPUT«dd bd b5 db 19 d0 49 1c 8 1␤»
timotimo that's certainly not it
m: say 0xff_00_11_22_33_44_55_66_77.polymod(256 xx *).fmt("%x")
camelia rakudo-moar 64343d: OUTPUT«77 66 55 44 33 22 11 0 ff␤»
timotimo now ... do i actually write LE or BE :\
i'd rather use nativecast :| 14:25
14:25 seatek left
viki tbrowder: git clone github.com/zoffixznet/perl6-CoreHa...s-Sourcery CHS; ./perl6 tools/install-dist.pl CHS 14:26
tbrowder: and yeah, as was mentioned, that way is being improved.
timotimo oh, i can just print a Buf[uint64] %) 14:27
oh, i can't
tbrowder thanks
viki tbrowder: basically, you say you tried R* and gave up and now you wish to add an installer of 3rd-party software to Rakudo, when R* is meant to be that. The proper approach would be to fix whatever issues you had with R* and improve that distro rather than have some 3rd-party stuff in Rakudo and some 3rd-party stuff in R* 14:29
timotimo slurp doesn't have :bin, but .slurp-rest does? 14:30
viki s: '.'.IO, 'slurp' 14:31
SourceBaby viki, Sauce is at github.com/rakudo/rakudo/blob/6434...th.pm#L460
viki s: '.'.IO, 'slurp-rest'
SourceBaby viki, Something's wrong: ␤ERR: Type check failed in binding to &code; expected Callable but got Nil (Nil)␤ in sub do-sourcery at /home/zoffix/services/lib/CoreHackers-Sourcery/lib/CoreHackers/Sourcery.pm6 (CoreHackers::Sourcery) line 42␤ in sub sourcery at /home/zoffix/services/lib/CoreHackers-Sourcery/lib/CoreHackers/Sourcery.pm6 (CoreHackers::Sourcery) line 33␤ in block <unit> at -e line 6␤␤
viki timotimo: they both do, from what I see 14:32
timotimo m: say slurp(:bin).perl
camelia rakudo-moar 64343d: OUTPUT«"Céad slán ag sléibhte maorga Chontae Dhún na nGall\nAgus dhá chéad slán ag an Eireagal ard ina stua os cionn caor is coll;\nNuair a ghluais mise thart le Loch Dhún Lúich’ go ciúin sa ghleann ina luí\nI mo dhiaidh bhí gleanntáin ghlas’ Gha…»
timotimo that's a string
viki s: &slurp 14:33
SourceBaby viki, Sauce is at github.com/rakudo/rakudo/blob/6434...rs.pm#L140
psch m: say &slurp.candidates>>.signature
camelia rakudo-moar 64343d: OUTPUT«((IO::ArgFiles:D $io = { ... }, :$bin, :$enc = "utf8", |c is raw) (Cool:D $path, :$bin = Bool::False, :$enc = "utf8", |c is raw))␤»
psch m: say slurp(:bin).WHAT
camelia rakudo-moar 64343d: OUTPUT«(Str)␤»
viki m: $*ARGFILES.slurp(:bin).^name.say
camelia rakudo-moar 64343d: OUTPUT«Str␤»
viki s: $*ARGFILES, 'slurp' 14:34
SourceBaby viki, Sauce is at github.com/rakudo/rakudo/blob/6434...es.pm#L124
viki well, there's ya problem
m: say dir 14:35
camelia rakudo-moar 64343d: OUTPUT«(".cpanm".IO ".local".IO ".npm".IO ".perl6".IO ".perlbrew".IO ".rcc".IO ".ssh".IO "Perlito".IO "evalbot".IO "log".IO "nqp-js".IO "p1".IO "p2".IO "perl5".IO "std".IO ".bash_history".IO ".bashrc".IO "mbox".IO ".lesshst".IO "evalbot.log".IO ".cpan".IO "dalek-…»
psch ah, not passing it on?
viki It's passing, but it's not excepting it
psch well, yes
viki m: "evalbot.log".IO.slurp(:bin).^name.say
camelia rakudo-moar 64343d: OUTPUT«Buf[uint8]␤»
viki *expecting 14:36
14:37 cibs left 14:38 bjz_ left, bstamour joined 14:39 cibs joined
viki So, is anyone fixing that already or am I clear to do it? 14:41
tbrowder viki: there are already issues on star addressing the problem--basically the problem of not being able to easily modify some of the variables such as DESTDIR-- looks like too much coupling with moarvm and nqp 14:43
viki Ah that, yeah
But those are present in Rakudo too :)
tbrowder gotta go--will look at racoon later 14:44
14:44 kyclark_ joined 14:46 ugjka is now known as blitter_monist
ZzZombo m: class A { has $!stuff } say A.^attributes(:local).^name 14:46
camelia rakudo-moar 64343d: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Strange text after block (missing semicolon or comma?)␤at <tmp>:1␤------> 3class A { has $!stuff }7⏏5 say A.^attributes(:local).^name␤ expecting any of:␤ infix␤ infix stopper…»
ZzZombo m: class A { has $!stuff }; say A.^attributes(:local).^name
camelia rakudo-moar 64343d: OUTPUT«List␤»
viki spectests the fix 14:47
ZzZombo m: say (1, 22/7, 42, 300).first: * > 5, :kv, :end; 14:49
camelia rakudo-moar 64343d: OUTPUT«(3 300)␤»
ZzZombo m: say (1, 22/7, 42, 300).first: * > 5, :p, :end; 14:50
camelia rakudo-moar 64343d: OUTPUT«3 => 300␤»
ZzZombo I'm not sure, but isn't 42 happens before 300 but > 5 as well? 14:51
doesn't* and is*
psch m: say (1, 2, 3, 4).first(:end)
camelia rakudo-moar fce1a2: OUTPUT«4␤»
psch m: say (1, 2, 3, 4).first()
camelia rakudo-moar fce1a2: OUTPUT«1␤»
ZzZombo fuck
I didn't mean to copy :end
dammit
viki m: my Blob[uint8] $x .= new: 1, 2, 3; my Blob[uint8] $y .= new: 4, 5, 6; my @blobs = $x, $y; dd [~] @blobs 14:59
camelia rakudo-moar fce1a2: OUTPUT«Blob[uint8].new(1,2,3,4,5,6)␤»
viki m: my Blob[uint8] $x .= new: 1, 2, 3; my Blob[uint8] $y .= new: 4, 5, 6; my @blobs = $x; dd [~] @blobs
camelia rakudo-moar fce1a2: OUTPUT«Cannot use a Buf as a string, but you called the Stringy method on it␤ in block <unit> at <tmp> line 1␤␤»
viki :(
This is mildly annoying 15:00
15:01 domidumont left
psch m: say ~(my Blob[uint8] $ .= new: 1,2,3) 15:01
camelia rakudo-moar fce1a2: OUTPUT«Cannot use a Buf as a string, but you called the Stringy method on it␤ in block <unit> at <tmp> line 1␤␤»
psch that's reduce being smart again apparently 15:02
viki Wouldn't it be calling infix ~ and not prefix? 15:03
s: &infix:<~>, \(my Blob[uint8] $x .= new: 1, 2, 3)
SourceBaby viki, Sauce is at github.com/rakudo/rakudo/blob/6434...ngy.pm#L20
viki s: &infix:<~>, \((my Blob[uint8] $x .= new: 1, 2, 3), (my Blob[uint8] $x .= new: 1, 2, 3))
SourceBaby viki, Sauce is at github.com/rakudo/rakudo/blob/6434...uf.pm#L669
psch well, there's no 1-ary infix:<~> that takes Blob
viki Right, seems there should be
psch is my actual guess
viki Just be unity
psch m: say (my Blob[uint8] $ .= new: 1,2,3).infix:<~> 15:04
camelia rakudo-moar fce1a2: OUTPUT«No such method 'infix:<~>' for invocant of type 'Blob[uint8]'␤ in block <unit> at <tmp> line 1␤␤»
ZzZombo m: class A { has $!stuff }; say A.perl
camelia rakudo-moar fce1a2: OUTPUT«A␤»
psch m: say (my Blob[uint8] $ .= new: 1,2,3).&infix:<~>
camelia rakudo-moar fce1a2: OUTPUT«Cannot use a Buf as a string, but you called the Stringy method on it␤ in block <unit> at <tmp> line 1␤␤»
ZzZombo m: class A { has $!stuff }; say (A,1).perl
camelia rakudo-moar fce1a2: OUTPUT«(A, 1)␤»
15:05 domidumont joined 15:08 Ven left
ZzZombo m: class A { has Str:D $!name='unnamed'; method set_name(Str:D $name ) returns ::?CLASS:D { $!name=$name;return 'asd'; } }; say A.new.set_name('Bob') 15:13
camelia rakudo-moar fce1a2: OUTPUT«Type check failed for return value; expected A but got Str ("asd")␤ in method set_name at <tmp> line 1␤ in block <unit> at <tmp> line 1␤␤»
viki ZzZombo: FWIW, camelia can take /msg and there's also #zofbot where you can use her 15:15
ZzZombo nice
15:15 Ven joined 15:17 Ven_ joined, Ven left
ZzZombo can I turn say $package.^attributes(:local) into names of attributes inline in ($m.name,$m,$package,$package.^attributes(:local),$attr).perl? 15:18
-say
psch m: class A { has $.a }; say A.^attributes>>.^name 15:19
camelia rakudo-moar fce1a2: OUTPUT«List␤»
psch m: class A { has $.a }; say A.^attributes>>.name
camelia rakudo-moar fce1a2: OUTPUT«($!a)␤»
psch m: class A { has $.a, $!b, @.c }; say A.^attributes>>.name
camelia rakudo-moar fce1a2: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Variable $!b used where no 'self' is available␤at <tmp>:1␤------> 3class A { has $.a, $!b7⏏5, @.c }; say A.^attributes>>.name␤»
psch m: class A { has ($.a, $!b, @.c) }; say A.^attributes>>.name
camelia rakudo-moar fce1a2: OUTPUT«($!a $!b @!c)␤»
15:21 kyclark_ left
ZzZombo hey, the trim.chars constraint somebody here suggested earlier doesn't work. 15:27
method set_name(Str:D $name where .trim.chars)
m: class A { has Str:D $!name='unnamed'; method set_name(Str:D $name .trim.chars) returns ::?CLASS:D { $!name=$name;return 'asd'; } }; say A.new.set_name('Bob')
camelia rakudo-moar fce1a2: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Malformed parameter␤at <tmp>:1␤------> 3e='unnamed'; method set_name(Str:D $name7⏏5 .trim.chars) returns ::?CLASS:D { $!nam␤ expecting any of:␤ constraint␤»
ZzZombo m: class A { has Str:D $!name='unnamed'; method set_name(Str:D $name where .trim.chars) returns ::?CLASS:D { $!name=$name;return 'asd'; } }; say A.new.set_name('Bob') 15:28
camelia rakudo-moar fce1a2: OUTPUT«Cannot convert string to number: base-10 number must begin with valid digits or '.' in '3⏏5Bob' (indicated by ⏏)␤ in method set_name at <tmp> line 1␤ in block <unit> at <tmp> line 1␤␤Actually thrown at:␤ in method set_name at <tmp> line…»
ZzZombo ^
viki ZzZombo: .trim.chars.so or *.trim.chars
ZzZombo m: class A { has Str:D $!name='unnamed'; method set_name(Str:D $name where *.trim.chars) returns ::?CLASS:D { $!name=$name;return 'asd'; } }; say A.new.set_name('Bob')
camelia rakudo-moar fce1a2: OUTPUT«Type check failed for return value; expected A but got Str ("asd")␤ in method set_name at <tmp> line 1␤ in block <unit> at <tmp> line 1␤␤»
15:28 canopus left
ZzZombo m: class A { has Str:D $!name='unnamed'; method set_name(Str:D $name where *.trim.chars) returns ::?CLASS:D { $!name=$name;return self } }; say A.new.set_name('Bob') 15:29
camelia rakudo-moar fce1a2: OUTPUT«A.new␤»
ZzZombo that * stuff is weird for sure
viki hm .so prolly won't work
ZzZombo: feel free to write it as { .trim.chars }
arnsholt ZzZombo: Less weird than you might think
It's basically convenience notation for a closure
ZzZombo that does it even stand for?
psch "*.chars" is similar to "{ $_.chars }" 15:30
viki ZzZombo: `where` smartmatches. plain .trim.chars calls .trim.chars on $_ and returns a number, such a smartmatch will expect the argument be a number. *.trim.chars / { .trim.chars } makes it a smartmatch against a block instead
arnsholt So *.foo.bar is basically "-> $x { $x.foo.bar }" or alternatively "{ $^x.foo.bar }" (equivalent, using the ^ twigil)
ZzZombo so does just placing it somewhere implicitly creates one?
15:31 cygx left
viki ZzZombo: not in all places, but basically yeah 15:31
m: say (* * *)(2, 5)
camelia rakudo-moar 77e9d4: OUTPUT«10␤»
viki m: say {$^a * $^b)(2, 5)
camelia rakudo-moar 77e9d4: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Missing block␤at <tmp>:1␤------> 3say {$^a * $^b7⏏5)(2, 5)␤ expecting any of:␤ statement end␤ statement modifier␤ statement modifier loop␤»
viki m: say {$^a * $^b}(2, 5)
camelia rakudo-moar 77e9d4: OUTPUT«10␤»
ZzZombo how does it determine the borders for the closure? 15:32
arnsholt Basically, a each * will become a closure argument, whenever it is used where a term is expected; unless the * term is the entirety of the expression
In the latter case, it becomes the special value Whatever
viki psch: you say similar.... what are the differences?
psch m: <abc def ghi>.map({ $_ ~~ /. ** 3/}).say 15:33
camelia rakudo-moar 77e9d4: OUTPUT«(｢abc｣ ｢def｣ ｢ghi｣)␤»
psch m: <abc def ghi>.map(* ~~ /. ** 3/).say
camelia rakudo-moar 77e9d4: OUTPUT«(｢abc｣ ｢def｣ ｢ghi｣)␤»
psch eh, something like that somewhere i think :/
viki: the difference, in the end, is that * builds a WhateverCode, while { } builds a Block
viki Ah 15:34
arnsholt Given that an expression involving * becomes a WhateverCode, not just a plain block, there's likely something different
psch viki: and well, if there's different types behind it it's probably because we want to distinquish them
arnsholt Not sure what though
psch m: <abc def ghi>.map(*.comb[*]).say # and, well, this kinda silly example 15:35
camelia rakudo-moar 77e9d4: OUTPUT«((a b c) (d e f) (g h i))␤»
psch m: <abc def ghi>.map({$_.comb[$_]}).say
camelia rakudo-moar 77e9d4: OUTPUT«Cannot convert string to number: base-10 number must begin with valid digits or '.' in '3⏏5abc' (indicated by ⏏)␤ in block <unit> at <tmp> line 1␤␤Actually thrown at:␤ in block <unit> at <tmp> line 1␤␤»
15:35 pierre_ left
psch mind, i'm aware that even if postcircumfix:<[ ]> wasn't Whatever-special it'd still be two arguments, while the topic usage is the same... 15:35
15:36 canopus joined 15:37 kyclark_ joined
ZzZombo that's enough Perl 6 for me today. CYA all. 15:37
viki \o 15:38
15:39 brrt left
viki How can I test :enc<Latin-1> works on slurp? 15:42
15:42 Ven_ left
viki or... well, any :enc 15:42
moritz you first write a temp file with a defined Blob 15:44
and then slurp, and test that the result is as expected
viki hmm
m: Blob[uint8].new(<10 20 30>).decode.perl.say 15:45
camelia rakudo-moar 77e9d4: OUTPUT«"\n\x[14]\x[1e]"␤»
viki m: Blob[uint8].new(<10 20 30>).decode('Latin-1').perl.say
camelia rakudo-moar 77e9d4: OUTPUT«"\n\x[14]\x[1e]"␤»
viki What sort of a blob?
