»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.
timotimo er, at least i think so 00:00
well, when you "but" or "does" them in at run time, they will
Xliff m: role A { method data { 1; }; }; class B does A { method data { 2; }; };
camelia ( no output )
Xliff m: role A { method data { 1; }; }; class B does A { method data { 2; }; }; B.new.data.say
camelia 2
00:01 graphene left
Xliff m: role A { method data { 1; }; }; class B { method data { 2; }; }; my $b but A = B.new; $b.data 00:01
camelia Cannot modify an immutable 'Any+{A}' type object
in block <unit> at <tmp> line 1
Xliff m: role A { method data { 1; }; }; class B { method data { 2; }; }; my $b = B.new but A; $b.data
camelia ( no output )
Xliff m: role A { method data { 1; }; }; class B { method data { 2; }; }; my $b = B.new but A; $b.data.say
camelia 1
Xliff \o/
00:02 graphene joined
Xliff That syntax has taken me long enough to finally grok, but I think I finally got it. 00:02
Class but (also) Role 00:03
00:03 p6bannerbot sets mode: +v graphene
Xliff m: role A { method data { 1; }; }; class B { method data { 2; }; }; my $b = B.new does A; $b.data.say 00:04
camelia 1
Xliff I think I prefer that.
timotimo++ # Thanks! Have a great night!
timotimo thanks, and same to you :) 00:05
00:06 kurahaupo joined, p6bannerbot sets mode: +v kurahaupo 00:10 graphene left 00:11 kurahaupo left, graphene joined 00:12 p6bannerbot sets mode: +v graphene 00:14 w_richard_w joined 00:15 p6bannerbot sets mode: +v w_richard_w 00:16 kurahaupo joined 00:17 p6bannerbot sets mode: +v kurahaupo 00:18 bolangi joined 00:19 p6bannerbot sets mode: +v bolangi, w_richard_w left 00:27 kurahaupo left 00:30 kurahaupo joined 00:31 p6bannerbot sets mode: +v kurahaupo 00:44 kurahaupo left 00:45 kurahaupo joined 00:46 p6bannerbot sets mode: +v kurahaupo 00:52 kurahaupo left 01:04 literal left 01:10 lizmat left 01:12 MasterDuke joined, p6bannerbot sets mode: +v MasterDuke, MasterDuke left, MasterDuke joined, herbert.freenode.net sets mode: +v MasterDuke, p6bannerbot sets mode: +v MasterDuke 01:32 graphene left 01:33 graphene joined 01:34 p6bannerbot sets mode: +v graphene 01:37 lizmat joined, p6bannerbot sets mode: +v lizmat 01:40 beeman left 01:41 lizmat left 02:09 Aceeri_ joined, p6bannerbot sets mode: +v Aceeri_
awwaiid hello peeps. In case you didn't see, we are hosting the DC-Baltimore Perl Workshop on April 6, 2019 in Silver Spring, MD (DC metro area). Y'all are invited! We are accepting talks through the end of the year; I'd love some perl6/raku/rakudo deliciousnesss -- CFP Press Ctrl-c to copy 02:19
bit.ly/dcbpw-cfp
mmm
copy/paste link fail. Oh well. Also see dcbpw.org/dcbpw2019/ 02:20
02:33 vike joined 02:34 p6bannerbot sets mode: +v vike 02:36 molaf left 02:48 molaf joined 02:49 p6bannerbot sets mode: +v molaf 03:01 bolangi left 03:02 bolangi joined 03:03 p6bannerbot sets mode: +v bolangi, someuser left 03:08 leont left 03:23 sivoais joined, p6bannerbot sets mode: +v sivoais 03:41 john_parr joined 03:42 p6bannerbot sets mode: +v john_parr 03:49 graphene left 03:50 graphene joined 03:51 p6bannerbot sets mode: +v graphene
SmokeMachine Is there any defined way to write browser modules? 04:25
04:27 kurahaupo joined 04:28 p6bannerbot sets mode: +v kurahaupo
holyghost SmokeMachine: start with the headers ? 04:29
SmokeMachine holyghost: what do you mean?
holyghost C headers API or something
what's your "browser" ? 04:30
e.g. you can do that for lynx, firefox etc. AFAIK
SmokeMachine holyghost: I mean something like perl6.github.io/6pad/#3d9c7dd4834c...9011ab6420 04:31
holyghost I cannot read that https 04:32
do you have a http link somwhere ?
SmokeMachine holyghost: I mean for Rakudo.js
holyghost I'm going a little bit of topic, mod_perl (5) for apache, and for js for I don't remember 04:33
There was talk about it last week
SmokeMachine I mean running modules in a browser 04:34
holyghost if you output to js ti should work
s/ti/it
but you have to have a perl6 plugin in firefox before using it 04:35
except the js I mean
See perlyroids in the year 2000+, an internet explorer plugin for perl5 04:36
by Adobe
SmokeMachine holyghost: have you seen Rakudo.js? 04:38
holyghost no
ah now I see 04:39
I don't know, but to use perl browserwise you have to load a plugin 04:40
SmokeMachine No, you haven’t...
