»ö« Welcome to Perl 6! | perl6.org/ | evalbot usage: 'p6: say 3;' or rakudo:, or /msg camelia p6: ... | irclog: irc.perl6.org or colabti.org/irclogger/irclogger_logs/perl6 | UTF-8 is our friend!
Set by moritz on 22 December 2015.
00:00 mcmillhj joined 00:04 mcmillhj left 00:15 benjikun joined
benjikun Does anyone know to run something when the user tries to quit (ctrl+c) 00:16
timotimo sure 00:19
benjikun How?
timotimo signal(SIGINT).tap({ say "oh no!"; exit }) 00:20
benjikun alright
sounds good
thank you
timotimo YW
it's a Supply that emits a value every time that signal is received
00:22 SpikeMaster joined 00:23 skids joined
benjikun I appreciate the quick response, thanks! 00:25
00:31 Cabanossi left 00:34 Cabanossi joined 00:41 BenGoldberg joined, Kyo91 joined 00:45 pharv_ left 00:46 Kyo91 left
benjikun m: my @things = <1 2 3 4>; @things.splice(0, @things.elems, ()); 00:47
camelia ( no output )
benjikun m: my @things = <1 2 3 4>; @things.splice(0, @things.elems, ()); say @things.perl;
camelia []
benjikun What are some other ways for clearing arrays lol 00:49
setting it to nil?
00:50 pharv joined
benjikun also for some reason spurt isn't sending anything to the file 00:56
does the file have to exist beforehand?
00:56 jaldhar joined 01:02 Cabanossi left 01:04 Cabanossi joined 01:05 coetry joined 01:06 coetry___ left 01:13 grondilu left 01:14 Kyo91 joined
jeromelanteri zengargoyle, yes, thank you 01:15
yoleaux 28 Jul 2017 19:14Z <zengargoyle> jeromelanteri: the Git::Wrapper thing has been solved! i made pull-request on Git::Wrapper with the fix.
28 Jul 2017 21:07Z <zengargoyle> jeromelanteri: was it you who wanted TagBar config for perl6? gist.github.com/77f253d6d721e4b628...5321b31e9f
jeromelanteri yoleaux, thank you to
01:15 benjikun left
HoboWithShotgun "No such method 'elements' for invocant of type 'