»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.
vrurg .tell SmokeMachine pulp is interesting, but aren't you trying to re-implement Sparrow6? 01:50
yoleaux vrurg: I'll pass your message to SmokeMachine.
vrurg .tell SmokeMachine and I'm sorry, but still can't find time to test your Red fixes. :( 01:51
yoleaux vrurg: I'll pass your message to SmokeMachine.
antoniogamiz morning o/ 07:24
how frequently is the documentation rebuilt?
Geth doc: antoniogamiz self-assigned Avoid whitespace at the beginning of an indexed word github.com/perl6/doc/issues/2801
antoniogamiz self-assigned Immediately view small changes in browser without rebuilding all pages github.com/perl6/doc/issues/2668
 07:35
SmokeMachine . 08:09
yoleaux 01:50Z <vrurg> SmokeMachine: pulp is interesting, but aren't you trying to re-implement Sparrow6?
01:51Z <vrurg> SmokeMachine: and I'm sorry, but still can't find time to test your Red fixes. :(
SmokeMachine .tell vrurg isn't Sparrow6 a replacement for puppet, ansible and stuff like that? pulp is much simpler! and about Red, no problem... take your time! 08:12
yoleaux SmokeMachine: I'll pass your message to vrurg.
SmokeMachine .tell vrurg thanks! 08:15
yoleaux SmokeMachine: I'll pass your message to vrurg.
El_Che Looking at our gitea instance at work, I realize I starting doing something without realizing it. This is the shebang I have been using: #!/usr/bin/env rakudo 08:46
moritz I've been using env perl6 for a while 08:47
El_Che I used to have that until some month ago
s
It seems I moved mentally from the Perl 6 name and Raku never could replace Rakudo in my head :) 08:48
antoniogamiz m: my $output = run("perl6", "-e", "say 'hello';", :out).out.slurp-rest;say $output; 09:03
camelia hello
antoniogamiz why the output is not stored in $output? mmm 09:04
I have also tried docs.perl6.org/routine/stdout but I dont get to save the output in a variable
antoniogamiz ah I have found the solution => docs.perl6.org/type/Proc 09:08
El_Che See, you just needed time 09:15
:)
antoniogamiz :D 09:16
Geth doc: b2fda75fc1 | Chloé++ | doc/Language/operators.pod6
Clarify infix Z behavior

  - What if one list is shorter? Resolve ambiguity.
   - What does the returned Seq contain exactly?		 09:17
synopsebot Link: doc.perl6.org/language/operators
doc: e71d63b1e6 | (Juan Julián Merelo Guervós)++ (committed using GitHub Web editor) | doc/Language/operators.pod6
Merge pull request #2902 from chloekek/infix-z-clarify

Clarify infix Z behavior
antoniogamiz mmm it looks like documentation is not being rebuilt 09:40
I made a change yesterday and it hasn't appear yet 09:41
El_Che it used to be a cron job every 10 minutes or so
.tell jjmerelo 11:40 < antoniogamiz> mmm it looks like documentation is not being rebuilt 09:42
yoleaux El_Che: I'll pass your message to jjmerelo.
rfold The documentation online seems to be 10 days old. 09:51
lizmat weekly: blogs.perl.org/users/athanasius/201...e-016.html 11:02
notable6 lizmat, Noted!
MasterDuke antoniogamiz: not sure what you meant, but the output was/is in $output 11:51
uzl .tell sena_kun I've written my question from yesterday more thoroughly here: gitlab.com/snippets/1875120 14:21
yoleaux uzl: I'll pass your message to sena_kun.
rfold TIL »ö« is Camelia, not a hyper amazed face 14:45
uzl rfold: never thought of it that way ;-). Isn't there an ASCII version too?! 14:47
rfold >>oe<<
Or ^ :o v (some assembly required) 14:48
El_Che Ƹ̵̡Ӝ̵̨̄Ʒ 14:49
could be an alternative utf8 vaiant
uzl .tell sena_kun I've added some sort of solution to the question I raised: gitlab.com/uzluisf/raku-text-slugi...sing-tests 14:51
yoleaux uzl: I'll pass your message to sena_kun.
