»ö« Welcome to Perl 6! | perl6.org/ | evalbot usage: 'p6: say 3;' or /msg camelia p6: ... | irclog: irc.perl6.org or colabti.org/irclogger/irclogger_log/perl6 | UTF-8 is our friend! 🦋
Set by Zoffix on 25 July 2018.
00:05 ryn1x joined, p6bannerbot sets mode: +v ryn1x 00:09 ryn1x left
timotimo you want to have something like $my-instance.attribute-name($value-it-should-get)? 00:10
00:11 ryn1x joined 00:12 p6bannerbot sets mode: +v ryn1x 00:24 beeman left
jdv79 kinda 00:25
more like $instance.rando_meth($val_for_baz_attr) and the sig of the rando_meth handles binding that to $!baz 00:26
similar to the common BUILD pattern where a named arg is bound directly to a attr 00:27
00:29 beeman joined, p6bannerbot sets mode: +v beeman
timotimo ah 00:29
yeah, all you need is to give it the right name and just not put a : in front
i think that ought to do it
00:30 Kaiepi left 00:43 MasterDuke joined, p6bannerbot sets mode: +v MasterDuke, MasterDuke left, MasterDuke joined, herbert.freenode.net sets mode: +v MasterDuke, p6bannerbot sets mode: +v MasterDuke
jdv79 what's the right name if its a positional passed as a literal 00:44
timotimo i'm not sure i understand? 00:45
m: class Test { has $.attribute; method sets-the-attr($!attribute) { say "set it!"; say $.attribute } }; Test.new.sets-the-attr(1)
camelia set it!
1
jdv79 ah. got it. 00:46
thanks
timotimo YW 00:47
00:52 cydf left 01:07 Kaiepi joined, p6bannerbot sets mode: +v Kaiepi 01:15 kurahaupo left 01:16 kurahaupo joined, p6bannerbot sets mode: +v kurahaupo
jdv79 lets say i have an array attr (@.baz) and i want to slip in a hash for the constructor args. is there a way to get %args<baz> to slip in properly? 01:35
timotimo hm, normally signature application is "binding", not sure if @-sigiled parameters do things differently 01:36
jdv79 it looks like it ends up as the first elem in the attr array 01:37
aka $!baz has [[1,2,3]]
rather @!baz 01:38
if i make the attr a scalar it doesn't double it up like so
timotimo perhaps it does do assignment semantics, but your array has a scalar container around it? 01:39
jdv79 i believe it does
01:42 holyghost joined
timotimo in that case it would appear you can't just have it like that 01:42
01:43 p6bannerbot sets mode: +v holyghost
jdv79 m: class Foo { has @.a; method b (:@!a) {} }; my @foo = 1,2,3; my %args = :a(@foo); my $f = Foo.new(|%args); $f.say 01:45
camelia Foo.new(a => [[1, 2, 3],])
01:46 molaf left
jdv79 so if to be able to slip constructor args the class can't use anything but scalar attrs 01:47
m: class Foo { has @.a; method b (:@!a) {} }; my @foo = 1,2,3; my %args = :a(@foo); my $f = Foo.new(|%args,:a(@foo)); $f.say 01:48
camelia Foo.new(a => [1, 2, 3])
jdv79 that's unfortunate
i thought this was a case were a slip would workk 01:49
timotimo slip only works if something is there to iterate, apparently this doesn't have that 01:54
SmokeMachine How can I get the path of a module’s file programaticaly? 01:55
jdv79 seems maybe inconsistent that its possible to have it work as expected without slipping the method args in and not when the slipping. 01:56
jnthn: is that on purpose and/or necessary? 01:57
timotimo oh, sorry, i should have said that slipping arguments will unpack lists to become positional args or named args depending on positional vs associative or something 02:00
maybe also iterable
02:03 finanalyst joined 02:04 p6bannerbot sets mode: +v finanalyst 02:13 hph^ joined 02:14 p6bannerbot sets mode: +v hph^
holyghost hello 02:27
I'm back to work in a few hours 02:29
sacomo hi all 02:54
02:57 newbie23 joined
sacomo is there a way to interpolate a string regex, e.g. my $r = "'some part' \\s (\\d+) \\s 'more string' (\\w+)", into a regex for match? $str ~~ / $r / 02:57
02:58 p6bannerbot sets mode: +v newbie23
buggable New CPAN upload: AI-Agent-Music-0.2.4.tar.gz by HOLYGHOST cpan.metacpan.org/authors/id/H/HO/...2.4.tar.gz 03:02
New CPAN upload: Bayes-Learn-0.1.10.tar.gz by HOLYGHOST modules.perl6.org/dist/Bayes::Learn...:HOLYGHOST
lookatme_q sacomo, EVAL ?? 03:06
or /<$r>/ 03:07
m: my $r = "'some part' \\s (\\d+) \\s 'more string' (\\w+)"; say "some part 123 more string foo" ~~ /<$r>/ 03:08
camelia Nil
lookatme_q m: my $r = "'some part' \\s (\\d+) \\s 'more string' (\\w+)"; say "some part 123 more stringfoo" ~~ /<$r>/ 03:09
camelia ｢some part 123 more stringfoo｣
lookatme_q ^^ sacomo
m: my $r = "'some part' \\s (\\d+) \\s 'more string' (\\w+)"; say "some part 123 more stringfoo" ~~ /<$r>/; say $/
camelia ｢some part 123 more stringfoo｣
｢some part 123 more stringfoo｣
buggable New CPAN upload: AI-Agent-0.2.3.tar.gz by HOLYGHOST modules.perl6.org/dist/AI::Agent:cpan:HOLYGHOST 03:12
sacomo lookatme_q: thanks. but I think that is only returning orig... 03:15
03:16 ufobat_ joined 03:17 p6bannerbot sets mode: +v ufobat_ 03:20 ufobat left
buggable New CPAN upload: Bayes-Learn-0.1.11.tar.gz by HOLYGHOST modules.perl6.org/dist/Bayes::Learn...:HOLYGHOST 03:32
holyghost Bayes::Learn is on github, github.com/theholyghost2
It should be debugged
If anyone wants to make a pulle request, it's nice code to read and learn from 03:33
s/pulle/pull
lookatme_q sacomo, what do you mean ?? 03:42
m: my $r = "'some part' \\s (\\d+) \\s 'more string' (\\w+)"; "some part 123 more stringfoo" ~~ /(<$r>)/; say $/
camelia ｢some part 123 more stringfoo｣
0 => ｢some part 123 more stringfoo｣
lookatme_q It matched obviously
03:43 Guest79068 joined
sacomo yeah, that matched... what I am trying to do is grab the captures also, for example: 03:44
m: say "some part 123 more" ~~ /'some part ' (\d+) ' more'/
camelia ｢some part 123 more｣
0 => ｢123｣
03:44 p6bannerbot sets mode: +v Guest79068
lookatme_q sacomo, IDK why it not captured 03:46
03:48 Guest79068 left
sacomo hmm 03:51
lookatme_q sacomo, but EVAL should work 03:53
sacomo yeah, ok thanks 03:54
03:55 khisanth_ left 04:07 w_richard_w joined 04:08 p6bannerbot sets mode: +v w_richard_w, khisanth_ joined, Itaipu left 04:09 p6bannerbot sets mode: +v khisanth_, Itaipu joined 04:10 p6bannerbot sets mode: +v Itaipu 04:12 Cabanossi left 04:13 Cabanossi joined 04:14 p6bannerbot sets mode: +v Cabanossi
buggable New CPAN upload: IO-Glob-0.6.tar.gz by HANENKAMP modules.perl6.org/dist/IO::Glob:cpan:HANENKAMP 04:22
Geth ecosystem: ccworld1000++ created pull request #417:
add GtkPlus [On the way]		 04:30
holyghost I've made some music from remixed from a frined of mine : here's the 2 sampled sounds : 04:34
www.mediafire.com/file/yg9evcxqgocs...1.mp3/file
www.mediafire.com/file/k71a94bc1xqq...4.mp3/file
It's industrial
I have programmed a simple sample recognizer with the use of Matrixes to interpolate 04:36
you just enlarge the 8-bit sample with a matrix, they'll come when Mathx::Matrix exists
Further on we might have Game::Markov::Sound 04:37
future music 04:38
buggable New CPAN upload: HTTP-Request-Supply-0.1.2.tar.gz by HANENKAMP cpan.metacpan.org/authors/id/H/HA/...1.2.tar.gz 04:42
holyghost ok, AI-Agent and AI-Agent-Music compile, I need tests 04:52
next is Bayes::Learn then Game::Markov 04:53
buggable New CPAN upload: Path-Router-0.2.tar.gz by HANENKAMP cpan.metacpan.org/authors/id/H/HA/...0.2.tar.gz 05:02
05:05 Itaipu left 05:09 Itaipu joined 05:10 p6bannerbot sets mode: +v Itaipu 05:12 ryn1x left
ZzZombo New release? Yes! 05:14
"panda-sub which pointed user to zef now removed" -- looks like a typo. I'd fix it myself, but WTF Github, I can't? 05:21
holyghost There's Mathx::Matrix on github.com/theholyghost2 05:32
a start with VectorN, MatrixNM
05:34 kurahaupo_ joined 05:35 p6bannerbot sets mode: +v kurahaupo_ 05:37 vrurg left 05:38 vrurg joined, kurahaupo left 05:39 p6bannerbot sets mode: +v vrurg, hdch left
holyghost If you want to make a non-initialzed nested list, maybe it would be handy to nest while pushing or setting 05:52
05:53 kurahaupo_ left
holyghost m: Class A { has @!l; sub set($y,$value) { push(@!l[$y] = $value; } } 05:53
camelia 5===SORRY!5=== Error while compiling <tmp>
You cannot declare attribute '@!l' here; maybe you'd like a class or a role?
at <tmp>:1
------> 3Class A { has @!l7⏏5; sub set($y,$value) { push(@!l[$y] = $v
expecting any of:
…
05:54 kurahaupo joined, p6bannerbot sets mode: +v kurahaupo
holyghost m: Class A { has @!l; sub set($y,$value) { push(@!l[$y], $value); } } 05:56
camelia 5===SORRY!5=== Error while compiling <tmp>
You cannot declare attribute '@!l' here; maybe you'd like a class or a role?
at <tmp>:1
------> 3Class A { has @!l7⏏5; sub set($y,$value) { push(@!l[$y], $va
expecting any of:
…
05:58 lizmat left
holyghost It's not Lisp or smalltalk anyway :-) 05:59
In C++ you have to use reserve 06:01
buggable New CPAN upload: DOM-Tiny-0.5.1.tar.gz by HANENKAMP modules.perl6.org/dist/DOM::Tiny:cpan:HANENKAMP 06:02
holyghost e.g. in the vector STL class
#include <vector>
.tell jmerelo I have a module ready on github (Mathx-Matrix), I won't upload to CPAN before I have tests so you never have to worry anymore 06:05
yoleaux holyghost: I'll pass your message to jmerelo.
holyghost .tell My apologies for broken CPAN modules (with time)
yoleaux holyghost: I'll pass your message to My.