15:46 kyclark_ left
viki m: Blob[uint8].new(174).decode.perl.say 15:47
camelia rakudo-moar 77e9d4: OUTPUT«Malformed UTF-8 at line 1 col 1␤ in block <unit> at <tmp> line 1␤␤»
15:48 Ven joined
viki m: Blob[uint8].new(174).decode('Latin-1').perl.say 15:48
camelia rakudo-moar 77e9d4: OUTPUT«"®"␤»
viki Got it. Thanks
15:53 kurahaupo__ joined 15:55 kurahaupo__ is now known as kurahaupo 16:05 Ven left 16:15 blitter_monist is now known as UgJkA 16:17 domidumont left 16:19 sufrostico joined 16:24 bstamour left 16:25 g4 left 16:26 kyclark_ joined 16:27 kyclark_ left
viki kicks dalek 16:31
oh, I just had both IRC screens in #perl6 and not one in #perl6-dev 16:32
viki unkicks dalek
16:39 kyclark_ joined
nine Very much worth reading: blog.aurynn.com/contempt-culture 16:41
psch the font for the italic bits looks really off here oO 16:43
...in either browser too
not to mention that it's really exhausting to read with all of those there, hooray for devtools that let me easy adjust css \o/ 16:44
ilmari it's using Qucikcsand Italic
16:45 kyclark_ left
ilmari which is the italic version of the one used for the headlines 16:45
psch i find the topic a really difficult though 16:48
because, honestly? i actually do think there's a lot of things about php that are objectively bad 16:49
granted, they cleaned up with php7, and i'm at least partly resting on an opinion if formed during working with it about two years ago
s/ if f/ i f/ 16:50
viki psch: isn't that how many folks that say Perl 6 sucks operate? They judge it based on some mailsender script they say in the '90s 16:51
huggable: php sucks :is: blog.aurynn.com/contempt-culture 16:52
huggable viki, Added php sucks as blog.aurynn.com/contempt-culture
viki s/say/saw/;
nine psch: the point is that telling people that the language they use and that makes them feel powerful is objectively bad has probably never attracted anyone to the alternatives you propose. Advertising cool things demonstrating features of your faviourite language however makes people curious.
psch viki: is it? i worked with php 5.1x (iirc) for about 8 months, building a CRUD application for my then-employer
viki: i'm not going into #php (or wherever) and telling people they should do something else 16:53
nine: if that's the point, then i obviously missed it 16:54
nine psch: to be convinced by it, people have to draw a conclusion by themselves. I.e. show them how simple and powerful something in your language is and they will wish they had that when they have to excert much greater effort doing the same in their language. 16:55
psch: telling them their language is bad because X is so much more tedious than in another language just makes you a hater and one of "them" vs. "us".
psch nine: but that's absolutely not what i'm reading in that post. the post, to me, is about "please everyone grow out of the inherit tribalism thing that humans have been doing for thousands of years" 16:56
which is a nice sentiment, but doesn't really help achieve that
huf so what is that blogpost saying? that there are no objectively bad langauges? or that pointing them out is useless?
viki wow, nine++ very nice article
huf or perhaps that objectively bad langauges got that way because their communities lacked people who knew how to make it better? 16:57
nine This part sums it up quite nicely: "It’s 2015, and I saw a presenter at a Python conference make fun of Java. How would that feel to people trying to move from Java into something else? I wouldn’t feel welcome, and I’d have learned that the idea that the Python community is welcoming wasn’t true."
Talking down on other languages doesn't help my cause (promoting Perl). It just makes me feel cool and elitist. Trouble is that an elite is by definition a rather small group. 16:59
16:59 skids joined
psch oh, i get that, there's lots of practical conclusions to be drawn from "humans like small groups but that alienates other humans and alienating others is bad" 17:00
huf and also that people dont actually use their brains much when picking technologies
psch i just don't find lots of inherit value in blog posts that state that
huf it's nearly pure feeling
psch anyway, this is getting to me a bit too much apparently, sorry :)
huf ... i suppose that happens in the brain too
psch &
17:00 Khisanth left
viki hopes there aren't any logs of #css from a decade ago :) 17:02
viki was elitist of the elite >:)
viki is looking for a really fast way to determine if stuff matched by d contains any combiners :/ 17:03
huf i moved from the usual teenage elitism straight to complete nihilist despair i think :) 17:04
viki needs it in straight nqp too 17:05
17:13 Khisanth joined
viki No takers? :) 17:15
I wonder how insane it is to hardcode all Nd chars as a character class.
m: say "7\x[308]" ~~ /^ \d+ $/; 17:19
camelia rakudo-moar 15f51a: OUTPUT«｢7̈｣␤»
viki Maybe not very insane, because it's much faster than \d+ too 17:20
m: say "{.5811738 / .2907747}x faster" 17:21
camelia rakudo-moar 15f51a: OUTPUT«1.9987083x faster␤»
17:26 khw left
viki oh never mind, it's slower :( 17:26
17:26 khw joined 17:30 cdg joined
viki m: gist.github.com/zoffixznet/9aea90b...17678911f3 17:31
camelia rakudo-moar 15f51a: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Unrecognized backslash sequence: '\d'␤at <tmp>:14␤------> 3say "\7⏏5d+ way: {$dd}s";␤ expecting any of:␤ argument list␤ double quotes␤ term␤»
viki m: gist.github.com/zoffixznet/cfe0332...8839b5ebe8
camelia rakudo-moar 15f51a: OUTPUT«\d+ way: -1.7459995s␤literal way: -1.5470691s␤diff: 0.113934925656422%␤»
viki m: gist.github.com/zoffixznet/1372003...3b88c575ac 17:32
camelia rakudo-moar 15f51a: OUTPUT«\d+ way: 1.6447186s␤literal way: 1.54817305s␤diff: 0.0587003202616372%␤»
viki damn, that's not even all the Nd chars... 17:37
Probably a good sign that the plan should be abandoned as it's hard to find bugs in