Altai-man_ uzl, o/ 14:52
oops
sena_kun looks like my nickname was stolen by a timeout... 14:53
yoleaux 14:21Z <uzl> sena_kun: I've written my question from yesterday more thoroughly here: gitlab.com/snippets/1875120
14:51Z <uzl> sena_kun: I've added some sort of solution to the question I raised: gitlab.com/uzluisf/raku-text-slugi...sing-tests
sena_kun anyway, uzl, right, you are asking the correct question and the solution is not bad for Cyrillic character. There is even a sub for such a transition, I think, let me check... 14:53
docs.perl6.org/type/Str#method_trans <- this one 14:54
m: "хихихи".trans(["х", "и"] => ["h", "i"]).say 14:55
camelia hihihi
sena_kun I see you already have hashes with substitutions, so likely can adapt the code to use `trans` method.
Voldenet after writing some async code, I truly believe that Promise.in is an awkward name for an async version of sleep 14:56
sena_kun but the more important part is what to do with languages that don't have some sort of alphabet, or have symbols that go beyond one, e.g. Chinese, Japanese and so on. I have no idea how to do that. 14:57
Voldenet, "A Promise that completes _in_ 5 seconds", is it awkward?
uzl sena_kun: Oh, I see. As of now, I just pass whatever character to char2ascii and return it if it's in the hash. Otherwise, return the same passed character. 14:58
sena_kun www.jeremyselier.com/entry/uslug.j...-generator <- here they don't convert Unicode into ASCII. 14:59
uzl Do you think it's necessary to do it with .trans? Or is char2ascii just enough?
sena_kun uzl, you can do either way, but I suspect that .trans is much more effective. Or it should be. 15:00
Voldenet sena_kun: when you read it in async code, it reads slightly worse than "say 'a'; await delay(.1); say 'b'"
(I've just named it delay in my code, seems less awkward)
uzl sena_kun: Oh, I great. I'll read the article. The only thing about .trans is that I must know the characters a priori or that's what I think. 15:01
sena_kun I wonder why are you required to use async one if you still await it just there, why not just `sleep`. And if you want to actually do some stuff on a timer, then `Promise.in(5).then(...)` reads not so bad. 15:02
Voldenet $*SCHEDULER.cue({ … }, :in(.5)) 15:03
uzl For instance, if I wanted to transliterate "Компьютер", then I'd need each character's romanized version before using .trans. With char2ascii, I just give it a character and return whatever it has for it.
sena_kun uzl, github.com/cocur/slugify/pull/195/files <- here they are using a dictionary of words. 15:04
and the results promise to be pretty useless, I believe. 15:05
uzl, you can do basically the same with .trans. of course, it won't work for Chinese. 15:06
uzl What would be the reason for that? 15:07
sena_kun Or, well, technically you can use a whole dictionary of symbols, but ugh. This is tricky.
sena_kun uzl, what is the purpose of your module? Do you want it for some specific usecase for people to use? 15:08
uzl uzl: Don't worry! I'll look at the link you provided but instead of using a json file which then needs to be parsed, I'll just use a hash. 15:09
sena_kun: I need it for something I'm doing but I figured other people might find it useful so I packaged it as a module instead. 15:10
uzl I'll look at the link you provided but instead of using a json file which then needs to be parsed, I'll just use a hash right in the code. 15:11
sena_kun uzl, can you set some scope of the project? If you want to convert _any_ human writing system that is in Unicode into ASCII, it seems like a research-level project. 15:12
this thing gets complicated fast 15:13
uzl Oh, not at all! That'd be too much. I'll add Chinese if only to pass the test that is still failing. 15:14
sena_kun for example, in Japanese "体" can has ASCII of "karada", "katachi", also "tai" and "tei" depending on the situation. And it is a very simple case, there are kanji with more reading than those entry-level ones. 15:15
uzl sena_kun: Anyway... Thanks for the all the help :)! 15:16
sena_kun if you certainly need to support Chinese, then adopting a dictionary is your way. Don't forget to check licenses too, because this is pretty important.
