ยปรถยซ 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! ๐Ÿฆ‹
Set by Zoffix on 25 July 2018.
Xliff_ m: my @a["a"] = 1; @a.gist.say 00:00
camelia Cannot convert string to number: base-10 number must begin with valid digits or '.' in '3โ5a' (indicated by โ)
in block <unit> at <tmp> line 1
Xliff_ Well, there goes THAT commit message.
24c8e172f Make 'my @a["a"]' a bit more awesome 00:01
Could someone explain? :)
mornfall so a LAST within the whenever for the supply i do not want to wait for
jnthn mornfall: Yes 00:02
Also if you're multiple different supplies you want to treat this way, you can probably factor it out into a sub 00:03
mornfall no worky :\ 00:08
i have both a QUIT and a CATCH in the react and (some) exceptions still kill the program 00:09
00:09 felher joined
jnthn mornfall: Dunno, I need to sleep now, I can have a look at it some more another time if you've a small reproduction of the problem. 00:09
mornfall 'a react block:g nnnn in (the block with the QUIT + CATCH) died because of the exception'
00:10 p6bannerbot sets mode: +v felher, felher left
Xliff_ Why is this an error? 00:13
Calling pango_font_map_list_families(Pango::Raw::Types::PangoFontMap, NativeCall::Types::CArray[NativeCall::Types::CArray[NativeCall::Types::Pointer[Pango::Raw::Types::PangoFontFamily]]]) will never work with declared signature (Pango::Raw::Types::PangoFontMap $fontmap, NativeCall::Types::CArray[NativeCall::Types::CArray[NativeCall::Types::Pointer[Pango::Raw::Types::PangoFontFamily]]] $families)
Or have I been working with this code too long?
As far as I can tell, those signatures match. 00:14
00:20 mowcat left
Xliff_ Oh hell. 00:21
Pointer[repr('CPointer')]] throws that error?
00:35 FieldsaBB joined 00:36 p6bannerbot sets mode: +v FieldsaBB 00:58 FieldsaBB left 01:23 ryn1x joined 01:24 p6bannerbot sets mode: +v ryn1x 01:37 MasterDuke joined, p6bannerbot sets mode: +v MasterDuke 01:38 MasterDuke left, MasterDuke joined, herbert.freenode.net sets mode: +v MasterDuke, p6bannerbot sets mode: +v MasterDuke
timotimo oh damn you, wordpress 02:02
it ate a bunch of stuff in scimon's post
well, really only the <d> from the infix:<d> definitions 02:05
02:38 evalable6 left, finanalyst joined 02:39 p6bannerbot sets mode: +v finanalyst 02:42 evalable6 joined 02:43 p6bannerbot sets mode: +v evalable6 02:49 zacts joined, p6bannerbot sets mode: +v zacts 02:50 vike1 joined 02:51 p6bannerbot sets mode: +v vike1, vike left 02:52 leont left 03:19 ufobat_ left
ryn1x . 03:22
buggable New CPAN upload: Graphics-TinyTIFF-0.0.2.tar.gz by RYNIX modules.perl6.org/dist/Graphics::Ti...cpan:RYNIX 03:24
Xliff_ timotimo: Can you nativecast a Blob[uint8] to a CArray[uint8] and vice versa? 04:01
I know you can use NativeCall::Blob to convert between the two, but would rather a cast be possible.
Would prefer to bind the CArray to the Blob, really. 04:02
04:08 Cabanoss- joined, p6bannerbot sets mode: +v Cabanoss- 04:09 molaf joined, p6bannerbot sets mode: +v molaf 04:10 Cabanossi left 04:11 ufobat joined 04:12 p6bannerbot sets mode: +v ufobat 04:14 TreyHarris left 04:16 mowcat joined 04:17 p6bannerbot sets mode: +v mowcat 04:34 liamr4c left 04:36 TreyHarris joined, p6bannerbot sets mode: +v TreyHarris 04:37 lookatme_q left 04:38 lookatme_q joined, p6bannerbot sets mode: +v lookatme_q 04:39 liamr4c joined, p6bannerbot sets mode: +v liamr4c 04:40 ryn1x left
Xliff_ m: use NativeCall; my $c = CArray[uint8].new(1, 2, 3, 4); my $b = nativecast(Blob[uint8], $a); $b.^name.say 04:50
camelia 5===SORRY!5=== Error while compiling <tmp>
Variable '$a' is not declared
at <tmp>:1
------> 3 3, 4); my $b = nativecast(Blob[uint8], 7โ5$a); $b.^name.say
Xliff_ m: use NativeCall; my $c = CArray[uint8].new(1, 2, 3, 4); my $b = nativecast(Blob[uint8], $c); $b.^name.say
camelia Internal error: unhandled target type
in sub nativecast at /home/camelia/rakudo-m-inst-1/share/perl6/sources/947BDAB9F96E0E5FCCB383124F923A6BF6F8D76B (NativeCall) line 680
in block <unit> at <tmp> line 1
04:55 kktt joined 04:56 p6bannerbot sets mode: +v kktt, molaf left 05:00 zacts left 05:22 MasterDuke left 05:24 daotoad joined 05:25 p6bannerbot sets mode: +v daotoad 05:45 robertle left 05:47 lookatme_q left, lookatme_q joined 05:48 p6bannerbot sets mode: +v lookatme_q 05:49 curan joined, p6bannerbot sets mode: +v curan
holyghost I've got an argument list with byte arrays working for my C++ perl6, it's based on Objective-C's NSData 05:51
05:54 w_richard_w joined 05:55 p6bannerbot sets mode: +v w_richard_w 06:17 daotoad left, daotoad joined 06:18 p6bannerbot sets mode: +v daotoad 06:20 jmerelo joined, p6bannerbot sets mode: +v jmerelo
jmerelo releasable6: status 06:20
yoleaux 18 Dec 2018 21:02Z <moritz> jmerelo: would you find something like gist.githubusercontent.com/moritz/...aving.pod6 helpful in the p6 docs?