.seen pmurias 04:42
yoleaux I saw pmurias 19 Nov 2018 21:43Z in #perl6: <pmurias> * survey
holyghost I don't know much about js but you can load it in your browser then use it's API
SmokeMachine .tell pmurias would you mind of giving me your opinion about perl6.github.io/6pad/#3d9c7dd4834c...9011ab6420 ? 04:43
yoleaux SmokeMachine: I'll pass your message to pmurias.
SmokeMachine holyghost: my question was to know if is there something different I should do on my module if it will depend on js... 04:44
holyghost now I understand, 04:45
SmokeMachine .tell pmurias is there something different I should do on my module if it will depend on js...
yoleaux SmokeMachine: I'll pass your message to pmurias.
holyghost I don't know really
SmokeMachine Does anyone know? 04:46
tyil I'm trying to use a react/whenever block for Proc::Async, but when I run my code, I get the error I should tap stderr/stdout before running the process p.tyil.nl/fdtn 04:48
04:54 Cabanoss- joined 04:55 p6bannerbot sets mode: +v Cabanoss- 04:56 Cabanossi left
lookatme_q tyil, you should put the stderr/stdout tap before start 05:13
tyil I have that now, and getting a moarvm panic 05:14
lookatme_q tyil, yeah, I got too 05:18
What happened ? :/ 05:19
tyil :/
if I knew I wouldn't be asking :p
lookatme_q Maybe a bug 05:25
I manual start the Proc::Async outside the react block, it works 05:26
05:40 Bucciarati left, avar left 05:41 Bucciarati joined, avar joined, avar left, avar joined, p6bannerbot sets mode: +v avar 05:42 p6bannerbot sets mode: +v Bucciarati, p6bannerbot sets mode: +v avar
tyil lookatme_q: can you show me what you mean? 05:46
05:52 Sound joined, p6bannerbot sets mode: +v Sound
tyil it seems to not panic if I dont return inside the whenever block 05:53
05:57 finanalyst joined 05:58 p6bannerbot sets mode: +v finanalyst
finanalyst could anyone here tell me how to use %?RESOURCES, mentioned at docs.perl6.org/language/variables#..._variables ? I have default files in a module that I want to access when an object is built 06:01
tyil finanalyst: put them into the resources directory, and reference them in the META6.json in the resources list 06:02
then you can use %?RESOURCES<relative/path/from/resources/dir> to access them
finanalyst tyil: thanks 06:11
tyil: do I need to reference each file in META6.json, or can I use a * ? 06:12
tyil I believe you'll need to add each file, no globbing
holyghost finanalyst: reference every file 06:20
it does not parse too well
06:20 Sound left 06:21 ufobat joined 06:22 p6bannerbot sets mode: +v ufobat 06:24 sauvin joined, p6bannerbot sets mode: +v sauvin 06:27 jmerelo joined, p6bannerbot sets mode: +v jmerelo
holyghost hi jmerelo 06:29
I am working on perl6 sound software, I have a WAV parser and am coding the GUI of a Wave Editor. I need to build a player too (I guess based on the linux sound C headers) 06:30
I'll put it on github when it's ready 06:31
I built some widgets today, need an OOP widget tree 06:32
problem is I know that using design patterns (gang of four book) works but I cannot use the papers at the end of the chapters because of patents, so it'll be more common 06:35
The first module should become Soundx::Tina (WAV things) 06:38
jmerelo Software patents do not apply in Europe and it would be stupid to include stuff in books that can't be used 06:55
holyghost: hi
07:19 jcarty joined 07:20 p6bannerbot sets mode: +v jcarty 07:23 domidumont joined 07:24 p6bannerbot sets mode: +v domidumont 07:26 someuser joined 07:27 p6bannerbot sets mode: +v someuser 07:29 ChoHag joined 07:30 p6bannerbot sets mode: +v ChoHag 07:38 domidumont left 07:39 graphene left 07:40 graphene joined 07:41 p6bannerbot sets mode: +v graphene 07:44 Bucciarati left 07:45 Bucciarati joined 07:46 p6bannerbot sets mode: +v Bucciarati
andrzejku .seen loren 07:47
yoleaux I saw loren 8 Jul 2016 13:51Z in #perl6: <loren> e, ^_^ .. haha don't mind
andrzejku .seen lookatme_q
yoleaux I saw lookatme_q 05:26Z in #perl6: <lookatme_q> I manual start the Proc::Async outside the react block, it works
07:50 khatar joined 07:51 p6bannerbot sets mode: +v khatar 07:52 abraxxa joined 07:53 p6bannerbot sets mode: +v abraxxa 07:54 abraxxa left 07:57 abraxxa joined, zakharyas joined 07:58 p6bannerbot sets mode: +v abraxxa, p6bannerbot sets mode: +v zakharyas 07:59 rindolf joined, domidumont joined, p6bannerbot sets mode: +v rindolf, abraxxa left 08:00 p6bannerbot sets mode: +v domidumont 08:01 abraxxa joined 08:02 robertle joined, p6bannerbot sets mode: +v abraxxa 08:03 p6bannerbot sets mode: +v robertle 08:14 coet[work] joined, p6bannerbot sets mode: +v coet[work]
lookatme_q andrzejku, I am busying :) 08:23
09:01 lizmat joined, p6bannerbot sets mode: +v lizmat 09:08 finanalyst left, finanalyst joined 09:09 p6bannerbot sets mode: +v finanalyst 09:12 finanalyst left 09:13 finanalyst joined, p6bannerbot sets mode: +v finanalyst 09:21 scimon joined 09:22 p6bannerbot sets mode: +v scimon 09:23 vike left 09:29 vike joined 09:30 p6bannerbot sets mode: +v vike 09:43 vike left 09:48 vike joined 09:49 p6bannerbot sets mode: +v vike 10:04 bolangi left 10:09 coet|login joined, p6bannerbot sets mode: +v coet|login 10:10 coet[work] left, coet|login is now known as coet[work] 10:16 bolangi joined 10:17 p6bannerbot sets mode: +v bolangi
Geth doc: 1e74f97826 | (Elizabeth Mattijsen)++ | doc/Language/glossary.pod6
Added lemma about gradual typing
 10:31
synopsebot_ Link: doc.perl6.org/language/glossary
10:33 ChoHag left 10:46 tobs` joined, p6bannerbot sets mode: +v tobs` 10:48 finanalyst left 10:50 finanalyst joined 10:51 p6bannerbot sets mode: +v finanalyst 10:56 ChoHag joined 10:57 p6bannerbot sets mode: +v ChoHag 10:58 leont joined 10:59 p6bannerbot sets mode: +v leont 11:25 finanalyst left 12:00 kurahaupo left, kurahaupo joined 12:01 p6bannerbot sets mode: +v kurahaupo 12:06 ChoHag left 12:09 zakharyas left, molaf left 12:11 khisanth_ left, ChoHag joined 12:12 p6bannerbot sets mode: +v ChoHag 12:14 christopher joined, christopher is now known as Guest41175 12:15 Guest41175 left
tbrowder I just read about the quantum ide qiskit from ibm and also found a list of quantum simulator software in many languages, including perl 5. is anyone working on one in p6? 12:29
12:33 ChoHag left
jmerelo tbrowder: not that I know of. 12:34
tbrowder any person here know enough about quantum computing to add a quantum-centric task on rosettacode? 12:36
jmerelo tbrowder: I'm kind of skeptic on that kind of thing... 12:39
tbrowder damian has demoed some kind of quantum mockup...something to help understand the thoeory and such 12:41
jmerelo tbrowder: I'm OK with the theory, I'm not so sure about the "computing" part 12:42
tbrowder :-/
i know what yoy mean—it’s the end of the world when q starts beaking pki all over 12:43
*breaking
12:47 kensanata joined, p6bannerbot sets mode: +v kensanata 12:53 molaf joined 12:54 p6bannerbot sets mode: +v molaf 13:05 jmerelo left 13:10 coet[work] left 13:12 rindolf left
lizmat wow: example of how providing type information will let your code run faster: 13:21
m: sub a($a) { }; for ^10_000_000 -> int $_ { a 42 }; say now - INIT now
camelia 5.9543976
lizmat m: sub a(Int $a) { }; for ^10_000_000 -> int $_ { a 42 }; say now - INIT now
camelia 0.1108427
lizmat m: sub a($a) { }; for ^10_000_000 -> int $_ { a 42 }; say now - INIT now
camelia 5.87832916
lizmat m: sub a(Int $a) { }; for ^10_000_000 -> int $_ { a 42 }; say now - INIT now
camelia 0.1038929
lizmat m: sub a($a) { }; my $a = 42; for ^10_000_000 -> int $_ { a $a }; say now - INIT now 13:22
camelia 0.3800129
lizmat m: sub a(Int $a) { }; my $a = 42; for ^10_000_000 -> int $_ { a $a }; say now - INIT now 13:23
camelia 0.1804015
lizmat the last two are probably more real world examples
13:24 beeman joined, p6bannerbot sets mode: +v beeman 13:32 ChoHag joined
jnthn When it's `Int $a`, then we don't emit code to produce the Scalar container. Thanks to lexical to local lowering of `$a`, that probably means spesh can see the box operation is dead, so doesn't bother. 13:32
13:33 p6bannerbot sets mode: +v ChoHag
jnthn Whereas when we have to allocate the container 'cus the thing passed may be Iterable, we can't avoid the boxing, so it's two more allocations 13:33
This is one of the things the work on EA/SR is aimed at tackling
lizmat well, it's already doing quite well.. :-)
13:34 zakharyas joined 13:35 p6bannerbot sets mode: +v zakharyas
lucasb EA/SR? 13:39
jnthn Escape Analysis, Scalar Replacement 13:40
13:42 sena_kun joined 13:43 p6bannerbot sets mode: +v sena_kun 13:44 loops left
lizmat yeah, --profile shows that the Int case doesn't do any allocations at all 13:46
lucasb ah, thanks! 13:57
dogbert2_ m: sub a($a) { }; for ^10_000_000 -> int $_ { a 42 }; say now - INIT now 14:12
camelia 6.024947628
dogbert2_ heh
that example runs a lot faster on my 32 bit system
lizmat dogbert2_: oddly enough, it does for me as well locally 14:13
dogbert2_ strange indeed 14:14
lizmat it's more like only 2x as slow for me
dogbert2_ same here
what might be the cause of that discrepancy 14:15
14:16 loops joined
lizmat I have no idea 14:16
14:17 p6bannerbot sets mode: +v loops 14:19 lizmat left
mornfall can i still catch exceptions with "or foo"? 14:22
if so, how do i get at the error?
timotimo doesn't "orelse" have exception catching semantics? 14:24
mornfall i mean, jget ... or fail( "stuff broke" ); seems to do what i mean 14:25
*but* i don't know how to get at the exception object that way
timotimo m: sub doesnt-work { die "oh no" }; doesnt-work orelse say "message: $_" 14:27
camelia 5===SORRY!5=== Error while compiling <tmp>
Undeclared routine:
orelse used at line 1
timotimo m: sub doesnt-work { die "oh no" }; doesnt-work() orelse say "message: $_"
camelia oh no
in sub doesnt-work at <tmp> line 1
in block <unit> at <tmp> line 1
timotimo m: sub doesnt-work { die "oh no" }; try doesnt-work() orelse say "message: $_"
camelia ( no output )
timotimo m: sub doesnt-work { die "oh no" }; (try doesnt-work()) orelse say "message: $_"
camelia Use of Nil in string context
message:
in block at <tmp> line 1
timotimo m: sub doesnt-work { die "oh no" }; (try doesnt-work()) orelse say "message: $!"
camelia message: oh no
timotimo that's how ^
14:28 ChoHag left
timotimo m: sub doesnt-work { die "oh no" }; (try doesnt-work()) or say "message: $_" 14:28
camelia Use of uninitialized value of type Any in string context.
Methods .^name, .perl, .gist, or .say can be used to stringify it to something meaningful.
message:
in block <unit> at <tmp> line 1
timotimo m: sub doesnt-work { die "oh no" }; (try doesnt-work()) or say "message: $!"
camelia message: oh no
timotimo or that way
mornfall i tried $! but something failed... lemme try again
(i mean, that's the first thing i tried)
yeah, $! seems to be Nil for typed exceptions :\ 14:29
timotimo does the sub in question fail or throw?
m: sub doesnt-work { fail "failed" }; (try doesnt-work()) or say "message: $!" 14:30
camelia message: failed
timotimo m: sub doesnt-work { fail "failed" }; (doesnt-work()) or say "message: $!"
camelia Use of uninitialized value of type Any in string context.
Methods .^name, .perl, .gist, or .say can be used to stringify it to something meaningful.
message:
in block <unit> at <tmp> line 1
timotimo ^- if you put a "try", the failure is fatalized
jnthn m: sub doesnt-work { fail "failed" }; doesnt-work() orelse say "message: $_" 14:31
camelia message: (HANDLED) failed
in sub doesnt-work at <tmp> line 1
in block <unit> at <tmp> line 1
jnthn m: sub doesnt-work { fail "failed" }; doesnt-work() orelse say "message: $_.exception()"
camelia message: failed
mornfall thanks (clearly or and orelse behave differently for 'fail') 14:35
timotimo orelse is there for exceptiony things
14:36 khatar left 14:39 graphene left
timotimo as is andthen 14:39
14:40 graphene joined 14:41 p6bannerbot sets mode: +v graphene 14:42 finanalyst joined, finanalyst left 14:45 Merfont left
mornfall any ideas why jget from WWW would give me this? Cannot use a Buf as a string, but you called the Str method on it 14:51
timotimo is there a way to specify an encoding? perhaps the server doesn't serve a Content-Type or something? 14:56
15:01 jcarty left
mornfall timotimo: indeed, using 'get' seems to do okay and JSON::Tiny fails on the result... thanks for the tip 15:04
15:09 MasterDuke left
ufobat how could I pass random key=value paris to a p6-script, what would be a good signature for MAIN() 15:10
15:14 molaf left
moritz have you tried *%options? 15:14
$ perl6 -e 'sub MAIN(*%options) { say %options.perl }' --foo=bar
{:foo("bar")}
ufobat i've tried it without * which didnt work 15:17
15:18 uzl joined 15:19 p6bannerbot sets mode: +v uzl 15:21 uzl left, uzl joined 15:22 p6bannerbot sets mode: +v uzl
ufobat does this work for you? perl6 -e 'my %*SUB-MAIN-OPTS = {:named-anywhare}; sub MAIN("a", :$b) {}' a -b=1 15:24
15:25 ZofBot joined, ChanServ sets mode: +v ZofBot, p6bannerbot sets mode: +v ZofBot 15:26 ZofBot_ joined, ChanServ sets mode: +v ZofBot_, p6bannerbot sets mode: +v ZofBot_, ZofBot__ joined, ChanServ sets mode: +v ZofBot__, ZofBot__ left, ZofBot_ left, ZofBot left, ZofBot joined, ChanServ sets mode: +v ZofBot, p6bannerbot sets mode: +v ZofBot
uzl m: class A {has $!id; method get-id { $!id }; submethod TWEAK { $!id = "Class A" }}; class B is A { has $!id; submethod TWEAK { $!id = "class B" }}; say B.new.get-id; 15:28
camelia Class A
15:28 clarkema joined
uzl m: class A {has $!id; method get-id { $!id }; submethod TWEAK { $!id = "Class A" }}; class B is A { has $!id; submethod TWEAK { $!id = "class B" }}; say B.new.^attributes; 15:29
camelia (Mu $!id Mu $!id)
15:29 dotdotdot left
uzl m: class A {has $!id; method get-id { $!id }; submethod TWEAK { $!id = "Class A" }}; class B is A { has $!id; submethod TWEAK { $!id = "class B" }}; say B.new.^attributes(:local); 15:29
camelia (Mu $!id)
15:29 p6bannerbot sets mode: +v clarkema
uzl How can I get the value of $!id in the class B instead of A? 15:29
moritz uzl: inside the class body, just use $!id 15:31
ufobat but i'd say within the method get-id you cant get it, because get-id of A cant see the private attributes of subclasses 15:32
sena_kun my $attr = B.new.^parents[0].^attributes(:local); $attr.get_value(B.new);
SmokeMachine m: class A {has $!id; method get-id { $!id }; submethod TWEAK { $!id = "Class A" }}; class B is A { submethod TWEAK { self.^attributes.first(*.name eq q[$!id]).set_value: self, "class B" }}; say B.new.get-id; # I know... its terrible...