17:48 obfusk left, obfusk_ joined 17:51 zakharyas joined 18:01 wamba joined
ugexe zef can now take named options after positionals (`--force install Foo`, `install --force Foo` and `install Foo --force` all do the same thing) by moving all options to the front of @*ARGS. Anyone have any ideas for when that would not work as expected? (other than a module name starting with `-`) 18:08
timotimo when you have named options that optionally take an argument 18:09
or when you have named options that always take an argument and you forget to move the argument along
ugexe you mean using `--target xxx` vs `--target='xxx'`? 18:11
timotimo yes 18:12
ugexe hmm. any ideas? maybe take the original @*ARGS and turn `--target xxx` into `--target=xxx` before the naive reordering? 18:14
18:14 firstdayonthejob joined
ugexe er shit it wont be setup right until its reordered 18:14
timotimo fair enough. if you can tell when xxx belongs to target or not
japhb Does MAIN now allow named arguments to not have '=' between key and value? Way back when, that was a hard requirement. 18:15
japhb hasn't checked the code recently
ugexe japhb: you're right 18:16
18:33 jameslenz left 18:42 wamba left 18:45 setty1_ left, AlexDaniel joined 18:46 setty1 joined 18:47 trnh joined
viki m: dd ^10 .map: {$^a + $^b} 18:51
camelia rakudo-moar 15f51a: OUTPUT«(1, 5, 9, 13, 17).Seq␤»
viki m: dd ^10 .map: -> ($a, $b) {$a + $b}
camelia rakudo-moar 15f51a: OUTPUT«Too few positionals passed; expected 2 arguments but got 0 in sub-signature␤ in block <unit> at <tmp> line 1␤␤»
viki Kinda sucks that doesn't work. I was hoping to be able to handle uneven lists in my map
m: dd ^9 .map: {$^a + $^b}
camelia rakudo-moar 15f51a: OUTPUT«Too few positionals passed; expected 2 arguments but got 1␤ in block <unit> at <tmp> line 1␤␤»
jnthn dd ^10 .map: -> $a, $b {$a + $b} 18:52
m: dd ^10 .map: -> $a, $b {$a + $b}
camelia rakudo-moar 15f51a: OUTPUT«(1, 5, 9, 13, 17).Seq␤»
jnthn m: dd ^9 .map: -> $a, $b? {$a + $b} 18:53
camelia rakudo-moar 15f51a: OUTPUT«Cannot resolve caller infix:<+>(Int, Mu); none of these signatures match:␤ ($x = 0)␤ (\a, \b)␤ (Real \a, Real \b)␤ (Int:D \a, Int:D \b --> Int:D)␤ (int $a, int $b --> int)␤ (Num:D \a, Num:D \b)␤ (num $a, num $b --> num)…»
jnthn m: dd ^9 .map: -> $a, $b? {$a + ($b // 0)}
camelia rakudo-moar 15f51a: OUTPUT«(1, 5, 9, 13, 8).Seq␤»
viki Oh damn
jnthn++
jnthn The extra parens turns it into an unpack, which is probalby not what you meant :)
viki Yeah :)
18:55 zakharyas left
timotimo masak: it looks like your latest tweet has its link b0rked by the em-dash in it 18:56
at least in my tweetdeck it ends up not clickable
not clickable in its entirety, i mean. just the beginning part
tailgate Is there more documentation to NQP than just what is in the github repo? 19:00
timotimo there's the internals workshop course by jnthn 19:01
it's pretty good
github.com/edumentab/rakudo-and-nq...als-course
tailgate thanks
timotimo see the pdf file linked from the readme
tailgate ooh, this is really nice
19:10 labster joined
viki huggable: \d :is: Match \d without combiners: gist.github.com/zoffixznet/e0cc87a...e-p6-p6-L3 19:13
huggable viki, Added \d as Match \d without combiners: gist.github.com/zoffixznet/e0cc87a...e-p6-p6-L3
19:21 RabidGravy joined 19:25 Tonik joined, espadrine joined 19:35 domidumont joined
AlexDaniel Just in case anybody missed the appeare 19:37
appearance of the new bot* :)
u: 🦋
unicodable6 AlexDaniel, U+1F98B BUTTERFLY [So] (🦋)
AlexDaniel u: {.uniname.words == 13}
unicodable6 AlexDaniel, U+0753 ARABIC LETTER BEH WITH THREE DOTS POINTING UPWARDS BELOW AND TWO DOTS ABOVE [Lo] (ݓ)
AlexDaniel, U+FBF9 ARABIC LIGATURE UIGHUR KIRGHIZ YEH WITH HAMZA ABOVE WITH ALEF MAKSURA ISOLATED FORM [Lo] (ﯹ)
AlexDaniel, U+FBFA ARABIC LIGATURE UIGHUR KIRGHIZ YEH WITH HAMZA ABOVE WITH ALEF MAKSURA FINAL FORM [Lo] (ﯺ)
AlexDaniel, gist.github.com/b8e6eb4b8ba7e67f52...5d46a17fb7
AlexDaniel u: {.uniname.words == 0} 19:38
unicodable6 AlexDaniel, U+0080 [Cc] ()
AlexDaniel, U+0081 [Cc] ()
AlexDaniel, U+0084 [Cc] ()
AlexDaniel, U+0099 [Cc] ()
jnthn u: {.uniname.contains('BEER')}
unicodable6 jnthn, U+1F37A BEER MUG [So] (🍺)
jnthn, U+1F37B CLINKING BEER MUGS [So] (🍻)
AlexDaniel you can just do this:
u: beer
unicodable6 AlexDaniel, U+1F37B CLINKING BEER MUGS [So] (🍻) 19:39
AlexDaniel, U+1F37A BEER MUG [So] (🍺)
AlexDaniel viki: by the way… if I do $message.reply(…) rapidly, the messages are sometimes in a different order. Is it supposed to be this way? 19:42
viki: for example, I had to add sleep(…) to get the expected result here: github.com/perl6/whateverable/blob...le.p6#L240 19:43
viki I have the same issue with NeuralAnomaly
geekosaur irc doesn't guarantee delivery order; it could conceivably be reordered by the server for whatever reason 19:44
viki I... not sure if it's supposed to be that way... I kinda figured the async nature of IRC was the cause
AlexDaniel sooo… the messages are actuall ordered correctly on the IRC::Client side? 19:45
geekosaur especially if testing with freenode it's entirely possible that message relay between servers follows different paths for different messages because at the time one message is sent a particular link looks not busy
AlexDaniel y*
19:46 Tonik left
AlexDaniel viki: perhaps you can add a configurable $message-delay to IRC::Client then? 19:46
viki NeuralAnomaly: source 19:47
NeuralAnomaly viki, See: github.com/zoffixznet/na
geekosaur AlexDaniel, you could try adding an output to stdout or file along with the message send to verify they're ordered correctly in the client. but scrambled message order on IRC is quite normal in multi-server configurations like freenode
viki AlexDaniel: well, I have a sleep of 3 seconds here and they still come outta order: github.com/zoffixznet/na/blob/mast...se.pm6#L99 19:48
AlexDaniel huh
viki well... tentatively 19:50
AlexDaniel yeah, that sounded suspicious 19:51
3 second delay should be enough… normally