06:06 Summertime left
kurahaupo holyghost: tell whom? 06:06
holyghost .tell jmerelo My apologies for broken CPAN modules (within time) 06:07
yoleaux holyghost: I'll pass your message to jmerelo.
holyghost kurahaupo: thanks
06:07 Summertime joined, p6bannerbot sets mode: +v Summertime
Geth doc: e2130a225e | (JJ Merelo)++ | doc/Type/Any.pod6
Adds better example for Any.roll

Eliminates the class example, and adds pointers. Refs #1731		 06:09
synopsebot Link: doc.perl6.org/type/Any
Geth doc: 54ea17ee87 | (JJ Merelo)++ | doc/Type/Any.pod6
Adds better example for Any.pick using Ranges

Eliminates the class example too; refs #1731
doc: d6548d0919 | (JJ Merelo)++ | doc/Type/Any.pod6
Improves descripton of prepend
06:12 jmerelo joined 06:13 p6bannerbot sets mode: +v jmerelo, kurahaupo_ joined
jmerelo O/ 06:13
yoleaux 11 Nov 2018 15:04Z <SmokeMachine> jmerelo: if you have some time could you please comment? (it's far from complete yet) gist.github.com/FCO/33be94f5ecfab7...etsanta-md
06:05Z <holyghost> jmerelo: I have a module ready on github (Mathx-Matrix), I won't upload to CPAN before I have tests so you never have to worry anymore
06:07Z <holyghost> jmerelo: My apologies for broken CPAN modules (within time)
06:14 p6bannerbot sets mode: +v kurahaupo_
jmerelo .tell SmokeMachine to please make a PR for the day you want too. I'll try to take a look today. 06:15
yoleaux jmerelo: I'll pass your message to SmokeMachine.
06:16 kurahaupo_ left, kurahaupo left 06:23 kurahaupo joined 06:24 p6bannerbot sets mode: +v kurahaupo
holyghost jmerelo: If you want to take a look at AI::Agent in the future, it compiles 06:25
jmerelo: so does AI::Agent::Music 06:26
jmerelo holyghost: does it have tests?
holyghost not yet
jmerelo holyghost: I think I'll pass, then. 06:27
holyghost ok
jmerelo: It's a base class anyway, I need to write a parser for agent-agent talk 06:28
especially the music agent system
You can or cannot use it
jmerelo holyghost: you say base class as if base classes didn't need tests... Everything needs a test 06:29
holyghost jmerelo: I know now 06:30
holyghost reads docs.perl6.org/language/testing 06:32
06:35 vike1 left 06:38 shadowpaste left 06:40 nebuchadnezzar left, shadowpaste joined, nebuchad` joined, p6bannerbot sets mode: +v shadowpaste 06:41 p6bannerbot sets mode: +v nebuchad` 06:50 vike1 joined 06:51 p6bannerbot sets mode: +v vike1 06:52 nebuchad` is now known as nebuchadnezzar
holyghost what's wrong with this ? pastebin.com/GiaY6Q1L 06:53
prove --exec perl6 -r t -I lib/ 06:54
t/make-agent.t .. ===SORRY!=== Error while compiling /home/erana/perl6/AI-Agent/t/make-agent.t
Undeclared name:
Agent used at line 15
Confused 06:56
at /home/erana/perl6/AI-Agent/t/make-agent.t:15
------> my $ag = AI:Agent:⏏:Agent.new(0);
jmerelo holyghost: double :: between AI and Agent 06:57
holyghost t/make-agent.t .. Could not find symbol '&Agent' 06:58
06:58 lizmat joined, p6bannerbot sets mode: +v lizmat
holyghost ^-- double 06:58
jmerelo holyghost: because you need to get the syntax right. See above.
you are writing AI:Agent, it should be AI::Agent 06:59
holyghost pastebin.com/PTtHUw90
$ prove --exec perl6 -r t -I lib/
t/make-agent.t .. Could not find symbol '&Agent'
in block <unit> at t/make-agent.t line 15
jmerelo holyghost: it would be better if you read the documents. 07:00
holyghost: use lib is for declaring paths to lib, not to use an actual lib
holyghost: in that code, you're not even running a test. You need to actually use some test routine like "ok" or "is" 07:01
holyghost: you should read thoroughly some Perl 6 book. 07:02
holyghost ok
jmerelo jmerelo: please start with the test documentation you've pointed to above. If you don't understand anything, please ask here or in StackOverflow or in both places 07:06
holyghost Default constructor for 'AI::Agent::Agent' only takes named arguments 07:07
jmerelo holyghost: default constructor for all classes do 07:08
holyghost What's a named argument ?
moritz m: sub f(:$x) { say $x }; f x => 42; 07:09
camelia 42
jmerelo holyghost: ... and that is why you need to read some book and/or documentation and/or both
holyghost: docs.perl6.org/type/Signature#inde...d_argument
holyghost t/make-agent.t .. Dubious, test returned 255 (wstat 65280, 0xff00) 07:15
Failed 1/1 subtests
I got it cornered 07:16
now the ok system
in an hour
jmerelo: thanks for the help, I only have the beastie book
So it just fails when not returning 1; 07:17
jmerelo: pastebin.com/9MSSYBCc 07:19
07:19 domidumont joined
holyghost Needs the test system further on 07:19
07:20 p6bannerbot sets mode: +v domidumont
holyghost t/make-agent.t .. ok 07:24
All tests successful.
jmerelo: I've got my first test :-) 07:25
07:39 robertle joined
holyghost jmerelo: I uploaded with the main test for a n Agent dispatch to CPAN 07:39
07:39 jmerelo left, p6bannerbot sets mode: +v robertle
buggable New CPAN upload: AI-Agent-0.2.4.tar.gz by HOLYGHOST modules.perl6.org/dist/AI::Agent:cpan:HOLYGHOST 07:42
holyghost ^-- this should give you a subclass for dispatchable messages to agents 07:43
07:44 xinming left 07:46 xinming joined, p6bannerbot sets mode: +v xinming 07:51 kdr22 left 08:00 abraxxa joined 08:01 p6bannerbot sets mode: +v abraxxa 08:02 w_richard_w left 08:05 kdr22 joined 08:06 p6bannerbot sets mode: +v kdr22 08:12 yqt joined 08:13 p6bannerbot sets mode: +v yqt 08:17 Itaipu left 08:22 rindolf joined 08:23 Itaipu joined, p6bannerbot sets mode: +v rindolf 08:24 p6bannerbot sets mode: +v Itaipu, zakharyas joined 08:25 p6bannerbot sets mode: +v zakharyas 08:30 kensanata joined, p6bannerbot sets mode: +v kensanata
holyghost What's wrong with the apply of arguments, they are stationary : pastebin.com/DLG80QYs 08:32
I need some kind of binding 08:35
08:37 dalek left 08:43 dakkar joined 08:44 sena_kun joined, p6bannerbot sets mode: +v dakkar 08:45 p6bannerbot sets mode: +v sena_kun 08:46 kdr22 left 08:47 kdr22 joined, Itaipu left 08:48 p6bannerbot sets mode: +v kdr22 08:52 Itaipu joined 08:53 p6bannerbot sets mode: +v Itaipu 08:57 kdr22 left, kdr22 joined 08:58 p6bannerbot sets mode: +v kdr22 08:59 parv joined 09:00 p6bannerbot sets mode: +v parv 09:01 fatguy joined, p6bannerbot sets mode: +v fatguy 09:03 cydf joined 09:04 p6bannerbot sets mode: +v cydf 09:15 rindolf left 09:19 Itaipu left 09:21 rindolf joined 09:22 p6bannerbot sets mode: +v rindolf 09:23 Itaipu joined 09:24 p6bannerbot sets mode: +v Itaipu, dogbert11 left 09:34 fatguy left
buggable New CPAN upload: AI-Agent-0.2.5.tar.gz by HOLYGHOST modules.perl6.org/dist/AI::Agent:cpan:HOLYGHOST 09:42
09:42 Freeaqingme1 joined, Freeaqingme1 left 09:47 tobs` joined
holyghost .tell jmerelo I have HashedAgent and Agent tested, you can use the ask routines. If you want to try SFY 09:47
yoleaux holyghost: I'll pass your message to jmerelo.