if you don't, just remove the test. ;)
uzl Sure, will do! Thanks! 15:17
uzl Have a nice day. 15:17
Voldenet sena_kun: Promise.in doesn't do the same stuff sleep does, consider this 15:18
m: for ^30 { $*SCHEDULER.cue({ say $_; await Promise.in(.5); say $_ }) }; sleep 1
camelia 0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
2…
Voldenet m: for ^30 { $*SCHEDULER.cue({ say $_; sleep(.5); say $_ }) }; sleep 1
camelia 0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
0
29
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
2…
Voldenet the output is different
Voldenet because Promise.in just tells the scheduler to return to the code after some time, sleep actually puts threads to sleep 15:26
sena_kun I still have a feeling that `Promise.in` has more than one usage patterns, and using it as a delayer is not the single one. 15:27
e.g. github.com/croservices/cro/blob/3c...s.pm6#L112 or github.com/croservices/cro/blob/54...r.pm6#L135
I am not too familiar with .cue, though, so just ignore me, please. :) 15:28
mst Voldenet: also commonly used for timeouts
Voldenet mst: that's what I'm using it for - timeouts
mst right, over in perl5 where we've got a few more years experience, we tend to refer to 'succeeds after N seconds' as a delay and 'fails after N seconds' as a timeout 15:29
so when you said 'as a delayer' I assumed the former
Voldenet Actually, it's a task for failing a connection when it hasn't authenticated itself for N seconds, so after the delay it checks some condition and drops the connection 15:33
so it's kind of a timeout, but the other way around
louisono hello 16:24
timotimo o/
louisono I'm currently learning perl6, is there someone here that can answer a question or two? 16:25
timotimo i'm sure many in here will be happy to do that 16:26
i might go AFK at a moment's notice, though
but if you ask your questions right away, others can read it when they look over to the chat 16:27
louisono ok so I'm trying to use phasers 16:28
and, inside a while loop I have written a LAST phaser but its content is run before everything else, even before the FIRST phaser 16:29
sena_kun m: while True { LAST { say 'last!' }; say 42; last; } 16:30
camelia 42
last!
sena_kun louisono, can you provide a piece of your code?
louisono here it is
glot.io/snippets/fe2w2mcbm2
sena_kun you can run small snippets here with `m:` prefix.
timotimo m: while True { LAST { say "[ast" }; say 42; last; } 16:31
camelia 42
[ast
louisono I noticed that I get the expected behaviour when the while loop isn't inside the MAIN subroutine though
timotimo m: sub MAIN { while True { LAST { say "[ast" }; say 42; last; } }
camelia [ast
42
timotimo hm, there's gotta be some additional component 16:32
louisono what do you mean? 16:33
Voldenet louisono: when you put something on the end of the MAIN it works, it's strange 16:55
m: sub MAIN { my $n = 12; while $n <= 18 { say $n; $n += 3; FIRST { say "starting" }; LAST { say "ending" }; }; 1 }
camelia WARNINGS for <tmp>:
starting
12
15
18
ending
Useless use of LOOP_BLOCK_1 symbol in sink context (line 1)
Voldenet m: sub MAIN { my $n = 12; while $n <= 18 { say $n; $n += 3; FIRST { say "starting" }; LAST { say "ending" }; }; }
camelia ending
starting
12
15
18
Voldenet and you get the funny little warning 16:56
louisono haha that's weird indeed 16:57
anyway thanks for pointing out a way to avoid the unexpected behaviour I encountered in the first place 16:58
Voldenet "LAST" phaser doesn't seem to be most used feature though, I'm pretty sure I've never used it 16:59
louisono ok I was just curious since I'm still in the learning phase where I "play" with the language and try things out 17:00
pepeperl Hi. Can I ask what is the prefered way to install perl 6 that provides a package manager? is it rakudo star? And if yes, would it provide zef as the package manager? There are a couple of options on the downloads page, not cear what's the primary / preferred one. Thx 19:23
moritz pepeperl: what operating sytem are you using? 19:24
pepeperl debian
moritz then I recommend nxadm.github.io/rakudo-pkg/#os-repositories
and also read nxadm.github.io/rakudo-pkg/#zef-mo...gular-user 19:25
pepeperl global zef installing packages in $HOME, got it :) 19:26
moritz could you provide a tl;dr on rakudo-pkg vs rakudo star? 19:28
moritz pepeperl: rakudo-pkg is debian packaged 19:29
and is only rakudo + zef, afaict
rakudo star also installs some other modules
pepeperl ok I'll just give it a try with your suggestion. Thanks for the advice 19:31
pepeperl hmm... I've ran the add-perl6-to-path script and confirmed that perl6 -v is working, but the install-zef-as-user script is failing with a bunch of "Failed to update cpan/p6c mirror" erors. Any advice? 19:44
pepeperl looked in the script, not many cues 19:46
lizmat hmmm... I wonder if that zef is still using the broken links ?? 19:47
El_Che ^^
tyil matiaslina: .
pepeperl lizmat maybe? pastebin.com/uAnmzwkc 19:49
El_Che hi, rakudo-pkg packager here
pepeperl El_Che hi
lizmat ecosystem-api.p6c.org/projects1.json works for me 19:50
pepeperl my browser gets it fine too
El_Che weird
lizmat pepeperl: if you paste that URL in a browser, does it do something for you ?
El_Che what the script does is install a vanilla zef from github according to the zef instructions. No magic there 19:51
pepeperl lizmat yes, as I just said, it shows the json fine
El_Che pepeperl: try curling the url on the same shell you're running the script
(or wget)
pepeperl good point 19:52
just installed debian 10 and it ended upobeing rather spartan, like no lspci spartan 19:53
also no curl
it works afterinstalling curl
oh this is going to be causing me a lot of unrelated headaches... 19:54
El_Che pepeperl: good to know, by the way 20:00
El_Che pepeperl: github.com/nxadm/rakudo-pkg#what-a...akudo-star 20:01
zo in short, with rakudo-pkg you get the runtime and the pkg manager in a OS package, with rakudo start you get a selection of preinstalled modules and the runtime, all to be compiled locally 20:02
my mindset is getting the base and installing what I need
pepeperl yeah, it does feel leaner. 20:03
El_Che I need to find the time to release Windows and MacOS pakages, but Linux is covered so far :) 20:04
and macos also by homebrew since they added a rakudo pkg next to rakudo-star 20:05
TreyHarris I _really_ need to move this virtual onto an SSD, I forgot to start restoring it before bed after the power came back and it only just now came back online 20:15
I would've been fine except the power blipped about twenty minutes before the 4-hour outage and it didn't have time to finish fsck in the interim 20:16
I stopped bothering with a UPS after we went ten years without an outage in Manhattan... my last UPS died just last year. whoops. 20:17
tbrowder TreyHarris: is the emacs perl6-mode repo current now with your updates? 23:35
Geth doc: d2a7c31559 | Coke++ | doc/Language/syntax.pod6
whitespace
 23:38
synopsebot Link: doc.perl6.org/language/syntax
tbrowder q 23:39
tbrowder .ask TreyHarris ^^^? 23:40
yoleaux tbrowder: I'll pass your message to TreyHarris.

Powered by Cro and the Raku® Programming Language.

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

Thank you!