I do understand that in some conditions it may not be so, but come on… 19:52
geekosaur I would think. even, shorter; pastes into a channel usually don't get reordered even at the beginning when the first few lines often get sent quickly (but with a short delay between)
19:55 rindolf left
viki Hm, well, here's the sequence of events: I send commands to a remote shell, awaiting each *write* to complete. Asynchronously, I receive their output and if it matches a pattern, I send a message to IRC. After I send the last command, I sleep for 3 seconds then send `exit` to remote shell and wait for my ssh Proc::Async to exit. I then print a message based on the exit code of Proc::Async.... Now the message 19:57
generated by the output of one of the commands I sent to remote shell somehow ends up AFTER the message that I send after shutting down ssh
Ahhh 19:58
The commands get written right away, but it takes them like an hour to run, so those 3second sleep does nothing at all. 19:59
AlexDaniel so just a tiny delay will fix it for most of the cases 20:00
viki And yeah, if you're using the same IRC::Client::Message.reply to send output, then it should be sequentially sent: github.com/zoffixznet/perl6-IRC-Cl...t.pm6#L328
(most likely the case is the same if you reply by returning from event handlers) 20:01
Filters tho, start a new promise, so there may be asynchrony there
kinda weird that they do :/ probaby should be redesigned 20:02
20:03 viki-telnet joined
viki-telnet 1 20:04
2
3
4
20:04 viki-telnet left
viki Well, that was a copy-paste of a chunk of text into telnet session... Seems in order 20:04
So *shrug*
Do .prints to IO::Socket::Async get sent sequentially? 20:06
20:17 bjz joined 20:30 domidumont left
AlexDaniel viki: maybe not 20:31
timotimo you should probably either await, or .then 20:32
AlexDaniel viki: um, but you should probably try it several times
because sometimes it is indeed in order
20:32 sufrostico left 20:33 cdg left 20:34 sufrostico joined, rindolf joined
RabidGravy If I have something like "class Foo { class Bar { } }" is there a part of composition of Foo where it gets to know of Bar, or do I have to grovel round in the stash somehow? 20:36
also related if I have to do some hack to find out about Bar, would it be completely composed at the time that Foo is composed? 20:38
timotimo just "is export" should do it 20:39
moritz a class gets composed right after the closing curly is parsed
viki m: use nqp; say nqp::sprintf("%f", 42e0)
camelia rakudo-moar 15f51a: OUTPUT«Directive f not applicable for type VMNull␤␤»
viki What does that mean?
timotimo somehow you got a null in there 20:40
viki hm
m: use nqp; my $a := nqp::list; nqp::push($a, 42e0); say nqp::sprintf("%f", $a)
camelia rakudo-moar 15f51a: OUTPUT«42.000000␤»
viki well, this works for me..
RabidGravy moritz, right so by the time the compose of Foo is called the Bar will be fully composed 20:41
timotimo well, nqp ops don't really support var args 20:42
so it makes sense that you've got to pass a list
moritz RabidGravy: yes 20:43
viki Noted.
RabidGravy cool, I'm toying with a hack where the classes form part of the definition of the outer one 20:44
timotimo loadaverage.org/group/perl - wow so much content 20:46
20:49 nowan left
viki I have another question: are there any other equality operators besides iseq_n and eqaddr? I need to somehow differentiate between 0e0 and -0e0 20:49
nqp: nqp::say(nqp::iseq_n(0e0, -0e0)); 20:50
camelia nqp-moarvm: OUTPUT«1␤»
viki nqp: nqp::say(nqp::eqaddr(0e0, 0e0));
camelia nqp-moarvm: OUTPUT«0␤»
timotimo we usually have special ops to ask for special floats
like isnan
viki don't see one for this case 20:51
viki relocates
20:52 nowan joined 21:01 cdg joined 21:03 cdg left 21:04 cdg joined
viki nqp: nqp::say(nqp::islt(nqp::atan2(-0e0, -1e0), 0)) 21:19
camelia nqp-moarvm: OUTPUT«No registered operation handler for 'islt'␤ at gen/moar/stage2/QAST.nqp:1596 (/home/camelia/rakudo-m-inst-2/share/nqp/lib/QAST.moarvm:compile_op)␤ from gen/moar/stage2/QAST.nqp:5874 (/home/camelia/rakudo-m-inst-2/share/nqp/lib/QAST.moarvm:compile_node)␤ …»
viki nqp: nqp::say(nqp::islt_n(nqp::atan2(-0e0, -1e0), 0e0)) 21:20
camelia nqp-moarvm: OUTPUT«No registered operation handler for 'atan2'␤ at gen/moar/stage2/QAST.nqp:1596 (/home/camelia/rakudo-m-inst-2/share/nqp/lib/QAST.moarvm:compile_op)␤ from gen/moar/stage2/QAST.nqp:5874 (/home/camelia/rakudo-m-inst-2/share/nqp/lib/QAST.moarvm:compile_node)␤…»
viki bah
man the docs need a bit of tweaking
nqp: nqp::say(nqp::islt_n(nqp::atan2_n(-0e0, -1e0), 0e0))
camelia nqp-moarvm: OUTPUT«1␤»
21:21 bjz left
viki nqp: nqp::say(nqp::islt_n(nqp::atan2_n(0e0, -1e0), 0e0)) 21:21
camelia nqp-moarvm: OUTPUT«0␤»
viki ehehe, hax :)
I guess I'll use that, with a XXX next to it 21:22
21:28 zakharyas joined
viki or learn to add an op to nqp... tho that sounds hard 21:31
timotimo not that hard 21:33
op goes in oplist.txt, run update_ops.p6, put an implementation into interp.c, add it to nqp's QASTCompilerMAST 21:34
on the JVM, you'd write a method into the SomethingSomethingOperationsFactory class and add a little description to QASTOperationsJAST or whatevr it's called 21:35
viki cool, I'll learn to do that then
21:52 Lucas_One left 21:53 trnh left 21:57 TEttinger joined 22:12 bjz joined
MasterDuke viki: your prop-Nd-only token, is it different from <:Nd>? 22:12
timotimo yes 22:13
because when you create a synthetic from a :Nd and some combiner, the synthetic will end up being :Nd, too
MasterDuke huh, do combiners have a named property? 22:15
timotimo a what now? 22:16
AlexDaniel eval: my @x = <a b c d>; dd @x[Mu] 22:17
evalable6 AlexDaniel, rakudo-moar 15f51a5: OUTPUT«(exit code 1) Cannot resolve caller postcircumfix:<[ ]>(Array, Mu); none of these signatures m…» 22:18
AlexDaniel, Full output: gist.github.com/deaf9bc9734c2fe08a...53daf19d07
AlexDaniel 65 suggestions, great!