releasable6 jmerelo, Next release will happen when it's ready. 0 blockers. 85 out of 85 commits logged
jmerelo .tell To tell you the truth, it might be better if you used a real example. If you interpret A, B and C as statements, it might be a bit confusing, because when using that module, A and C are going to run at runtime, B at compile time... The general idea is OK, though... 06:25
yoleaux jmerelo: I'll pass your message to To.
holyghost hi jmerelo, see above 06:27
hello btw
jmerelo hi, seeing it now 06:28
holyghost I've put some more work into my C++ perl6, it's called UnterPaganPerl6 06:29
jmerelo UnterPagan? 06:30
holyghost I'd be glad to make another implementation in full STL
from there
it's a prototype
jmerelo perl6 is mostly implemented in Perl6. The virtual machine, MoarVM, is implemented in C. 06:31
holyghost I know, I have made some classes for vm, architecture etc, it's all in its beginning stages 06:32
I want to port it to several architectures, it's more a compiler than something else 06:33
jmerelo holyghost: before you go any further, you might want to consider the fact that it took 15 years for a lot of people to take Perl6 from the beginning stages to the Christmas release.
holyghost ok
06:34 domidumont joined
holyghost I want to use the OOP 06:34
jmerelo holyghost: so you might want to pursue other venues that might take you closer to fruition.
06:34 p6bannerbot sets mode: +v domidumont
holyghost of perl6 to make a full OO implementation 06:34
sorry, 06:35
I mean I will only implment OO stuff from perl6
then people can go on from there
I know there's a standard however
06:35 lookatme_q left 06:36 lookatme_q joined
jmerelo For instance, to write documentation and examples for github.com/theholyghost2/Bayes-Learn, move the code to lib/Bayes/Learn.pm (which is the customary way of doing it), adding continuous integration to the repo, write tests for all functions... 06:36
06:37 p6bannerbot sets mode: +v lookatme_q
holyghost I'll do that in the future 06:37
jmerelo There's 007 by masak too github.com/masak/007 Extremely interesting, experimental language.
holyghost ok, I'll put it on my list 06:39
06:40 robertle joined 06:41 p6bannerbot sets mode: +v robertle 06:42 nelsonhb joined 06:43 p6bannerbot sets mode: +v nelsonhb
buggable New CPAN upload: PDF-Class-0.3.3.tar.gz by WARRINGD modules.perl6.org/dist/PDF::Class:cpan:WARRINGD 06:44
holyghost HAND jj, I'm afk for while 06:48
06:49 Cabanoss- left
jmerelo holyghost: see you! 06:49
06:50 Cabanossi joined 06:51 p6bannerbot sets mode: +v Cabanossi 06:56 eseyman left 07:01 woolfy joined, woolfy left 07:04 eseyman joined, p6bannerbot sets mode: +v eseyman 07:08 domidumont1 joined 07:09 p6bannerbot sets mode: +v domidumont1, jmerelo left 07:11 domidumont left 07:12 alixir12 joined 07:13 p6bannerbot sets mode: +v alixir12 07:15 andrzejku joined 07:16 p6bannerbot sets mode: +v andrzejku 07:17 w_richard_w left 07:18 alixir12 left 07:26 sno left 07:34 rindolf joined 07:35 p6bannerbot sets mode: +v rindolf 07:45 sno joined, p6bannerbot sets mode: +v sno 07:55 rindolf left 08:00 robertle left 08:03 nadim_ joined 08:04 p6bannerbot sets mode: +v nadim_ 08:07 rindolf joined, dct joined, p6bannerbot sets mode: +v dct 08:08 p6bannerbot sets mode: +v rindolf 08:10 dogbert17 left 08:16 domidumont1 left 08:33 robertle joined, p6bannerbot sets mode: +v robertle 08:38 nadim_ left, finanalyst left 08:40 finanalyst joined 08:41 p6bannerbot sets mode: +v finanalyst 08:43 robertle left 08:55 domidumont joined 08:56 p6bannerbot sets mode: +v domidumont 08:57 cjkinni left, cjkinni joined 08:58 p6bannerbot sets mode: +v cjkinni 08:59 kktt left 09:00 daotoad left 09:02 dakkar joined, p6bannerbot sets mode: +v dakkar 09:05 aindilis left 09:14 robertle joined, sno left, aindilis joined, p6bannerbot sets mode: +v robertle 09:15 p6bannerbot sets mode: +v aindilis 09:16 sno joined 09:17 p6bannerbot sets mode: +v sno 09:19 mowcat left 09:33 zakharyas joined 09:34 p6bannerbot sets mode: +v zakharyas 09:35 zakharyas1 joined, p6bannerbot sets mode: +v zakharyas1 09:37 zakharyas left 09:46 aindilis left 09:53 tobs` joined 09:54 p6bannerbot sets mode: +v tobs`
ufobat whenever i do something in perl6 i find something that is not competly satisfying me 10:12
jnthn I get that with pretty much every language/technology. :) 10:14
ufobat when i started with perl5 the reason why i wasnt happy was alywas "me being stupid" 10:15
10:15 averna joined
ufobat now i do PR and issues :-) 10:15
but i still feel stupid :-) 10:16
10:16 p6bannerbot sets mode: +v averna
jnthn feels like his brain already left for Christmas vacation this week... :) 10:17
ufobat 2 more days of work still left for me :-) 10:18
i am on vacation till 6th Jan!