09:48 p6bannerbot sets mode: +v tobs`, kerframil joined, p6bannerbot sets mode: +v kerframil 10:05 hamhu3_ joined 10:06 p6bannerbot sets mode: +v hamhu3_ 10:07 hamhu3 left 10:09 Itaipu left 10:11 Itaipu joined
buggable New CPAN upload: Mathx-Stat-0.1.14.tar.gz by HOLYGHOST modules.perl6.org/dist/Mathx::Stat:...:HOLYGHOST 10:12
10:12 p6bannerbot sets mode: +v Itaipu
holyghost ok, enough work, I am going to grab a beer 10:20
10:21 Sound joined, p6bannerbot sets mode: +v Sound
holyghost Mathx::Stat has 2 tests, but I need to configure the outcome of the "ok"'s 10:21
I am getting there , slowly
jmerelo, lizmat : Hope you like the new github modules
s/new/updated 10:27
lizmat and now for something completely different: jnthn.net/all-the-czech-railways/#2018
holyghost jnthn, lizmat: I like the beer mentioning 10:29
10:30 rindolf left
holyghost <-- old fart, linux this month 10:31
elder
I don't remember, I joined linux in november 1998 with 7 debian floppies 10:32
I burned them myself with rawrite.exe in win 98 10:34
root.bin/root.flp for example 10:36
10:39 fatguy joined, p6bannerbot sets mode: +v fatguy
fatguy when i .parse grammar with action to a class, looks like it also pass non matched string to the method since i got 'Use of Nil in string context' , how to avoid ? 10:40
code pastebin.com/dR9b2pe6 10:41
10:42 ChoHag left
sena_kun fatguy, can you also provide `sar30` file? at least partly, at the point where warnings are present. 10:43
10:46 rindolf joined, p6bannerbot sets mode: +v rindolf
fatguy +sena_kun: ufile.io/6yhae 10:51
10:51 Sound left 10:56 domidumont left
sena_kun well, the thing is, as stated, in lines 25 and 26. when I did a simple `say $/` before those lines, I saw that not every parsed structure has `PM|AM` present. `$/[$UnknownThing]` evaluates to Nil, and when you try to `Nil eq "foo"`, it warns you that Nil is used in string context. 10:56
to avoid that, you may want to use something like `if $/{3}:exists && $/[3] eq 'AM';`. It's the first thing that comes from my mind, but it's ugly. :)
personally, I would create a new token, like `token am-or-pm { 'AM' | 'PM' }`, and then used it as `<am-or-pm>` instead of `('AM')`. then in action method you can do like `with $<am-or-pm> { make self.foo($/) }`. } 10:58
*instead of `AM` and `PM` notes, of course, without parens 10:59
fatguy, ^
10:59 ChoHag joined 11:00 p6bannerbot sets mode: +v ChoHag
sena_kun or, rather, `with $<am-or-pm> { if } 11:01
ugh
*when `PM` { foo }; when 'AM' { bar } 11:02
fatguy +sena_kun: thanks for your advise, let me try it :) 11:05
sena_kun fatguy, you are welcome!
holyghost lizmat: I really like the railway link :-) 11:07
11:08 finanalyst left 11:23 nwellnhof joined, p6bannerbot sets mode: +v nwellnhof
nwellnhof p6: say <0>.comb 11:23
camelia (0)
nwellnhof OK, this was fixed: github.com/rakudo/rakudo/issues/2010 11:24
11:24 nwellnhof left
holyghost jnthn: thanks for relieving on chezch railways :-) 11:36
jnthn :)
holyghost really like the winter photos 11:40
I'm in since "de snoezige snowijt", a strip from Vandersteen 11:41
about 30 years ago
It was about "Flip, het winterkonkje"
Flip, the winter king, a bird 11:42
El_Che holyghost: are you from .be?
holyghost yes
11:42 pecastro joined
holyghost The Netherlands 11:42
lizmat .be is not the Netherlands ? 11:43
holyghost it was :-)
El_Che not since we threw William I over the border :)
11:43 p6bannerbot sets mode: +v pecastro
holyghost we ventured with some on with some musketeers on foot 11:43
I don't remember the name 11:44
split off in 1930
s/19/18
anyway, I'm a Belgian, Flemish in Brussels 11:45
not a Walloon :-)
lizmat holyghost: then maybe you should join El_Che for a Brussels.pm meeting :-)
El_Che I used to live close to Vandersteen :)
holyghost You're bruxelles.pm.org, El_Che ? 11:46
El_Che no, but I know them and sometimes I drop by
holyghost ok, I work for them
I get some 100s of euros each time I release a package 11:47
I'm not a system hacker
El_Che: you're from Antwerp ? 11:48
El_Che yes
holyghost I cannot CTCP you
nazi-town in Belgium :-) 11:49
I work for Alexandre
I wanted to go to a meeting decades ago, also to Munich YAPC
I'll join a meeting once but I need more time to code really 11:50
I'm not a meeting fellow 11:51
gnoll statutes :-)
glad to know where you're from 11:52
I have a music/graphics codirector in Antwerp
for my band "Lost Momentum" 11:53
lizmat: there's 2 links up above 11:54
www.mediafire.com/file/k71a94bc1xqq...4.mp3/file
That's my latest music with her 11:55
El_Che: It's a rework from the group "Implant", they're Belgian
Rage is one of their songs 11:56
I tried to mix with Juno reactor, but it isn't easy
It's simple but I work on it from time to time in the weekend 11:58
As I said I'm going to write some perl6 sound modules
that's something a reverse engineer can parse :-)
lizmat: It's not metal, I'm sorry :-) 12:01
Anyway, El_Che I'm engoying a Chimay Blue 12:02
if you like that beer, it's brussels one
El_Che sorry, somewhat absent, wat working at some code for work 12:03
holyghost ok, nm
lizmat holyghost: also distracted... and it's woolfy who's very much into metal 12:07
lizmat is more of a prog rock person
12:07 Itaipu left
holyghost woolfy ? 12:07
12:07 kudzo__ joined
holyghost lizmat: sure 12:08
12:08 p6bannerbot sets mode: +v kudzo__
holyghost lizmat: I like new wave which a friend of mine has delivered me 12:08
lizmat: Covenant for example, it's not funny however 12:09
12:09 Itaipu joined
holyghost lizmat: I prolly cannot talk to woolfy about Metallica, the black album 12:09
12:09 p6bannerbot sets mode: +v Itaipu
holyghost lizmat: I also love "the matrix" film sounds 12:10
woolfy : see above for good metal from Keanu Reeves
12:11 kudzo_ left 12:18 finanalyst joined 12:19 p6bannerbot sets mode: +v finanalyst
holyghost lizmat: Is woolfy older than 30 ? else I'll tell him some good sounds 12:21
lizmat one good and one bad assumption :-) 12:22
holyghost lizmat: If he's young I'd like to tell him about new wave, terefrom 12:23
12:23 domidumont joined
El_Che colder... 12:23
:)
holyghost s/terefrom/therefrom
lizmat two bad assumptions :-)
12:24 p6bannerbot sets mode: +v domidumont
holyghost .tell woolfy we need to talk about music, new wave and metal 12:24
yoleaux holyghost: I'll pass your message to woolfy.
holyghost so there :-)
lizmat I'll check what the old girl can do for you :-)
holyghost girl ?
woolfy is a girl ? 12:25
El_Che warmer
holyghost lol
old women make me fry :-)
El_Che and now she'll hit you
holyghost anyway, I'll tell her about Covenant, Sepultura and Industrial/Dark wave 12:26
12:26 woolfy joined
holyghost hello 12:26
Can you see this backlog ?
no because I don't run BitchX :-) 12:27
El_Che there are logs
holyghost woolfy : check the backlog
woolfy: we need to talk, kibologist 12:28
Chaos A.D.
El_Che holyghost: it sounds you're 40+
I remember seeing them around that time 12:29
holyghost El_Che: yes
Back in the day I used a 28.8k modem
external
in 1995 a simple winsoft modem
El_Che 14.4 internal :)
holyghost coolness 12:30
I'm 41
years old
Kaiepi o/
holyghost hi
Kaiepi: tell us your secrets :-) 12:31
Kaiepi well i'm 22 and a hs dropout
got my ged though
SmokeMachine . 12:32
yoleaux 06:15Z <jmerelo> SmokeMachine: to please make a PR for the day you want too. I'll try to take a look today.
holyghost Kaiepi: ged == ?
Kaiepi equivalent to a hs diploma
holyghost ok
SmokeMachine .tell jmerelo sorry, I didn't get it... PR for what? 12:33
yoleaux SmokeMachine: I'll pass your message to jmerelo.
holyghost I graduated in 2015, Ph.D in physics and Computer Science
in 2002 Mr. CS
Kaiepi nice
i plan on going to uni for cs once i get my shit together
holyghost sure
Better read some books then :-) 12:34
Kaiepi got mental health issues and addictions i need to get sorted out first
what sort of books?
holyghost the wizard book. the dragon book
(TNHD hacker dict)
Kaiepi ah sicp 12:35
holyghost coolness, I got one there :-)
Kaiepi dragon book being Compilers: Principles, Techniques, and Tools?
sena_kun yes 12:36
holyghost yacc and lex
the wizard book, scheme meta-evals 12:37
and an intro to scheme
You can then program emacs somewhat :-)
12:38 woolfy left, titsuki left
Kaiepi i'm a vim guy 12:39
12:39 finanalyst left
holyghost me too, elvis and BSD vi 12:39
Kaiepi: anyway take care to grep book in TNHD 12:40
it's not stupid, old yes, but good for UNIX->Linux 12:41
s/UNIX/UNIX</
holyghost misses woolfy 12:43
I thought I could talk about music, with my sounds, I need remixers :-)
lizmat holyghost: if you want to talk about music, this is probably not the right channel 12:44
Kaiepi tnhd?
12:45 zakharyas left
holyghost Kaiepi: The New Hacker Dictionary 12:47
lizmat: ok
Kaiepi: if you need help for C++, I'm here
Kaiepi: To tell secrets: Effective C, More Effective ++, Effective STL should you get started also C++faqs online 12:49
Kaiepi ah
holyghost lizmat: I like meta-chat, but music prolly isn't the right channel
Kaiepi: so you need those books :-) 12:50
lizmat: unless I meta-magic it into a package with a remix :-)
12:50 Itaipu left
holyghost lizmat: See above for sound software from me 12:51
12:51 Itaipu joined
holyghost Kaiepi: I don't like amazon but you can get them 2nd hand for 5$ or so 12:52
12:52 p6bannerbot sets mode: +v Itaipu 12:56 fatguy left 13:21 haukex joined 13:22 p6bannerbot sets mode: +v haukex 13:29 finanalyst joined, p6bannerbot sets mode: +v finanalyst 13:36 damaxi joined, damaxi is now known as andrzejku, p6bannerbot sets mode: +v andrzejku
andrzejku someone saw araraloren here? 13:41
araraloren are you there?
/!seen araraloren
!seen araraloren
timotimo .seen araraloren 13:42
yoleaux I saw araraloren 4 Nov 2018 16:06Z in #perl6: <araraloren> Zoffix okay, thanks :)
El_Che .seen araraloren
yoleaux I saw araraloren 4 Nov 2018 16:06Z in #perl6: <araraloren> Zoffix okay, thanks :)
andrzejku :(
I miss my friend
.seen loren
yoleaux I saw loren 8 Jul 2016 13:51Z in #perl6: <loren> e, ^_^ .. haha don't mind
13:45 jmerelo joined 13:46 p6bannerbot sets mode: +v jmerelo
jmerelo O/ 13:47
yoleaux 09:47Z <holyghost> jmerelo: I have HashedAgent and Agent tested, you can use the ask routines. If you want to try SFY
12:33Z <SmokeMachine> jmerelo: sorry, I didn't get it... PR for what?
jmerelo .tell SmokeMachine I see you did it already. Great! 13:49
yoleaux jmerelo: I'll pass your message to SmokeMachine.
SmokeMachine .
yoleaux 13:49Z <jmerelo> SmokeMachine: I see you did it already. Great!
SmokeMachine jmerelo: :(
jmerelo: :)
sorry, wrong key...
jmerelo SmokeMachine: hi!
SmokeMachine jmerelo: hi! :) 13:50
jmerelo We could be like that the whole day...
SmokeMachine :)
jmerelo I'll take a look at the article. Did someone add you to the WordPress blog? 13:51
SmokeMachine jmerelo: that's not my first perl6 advent calendar post...
jmerelo SmokeMachine: ah, great, so you're in already
SmokeMachine jmerelo: this was my first one: perl6advent.wordpress.com/2016/12/...e-testing/ 13:52
jmerelo .tell holyghost great to see you're advancing, but while these tests exist, they don't really test all functionality and what $ag is supposed to do. You should check for the return value, change of state or something.
yoleaux jmerelo: I'll pass your message to holyghost.
jmerelo SmokeMachine: I remember that one. Very interesting. Didn't remember it was yours, though... 13:53
13:54 sjoshi joined
SmokeMachine thanks 13:54
13:55 p6bannerbot sets mode: +v sjoshi 13:56 scimon joined 13:57 p6bannerbot sets mode: +v scimon
holyghost jmerelo: I'm here 13:58
yoleaux 13:52Z <jmerelo> holyghost: great to see you're advancing, but while these tests exist, they don't really test all functionality and what $ag is supposed to do. You should check for the return value, change of state or something.
holyghost jmerelo: I'll extend test functionality 13:59
I almost fell asleep
jmerelo holyghost: would it be possible that you refrain from uploading new versions to CPAN while your tests are not completed? 14:00
holyghost no if I comprehend
cpan and github are synced
I was working on Mathx::Stat tests (2) 14:01
anyway, the base class dipatch of the Agent.pm6 works
s/dipatch/dispatch
jmerelo holyghost: how come? You have to specifically upload to cpan. That's what I do, at least. New version, make dist, upload to CPAN.
holyghost eco: AI::Agent 14:02
buggable holyghost, Failed to decode result JSON:
jmerelo holyghost: works as in it's thoroughly tested in any aspect and I also provide an example so that people can use it?
holyghost jmerelo: works in the meaning,
with an example yes in t/make-agent.t 14:03
it's a superclass however, which I might need to extend to test
timotimo yes, definitely
holyghost That's what AI-Agent-Music is used in
jmerelo holyghost: that's not an example, it's a test, it just tests that it does not crash, it does not test for actual state change or for the actual functionality 14:04
timotimo you can write tests that look like code you'd expect your users to write
plus a bunch of verification code that everything does what you expect
holyghost sure, I'll have to extend then
jmerelo timotimo: from my point of view, tests in many cases can be used as examples, if they clearly express intent beyond the syntax. That's not the case here.
holyghost with my excuses, I wrote a simple test _script_, I might want tot use OOP for it 14:05
s/tot/to
jmerelo timotimo: "my $ag = AI::Agent::Agent.new( x => 0 );" does not say what x is for, or what kind of values it can take, or anything. The other test uses a dummy message, with an empty %optargs, and is also OK. You are none the wiser about what that object is supposed to do. 14:06
holyghost: it's also customary to have the base class with the same name as the module. In your case, it would probably be the best to create an lib/AI/Agent.pm6 which would hold the base class (which if I understand is now in AI::Agent::Agent) 14:07
holyghost jmerelo: note my $msg = "ask about"; 14:08
jmerelo: it's not an empty message 2nd time
Ulti if I want to find all the substrings that match a given grammar is there a nice way to do that? 14:09
14:09 jmerelo left
Ulti parse is like the entire string is going to match TOP rather than looking for all the matches within like a regex 14:09
sena_kun Ulti, with overlapping substrings possible? 14:10
Ulti preferably not but I'd take that
jnthn Ulti: Probably $str.match(/<TheGrammar::TOP>/, :g) does it
sena_kun well, I don't know about overlapping ones(well, except for bruteforce char-by-char), but, oh, I'm late. :)
Ulti ok so thats like taking TOP out as a regex 14:11
14:11 holyghost left
jnthn and I think pass :ov to get overlapping ones if you did want them 14:11
Ulti thanks 14:12
14:19 holyghost joined 14:20 p6bannerbot sets mode: +v holyghost
holyghost timotimo: you would like me to write a full test (OO) for the AI::Agent::Agent class ? 14:24
to interleave 14:25
timotimo sorry, i'm too busy with other stuff to write the tests for you 14:27
holyghost ok
timotimo but i imagine you would have a "class TestAgent is AI::Agent::Agent { ... }" implement some methods, and then do some stuff with it and see if what it did matches with what it was supposed to do 14:28
holyghost timotimo: with all due respect those are overridden methods 14:29
timotimo: the sole keyword as a message to self.dispatch is "ask about"
timotimo: which answers the correct string 14:30
to dispatch an agent I would have 2 agents which just return a &self.ask_agentmethod, which means it has to parse the dispatched agent 14:31
timotimo OK, then the tests could also be "try creating a class that doesn't re-implement all necessary methods and make sure that it throws an exception"
holyghost that's not the point of AI::Agent::Agent, it is a base class for the parsing stuff
timotimo: that's already in 14:32
the 2 tests
timotimo OK
holyghost else you just call a method on an instance which constructs nicely in t/
timotimo i'm not the best at testing code, either
holyghost it's not my kind of thing also 14:33
timotimo but for every piece of code you have in the module, you should try to find a way to test that it does what you hope
holyghost that's ok except parsing other agents return a &$agent.dispatch 14:34
that's incomputable in a test
where dispatch is the double dispatched agent's dispatch method
with an ok syntax 14:35
it's a ref
scalar vs. ref
14:36 kerframil left
holyghost anyway, the dispatch proc works, so there shouldn't be much of a problem anymore 14:37
dispatch of "ask about" 14:38
message
14:38 pmurias joined, p6bannerbot sets mode: +v pmurias
holyghost dispatch of message "agent" gives &$agent.dispatch 14:39
that's all the base class does
So I'm half right :-)
pmurias: check out AI::Agent, my 1st 2 tests 14:40
14:42 Itaipu left 14:43 broquaint left, broquaint joined 14:44 p6bannerbot sets mode: +v broquaint
holyghost hail to broquaint 14:44
14:46 Itaipu joined 14:47 p6bannerbot sets mode: +v Itaipu
holyghost lizmat: hail to Itaipu ! 14:52
14:53 reportable6 left, reportable6 joined
holyghost woolfy: hail to Itaipu ! 14:53
14:53 p6bannerbot sets mode: +v reportable6
holyghost Oh, they're not online :-) 14:55
14:55 undersightable6 left, undersightable6 joined 14:56 p6bannerbot sets mode: +v undersightable6 14:58 kurahaupo left, kurahaupo joined 14:59 p6bannerbot sets mode: +v kurahaupo 15:02 zakharyas joined
holyghost hail to zakharyas 15:02
15:03 p6bannerbot sets mode: +v zakharyas
holyghost hello, what you've been up to zakharyas ? 15:05
I remember that zakharyas was the name of an elder necrarch, zakharyas the everliving. An vampire 15:07
jdv79 anyone know about this?: nopaste.xyz/?285372ce79885ca5#Y8Bo...jmxj9Ae6I= 15:09
15:13 sergot left, dalek joined, ChanServ sets mode: +v dalek 15:14 Geth_ joined, synopsebot_ joined, p6lert_ joined, ChanServ sets mode: +v Geth_, ChanServ sets mode: +v synopsebot_, synopsebot left, Geth left, p6bannerbot sets mode: +v dalek, p6bannerbot sets mode: +v synopsebot_, p6bannerbot sets mode: +v p6lert_ 15:15 parv left, p6lert left, SourceBaby left
Geth_ whateverable: 6216854099 | (Aleks-Daniel Jakimenko-Aleksejev)++ | xbin/Undersightable.p6
Remove 6lang.party

It's no longer a thing		 15:15
jdv79 .tell jnthn i'm not sure who else to ask - nopaste.xyz/?285372ce79885ca5#Y8Bo...jmxj9Ae6I= 15:17
yoleaux jdv79: I'll pass your message to jnthn.
15:18 Sgeo__ joined 15:19 p6bannerbot sets mode: +v Sgeo__ 15:20 Sgeo_ left
holyghost jdv79: if you paste in pastebin.com I might be of some help 15:21
I don't parse upper SSL
lizmat m: dd my %args = :a([1,2,3]) # jdv79 15:22
camelia Hash %args = {:a($[1, 2, 3])}
lizmat when you put the array into the hash, it gets itemized
because the Array object is put into a container 15:23
because a Hash is mutable
m: my %args is Map = a => [1,2,3]
camelia ( no output )
lizmat m: dd my %args is Map = a => [1,2,3] 15:24
camelia Map.new((:a([1, 2, 3])))
jdv79 i realize that. my question is is there a way to get around that in this case.
lizmat a Map doesn't have that issue
m: class C { has @.a }; my @foo = 1,2,3; my %args is Map = :a(@foo); my $c = C.new(|%args); $c.say
camelia C.new(a => [1, 2, 3])
15:25 Manifest1 joined, p6bannerbot sets mode: +v Manifest1
holyghost m: class C { has @.a; }; my bar (%args) { } my @foo = 1,2,3; my %args is Map = :a(@foo); my $c = C.new(bar(|%args)); $c.say 15:29
camelia 5===SORRY!5===
Type 'bar' is not declared
at <tmp>:1
------> 3class C { has @.a; }; my bar7⏏5 (%args) { } my @foo = 1,2,3; my %args i
Malformed my
at <tmp>:1
------> 3class C { has @.a; }; my7⏏5 bar (%args) { } my @foo = 1…
jdv79 lizmat: ah. i see it now. its buried in the Signatures doc page
holyghost m: class C { has @.a; }; my bar (%args) { } my @foo = 1,2,3; my %args is Map = :a(@foo); my $c = C.new(C.bar(|%args)); $c.say 15:30
camelia 5===SORRY!5===
Type 'bar' is not declared
at <tmp>:1
------> 3class C { has @.a; }; my bar7⏏5 (%args) { } my @foo = 1,2,3; my %args i
Malformed my
at <tmp>:1
------> 3class C { has @.a; }; my7⏏5 bar (%args) { } my @foo = 1…
holyghost lizmat: is they're a wat to apply sub bar to %args ?
s/wat/way
to be honest, I'm looking for simple "apply" routine 15:31
jdv79 oof. maybe that shoudl be on the Slip page too. idk. doc'ing such a "deep" lang seems hard.
15:31 Itaipu left, pmurias left
jdv79 thanks 15:31
lizmat jdv79: yw 15:32
15:32 Itaipu joined
lizmat holyghost: perhaps start by making "bar" a method and making sure the number of opening curlies matches the number of closing curlies? 15:32
15:33 p6bannerbot sets mode: +v Itaipu
holyghost m: class C { has @.a; }; my bar (%args) { }; my @foo = 1,2,3; my %args is Map = :a(@foo); my $c = C.new(C.bar(|%args)); $c.say 15:34
camelia 5===SORRY!5===
Type 'bar' is not declared
at <tmp>:1
------> 3class C { has @.a; }; my bar7⏏5 (%args) { }; my @foo = 1,2,3; my %args
Malformed my
at <tmp>:1
------> 3class C { has @.a; }; my7⏏5 bar (%args) { }; my @foo = …
holyghost m: class C { has @.a; my bar (%args) { }; }; my @foo = 1,2,3; my %args is Map = :a(@foo); my $c = C.new(C.bar(|%args)); $c.say 15:35
camelia 5===SORRY!5===
Type 'bar' is not declared
at <tmp>:1
------> 3class C { has @.a; my bar7⏏5 (%args) { }; }; my @foo = 1,2,3; my %a
Malformed my
at <tmp>:1
------> 3class C { has @.a; my7⏏5 bar (%args) { }; }; my @foo = …
holyghost m: class C { has @.a; method bar (%args) { }; }; my @foo = 1,2,3; my %args is Map = :a(@foo); my $c = C.new(C.bar(|%args)); $c.say
camelia Too few positionals passed; expected 2 arguments but got 1
in method bar at <tmp> line 1
in block <unit> at <tmp> line 1
holyghost lizmat: ^-- 15:36
I've solved it with a : 15:38
method dispatch($message, %optargs) {
unless $!done {
self.dispatch($message, %optargs) for %!dependencies;
$!done = True;
}
}
lizmat: same thing
method BUILD() {
self.add-dependency(&self.ask);
self.add-dependency(&self.agent);
}
method add-dependency($dependency) {
%.dependencies{$dependency.name => $dependency};
}
sorry about the flood, just showing 15:39
El_Che holyghost: use a paste service 15:40
holyghost I'm sorry
lizmat holyghost: the "bar" method expects an Associative. You're flattening the hash in the call to named parameters. dispatch will fail, because you don't pass a hash as a positional
either don't slip the hash in the call, or make the parameter of "bar" a slurpy hash
holyghost pastebin.com/YaP9wAQT 15:41
jnthn .tell jdv79 Do `%foo<a> := @foo` when building the thing to not have the Scalar containers in the first place, or `|%h.Map` to get rid of them 15:42
yoleaux 15:17Z <jdv79> jnthn: i'm not sure who else to ask - nopaste.xyz/?285372ce79885ca5#Y8Bo...jmxj9Ae6I=
jnthn: I'll pass your message to jdv79.
scimon Hey all
lizmat scimon o/ 15:43
timotimo holyghost: i don't think "&self.ask" is what you think it is
greetings scimon
lizmat continues working on the P6W
holyghost timotimo: I should know my refs 15:45
it dispatches in method dispatch correctly
timotimo you know that perl6 doesn't have refs any more, right?
holyghost method refs should work ? 15:46
timotimo not like you wrote in your code
did you ever try to see if %.dependencies contains what you expected? 15:47
in that code, i mean?
holyghost eco: AI::Agent
buggable holyghost, AI::Agent 'AI network Agent system, Music Link Agent': cpan.metacpan.org/authors/id/H/HO/H...2.4.tar.gz 1 other matching results: modules.perl6.org/s/AI%3A%3AAgent
holyghost ^-- check those tests 15:48
oh AI-Agent has disappeared
maybe try zef AI::Agent
s/zef/zef install 15:49
the tests of AI-Agent work
jdv79 jnthn: thanks! 15:50
yoleaux 15:42Z <jnthn> jdv79: Do `%foo<a> := @foo` when building the thing to not have the Scalar containers in the first place, or `|%h.Map` to get rid of them
timotimo holyghost: how can you be sure it does anything at all? 15:51
holyghost timotimo: This is my test file: pastebin.com/Pi8JKurh 15:52
timotimo that's what i'm looking at right now 15:53
what do you make of the 8 warnings it outputs? 15:54
holyghost It passes 2 tests on mu machine
s/mu/my
timotimo: nothing because of the ok $hag.dispatch ...
timotimo ?
holyghost it returns the correct value, AI::Agent::HashedAgent should be used as a superclass as I mentioned before 15:55
it dispatches a correct $msg
timotimo how do you know? 15:56
holyghost you can then call it by AI::Agent::HashedAgent.ask
15:56 cydf left
timotimo the tests don't seem to test HashedAgent.ask at all, though? 15:57
holyghost timotimo: pastebin.com/DXusjsAG
timotimo i think you're misunderstanding what "ok" does 15:58
holyghost timotimo: no because it has not been superclassed
timotimo ok is a passed test if the result is true-ish
and the text after that is what it shall output as the "name" of the test
see what happens when you change the text after the "ok $ag.dispatch($msg, %optargs)" line
holyghost $ag and $hag are different instances 16:00
timotimo yeah
holyghost so is the ok return value
timotimo try it with both
16:00 kurahaupo left, domidumont left
holyghost my $num-tests = 2; 16:00
16:00 kurahaupo joined
timotimo i have another suggestion 16:01
16:01 kurahaupo left
timotimo try what happens when you dispatch the same message a second time on the same instance 16:01
16:01 kurahaupo joined
timotimo with a second test that looks just the same 16:01
and adjust the $num-tests to 3, of course
you will find that the second time you dispatch the same message it will fail
buggable New CPAN upload: Template-Anti-0.5.2.tar.gz by HANENKAMP modules.perl6.org/dist/Template::An...:HANENKAMP 16:02
16:02 p6bannerbot sets mode: +v kurahaupo
timotimo it's because your test tests whether the return value of the dispatch method is truthy 16:02
a method without an explicit "return" will return the value of the last statement in it
16:02 jast left
timotimo the last statement in your dispatch method is the "unless" block, which does the same thing: it returns the value of its last statement inside, which is $!done = True. the result of that is True, so dispatch returns True the first time. the second time the unless block doesn't run and the return value is Nil 16:03
16:04 troys joined
holyghost ok 16:04
timotimo and i'll tell you right now, the dependencies attribute does not have in it what you expect, you should check it, perhaps in the BUILD submethod (please make it a submethod)
16:04 p6bannerbot sets mode: +v troys, jast joined 16:05 jmerelo joined, p6bannerbot sets mode: +v jast 16:06 p6bannerbot sets mode: +v jmerelo
timotimo perhaps dd it for more clarity than you get when you just "say" it 16:06
AFK
16:09 Itaipu left 16:12 Itaipu joined, ExtraCrispy joined, p6bannerbot sets mode: +v ExtraCrispy, p6bannerbot sets mode: +v Itaipu
holyghost timotimo: thx, I need to redesign my hash 16:13
timotimo oh, another thing: "zef test" will not output your test's output, so you can more easily miss stuff, like debug output. i recommend running the tests with "perl6 -I lib t/foo.t" instead so you can see everything
i see the docs have nothing about methodrefs 16:14
but methodrefs don't exist in the same way as they did in perl5 16:15
holyghost timotimo: ?
timotimo you either use a little curly block as a closure over self, or you get the method from the object with .^find_method("methodname")
jmerelo timotimo: well, refs in general.
timotimo jmerelo: refs don't exist, but methodrefs exist even less 16:16
they don't-exist very strongly
they un-exist to a higher degree than simply not being there
jmerelo timotimo: they don't++ exist
timotimo: they don't even meta-exist
timotimo they (exist := -Inf)
jmerelo lols 16:17
holyghost +1
haukex Hi all! Thought you might be interested: webperl.zero-g.net/perl6.html
timotimo AFK
holyghost jmerelo: so method refs are unofficial ?
coolness, I'll look it up later on, method refs as timotimo says should be basic perl6 16:18
as in C++ STL 16:19
jmerelo holyghost: you can simply, if you use the correct syntax, store a callable that calls an object method in a container. But the whole pointer/reference thing does not exist in Perl 6
holyghost jmerelo: by usinng @!dependencies ? 16:21
the above is simply offensive
jmerelo holyghost: not exactly. Let's just say you can do it.
holyghost that's a cored system AFAIK 16:22
anyway, I just wanted a hash of methods
not an array :-| 16:23
smalltalk uses a dictionary for class methods 16:24
16:24 Itaipu left, cydf joined
holyghost anyway, I'll have to redesign 16:24
16:24 Itaipu joined 16:25 p6bannerbot sets mode: +v cydf
holyghost jmerelo: with all due respect, AI::Agent::Agent is usable 16:25
not AI::Agent::HashedAgent
16:25 p6bannerbot sets mode: +v Itaipu
jmerelo holyghost: I'm not saying otherwise 16:25
holyghost np 16:26
jmerelo What I'm saying is that it's costumary to have the main class in a module to have the same name as the module.
holyghost ok
jmerelo You have no AI::Agent actual class, nor an AI/Agent.pm6 file which includes documentation and somesuch
Please not I'm not saying that it's usable either. Just that if you want to build a module, that's the usual thing. 16:27
16:27 pmurias joined, p6bannerbot sets mode: +v pmurias, vike1 left
holyghost I will take a look at it tomorrow, it's 5:29 pm here, I've started coding since 3am 16:27
jmerelo If people install an X::Y, they will want to do an use X::Y to use it, not an X::Y::Y (or whatever).
pmurias haukex: super cool :) 16:28
exit
16:28 pmurias left 16:29 pmurias joined, p6bannerbot sets mode: +v pmurias, TimToady left, TimToady joined, tolkien.freenode.net sets mode: +v TimToady, p6bannerbot sets mode: +v TimToady
haukex pmurias: oh, hi! Rakudo.js is super cool ;-) I just sent you an email a second ago 16:30
16:31 andrzejku left
pmurias haukex: I have heard of webperl recently, I was even thinking how interop between the too should work 16:32
jmerelo Hi, TimToady :-) 16:33
haukex pmurias: I already gave that a try :-) github.com/haukex/webperl/blob/mas...test6.html
TimToady is officially back from vacation, but completely trashed due to helping SoCal family deal with wildfire evacuations (nothing of ours burnt down, thankfully), and now I get to breathe all the smoke in NorCal, so it's gonna be a while before I'm back in sync enough to act like a BDFL...and maybe that's a good thing :)
yoleaux 7 Nov 2018 12:07Z <Zoffix> TimToady: please respond to github.com/perl6/user-experience/issues/33
holyghost TimToady: wb 16:34
jmerelo TimToady: you've literally been in hell and back...
haukex pmurias: I'll be AFK for dinner now, hopefully back later 16:35
El_Che TimToady: there priorities in life. It's good you know yours. I hope you and your family and friends are well.
holyghost is AFK
16:41 vike1 joined 16:42 p6bannerbot sets mode: +v vike1
timotimo haukex: oooh nice 16:49
so, can someone explain to me why this happens: 16:50
m: class test { method a { say &self } }; test.a
camelia Nil
timotimo we generate an ifnull for the lexical &self, and if it's null we return Nil 16:51
16:51 zakharyas left
timotimo but for other names it doesn't do that 16:51
m: class test { method a { say &foobar } }; test.a
camelia 5===SORRY!5=== Error while compiling <tmp>
Undeclared routine:
foobar used at line 1
timotimo unsurprisingly
greppable6: &self
greppable6 timotimo, Found nothing!
16:52 reportable6 left, reportable6 joined
timotimo i don't seem to find any mention of "&self" in the spec tests 16:52
16:52 p6bannerbot sets mode: +v reportable6 16:53 lucasb joined 16:54 p6bannerbot sets mode: +v lucasb
SmokeMachine m: my \bla = 1; say &bla 16:55
camelia Nil
SmokeMachine timotimo: that's it? ^^
m: class test { method a(\foobar:) { say &foobar } }; test.a 16:56
camelia Nil
holyghost timotimo, jmerelo : $!done stays True while reloading HashedAgent.dispatch, so 2 tests do not fail 16:57
SmokeMachine timotimo: isnt & "working" with any sigilless var?
lizmat is glad to hear TimToady is ok 16:58
timotimo weird, because the lexical lookup is still for &bla in that case 16:59
pmurias rakudo.js update: blogs.perl.org/users/pawel_murias/2...tests.html 17:01
timotimo can't read that right now because the cat is expecting to get fed and is standing in front of my screen 17:02
pmurias: "exciting" instead of "exiting" in the last line :)
pmurias++ 17:03
lucasb github.com/haukex/webperl/blob/mas...js#L31-L41
^^ Raku =^.^=
17:05 scotticles joined 17:06 p6bannerbot sets mode: +v scotticles
scotticles currently looking at learning perl6, wanting to do some web stuff is Cro the best option to build on? 17:06
17:06 Itaipu left
Geth_ doc: a8a25c0e42 | Coke++ | doc/Type/Any.pod6
Statements need to end with ;
 17:06
synopsebot_ Link: doc.perl6.org/type/Any
jmerelo scotticles: probably. Best of luck, and come back here if you need help
MasterDuke scotticles: there's also #cro 17:07
timotimo cro is pretty good, though it doesn't yet bring its own templating and such (there is something experimental in a repo called cro-webapp, though) 17:08
scotticles ah ok, thanks, also is DBI still best to use the p5 include to handle databases or is there something for perl6
sena_kun scotticles, it may be. but don't assume Cro is web-framework like django/dancer/etc, it is more of microsevice building tool(with nice modules built for HTTP), so things like database access and templating are not out of the box.
timotimo Db::Pg or what it's called if you have a postgres database
otherwise DBIish will do what you need
DBIish also does postgres 17:09
but DB::Pg does it better
jmerelo scotticles: there's DBIish, which includes several databases, and they there's RED, an ORM by SmokeMachine which gives you a high-level interface
scotticles yeah i should use postgres for this project, its been on my todo list for a long time of learning it vs using mysql
thanks this will get me going
kensanata I've been using Cro with Template::Mustache and that worked very well.
17:10 Itaipu joined 17:11 sjoshi left, finanalyst left, p6bannerbot sets mode: +v Itaipu
SmokeMachine scotticles: if you'd like to use a ORM, I think Red is a good choose: github.com/FCO/Red 17:11
jmerelo SmokeMachine: :-) see above ^^^ 17:12
SmokeMachine jmerelo: I saw it! thanks! :)
jmerelo SmokeMachine++
scotticles havn't considered an ORM , so used to writing sql. But i'll look into Red, might as well. 17:13
SmokeMachine jmerelo: sorry being boring you.... but had you had time to see the current version of the post? 17:14
timotimo i've got a project right now where i use cro + sqlite in the backend and react.js in the frontend
the one thing i wish would be better is compile times whenever i update the component that currently has all the database queries in it 17:15
i might build a middleware of some kind that lets me edit the queries in the browser and re-run if i don't like the results :D
SmokeMachine timotimo: why dont you try to use Red? 17:16
timotimo because i mostly just pull stuff out of the DB and pass it on as json :)
jmerelo SmokeMachine: I have a deadline for a conference today, and I'm really busy with a couple of papers... Check out my GH activity.
Plus classes...
17:16 Itaipu left, rindolf left
timotimo in the future i may even let sqlite generate the json directly in some cases 17:16
SmokeMachine jmerelo: sorry... 17:17
timotimo but i'd be interested to hear from you if you can see much benefit that RED could give based on my code
i'll give you a link
17:17 Itaipu joined
SmokeMachine timotimo: :) 17:18
timotimo github.com/timo/moarperf/blob/mast...b.pm6#L123 - for example
17:18 p6bannerbot sets mode: +v Itaipu
pmurias timotimo: thanks for spoting the typo ;) 17:19
SmokeMachine timotimo: yes... I dont think Red can help you yet... but it will in the near future! :) 17:20
timotimo wow, how? 17:21
SmokeMachine timotimo: there are some of your queries that could be beneficied by Red... the simpler ones... 17:24
timotimo right 17:25
do you happen to have any experience with sqlite and recursive queries? :)
SmokeMachine timotimo: for example: www.irccloud.com/pastebin/BqTbD7SO/
timotimo actually, i'm done with that particular can of worms 17:26
and i won't revisit that any time soon :D
SmokeMachine timotimo: if calls where a Red model, could be written as: `Calls.^all.grep(*.id == $id).map: { .id, .parent, .routine }` 17:27
17:27 domidumont joined
timotimo does that build an efficient sql query? 17:28
SmokeMachine yes, the same you wrote...