MasterDuke like <:Nd> is digits
22:21 girafe joined, zakharyas left
AlexDaniel m: my @x = <1 2 3>; my $z = -999999999999999999; say @x[$z] 22:25
camelia rakudo-moar 15f51a: OUTPUT«Index out of range. Is: -999999999999999999, should be in 0..Inf␤ in block <unit> at <tmp> line 1␤␤Actually thrown at:␤ in block <unit> at <tmp> line 1␤␤»
AlexDaniel m: my @x = <1 2 3>; my $z = -9999999999999999999; say @x[$z]
camelia rakudo-moar 15f51a: OUTPUT«(Any)␤»
22:37 TimToady left 22:39 TimToady joined 22:40 nowan_ joined
timotimo food and movie time 22:41
22:43 nowan left 22:51 canopus left
kalkin- m: my @a = 'a' ... 'ﭏ'; say @a 23:01
and it hangs
camelia rakudo-moar 15f51a: OUTPUT«(timeout)» 23:02
kalkin- m: my @a = 'a' .. 'ﭏ'; say @a # but this works "out"
camelia rakudo-moar 15f51a: OUTPUT«Error encoding UTF-8 string: could not encode codepoint 55296␤ in block <unit> at <tmp> line 1␤␤»
23:03 canopus joined
AlexDaniel what this kind of range actually does? 23:04
is it this weird stringy thing that barely makes any sense?
kalkin- Well at least I expected it not to timeout, but actually I thought the range would represent all unicode chars for a to aleph 23:05
Dunno, may be the second error is valid, but why did it timeout in the case of a lazy range?
viki It's not a lazy range, but a sequence operator. 23:06
m: say ("a" ... "ﭏ")
camelia rakudo-moar 15f51a: OUTPUT«(a b c d e f g h i j k l m n o p q r s t u v w x y z { | } ~ ^H   ¡ ¢ £ ¤ ¥ ¦ § ¨ © ª « ¬ ­ ® ¯ ° ± ² ³ ´ µ ¶ · ¸ ¹ º » ¼ ½ ¾ ¿ …»
viki m: say "ﭏ".ords 23:07
camelia rakudo-moar 15f51a: OUTPUT«(64335)␤»
kalkin- viki: ahh ok sequence :)
viki well, I'm sure it'll stop once it reached 64335
hm 23:08
espadrine I can see why this character might be annoying www.fileformat.info/info/unicode/ch.../index.htm
23:09 RabidGravy left
viki espadrine: huh? 23:09
espadrine: you need better font on your system :)
espadrine the error blames it on character 55296, aka U+D800
viki .u ﭏ
yoleaux U+FB4F HEBREW LIGATURE ALEF LAMED [Lo] (ﭏ)
viki espadrine: oh that
23:10 salparadise left
viki m: print ("a" ... "z")[*-1] 23:10
camelia rakudo-moar 15f51a: OUTPUT«z»
viki s: &infix:<...>, \("a", "z") 23:11
SourceBaby viki, Sauce is at github.com/rakudo/rakudo/blob/15f5...rs.pm#L437
viki kalkin-: what were you trying to do anyway?