camelia class B
15:32 coet[work] joined, p6bannerbot sets mode: +v coet[work]
ufobat B doesn't see $!id of A and A doesnt see $!id of B, they're both something different 15:33
sena_kun using MOP, but that is a bad thing (though I hope you __really__ want that).
m: class A { has $!id; method get-id { $!id }; submethod TWEAK { $!id = "Class A" }}; class B is A { has $!id; submethod TWEAK { $!id = "class B" }}; my $attr = B.new.^parents[0].^attributes(:local)[0]; say $attr.get_value(B.new); 15:34
camelia Class A
ufobat you could make $.id public?
15:35 uzl left
moritz or you could trust the parent class 15:35
15:35 tobs` left
SmokeMachine m: class A {has $.id is rw = "Class A"; method get-id { $!id }; class B is A { submethod TWEAK { $.id = "class B" }}; say B.new.get-id; # I know... its terrible... 15:36
camelia 5===SORRY!5=== Error while compiling <tmp>
'A::B' cannot inherit from 'A' because 'A' isn't composed yet (maybe it is stubbed)
at <tmp>:1
15:37 dotdotdot joined 15:38 p6bannerbot sets mode: +v dotdotdot 15:39 uzl joined, p6bannerbot sets mode: +v uzl
uzl sorry, i got disconnected. 15:40
moritz: I meant accessing it through the inherited `get-id` method.
jnthn m: class A { has $.id = 1; }; class B is A { has $.id = 2 }; given B.new { say .id; say .A::id } 15:42
camelia 2
1
moritz uzl: the inherited get-id method can't access it 15:43
uzl I was backlogging and read the responses. Thanks! 15:47
15:47 uzl left 15:58 coet|login joined, coet|login left, abraxxa left 16:03 graphene left 16:04 graphene joined 16:05 p6bannerbot sets mode: +v graphene, coet[work] is now known as orejotas, orejotas left, orejotas joined, barjavel.freenode.net sets mode: +v orejotas, p6bannerbot sets mode: +v orejotas 16:06 khatar joined, khatar left, rindolf joined, p6bannerbot sets mode: +v rindolf, orejotas is now known as coet 16:07 coet left, coet joined, barjavel.freenode.net sets mode: +v coet, p6bannerbot sets mode: +v coet, eiro1 joined 16:08 p6bannerbot sets mode: +v eiro1 16:09 coet is now known as coet[work] 16:12 ChoHag joined 16:13 Kaiepi joined, p6bannerbot sets mode: +v ChoHag 16:14 p6bannerbot sets mode: +v Kaiepi, eiro1 left, eiro joined, coet[work] left, pampolut joined, p6bannerbot sets mode: +v pampolut, pampolut left 16:15 coet[work] joined, p6bannerbot sets mode: +v coet[work], coet[work] left
mornfall can I somehow ask that a thing happens at exit? sort of like with END {}, but I only want to run the thing if I get to a certain point of the program 16:15
16:15 p6bannerbot sets mode: +v eiro, coet[work] joined, p6bannerbot sets mode: +v coet[work], coet[work] left
mornfall what I actually want: I have a Proc::Async running in the background and I want to clean it up if someone calls exit(1) in the middle of the script 16:16
it's a bit problematic because the program is actually running under a different UID, so Proc::Async won't be able to kill it even if it wanted to
16:16 coet[work] joined, p6bannerbot sets mode: +v coet[work]
mornfall so I need my own piece of code to run 16:16
well i guess i can do a thing like END { cleanup if ( $pid ); } 16:17
Xliff mornfall: I was about to suggest something like that. 16:18
timotimo there's a module for that, afaik 16:23
oh, the bots aren't there?
16:24 kensanata left
mornfall another thing: how do i undef a variable? $pid = undef tells me to use Any, but Any.new gives me something that evaluates to true? 16:24
timotimo you're supposed to assign Any, not Any.new :) 16:25
mornfall as in the type?