10:18 averna left, lookatme_q left 10:19 lookatme_q joined
jnthn Don't think I'll be off until the 6th, but yeah, next week I certainly am :) 10:19
10:20 p6bannerbot sets mode: +v lookatme_q
ufobat hope you will enjoy it! 10:21
my wife and my 2 children are going home to my moms house for a few days, there will be way to less space for each of us. lets see who first gets anoyed. ;) 10:22
buggable New CPAN upload: IoC-0.0.3.tar.gz by UFOBAT modules.perl6.org/dist/IoC:cpan:UFOBAT 10:24
ufobat buggable++ :) 10:25
tbrowder jmerelo: i think one way to get p6pod parser only, per jnthnโ€™s thoughts, might be to split the pod grammar/actions into PodGrammar.nqp and PodActions.nqp (which are already there) and then another tool chain/grammar can use that common code for parsing without all of the current p6 code. 10:27
llfourn o/ #perl6! is there a way in the language to make :D the default on class attributes and/or sub params? 10:30
10:31 uvelichitel joined, uvelichitel left
jnthn m: use parameters :D; sub foo(Int $x) { }; foo(Int) 10:33
camelia 5===SORRY!5=== Error while compiling <tmp>
use parameters not yet implemented. Sorry.
at <tmp>:1
------> 3use parameters :D7โ5; sub foo(Int $x) { }; foo(Int)
jnthn No :)
But that's how it'll look once somebody implements it :) 10:34
llfourn jnthn: heh thanks. I thought I remembered this being a TODO a while ago :)
I've been writing a lot of rust recently and not knwoing whether things will be defined gives me a weird feeling in my stomach now :\ 10:35
jnthn It's one of those things that I know I could worry about in theory, but rarely bites me in practice. 10:37
llfourn nod. I think I just have to let go of that feeling and only use :D when I actually want to do a runtime check. 10:41
lizmat jnthn: about implementing use parameters :D 10:44
isn't that just a matter of having "use parameters :D" set a dynamic variable and have the parameter parsing look at that ? 10:45
jnthn: or am I missing something (less) obvious ? 10:46
10:46 domidumont left
jnthn Well, devil will be in the details 10:48
What does `Str @foo` mean under that pragma?
Geth gtk-simple: aed9855fb9 | (Jonathan Stowe)++ | 2 files
The GTK::Simple::TextView.text needs an 'is rw' to work

Fixes #87
10:49 Cabanossi left, kurahaupo joined, aindilis joined 10:50 p6bannerbot sets mode: +v kurahaupo, andrzejku_ joined, Cabanossi joined, p6bannerbot sets mode: +v aindilis, Averna joined, p6bannerbot sets mode: +v andrzejku_ 10:51 p6bannerbot sets mode: +v Cabanossi, p6bannerbot sets mode: +v Averna, Averna left 10:52 Averna joined, finanalyst left, p6bannerbot sets mode: +v Averna, andrzejku left, andrzejku_ is now known as andrzejku 10:53 finanalyst joined
lizmat jnthn: I assume sub a(Str @foo) ? I guess Positiona[Str:D] ? 10:53
*Positional
10:53 p6bannerbot sets mode: +v finanalyst
jnthn Or is it Positional:D[Str] ? 10:54
Note that since these are nominal types, the array being passed would have had to been declared as an Array[Str:D] to match if we take the Positional[Str:D]
lizmat indeed.... so I guess even more "useless" than it is nonw 10:55
*now
10:57 finanalyst left 10:58 finanalyst joined
jnthn How is it useless now? 10:58
But yes, it would maybe put a bit too much of a burden on the caller, if folks already think having to be consistent with what they pass is a burden now
10:58 p6bannerbot sets mode: +v finanalyst 10:59 Averna left, Averna joined 11:00 Averna left
lizmat m: sub a(Str @a) { dd @a }; a <a b c d> # jnthn people repeatedly think that should work 11:01
camelia Type check failed in binding to parameter '@a'; expected Positional[Str] but got List ($("a", "b", "c", "d"))
in sub a at <tmp> line 1
in block <unit> at <tmp> line 1
lizmat m: dd List ~~ Positional[Str] 11:03
camelia Bool::False
lizmat m: dd List[Str] ~~ Positional[Str]
camelia 5===SORRY!5=== Error while compiling <tmp>
List cannot be parameterized
at <tmp>:1
------> 3dd List[Str]7โ5 ~~ Positional[Str]
lizmat now, if we could parameterize List *and* <a b c d> would codegen as a List[Str], we'd be in business 11:04
jnthn Except all the ways we'd be out of business :)
lizmat well, yeah 11:05
jnthn Tests like `is-deeply $foo.bar(), <a b c d>` for example would suddenly start failing.
