»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.
melezhik2 Bird - lightweight alternative to chef inspec and goss, written in Perl6. Just started to play with it. github.com/melezhik/bird 00:57
holyghost Hi 01:31
comborico1611 Does anyone know of a video discussion of using perl6 as a shell alernative? 01:37
holyghost It'd also be nice if perl6 was ported to PSP or PS linux. I'll look into that later on 01:47
discord6 <Tyler (Aearnus)> ps linux as in for the ps3? 01:48
holyghost yes
holyghost Also SGI mips would love to have a perl6 for that 01:50
holyghost I am going to get an Octane machine for that one, there's no video driver for the gfx cards last time I checked in linux (2004) 01:53
SGI made its own linux kernel for it
discord6 <Tyler (Aearnus)> that'd be pretty interesting. i know the moarvm folks have been really good about platform portability so it shouldn't be too hard 01:56
holyghost indeed, portable C is more or less welcome 01:57
holyghost It'll be for Christmas by my account 02:04
discord6 <Tyler (Aearnus)> hahaha. well what version of gcc runs on ps Linux? 02:07
holyghost I don't own a PS3 anymore, I gave mine to my children. gcc 4 minimum I think 02:08
discord6 <Tyler (Aearnus)> Hmm
holyghost you can probably compile one
or crosscompile
discord6 <Tyler (Aearnus)> Unrelated, but I wonder how easy porting to the Nintendo switch would be. They've already ported Ruby to it, so... 02:09
holyghost cool, I hear it's a good system
I'm a bit of a i386, powerpc and mips dude 02:11
holyghost I don't know much japanese hardware except maybe for the ARM 02:11
discord6 <Tyler (Aearnus)> it's an Nvidia tegra with ARM 02:12
holyghost which is good 02:13
holyghost We can use perl6 on ps and switch to develop games of course 02:17
holyghost they're more or less like full capable systems with ram, video ram and disk space 02:18
else you need a latch hex in perl6 (from C) then calibrate and you're done 02:19
discord6 <Tyler (Aearnus)> that was the big thing behind www.rubymotion.com/ on console 02:21
<Tyler (Aearnus)> and it works well (it's what I was mentioning earlier) so it's a really good idea
Elronnd I would probably use perl6 for games if it were possible to embed 02:38
I want to use something that has native compilation and static typing for the main loop/startup/all that, for performance 02:39
cpan-p6 New module released to CPAN! Async::Workers (0.0.6) by 03VRURG 03:15
antoniogamiz noisegul_: what did you need? 07:13
noisegul_: I will make the release with the fixed versions in a moment 07:14
antoniogamiz mmm $! cannot be used to access one attribute from a parent class¿ 07:51
xiaomiao hrm. are there proper nqp tarballs, or do I need to cook my own now? (missing 3rdparty/ bits, and configure calls 'git' for no reason) 08:06
antoniogamiz I think I'm getting how oop works in perl6... => gist.github.com/antoniogamiz/3b9fa...0b0506a307 08:07
Geth doc/master: 4 commits pushed by (JJ Merelo)++ 08:18
AlexDani` xiaomiao: yes, here github.com/perl6/nqp/releases 08:35
AlexDaniel xiaomiao: as for “proper”, yes there's an issue in configure 08:35
xiaomiao: we might have 2019.07.1 with that fixed 08:36
AlexDaniel xiaomiao: it should actually build fine, even though there are some errors about `git`. Or are you seeing some other errors? 08:36
antoniogamiz is there a way to execute some code after an instance of a type object is created? 08:38
I mean, after bless or nextwith for instance
xiaomiao AlexDaniel: it would build, if it had network access and git :) 08:39
both assumptions that I find highly optimistic
AlexDaniel xiaomiao: can you show the log?
xiaomiao AlexDaniel: Can't exec "git": No such file or directory at Configure.pl line 15. 08:40
I can sed things into compliance, but that's not the best way to package things
AlexDaniel xiaomiao: and after that? 08:41
AlexDaniel xiaomiao: I'm asking because even though it complains about not being able to exec git, it should still continue 08:42
at least that's my understanding
but maybe I'm wrong 08:43
xiaomiao the tarball at github.com/perl6/nqp/releases/down....07.tar.gz is missing 3rdparty/nqp-configure 08:46
so of course it would try to quietly fetch that 08:47
AlexDaniel xiaomiao: to me it looks like it is there
discord6 <Tyler (Aearnus)> how do I create a zef compatible package? or do I just have to create Build.pm manually 08:48
AlexDaniel xiaomiao: I think you clicked on the wrong one :)
xiaomiao AlexDaniel: so either github is tricking me, or you're not looking at the same tarball
AlexDaniel xiaomiao: try the link you just sent here 08:49
xiaomiao yes, that's the one that fails
AlexDaniel but it has 3rdparty/nqp-configure with files in it
xiaomiao now I am properly confused :D 08:51
and now it doesn't abort and just spews 'harmless' warnings 08:52
well ok
your telekinesis skills impress me! ;) 08:53
AlexDaniel :)
xiaomiao: unfortunately there's no way to disable autogenerated source tars on github :(
so yes, people will sometimes get the wrong one 08:54
xiaomiao yes and github still can't into tarball, so ...
antoniogamiz <Tyler (Aearnus)>: you should create a META6.json file
xiaomiao quite funny how they haven't found the time to fix that bug in years
antoniogamiz <Tyler (Aearnus)>: see docs.perl6.org/language/modules
AlexDaniel it's fine in most cases because people have internet connection and git
discord6 <Tyler (Aearnus)> cool, got that META6.json file all created 08:56
xiaomiao AlexDaniel: unless you're one of them silly distro packagers who insist on weird constraints 08:59
AlexDaniel xiaomiao: in that case just make sure to get the right tar :)
we will fix these annoying warnings anyway
xiaomiao thanks 09:00
AlexDaniel xiaomiao: see github.com/rakudo/rakudo/issues/3067 for what will be in the point release
xiaomiao ... and now rakudo fails, it's a nice brain teaser :)
AlexDaniel xiaomiao: let me guess, something about /usr ? 09:01
xiaomiao /usr/include/moar/moar.h:27:10: fatal error: ffi.h: No such file or directory
xiaomiao that looks like incomplete include path to me :) 09:01
rfold When I run "make t/spec/S12-traits/parameterized.t" it says: "make: Nothing to be done for 't/spec/S12-traits/parameterized.t'." I did clone into t/spec and I installed Rakudo as per README.
"make spectest" does work. 09:03
AlexDaniel xiaomiao: hmmm, are you sure you're using the right tarball? :) 09:04
xiaomiao yes
AlexDaniel xiaomiao: I see github.com/MoarVM/MoarVM/issues/320
xiaomiao but libffi might not be in the expected place
AlexDaniel xiaomiao: weird 09:07
xiaomiao well, there might be more than one version present, so ...
discord6 <Tyler (Aearnus)> Is there any nice syntax to define dynamically scoped subroutines? 09:08
<Tyler (Aearnus)> my &*g = sub ($x) { $x.say } doesn't seem to work right
rfold I got it working by adding t/spec/*/*.t to the list of globs in Makefile.
AlexDaniel rfold: interesting, what's your system? 09:09
could it be some sort of difference in make or something?
rfold Oh I see what's wrong, t/spec is a symlink for some reason.
AlexDaniel hmm
rfold Must have done something wrong with the setup. 09:10
If I put the spec repo directly in there then it works.
tobs m: sub f { my &*g = sub ($x) { say "g: ", $x }; h $^x }; sub h { &*g.($^x) }; f 42 09:24
camelia g: 42
tobs <Tyler (Aearnus)>: it seems to work ^ 09:25
rfold p6: sub f(--> 2) { my &g = { 1 }; say g }; f 09:43
camelia 1
elcaro rfold: f still returns 2, but you're not doing anything with the return 11:44
m: sub f(--> 2) { my &g = { 1 }; say g }; say f 11:45
camelia 1
2
cpan-p6 New module released to CPAN! Gnome::GObject (0.13.11) by 03MARTIMM 12:44
Kaypie is it possible to create an IO::Handle instance with a given fd? 14:07
timotimo worst case - and not portably - you can open /proc/self/fd/$n 14:14
but you surely already knew that
Kaypie yeah, but i like being able to run my own code on my machine 14:16
there's a way to get the path with fcntl on mac it looks like, but that's not portable either 14:17
Kaiepi wait i can just use ptsname for what i'm doing to get the filename i want 14:18
uzl .tell jmerelo From the SO question about iteration, this is the example using the Iterator role (sprunge.us/Mz6GCl) but I'm still unable to iterate over the Scalar object. Am I doing something wrong? Thanks. 14:40
yoleaux uzl: I'll pass your message to jmerelo.