jkramer Or assign Nil
timotimo yeah
type objects are what perl6 has for undefined values
16:25 domidumont left
[Coke] you can think of the type objects like Any or Int type-specific undefs. 16:26
mornfall interesting
timotimo it has all kinds of cool consequences in the language 16:27
for example, when you write my FooBar $blep .= new, it'll work because the default value for $blep is the FooBar type object 16:31
so that'll call new on the type object and assign the result to the variable 16:32
mornfall any neat tricks to get get Proc::Async stdout/stderr as lines? 16:34
jnthn $proc.stdout.lines 16:35
16:35 robertle left
mornfall .tap, i just tried that (dunno if it worked yet) 16:35
jnthn `whenever $proc.stdout.lines -> $line { }`
mornfall \o/ almost (need to tack a newline at the end now) 16:36
cool stuff
i think i'll get used to this :p 16:37
timotimo can probably :!chomp in the .lines 16:38
mornfall .lines already eats the newlines (but i need them back)
timotimo yes
that's why you :!chomp instead of the default which is :chomp
mornfall oh wait... is !chomp supposed to un-chomp? 16:39
timotimo docs.perl6.org/routine/lines#class_Supply
yeah, :foo is foo => True and :!foo is foo => False
mornfall okay, another day :) there's only so much magic i can digest in one afternoon
ufobat is this a rakudo bug=
gist.github.com/ufobat/cea6b00f102...mmands-txt
timotimo the sub main opts need to be exported, too 16:41
mornfall if a Proc::Async dies of its own volition, am i guaranteed to get everything it spewed via .tap?
timotimo i'd assume when rakudo's not able to process enough, it'll block the process so it can't write data that won't arrive 16:42
but i'd have to verify
AFKBBL
ufobat timotimo, even with our %*SUB-MAIN-OPTS is export = ( 'named-anywhere' => True); it doesnt work 16:43
mornfall okay something is not working :| 16:53
16:54 ChoHag left 16:56 rindolf left 16:57 ExtraCrispy joined, p6bannerbot sets mode: +v ExtraCrispy 17:01 ChoHag joined 17:02 p6bannerbot sets mode: +v ChoHag
mornfall okay i think i need to somehow wait for the tapped supplies 17:05
17:06 TreyHarris joined, p6bannerbot sets mode: +v TreyHarris
mornfall okay I can get a Promise out of it, it seems 17:06
17:07 robertle joined 17:08 p6bannerbot sets mode: +v robertle 17:30 rindolf joined 17:31 p6bannerbot sets mode: +v rindolf 17:37 ufobat left 17:38 Ven`` joined 17:39 p6bannerbot sets mode: +v Ven``
mornfall okay i have no idea how to use react 17:43
or Supply.from-list
hm, that bit seems to work... i'm still puzzled though :\ 17:46
17:47 andrzejku_ joined, p6bannerbot sets mode: +v andrzejku_ 17:48 andrzejku_ is now known as andrzejku 17:50 molaf joined 17:51 p6bannerbot sets mode: +v molaf 17:56 andrzejku_ joined 17:57 p6bannerbot sets mode: +v andrzejku_ 17:59 andrzejku left, andrzejku_ is now known as andrzejku 18:01 ryn1x joined, p6bannerbot sets mode: +v ryn1x 18:02 beeman left
mornfall okay the captive program is buffering its output /o\ 18:04
can i ask Proc::Async to run the thing in a PTY?
or is there some other way to coax external programs to autoflush?
18:06 ryn1x left 18:15 coet[work] left
timotimo PTY is really, really, really fiddly to do 18:16
especially when crossplatforming something
18:16 Ven`` left 18:19 Ven`` joined, andrzejku left 18:20 p6bannerbot sets mode: +v Ven``
mornfall timotimo: i know, since the slave is a python script i figured i can just do python -u ... 18:20
nonetheless, a single 'whenever' block can run concurrently with itself? 18:21
because that's something i didn't expect
(i would also like to stop that from happening but didn't find a way yet)
throttle might be the thing, but it's... confusing 18:24
18:31 beeman joined, p6bannerbot sets mode: +v beeman 18:33 random_yanek joined, p6bannerbot sets mode: +v random_yanek, Ven`` left 18:38 joy_ joined, p6bannerbot sets mode: +v joy_ 18:41 andrzejku joined, p6bannerbot sets mode: +v andrzejku 18:48 Ven`` joined 18:49 p6bannerbot sets mode: +v Ven`` 18:56 scimon left, andrzejku_ joined 18:57 p6bannerbot sets mode: +v andrzejku_ 18:59 andrzejku__ joined, andrzejku left, andrzejku__ is now known as andrzejku 19:00 p6bannerbot sets mode: +v andrzejku, zakharyas left 19:02 andrzejku_ left
timotimo there's only ever one piece of code running per "react" or "supply" block 19:07
19:14 hamhu3 left 19:17 pmurias joined, p6bannerbot sets mode: +v pmurias
pmurias SmokeMachine: hi 19:18
yoleaux 04:43Z <SmokeMachine> pmurias: would you mind of giving me your opinion about perl6.github.io/6pad/#3d9c7dd4834c...9011ab6420 ?
04:45Z <SmokeMachine> pmurias: is there something different I should do on my module if it will depend on js...
pmurias SmokeMachine: super cool that you are building something with rakudo.js
SmokeMachine pmurias: hi!
pmurias: is there any best practices to crete modules for it? 19:19
pmurias SmokeMachine: not yet, just started thinking about it
I'll have to go through the module meta data spec and see how implementation dependend ones should be marked
and it would be cool if 6pad had a way of using modules, got to think how to hook that up too 19:20
19:21 perlbot joined
SmokeMachine that would be wonderful! 19:21
19:21 p6bannerbot sets mode: +v perlbot, perlbot left
pmurias is that our official META6.json spec: design.perl6.org/S22.html#META6.json 19:25
19:32 graphene left, sena_kun left 19:33 perlbot joined, p6bannerbot sets mode: +v perlbot, graphene joined 19:34 p6bannerbot sets mode: +v graphene 19:35 lizmat joined, p6bannerbot sets mode: +v lizmat, simcop2387 joined, p6bannerbot sets mode: +v simcop2387 19:37 pmurias left 19:38 pmurias joined, p6bannerbot sets mode: +v pmurias
lucs What do I need to change to make something like this, which works: 19:55
if $cond { $instance.foo } else { $instance.bar }
to something like this (which doesn't):
$instance.($cond ?? foo !! bar)
pmurias SmokeMachine: I'm thinking we need something like a :from style modifier as dependency 19:57
SmokeMachine: See design.perl6.org/S22.html#%3Afrom%3Cbin%3E
SmokeMachine: a :from<npm> seems sensible for js modules not sure right know what a good one for just raw EVAL stuff would be 19:58
SmokeMachine pmurias: I agree 19:59
tobs m: IntStr.new(10, "ten")."{ $_ ?? "Int" !! "Str" }"().say for True, False 20:03
camelia 10
ten
tobs lucs: this uses ."method-name" call syntax, which allows interpolation of expressions
the parentheses are required though 20:04
lucs tobs: Perfect. Thanks. 20:06
masak m: IntStr.new(10, "ten")."$(my $method)"().say given $method = $_ ?? "Int" !! "Str" for True, False 20:10
camelia 5===SORRY!5=== Error while compiling <tmp>
Missing semicolon
at <tmp>:1
------> 3ay given $method = $_ ?? "Int" !! "Str" 7⏏5for True, False
20:11 sena_kun joined
masak oh, `given` and `for` clash, maybe 20:11
m: (IntStr.new(10, "ten")."$(my $method)"().say given $method = $_ ?? "Int" !! "Str") for True, False
camelia 10
ten
masak not my best code... :P
20:11 p6bannerbot sets mode: +v sena_kun 20:12 irdr left
vrurg Have a question. Lets assume theres a class B inheriting from A. There is an instance of A (my $a = A.new). How do I create and instance of B which would be a clone of $a without manulally passing on all attributes from A? 20:13
20:14 TreyHarris left
vrurg Something like B.new( |$a.Hash )? 20:15
jnthn If they're all public attributes, probably B.new(|$a.Capture) will do it 20:16
vrurg Yes, they're. Thanks jnthn! 20:18
20:19 irdr joined 20:20 p6bannerbot sets mode: +v irdr
pmurias SmokeMachine: I'll have to play around with how code splitting in parcel works so it might be possible to precompile a lot of Perl 6 modules to allow them to be usable in 6pad 20:20
masak m: class C { has $.x; has $.y }; my $c = C.new(:x(1), :y(2)); say $c.Capture 20:21
camelia \(:x(1), :y(2))
20:22 andrzejku_ joined 20:23 p6bannerbot sets mode: +v andrzejku_
pmurias got to walk the dog (as I fear it might be going berserk on house interior) 20:23
&
20:25 andrzejku left, andrzejku_ is now known as andrzejku, AlexDaniel joined, p6bannerbot sets mode: +v AlexDaniel
El_Che weekly: news.ycombinator.com/item?id=18561587 lots op Perl 6 in the comments 20:27
notable6 El_Che, Noted!
El_Che ^-- lizmat
lizmat yeah, Go2 ... :-)
El_Che every language seems to have a Perl 6/Python 3 moment :) 20:28
mornfall how about php 6 :p 20:38
El_Che lizmat: it's look like they are jumping on the unicode wagon (1. #20706 General Unicode identifiers based on Unicode TR31) although not Perl6-crazy style :)
lizmat then they will find out that it is not enough :-) 20:39
masak El_Che: if identifiers are made public by making the first character uppercase, how would you make an identifier starting with a Chinese character public?