lizmat there's that :-) 11:06
but that problem exists even without 'use parameters :D'
so I'd not consider that a detail of this particular feature 11:07
ufobat jnthn, lizmat what do you both say for not having any specs/roast-tests regarding %?RESOURCES ? 11:10
lizmat ambivalent: feel that we should have tests for the standard CUR implementations that are provided with rakudo 11:11
ufobat my worries are not having no tests for the implementation, but more not having a defined interface and expected behaviour. 11:14
jnthn We probably should have some on %?RESOURCES; the CUR details need more careful thought/discussion, and are probably in the same camp as the MOP (e.g. we move cautiously, and don't try to spec every detail)
11:28 nadim_ joined 11:29 p6bannerbot sets mode: +v nadim_
El_Che ::::::::;;;@=;<ยดยดยดยดรกรกรกรกmร ``ร $$-- 11:30
ufobat m: ::::::::;;;@=;<ยดยดยดยดรกรกรกรกmร ``ร $$-- 11:33
camelia 5===SORRY!5=== Error while compiling <tmp>
Name component may not be null
at <tmp>:1
------> 3::7โ5::::::;;;@=;<ยดยดยดยดรกรกรกรกmร ``ร $$--
11:38 sno left 11:44 dogbert17 joined 11:45 p6bannerbot sets mode: +v dogbert17 11:48 finanalyst left 11:49 finanalyst joined, p6bannerbot sets mode: +v finanalyst
moritz after working with python "mixins" (basically just lean classes + multi inheritance) a bit, I can say without a doubt that I prefer Perl 6 roles :-) 11:51
11:57 lucasb joined, p6bannerbot sets mode: +v lucasb 12:02 finanalyst left 12:05 fluca1978 joined 12:06 p6bannerbot sets mode: +v fluca1978 12:13 zakharyas1 left 12:25 nelsonhb left 12:56 robertle left 12:59 fluca1978 left 13:28 reach_satori left 13:33 MorayJ joined 13:34 p6bannerbot sets mode: +v MorayJ 13:44 robertle joined, p6bannerbot sets mode: +v robertle 14:00 finanalyst joined, p6bannerbot sets mode: +v finanalyst 14:04 zakharyas joined, p6bannerbot sets mode: +v zakharyas 14:09 kensanata joined, p6bannerbot sets mode: +v kensanata 14:40 molaf joined, p6bannerbot sets mode: +v molaf 14:58 molaf left, nadim_ left 15:11 curan left, molaf joined 15:12 p6bannerbot sets mode: +v molaf
buggable New CPAN upload: Tomtit-0.0.6.tar.gz by MELEZHIK modules.perl6.org/dist/Tomtit:cpan:MELEZHIK 15:14
15:23 finanalyst left 15:26 sena_kun joined 15:27 melezhik joined, p6bannerbot sets mode: +v melezhik, p6bannerbot sets mode: +v sena_kun 15:28 domidumont joined 15:29 p6bannerbot sets mode: +v domidumont 15:35 Sgeo_ joined 15:36 p6bannerbot sets mode: +v Sgeo_ 15:38 Sgeo left 15:43 zakharyas left 15:44 pyrimidine joined 15:45 p6bannerbot sets mode: +v pyrimidine 15:46 hythm_ joined, p6bannerbot sets mode: +v hythm_
hythm_ p6: my @A; my $b; class C { has @.a; multi method new (:@A) { say 'new1'; self.bless(:@A) }; multi method new (:$b) { say 'new2';self.bless(a => @$b) } }; C.new: :$b 15:47
camelia new1
hythm_ Should not this dispatch to the second new method?
timotimo named arguments only work as tie-breakers
and since named arguments are optional by default, either one of them is fine 15:48
15:48 lizmat left
timotimo so it'll just take the first one that's defined 15:48
hythm_ Aah, makes sense, so just mark them as requured ir use pisitional args. Right? 15:49
Sorry for typo, (using phone)
melezhik how do I read a single line from stdout ? lines()?
sena_kun last?
but, get is better, I think 15:50
hythm_ s/requured/required/
sena_kun also prompt. :)
m: say prompt('Ehehe? ');
camelia Ehehe? ยปWann treffen wir drei wieder zusamm?ยซ
sena_kun m: my $a = get; say $a; 15:51
camelia ยปWann treffen wir drei wieder zusamm?ยซ
sena_kun also:
m: for lines() -> $a { say $_; last; }
camelia (Any)
melezhik get() works fine, thanks!