17:28 p6bannerbot sets mode: +v domidumont
timotimo cool 17:28
gotta look at the implementation at some point :)
17:29 sjoshi joined, dakkar left, p6bannerbot sets mode: +v sjoshi 17:30 scimon left
SmokeMachine timotimo: take a look at this sample: github.com/FCO/Red/wiki/Features-Sample 17:30
timotimo does it break down if you have something like .map: { .id, (.parent if .id > 10), .routine }, too?
SmokeMachine timotimo: currently yes... 17:31
timotimo then i have a clue how you might have implemented that :D
SmokeMachine :)
timotimo: would you like to guess? 17:32
timotimo you first pass an "information collector" object to the block that map gets 17:33
to find out what columns are asked for
SmokeMachine that's it...
timotimo i might have done the same thing
SmokeMachine not only to .map...
timotimo right
and you probably have implementations of &infix:<==> and friends, too 17:34
SmokeMachine timotimo: thats it!
timotimo you could even support user-provided infix operators with sql's custom functions :D
SmokeMachine timotimo: github.com/FCO/Red/blob/master/lib...rators.pm6
timotimo but that reaches into "terribly clever" territory 17:35
SmokeMachine timotimo: what do you mean with "terribly clever"? 17:36
timotimo: I have a lot to do with Red yet... but i think its going to become very good to use... 17:37
timotimo are you considering exhaustively mapping the decision states of such grep and map functions? since you already know all comparisons and computations that happen 17:39
tobs` Can I tell List.reduce to be "right-associative" somehow? As in I want <a b c>.reduce(&f) to compute f(a, f(b, c)) instead of f(f(a, b), c).
timotimo m: say &infix:<**>.assoc
camelia No such method 'assoc' for invocant of type 'Sub+{is-pure}+{Precedence}'
in block <unit> at <tmp> line 1
jmerelo tobs`: using reverse?
timotimo m: say &infix:<**>.^assoc
camelia No such method 'assoc' for invocant of type 'Perl6::Metamodel::ClassHOW'
in block <unit> at <tmp> line 1
timotimo how does assoc, hmm.
m: say &infix:<**>.^methods
camelia (prec is-pure soft <anon> yada <anon> <anon> <anon> perl <anon> wrap <anon> <anon> <anon> <anon> <anon> onlystar multi <anon> <anon> gist leave package <anon> <anon> cando candidates <anon> unwrap <anon> BUILDALL set_why <anon> has-phasers fire_phaser…
timotimo m: say &infix:<**>.prec
camelia {assoc => right, prec => w=}
timotimo ah, of course 17:40
SmokeMachine timotimo: sorry, I didnt understand... what?
timotimo m: sub f($a, $b) is assoc<right> { say "<$a, $b>"; $a ~ $b }; <a b c d>.reduce(&f)
camelia <c, d>
<b, cd>
<a, bcd>
timotimo tobs`: is that it?
SmokeMachine: in my example with .parent if .id > 10 17:41
jmerelo timotimo: cool
timotimo you could hook it up to return true the first and false the second time
tobs` heh, cool timotimo++
timotimo that way it would find that sometimes .parent can be found
SmokeMachine timotimo: I dont know how to solve that yet... but I was thinking in a slang...
timotimo: currently I only run it once... 17:42
tobs` jmerelo: good thought, but my f is not commutative, so it'd be a double reverse List.reverse.reduce({ f($^b, $^a) })
SmokeMachine but I think thats a good idea...
tobs` m: <a b c d>.reverse.reduce({ $^b ~ $^a }).say 17:43
camelia abcd
tobs` m: <a b c d>.reverse.reduce({ say $^b ~ $^a })
camelia cd
bTrue
aTrue
timotimo m: <a b c d>.reverse.reduce( &[R~] )
camelia ( no output )
timotimo m: <a b c d>.reverse.reduce( &[R~] ).say
camelia abcd
timotimo m: <a b c d>.reverse.reduce( [R~] ).say
camelia Cannot resolve caller reduce(Seq: Str); none of these signatures match:
(Any:U: &, *%_ --> Nil)
(Any:D: &with, *%_)
in block <unit> at <tmp> line 1
timotimo m: <a b c d>.reverse.reduce( &[~] ).say
camelia dcba
SmokeMachine timotimo: but had you liked how Red is writen? 17:44
timotimo reversing operators is built-in to perl6 :) ):
:) :)
SmokeMachine: have not looked deep enough, but i think it's cool
obviously that might change if i encounter some problems with that approach in real-life usage
SmokeMachine :) 17:45
timotimo: Now I see! i can get that .id.Bool was called and call the clousure again... 17:54
17:56 kurahaupo_ joined
timotimo well, really the result of .id > 10 will have .Bool called on it, but yeah, that's exactly it 17:57
17:57 p6bannerbot sets mode: +v kurahaupo_, kurahaupo left 17:58 kurahaupo_ left, rindolf joined, kurahaupo joined 17:59 p6bannerbot sets mode: +v rindolf, p6bannerbot sets mode: +v kurahaupo 18:02 robertle left 18:05 haukex left
jmerelo 6 days without a Perl 6 question in StackOverflow... stackoverflow.com/questions/tagged/perl6 18:09
I should start again the campaign to (also) ask your questions in SO...
18:09 haukex joined 18:10 p6bannerbot sets mode: +v haukex
SmokeMachine timotimo: yes... the result is a Red::AST... so... 18:10
jmerelo: maybe it means that the docs are getting "too" good... :P 18:11
jmerelo SmokeMachine: he. I wish
SmokeMachine just joking... 18:12
18:13 kurahaupo left
tobs` m: use Test; is-deeply set(1, |set(1)), Set(1, |Set(1)) 18:13
camelia not ok 1 -
# Failed test at <tmp> line 1
# expected: Set.new(1)
# got: Set.new(1 => Bool::True,1)
18:13 kurahaupo joined
lizmat m: dd (1,2).Set.Slip 18:13
camelia slip(2 => Bool::True, 1 => Bool::True)
18:14 p6bannerbot sets mode: +v kurahaupo
lizmat m: dd Set(1).Slip 18:14
camelia slip(1 => Bool::True,)
lizmat m: dd set(1).Slip 18:15
camelia slip(1 => Bool::True,)
tobs` m: dd Set.new(1, |Set.new(1))
camelia Set.new(1 => Bool::True,1)
tobs` m: dd Set(1, |Set(1))
camelia Set.new(1)
18:15 kurahaupo left 18:16 kurahaupo joined
tobs` so, the documentation is correct in saying set() is just Set.new(), but I don't understand the different behaviour with slip 18:16
lizmat m: dd Set(1).Slip
camelia slip(1 => Bool::True,)
18:16 kurahaupo left, kurahaupo joined
lizmat ah, I think I know what it is 18:16
m: dd (1 => True).Set
camelia Set.new(1)
18:17 p6bannerbot sets mode: +v kurahaupo
lizmat m: dd Set.new(1, 1=> True) 18:18
camelia Set.new(1 => Bool::True,1)
lizmat Set.new attempts to be smart on how it setifies its parameters
m: dd Set.new(1, 2 1=> True)
camelia 5===SORRY!5=== Error while compiling <tmp>
Unable to parse expression in argument list; couldn't find final ')' (corresponding starter was at line 1)
at <tmp>:1
------> 3dd Set.new(1, 27⏏5 1=> True)
expecting any of:
in…
lizmat m: dd Set.new(1, 2, 1=> True)
camelia Set.new(1,2,1 => Bool::True)
lizmat actually, set() is
m: dd set(1, 2, 1=> True)
camelia Set.new(1,1 => Bool::True,2)
lizmat hmmm 18:19
tobs` you mean Set() is?
lizmat ok, clearly I'm too dim atm
tobs` m: dd Set(1, 2, 1 => True)
camelia Set.new(2,1)
lizmat right, that one
m: dd Set(1, 2, 3, 1 => True)
camelia Set.new(3,2,1)
lizmat m: dd Set(2, 3, 1 => False) 18:20
camelia Set.new(3,2)
lizmat m: dd Set(2, 3, 1 => True)
camelia Set.new(1,3,2)
lizmat goes back to working on the P6W
tobs` thanks lizmat
18:20 holyghost left
tobs` it's even right there in the documentation of the Set() routine ._. 18:21
18:21 holyghost joined, scotticles left
tobs` when you think you found an undocumented corner case... 18:21
18:22 p6bannerbot sets mode: +v holyghost, Itaipu left 18:23 Itaipu joined 18:24 p6bannerbot sets mode: +v Itaipu, sauvin left 18:28 lucasb left 18:29 damaxi joined 18:30 p6bannerbot sets mode: +v damaxi 18:35 Itaipu left, Itaipu joined 18:36 p6bannerbot sets mode: +v Itaipu 18:37 kdr23 joined 18:38 p6bannerbot sets mode: +v kdr23 18:39 Itaipu left 18:40 kdr22 left 18:41 Itaipu joined, p6bannerbot sets mode: +v Itaipu 18:45 damaxi is now known as andrzejku 18:51 Itaipu left 18:55 noganex_ left 18:56 noganex joined 18:57 haukex left, p6bannerbot sets mode: +v noganex 18:58 ryn1x joined 18:59 p6bannerbot sets mode: +v ryn1x, ryn1x left 19:00 ryn1x joined 19:01 p6bannerbot sets mode: +v ryn1x 19:14 tobs` left 19:26 Mithaldu left 19:27 Mithaldu joined, kensanata left 19:28 p6bannerbot sets mode: +v Mithaldu 19:36 ryn1x left 19:37 molaf joined 19:38 p6bannerbot sets mode: +v molaf 19:41 domidumont left 19:43 natrys joined 19:44 p6bannerbot sets mode: +v natrys 19:49 pmurias left 20:04 sjoshi left 20:17 jmerelo left 20:18 yqt left
SmokeMachine why is ACCEPTS the oposite? I men: 20:22
m: my $a = 42 but role :: { method ACCEPTS(|c) { say "called { c }"; True } }; given $a { when "b" {}}
camelia ( no output )
SmokeMachine m: my $a = 42 but role :: { method ACCEPTS(|c) { say "called { c }"; True } }; given "b" { when $a {}}
camelia called b
SmokeMachine that means: if I want to do something when my object is used on a given/when, I cant... I have to change every other class to insert another ACCEPTS candidate (ok, I can only change Any...) 20:24
tobs SmokeMachine: in a given-when, you're trying to see if the topic falls into any of the when categories, via smartmatch 20:25
gfldex m: my $a = 42 but role :: { method ACCEPTS(|c) { say "called { c }"; True } }; $_ = "b"; when $a {}
camelia called b
tobs I think it makes sense to ask the *categories* if they accept the topic
I mean, that's not an explanation, but evidence that it makes sense to some :) 20:26
SmokeMachine the gfldex makes sense...
but it makes more dificult what I want to do... :( 20:27
gfldex this is the rare case where i make sense :)
SmokeMachine: what do you want to do?
SmokeMachine tobs: that makes sense... 20:28
gfldex: try to find whats happening inside a Callable... 20:29
gfldex: as I was talking with timotimo (colabti.org/irclogger/irclogger_lo...-12#l1102) but I was thinking how I could do that for given/when... 20:31
m: use MONKEY-TYPING; role R {}; augment class Any { multi method ACCEPTS(R $a) { say "called {self} $a"; True } }; Any.^compose; given R.new { when "b" {}} # :( 20:37
camelia ( no output )
masak moritz: does writing/modifying grammars ever stop feeling like an exercise in Spooky Action at a Distance? 20:42
(asking for a friend, who just busted his grammar *again*)
good news: my anonymous friend just found the problem -- it was a stray bit of whitespace in a `rule` 20:47
jnthn masak: I'm not moritz, but I dunno if I ever really felt like that... :) 20:50
20:50 Sound joined, p6bannerbot sets mode: +v Sound, kst joined 20:51 p6bannerbot sets mode: +v kst
jnthn masak: I mean, sure, a change I do might regress a test I wasn't expecting it to at some point along the way, but it doesn't spook me more than that happening in non-grammar code would :) 20:51
20:54 newbie23 left 20:55 newbie23 joined, p6bannerbot sets mode: +v newbie23 21:10 rindolf left 21:16 rindolf joined 21:17 p6bannerbot sets mode: +v rindolf
moritz masak: kinda, yes 21:18
SmokeMachine where CX::Next is handled? 21:23
better question: how can I create a sub that accepts next? 21:25
is that possible? 21:34
moritz m: sub f(&g) { g(); CONTROL { when CX::Next { say "next caught" } } }; f sub { next } 21:36
camelia next caught
moritz SmokeMachine: ^^ like that?
SmokeMachine moritz: thanks! I didnt know CONTROL... I was trying with CATCH!
moritz SmokeMachine: my pleasure 21:37
SmokeMachine CX::Next isnt resumable? 21:39
is there a way to resume it?
moritz m: sub f(&g) { g(); CONTROL { when CX::Next { say "next caught"; .resume } } }; f sub { next; say "after next" }
camelia next caught
This exception is not resumable
in block at <tmp> line 1
in sub at <tmp> line 1
in sub f at <tmp> line 1
in block <unit> at <tmp> line 1
21:44 kensanata joined, p6bannerbot sets mode: +v kensanata
kensanata Quick question before I go to bed. I just finished building rakudo star 2018.10. Strangely, 2018.6 ended up as "/home/alex/rakudo/install/bin//perl6" according to "which perl6" and I can find 2018.10 as "rakudo/bin/perl6" -- how did this happen? Did I miss a step somewhere or was the default changed? 21:47
kensanata changes ~/.bashrc and stops wondering 21:48
gfldex lizmat: have you considered to write a how-has-Perl 6-grown-in-the-last-12-months-post for the advent calendar? 21:49
kensanata cries as all the zef stuff breaks...
gfldex lizmat: I'm asking because I need the numbers of new modules until xmass 2018 for a post I might write. :) 21:50
21:51 dct joined, p6bannerbot sets mode: +v dct
lizmat gfldex:I haven't :-) 21:51
gfldex lizmat: please tell me when you reconsider
lizmat I will 21:52
sena_kun I am thinking about writing a post, but I have never made one(except literature, but that's another deal) and cannot really think of topic. I know stuff about Cro, a lot about Comma and am building some ASN.1 library right now. I wonder if something can hit the bar. 21:56
ugh, s/hit the bar/be useful and meaningful enough/ 21:59
SmokeMachine Oh! talking about Comma! jnthn, sena_kun I am mentioning Comma on my perl6 advent calendar post, is it OK? Not finished yet, but if you want to know how/why its being mentioned: gist.github.com/FCO/33be94f5ecfab7...etsanta-md (if you'd like to comment too, that OK :) ) 22:00
*that's OK 22:01
lizmat and another Perl 6 Weekly hits the Net: p6weekly.wordpress.com/2018/11/12/...st-diwali/ 22:02
sena_kun not sure about jnthn, but you have my Seal of Approval. :) Don't see any harm in such mention.
lizmat moritz: p6weekly.wordpress.com/2018/11/12/...st-diwali/ 22:03
notable6: weekly reset
notable6 lizmat, Moved existing notes to “weekly_2018-11-12T22:03:38Z”
timotimo \o/ 22:08
lizmat++
22:08 kensanata left
sena_kun lizmat++ 22:08
timotimo lizmat: did last squashathon's winner get a prize? 22:09
22:10 leont joined
lizmat well, I know the nick of the winner 22:11
I don't have an address yet to send Camelia to
22:11 p6bannerbot sets mode: +v leont
moritz lizmat++ 22:14
22:14 newbie23 left 22:15 newbie joined 22:16 p6bannerbot sets mode: +v newbie
jdv79 lizmat: did you do anything on the one arg rule/itemization/etc... in your articles? 22:16
22:16 dct left
jdv79 i don't see it 22:16
and do you plan to?
lizmat you mean the opensource.com articles ?
jdv79 yeah. i thought there'd be a bit in the containers artile but i don't see it
22:17 newbie left 22:18 newbie joined 22:19 p6bannerbot sets mode: +v newbie, andrzejku left
lizmat it's not really about containers, but about signatures / parameters 22:19
jdv79 cool p6w though. its getting quite long.
lizmat tell me about it
jdv79 yeah, that's true.
lizmat goes off for some R&R
22:19 newbie left 22:20 newbie joined, p6bannerbot sets mode: +v newbie 22:30 sena_kun left
jnthn SmokeMachine: Yeah, find by me, though we don't tend to write it fully capitalized (e.g. just Comma). It's named after a type of butterfly, rather than being an acronym. :) 22:33
*fine by me
SmokeMachine jnthn: fixing it...
22:52 ryn1x joined, p6bannerbot sets mode: +v ryn1x 23:02 lucasb joined, natrys left, p6bannerbot sets mode: +v lucasb 23:05 comborico1611 joined 23:06 p6bannerbot sets mode: +v comborico1611, ryn1x left 23:08 rindolf left 23:11 ryn1x joined 23:12 p6bannerbot sets mode: +v ryn1x 23:16 pecastro left 23:19 Sound left 23:30 jast left
lucasb so, 'my Foo @a' is an Array that does Positional[Foo], right? 23:38
23:38 jast joined
lucasb and 'my Foo %h' is a Hash that does Associative[Foo], ok? :) 23:38
but... 23:39
23:39 p6bannerbot sets mode: +v jast
lucasb 'my Foo $x' isn't a Scalar that does Itemable[Foo] 23:39
Scalar is a class. I mean "Itemable" for whatever the "missing" role was supposed to be named :) 23:40
IOW, did this design lost an opportunity to have some symmetry between Scalars, Positionals and Associatives?
it shows in this case: 23:41
m: constant Foo = 42; my Foo @x # ok, error, nonsense code
camelia 5===SORRY!5=== Error while compiling <tmp>
Can not parameterize Array with 42
at <tmp>:1
------> 032; my Foo @x # ok, error, nonsense code7⏏5<EOL>
expecting any of:
constraint
lucasb m: constant Foo = 42; my Foo $x
camelia 5===SORRY!5=== Error while compiling <tmp>
Variable definition of type Int (implicit : by pragma) requires an initializer
at <tmp>:1
------> 3constant Foo = 42; my Foo $x7⏏5<EOL>
expecting any of:
constraint
23:42 ryn1x left
lucasb if a Scalar was a parameterizable Itemable[::T], then it would have the same error as arrays, "Cannot parameterize Scalar with 42" 23:43
23:46 comborico1611 left
lucasb "constant Foo = 42; my Foo $x" doesn't make sense, but let's follow what the error message says. it requires an initializer, so let's give it one: 23:51
m: constant Foo = 42; my Foo $x = Any
camelia Type check failed in assignment to $x; expected Int but got Any (Any)
in block <unit> at <tmp> line 1
lucasb expects Int...
m: constant Foo = 42; my Foo $x = Int
camelia Type check failed in assignment to $x; expected 42 but got Int
in block <unit> at <tmp> line 1
lucasb ok, it changed its mind, wants 42 now :) 23:52
m: constant Foo = 42; my Foo $x = 42
camelia Type check failed in assignment to $x; expected type 42 cannot be itself (perhaps Nil was assigned to a :D which had no default?)
in block <unit> at <tmp> line 1
lucasb m: constant Foo = 42; my Foo $x = 21+21
camelia Type check failed in assignment to $x; expected 42 but got 42
in block <unit> at <tmp> line 1
23:57 lucasb left 23:59 Pllll joined, p6bannerbot sets mode: +v Pllll

Powered by Cro and the Raku® Programming Language.

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

Thank you!