El_Che yeah, they are pretty ambivalent about the go 2 changes
masak: "There are normalization questions which we need to answer" ? 20:40
20:40 zakharyas joined
El_Che masak: maybe they'll introduce twigils :) 20:41
20:41 p6bannerbot sets mode: +v zakharyas
masak this does not seem to be about normalization. it's about using the capitalization of the first character in the identifier as a `public`/`export` trait 20:44
and you can't do that with (e.g.) hanzi, because they do not have lowercase/uppercase 20:45
El_Che masak: I am completely ignorant (perfectly numb) about casing in Chinese
masak it's simple: there is none
El_Che then we have a problem :)
masak that's my point
maybe you simply can't use hanzi for API-level things 20:46
El_Che - new features: support for most languages, execept Chinese and some other "minor" languages :)
masak pfah :)
El_Che masak: It's weird to combine latin names and local alphabets 20:47
like articles do by adding random local words in the article
"So, he answered in Spanish, "Hey, compadre, where are you going" 20:48
grr
:)
masak my native language isn't English, but I do all my programming in English. which probably makes me biased in a way 20:49
El_Che or old 20:50
:)
20:51 lizmat left 20:53 lizmat joined, p6bannerbot sets mode: +v lizmat 20:57 lizmat_ joined, p6bannerbot sets mode: +v lizmat_
masak not that you're wrong, but how would using English make me old? :) 20:58
El_Che because in the olde days, that's all we had :) 20:59
20:59 lizmat left, kurahaupo_ joined 21:00 p6bannerbot sets mode: +v kurahaupo_
masak not sure what you mean 21:00
all mainstream languages have keywords in English -- I'd be interested to hear about even one exception 21:01
21:01 kurahaupo left
masak English, for better or worse, is the lingua franca (yes, I know) of computing 21:02
21:03 kurahaupo joined 21:04 p6bannerbot sets mode: +v kurahaupo 21:05 kurahaupo_ left 21:07 kurahaupo left
El_Che masak: you missed zoffix's utf8 tweets about crazy operators :) 21:08
masak: but, idd, I also use english vars and whatnot
s/idd/indeed/ 21:11
timotimo masak: i have one for you, but i forgot what it's called! 21:12
nas.sr/%D9%82%D9%84%D8%A8/ here it is
21:14 pmurias left
lucasb ah, the great قلب programming language 21:16
21:20 pmurias joined, p6bannerbot sets mode: +v pmurias 21:21 ranguard joined
El_Che manual.lenguaje-latino.org/si_sino_if_else.html 21:21
programming in spanish
21:21 ranguard left
El_Che Logo also has a Spanish version, iirc 21:22
timotimo well, i mean, there's also basic in german, isn't there? 21:23
El_Che baßic? 21:24
en.wikipedia.org/wiki/Non-English-..._languages 21:25
timotimo baſic
El_Che thinking of the Cold War, I would expect some Russian languages
21:26 zakharyas left
El_Che Someone here going to DockerCon next week? 21:27
21:28 lizmat_ left
AlexDaniel El_Che, robertle: 2018.11 was released some seconds ago :) v6.d is now the default language version 21:31
El_Che \œ/ 21:32
it's a typo, but consider it a Janus face looking at the past and the future at the same time :)
AlexDaniel: thx, running on travis now! 21:34
AlexDaniel squashable6: next 21:40
squashable6 AlexDaniel, ⚠🍕 Next SQUASHathon in 2 days and ≈12 hours (2018-12-03 UTC-12⌁UTC+14). See github.com/rakudo/rakudo/wiki/Mont...Squash-Day
Geth advent: Altai-man++ created pull request #3:
Claiming a spot for my second article		 21:47
lucasb cool, just noticed 5.28.1 was also released! 21:51
21:54 dct joined, p6bannerbot sets mode: +v dct
Geth advent: 184a254a1c | Altai-man++ (committed using GitHub Web editor) | perl6advent-2018/schedule
Claiming a spot for my second article
 22:01
advent: ce0746744e | (Aleks-Daniel Jakimenko-Aleksejev)++ (committed using GitHub Web editor) | perl6advent-2018/schedule
Merge pull request #3 from Altai-man/patch-1

Claiming a spot for my second article
sena_kun wonders how many words are "too many" for a blog post 22:04
22:05 Celelibi joined 22:06 p6bannerbot sets mode: +v Celelibi 22:14 sno joined, p6bannerbot sets mode: +v sno 22:18 sno left 22:22 Ven`` left 22:23 sno joined, graphene left 22:24 p6bannerbot sets mode: +v sno, sena_kun left, graphene joined 22:25 p6bannerbot sets mode: +v graphene 22:30 sena_kun joined, p6bannerbot sets mode: +v sena_kun 22:41 Khisanth joined, graphene left 22:42 p6bannerbot sets mode: +v Khisanth, graphene joined 22:43 p6bannerbot sets mode: +v graphene
tbrowder i have posted a draft of my Advent article here ,github.com/tbrowder/Perl6-Advent/b...-04.md> 22:46
i would appreciate any technical corrections (particularly the grammar stuff) or other suggestions 22:47
22:53 sena_kun left, perlbot left, simcop2387 left 22:57 simcop2387 joined, p6bannerbot sets mode: +v simcop2387
lucasb just a nitpick about capitalization, which you may disagree, I prefer to use "POD" for the Perl 5 variant and "Pod" for the Perl 6 variant of the "plain old documentation" formats :) 22:57
I think this case distinction helps to distinguish which variant is being talked about 22:59
El_Che lucasb: that's confusing 23:00
lucasb indeed it is 23:01
tbrowder i consciously used POD because it's really an acronym, but I'm amenable to changing. jmerelo: thoughts?
23:02 Celelibi left
El_Che if one needs an explanation of the difference between POD and Pod, it looks like a lost battle to me 23:03
damn, we suck at names
tbrowder jmerelo i think is pushing for "Pod 6" or "Pod6". we do use the ".pod6" file extension...
AlexDaniel haha let's call it rakud 23:04
tbrowder "rakupod?" 23:05
rakup
23:06 perlbot joined, p6bannerbot sets mode: +v perlbot
tbrowder "rakup" the balls, boys, and lets have a pool game! 23:06
23:07 lizmat joined, p6bannerbot sets mode: +v lizmat 23:08 Celelibi joined 23:09 p6bannerbot sets mode: +v Celelibi 23:13 andrzejku left, andrzejku joined 23:14 p6bannerbot sets mode: +v andrzejku 23:16 kerframil joined, p6bannerbot sets mode: +v kerframil 23:23 rindolf left 23:28 pmurias left 23:42 beeman 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!