sena_kun m: for lines() { .say; last; } 15:52
camelia ยปWann treffen wir drei wieder zusamm?ยซ
melezhik prompt is good too! 15:53
sena_kun :)
16:10 rindolf left 16:11 rindolf joined 16:12 p6bannerbot sets mode: +v rindolf, hythm_ left 16:13 jmerelo joined 16:14 p6bannerbot sets mode: +v jmerelo
jmerelo releasable6: status 16:21
releasable6 jmerelo, Next release will happen when it's ready. 0 blockers. 85 out of 87 commits logged
jmerelo, Details: gist.github.com/466dc6dd3eb9ce5901...0a460bdb81
16:22 MorayJ left 16:28 sno joined 16:29 p6bannerbot sets mode: +v sno
dakkar how bad an idea is it, to use nqp::serialize to store data to disk? 16:40
I have a (rather large) data structure, that I keep in memory for speed 16:41
I'd like it to be kept on disk as well, so I don't have to regenerate it at the next startup of the program
Xliff_ dakkar: Is it in a module?
timotimo serialization also has dependencies and such
16:42 Xliff_ is now known as Xliff
dakkar I've seen github.com/FROGGS/p6-Ser/blob/master/lib/Ser.pm 16:42
timotimo if you generate the data at compile time with BEGIN, it'll land in the precomp file if it's a module
dakkar it's runtime data, though
Xliff Yeah. ^^ That's where I was going.
dakkar www.thenautilus.net/cgit/MaildirIn...m6?h=bayes โ† this thing 16:43
it's an index of a set of mailboxes
with ~250k messages to scan, I'd like to avoid rebuilding it each time 16:44
timotimo what is it, not regenerate every startup, or must be regenerated all the time? :)
dakkar โ˜บ
I'll add a "this file was last seen at $time", so I can do a quick scan via stat and update it once it's been loaded 16:45
most email messages don't move around while the program is not running 16:46
or, if you prefer, updating the structure is quick, building it from scratch is slow
let me start againโ€ฆ 16:47
I have this data structure I want to store on disk 16:48
the serialisation format is not particularly important, as long as it's quick and not too large
nqp::serialize seems a sensible choice to me, but I don't know enough to be sure
opinions? 16:49
jnthn dakkar: Are you already aware that anything you serialize that way will be invalidated when you, for example, upgrade your Perl 6 version?
s/version/implementation/ 16:50
dakkar "implementation" as in, switch from Moar to JVM?
jnthn No
Just install a new version of Rakudo
Oh, but that'll do it also :)
timotimo i'd recommend going with JSON::Fast or MessagePack (whichever one is used in moar's remote debugger library)
jnthn I'd avoid using nqp::serialize, though. It's built specifically to help support module precompilation. 16:53
dakkar jnthn: interesting. I had mis-estimated the stability of the serialisation
so, the format is stable, but the representations of the objects I store may not be
timotimo: I'll have to see how slow those are โ˜บ 16:54
jnthn Well, you'll end up with cross-references to types in CORE.setting
And because nothing in Perl 6 has an absolute name, there's no way to write any kind of dynamic linking 16:55
dakkar ouch
jnthn So everything just references by index in another table.
Which, given what we need for precomp, is totally fine.
dakkar so any change in core invalidates all serialised objects
jnthn Yes. 16:56
And CORE is linked against other bits of the compiler, so any change there too :)
Xliff dakkar: If everything can be stored as records, might be easiest just to build a set of repr('CStructs') and use C to write/read from disk.
jnthn Precomp is just cache, so none of this is a problem: we can regenerate them whenever we need. But that's not so useful for serialization in general :) 16:57
dakkar Xliff: I'm not *that* desperate for speed (yet) ๐Ÿ˜œ
jnthn: (tangential) is there any risk, when deserialising something built from a different rakudo, to get out different objects, without realising?
Xliff dakkar: Not talking speed, really. Thinking more about stability. 16:58
If JSON won't work for you, something like that is probably the next way to go, unless you come up with your own serialization format.
Perl5 had Storable, but that was XS-backed. 16:59
dakkar hmm. do I want to write a Perl6 Serial ?โ€ฆ 17:00
Sereal, even
Xliff Well, if you have idea's. I'll throw my hat into the ring to help. ;)
jnthn dakkar: No, or at least, you'd have to be incredibly unlucky 17:02
(Like, SHA-1 collision unlucky)
dakkar jnthn: good enough ๐Ÿ˜œ
Xliff My eyes are definitely going bad. I first thought that smiley had fangs. 17:03
dakkar get a better font ๐Ÿฆ‡ 17:08
Xliff Now that's just being cheeky.
:)
dakkar emojipedia.org/vampire/ 17:09
and with that, I'll disappear (last workday for two weeks!) 17:10
as someone said, have an appropriate amount of fun
17:11 dakkar left 17:13 reach_satori joined 17:14 p6bannerbot sets mode: +v reach_satori
jmerelo Did you check out today's advent calendar entry on interactive desktop apps? perl6advent.wordpress.com/2018/12/...ktop-apps/ 17:19
.seen tobs 17:20
yoleaux I saw tobs 18 Dec 2018 17:52Z in #perl6: <tobs> beats me
Xliff .seen scimon 17:23
yoleaux I saw scimon 12 Dec 2018 16:38Z in #perl6: <scimon> Neat.
17:26 kensanata left
tobs` jmerelo: the post is in the works, will be ready in due time. 17:28
jmerelo tobs`: great :-) Checking, just in case :-)
m: say $*PID, $*PID.perl 17:31
camelia 2857228572
Altreus anything in p6 or on the roadmap like decorators in other languages? A syntactic construct that wraps a function around another function or method? 17:39
tadzik maybe traits are what you're looking for? 17:40
Altreus I was thinking, maybe I could adorn all my methods that return promises, to also create another method that awaits the original one
tadzik sub foo is route('/bar') { } like in Flask
Altreus hey I didn't actually think of that
sub foo is Promising { } 17:41
jmerelo Altreus: You can just use functions as any other variable, no special syntax to wrap a function around another. You just do it.
Altreus can you give an example before I say something silly? :)
jmerelo Altreus: there's also this docs.perl6.org/routine/%E2%88%98 for function composition 17:42
Altreus I'm just thinking that in python you do @somefunction\ndef anotherfunction: ...
jmerelo Altreus: which is not nice at all. 17:43
Altreus ah, antipattern is it
I'm also a fan of answers like this :)
tadzik every time I teach this I start by explaining that it's just anotherfunction = somefunction(anotherfunction)
jmerelo you can use f โˆ˜ g or simply call one from the other
Altreus can I do that to define a method? sorry if I'm being thick 17:44
I'm just trying to find the nicest way of reducing boilerplate
I'd like to have counterpart methods to everything that returns a Promise, to provide versions that await it
jmerelo m: sub foo ( $whatever ) { $whatever*2 }; sub bar ($whatever) { $whatever / 2 }; my $id = foo โˆ˜ bar; say $id(333)
camelia 5===SORRY!5=== Error while compiling <tmp>
Preceding context expects a term, but found infix โˆ˜ instead.
at <tmp>:1
------> 3tever) { $whatever / 2 }; my $id = foo โˆ˜7โ5 bar; say $id(333)
tadzik hmm
Altreus: I'd go for a metaclass that dupes all the methods that return a Promise 17:45
Altreus that sounds scary but learnable
jmerelo m: sub foo ( $whatever ) { $whatever*2 }; sub bar ($whatever) { $whatever / 2 }; my $id = &foo โˆ˜ &bar; say $id(333)
camelia 333
Altreus I played with metaclasses a little bit when I read the code for one of the orms
tadzik so, at class composition time: inspect every method you have, if it returns a promise, plug in a new one that awaits the old one
. o O ( can you have metaroles? )
Geth doc: 05cd7b6efa | (JJ Merelo)++ | doc/Language/variables.pod6
Improving the page by (slight) rewriting
 17:46
synopsebot Link: doc.perl6.org/language/variables
17:51 captain-adequate joined 17:52 p6bannerbot sets mode: +v captain-adequate 18:01 jmerelo left 18:02 melezhik left 18:09 oftl left
SmokeMachine m: multi trait_mod:<is>(Routine $r, :$promising) { my &orig := $r.clone; $r.wrap: -> |c { start { orig |c } } }; sub bla is promising { sleep 1; 42 }; my $a = bla; say $a; say await $a 18:10
camelia Promise.new(scheduler => ThreadPoolScheduler.new(initial_threads => 0, max_threads => 64, uncaught_handler => Callable), status => PromiseStatus::Planned)
42
18:13 ponbiki left 18:22 zakharyas joined 18:23 p6bannerbot sets mode: +v zakharyas 18:26 lucasb left 18:33 sauvin left 18:34 nadim_ joined
SmokeMachine Altreus: ^^ 18:35
18:35 p6bannerbot sets mode: +v nadim_ 18:40 sno left 18:42 sno joined 18:43 p6bannerbot sets mode: +v sno, markjreed joined 18:44 sno left, p6bannerbot sets mode: +v markjreed 19:10 zakharyas left 19:11 Scimon joined 19:12 p6bannerbot sets mode: +v Scimon
Scimon Hi 19:12
19:16 nadim_ left, Scimon left
SmokeMachine hi! 19:17
19:19 sno joined, oftl joined
Kaiepi i ordered the wrong type of ram bro my machine... 19:19
there is'tn and emoji that escpresses my rage
tadzik I feel your pain
19:19 p6bannerbot sets mode: +v sno
tadzik I fought my new ram sticks for like a week 19:19
19:20 p6bannerbot sets mode: +v oftl
tadzik I ended up downclocking them to 2100, otherwise they randomly reboot my system 19:20
Kaiepi i think i can gvie it as a preent
19:23 lizmat joined, p6bannerbot sets mode: +v lizmat 19:24 nadim joined 19:25 p6bannerbot sets mode: +v nadim 19:36 markjreed left 19:42 markjreed joined 19:43 p6bannerbot sets mode: +v markjreed 19:44 Sgeo__ joined
Geth doc: 6bee0a45cc | Coke++ | doc/Language/variables.pod6
whitespace
 19:45
synopsebot Link: doc.perl6.org/language/variables
19:45 p6bannerbot sets mode: +v Sgeo__ 19:48 Sgeo_ left 19:58 ponbiki joined, p6bannerbot sets mode: +v ponbiki 20:02 zakharyas joined 20:03 p6bannerbot sets mode: +v zakharyas
tony-o m: sub r { say "x"; }; &r.wrap({ say "wrap"; callsame; }); r; 20:10
camelia wrap
x
20:12 markjreed left
Xliff m: sub r { say "x"; }; &r.wrap({ say "wrap"; callsame; }); 20:16
camelia ( no output )
Xliff tony-o: Where is wrap documented? 20:17
SmokeMachine m: multi trait_mod:<is>(Routine $r, :$promising) { my &orig := $r.clone; $r.wrap: -> |c { Promise.start: &orig.assuming: |c } }; sub bla($int) is promising { sleep $int div 10; $int }; my $a = bla 42; say $a; say await $a
camelia Promise.new(scheduler => ThreadPoolScheduler.new(initial_threads => 0, max_threads => 64, uncaught_handler => Callable), status => PromiseStatus::Planned)
42		 20:18
tony-o Xliff: docs.perl6.org/type/Routine#method_wrap
Xliff Thanks.