23:11 ircuse joined
kalkin- viki: nothing in particular, just experimenting 23:11
viki m: sink eager 'a', *.succ ... 'ﭏ'
camelia rakudo-moar 15f51a: OUTPUT«(timeout)» 23:12
23:12 ircuse is now known as salparadise
viki Kinda embaraccing that I don't know what "a" ... "ﭏ" does 23:12
viki recalls ranting on a ticket about the current behaviour being fine as is :) 23:13
vendethiel m: sink eager 'a'..'ﭏ'
camelia ( no output )
vendethiel m: my @a = eager 'a'..'ﭏ'
camelia ( no output )
TEttinger viki: maybe chosen because it's FB4F in unicode? (hex)
viki TEttinger: I don't follow... what's chosen? 23:14
TEttinger oh sorry, wasn't sure if the ... was unclear or the second unicode character was not shown 23:15
viki m: .say for "a" ... "ﭏ" 23:17
kalkin-: unless Unicode has a zillion backslashes, I think you may have found a bug
camelia rakudo-moar 15f51a: OUTPUT«(timeout)a␤b␤c␤d␤e␤f␤g␤h␤i␤j␤k␤l␤m␤n␤o␤p␤q␤r␤s␤t␤u␤v␤w␤x␤y␤z␤{␤|␤}␤~␤^H␤␤␤␤␤␤␤␤␤␤␤␤␤␤␤␤␤␤␤␤␤␤␤␤␤…»
viki If I run the above, I see a backslash slip through every few seconds
m: .say for ("a" ... "ﭏ").grep: "̿ 23:19
camelia rakudo-moar 15f51a: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Bogus term␤at <tmp>:1␤------> 3.say for ("a" ... "ﭏ").grep:7⏏5 "̿␤ expecting any of:␤ argument list␤ infix␤ infix stopper␤ prefix␤ term␤»
viki m: .say for ("a" ... "ﭏ").grep: "̿"
camelia rakudo-moar 15f51a: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Bogus term␤at <tmp>:1␤------> 3.say for ("a" ... "ﭏ").grep:7⏏5 "̿"␤ expecting any of:␤ argument list␤ infix␤ infix stopper␤ prefix␤ term␤»
viki m: .say for ("a" ... "ﭏ").grep:{ "̿"}
camelia rakudo-moar 15f51a: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Bogus statement␤at <tmp>:1␤------> 3.say for ("a" ... "ﭏ").grep:{7⏏5 "̿"}␤ expecting any of:␤ prefix␤ term␤»
viki (╯°□°）╯︵ ┻━┻
If I run the above, I see a backslash slip through every few seconds
wtf
sorry, my terminal glitched 23:20
AlexDaniel eval: .ord.say for "a" ... "ﭏ" 23:22
ok that was a bad idea I guess
23:22 evalable6 left
espadrine that's like coughing and having the house fall down… 23:24
23:24 evalable6 joined
viki perl6 -e '.say for ("a" ... "ﭏ")[^4000]' | sort | uniq | wc -l 23:25
425
AlexDaniel yea
viki m: ("a" ... "ﭏ")[^4000].sort.unique.elems.say
camelia rakudo-moar 15f51a: OUTPUT«735␤»
viki s: "a", 'succ', \() 23:26
SourceBaby viki, Sauce is at github.com/rakudo/rakudo/blob/15f5...tr.pm#L270
psch uh, that Sequence end point looks really confusing here..?
AlexDaniel well, it's right to left…
psch oohh
viki psch: it's aleph one 23:27
psch nah, it's just aleph
viki *aleph
psch but the right to left makes more sense there :)
AlexDaniel u: m: ("a" ... "ﭏ")[^4000].sort.unique.elems.say
unicodable6 AlexDaniel, U+003A COLON [Po] (:)
AlexDaniel, U+006D LATIN SMALL LETTER M [Ll] (m)
AlexDaniel, U+0020 SPACE [Zs] ( )
AlexDaniel, gist.github.com/9f55ace1c5f04bd26b...361754aa5c
psch i'm still not completely certains what's to come out of that 23:28
the RTL marker is right behind the first double quote?
AlexDaniel .u "ﭏ"000] 23:29
yoleaux U+0022 QUOTATION MARK [Po] (")
U+0030 DIGIT ZERO [Nd] (0)
U+005D RIGHT SQUARE BRACKET [Pe] (])
viki .u ℵ
yoleaux U+2135 ALEF SYMBOL [Lo] (ℵ)
viki .u ℵ
yoleaux U+2135 ALEF SYMBOL [Lo] (ℵ)
AlexDaniel .u ﭏ"0
yoleaux U+0022 QUOTATION MARK [Po] (")
U+0030 DIGIT ZERO [Nd] (0)
U+FB4F HEBREW LIGATURE ALEF LAMED [Lo] (ﭏ)
viki .u "ﭏ")
yoleaux U+0022 QUOTATION MARK [Po] (") 23:30
U+0029 RIGHT PARENTHESIS [Pe] ())
U+FB4F HEBREW LIGATURE ALEF LAMED [Lo] (ﭏ)
psch yeah the U+2135 doesn't show here, but ﭏ does
...lamed o.o
anyway, that just caught my curiousitiy. i'll resign to probably-not-knowing, g'night o/ 23:31
viki m: for ("a" ... "ﭏ")[^4000] { state $x = .ord; say "Got {.ord} but has $x before" if $x > .ord; $x = .ord } 23:32
camelia rakudo-moar 15f51a: OUTPUT«Got 768 but has 831 before␤Got 768 but has 831 before␤Got 768 but has 831 before␤Got 768 but has 831 before␤Got 768 but has 831 before␤Got 768 but has 831 before␤Got 768 but has 831 before␤Got 768 but has 831 before␤Got 768 but has 831 befo…»
viki m: say 768.base: 16; say 831.base: 16
camelia rakudo-moar 15f51a: OUTPUT«300␤33F␤»
viki m: "\x[300]\x[33F]".uninames.say
camelia rakudo-moar 15f51a: OUTPUT«(COMBINING GRAVE ACCENT COMBINING DOUBLE OVERLINE)␤»
viki m: "\x[300]".succ.uniname.say
camelia rakudo-moar 15f51a: OUTPUT«COMBINING GRAVE ACCENT␤»
viki kyclark: would you be able to rakudobug this? And include link to the chat: irclog.perlgeek.de/perl6/2016-11-25#i_13630211 23:33
huggable: rakudobug
huggable viki, [email@hidden.address] or use perl6 query on rt.perl.org ; see github.com/rakudo/rakudo/#reporting-bugs
viki ^ just email to that address (the site option is currently disabled)
23:37 rindolf left 23:40 wamba joined, bjz left 23:41 firstdayonthejob left
viki oh heh, only now that I copy-pasted that code into my editor do I see the weirdness 23:47
23:54 espadrine left 23:57 UgJkA 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!