jnthn uzl: The point of a Scalar container is to make things a single item 14:49
*act as
So `.say for $obj` would be expected to do a single iteration whatever the $obj Scalar holds; it's the Scalar that determines that, not the contents of it 14:50
uzl jnthn: I'd like to say I knew that already ;-)! Thanks for the clarification. 14:54
So regardless of implementing the class with `Iterator`, assigning to Scalar means that I cannot iterate over the object instance (e.g., it's a single item)? 14:55
BTW, the code was in response to JJ's answer (stackoverflow.com/a/57122642/10824322) on SO. 14:56
jnthn uzl: Yes 14:57
Similarly, it means it won't flatten away in other contexts
uzl jnthn: Sorry for being too pedantic. Wouldn't that mean that the only way to get the iterable items from a class implementing those roles is during object construction? What if I want to keep the class around? 15:01
uzl I cannot, for example, create my object and then assign it to a Positional if I want its iterable items when I need them? 15:02
jnthn You can bind it to a Positional instead (my @a := YourClass.new), or just do `for |$obj` at the point you iterate it, or use a sigilless variable 15:05
Or put a method on the class that returns something iterable, rather than making it directly iterable 15:06
Exactly which is best depends on the problem at hand
cpan-p6 New module released to CPAN! Gnome::Gtk3 (0.17.2) by 03MARTIMM 15:25
rfold p6: my $a = [1, 2, 3]; for $a.self { .say } 15:44
camelia 1
2
3
rfold p6: my $a := [1, 2, 3]; for $a { .say } 15:45
camelia 1
2
3
timotimo p6: my $a := [1, 2, 3]; for $a<> { .say } 15:47
camelia 1
2
3
timotimo p6: my $a = [1, 2, 3]; for $a<> { .say }
camelia 1
2
3
timotimo that's the one i meant
people are going to space today, it seems 15:48
rfold Can I run spectest with fewer threads? 16:40
rfold Ah, TEST_JOBS=n environment variable. 16:43
Kaiepi currently working on a library that abstracts the annoying parts of using PTYs away from the user 18:37
antoniogamiz pty?
Kaiepi pseudo-terminal
antoniogamiz ah
Kaiepi it's a posix thing
antoniogamiz good luck with that then :D 18:37
Kaiepi nothing wrong with the custom thing i have for vim-easy-align for traits at all :^) media.discordapp.net/attachments/4..._scrot.png 18:43
cpan-p6 New module released to CPAN! App::MoarVM::ConfprogCompiler (0.0.8) by 03TIMOTIMO 18:54
Kaiepi ...turns out i can't just subclass IO::Handle at all for this 19:36
or at least i can, just i can't abstract out the parts where you deal with the child process' stdin/stdout/stderr because IO::Handle doesn't support creating new instances using fds 19:37
fuck 19:38
Geth ¦ problem-solving: AlexDaniel assigned to jnthn Issue Should IO::Handle instances be possible to construct using file descriptors? github.com/perl6/problem-solving/issues/73 20:57
discord6 <Tyler (Aearnus)> hey, so I have a module Tortoiseshell in Tortoiseshell.pm6. next to it there's a native folder, and in it a file GLFW.pm6 declaring Tortoiseshell::Native::GLFW. how do I use that library from Tortoiseshell.pm6? 21:36
Kaiepi is design.perl6.org/S03.html down? 21:49
sena_kun github.com/perl6/perl6.org/issues/131 21:50
noisegul Which umbrella term does Perl6 terminology use for methods/subroutines/routines. According to docs.perl6.org/language/functions it's 'functions', or is it 'routines'? 22:32
jnthn noisegul: "routine" is the general term, and the type structure even matches that :) 22:44
m: say Sub.^mro; say Method.^mro;
camelia ((Sub) (Routine) (Block) (Code) (Any) (Mu))
((Method) (Routine) (Block) (Code) (Any) (Mu))
noisegul jnthn: Alright, thank you for the clarification :) 22:50
Xliff What's the difference between a Block and a Routine? 23:02
sena_kun >A Routine is a code object meant for larger unities of code than Block. 23:07
sena_kun >The routine level is also the one at which multiness (multi subs and multi methods) are handled 23:07
Xliff jnthn: Any luck with IO::Socket::Async::SSL? 23:20
jnthn Xliff: Not had chance to look yet, and I'm a bit behind with PerlCon prep compared to where I'd like to be (and the "punishment" for not getting that done in the next week is that I have to do it during my vacation :)) 23:59