SmokeMachine docs.perl6.org/routine/wrap
Kaiepi i can't wait until my grant progoposal getps approved or denied 20:27
lizmat Kaiepi: don't expect that to happen this year... 20:34
20:34 Actualeyes joined
jnthn Kaiepi: I've been looking over it so I can give feedback. One thing that's not clear: will IO::Socket::Async get all of the features (UNIX sockets, raw sockets, sending descriptors, etc.) that are proposed? Also, where do you see breaking changes being needed? 20:35
20:35 p6bannerbot sets mode: +v Actualeyes
lizmat Kaiepi: voting doesn't start until after the 22nd... and that typically takes weeks... 20:35
20:40 domidumont left
SmokeMachine Kaiepi: have you added unix sockets to your proposal? 20:49
Kaiepi yes 20:53
the breaking chganges in IO::SOcket::Async has to do paricularlarly with .listen 20:55
yes SmokeMachine
Xliff m: my $c = CArray[uint8].new(0 xx 10000); say now - INIT.now
camelia 5===SORRY!5=== Error while compiling <tmp>
Whitespace required after keyword 'INIT'
at <tmp>:1
------> 3y[uint8].new(0 xx 10000); say now - INIT7โ5.now
Xliff m: my $c = CArray[uint8].new(0 xx 10000); say now - INIT now
camelia 5===SORRY!5=== Error while compiling <tmp>
Undeclared name:
CArray used at line 1. Did you mean 'Array', 'array'?
Xliff m: use NativeCall; my $c = CArray[uint8].new(0 xx 10000); say now - INIT now 20:56
camelia 0.0387783
Xliff m: use NativeCall; my $c = CArray[uint8].new(0 xx 600000); say now - INIT now
camelia 1.112439
Xliff Is there a faster way to initialize a CArray?
timotimo Kaiepi: does it have to break? wouldn't it be possible to differentiate between .new.listen and .listen?
Xliff m: use NativeCall; my $c = CArray[uint8].new(0 xx 6000000); say now - INIT now
camelia 6.44441923 20:57
Xliff m: (640*480).say 20:58
camelia 307200
Kaiepi m: m: say awaait IO::SocketLLNET.connect: ':host<google.com>, :80port 21:01
camelia 5===SORRY!5=== Error while compiling <tmp>
Unable to parse expression in single quotes; couldn't find final "'" (corresponding starter was at line 1)
at <tmp>:1
------> 3NET.connect: ':host<google.com>, :80port7โ5<EOL>
expecting โ€ฆ
21:02 i1nfusion joined, p6bannerbot sets mode: +v i1nfusion
tobs` Advent #20 is 95% finished. Now I just need 30 small illustrations and maybe something to eat :') 21:04
Kaiepi imt not wusre timotimo. which is part so why i wanna disccuss hous best to impolememnet it 21:06
tbrowder hi, #perl6 21:08
21:09 dct left 21:11 reach_satori left
Kaiepi timotimo, that's not the probme 21:11
tbrowder iโ€™m starting to port a p5 module (Date::Calc) and want to (1) preserve the original sub names and signatures as much as possible and (2) offer an alternative set of names and better signatures. 21:12
21:13 cbk joined, p6bannerbot sets mode: +v cbk
cbk what version of rakudo will support perl6.d 21:13
lizmat 2018.11 and higher default to it 21:14
is that the answer to your question ?
earlier versions supported it with "use 6.d.PREVIEW" 21:15
21:15 zakharyas left
tbrowder question: as a first cut, is there a better way to offer a different sub name other than, say, โ€œsub new-name($arg) { Old_Name($arg); }โ€? 21:15
cbk well i think it does. now can i just remove my .rakudobrew dir and install 2018.11?
lizmat cbk: probably, but I'm no rakudobrew expert 21:16
tbrowder: if they have the same interface, you can:
cbk lizmat, ok. Thank you. 21:17
lizmat m: constant &MAXIMAX = &max; dd MAXIMAX 42,666 # tbrowder
camelia 666
tbrowder lizmat: is there a special place for seeing butterfly modules and progress? 21:19
lizmat modules.perl6.org/t/CPAN5 21:20
Geth perl6.org: 033ba3eb27 | (Tom Browder)++ (committed using GitHub Web editor) | source/resources/index.html
update bok status
 21:25
tbrowder never could type... 21:26
lizmat: thnx, i always forget that, maybe a big butterfly image and link would help 21:29
21:30 robertle left 21:39 cbk left 21:44 reach_satori joined 21:45 p6bannerbot sets mode: +v reach_satori 21:50 Possum left 21:52 andrzejku_ joined 21:53 p6bannerbot sets mode: +v andrzejku_, Sgeo__ left 21:55 andrzejku left, andrzejku_ is now known as andrzejku 22:12 lucasb joined, p6bannerbot sets mode: +v lucasb, daxim left 22:23 tobs` left 22:25 Kaiepi is now known as Highpie 22:27 sena_kun left 22:34 kurahaupo left 22:35 Praise left 22:37 Praise joined, Praise left, Praise joined, p6bannerbot sets mode: +v Praise 22:39 kurahaupo joined
AlexDaniel u: ๐‘’e 22:39
unicodable6 AlexDaniel, U+1D452 MATHEMATICAL ITALIC SMALL E [Ll] (๐‘’)
AlexDaniel, U+0065 LATIN SMALL LETTER E [Ll] (e)
22:40 p6bannerbot sets mode: +v kurahaupo, kybr left 22:41 tobs` joined 22:42 p6bannerbot sets mode: +v tobs`
lucasb eeeeeeeeeeeeeeeeeeee 22:51
that github repo drama :) 22:52
23:02 Possum joined, p6bannerbot sets mode: +v Possum 23:03 rindolf left 23:09 nadim left
AlexDaniel lucasb: what drama? 23:11
AlexDaniel can't see any drama
lucasb AlexDaniel: I was refering to another thing :) linuxwit.ch/blog/2018/12/e98e/ 23:14
23:15 Highpie is now known as Highpi
lucasb maybe "drama" wasn't the best word. Anyway, just a confusion in GitHub. :) 23:15
github.com/eeeeeeeeeeeeeeeeeeeeeee...eeeeeeeeee
the repository is archived and we lost the opportunity of putting a Perl 6 version there :D 23:16
tobs Is today's article 20 minutes overdue or 40 minutes underdue? I don't know if I should "Publish immediately" or not. 23:18
AlexDaniel tobs: it's 23:27 UTC 23:27
tobs phew :)
that gives me time to fix all the instances of "<>" that were swallowed by the converter 23:28
AlexDaniel lucasb: giving users the ability to create long repo names and then falling apart when a repo with a long name is crated. What a shitshow
created* 23:29
what's the longest module possible module name in our ecosystem? Should I create module Foo::๐‘’๐‘’๐‘’๐‘’๐‘’๐‘’๐‘’๐‘’๐‘’๐‘’๐‘’ ?
s/module// 23:30
23:31 scovit joined, p6bannerbot sets mode: +v scovit
AlexDaniel I don't feel like doing it today, but it will be a nice addition to the collection: modules.perl6.org/search/?q=foo%3A%3A 23:31
scovit Hi, I am playing with the MOP, I am doing something like ClassObject.^add_method("a", -> $x { $x.say }); 23:32
Does anybody knows how to add a multi? 23:33
Juerd add_multi_method :) 23:35
AlexDaniel m: multi foo(Str, Int $x) { say โ€˜int multiโ€™ }; multi foo(Str, Str $x) { say โ€˜str multiโ€™ }; Str.^add_method("a", &foo); Str.a(โ€˜42โ€™); Str.a(โ€˜fooโ€™)
camelia str multi
str multi
AlexDaniel m: multi foo(Str, Int $x) { say โ€˜int multiโ€™ }; multi foo(Str, Str $x) { say โ€˜str multiโ€™ }; Str.^add_method("a", &foo); Str.a(42); Str.a(โ€˜fooโ€™)
camelia int multi
str multi
scovit Ok, thanks 23:36
add_multi_method would have been nice and easy to find ;) 23:37
AlexDaniel add_multi_method exists but I'm not sure how to get it work
jnthn Same as add_method 23:46
buggable New CPAN upload: PDF-ISO_32000-0.0.5.tar.gz by WARRINGD cpan.metacpan.org/authors/id/W/WA/...0.5.tar.gz 23:47
jnthn Oh, but if you're not keeping the MOP contract and calling add_multi_method without a compose call afterwards, it'll have no effect. :)
23:49 Cabanossi left 23:50 Cabanossi joined 23:51 p6bannerbot sets mode: +v Cabanossi
timotimo tobs: i'm a bit late, but it's not the converter that swallows the <>, it's the wordpress editor when you paste in html 23:56
it thinks you're trying to put html tags in, notices they're invalid, and just removes them ... or something like that?
tobs timotimo: yeah, I found that out meanwhile. I'm about to complete my proofreading and then I'm going to add all of them again once and for all. 23:57
The editor goes "&gt;" โ†’ ">" โ†’ "" when you review the article multiple times. 23:58

Powered by Cro and the Rakuยฎ Programming Language.

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

Thank you!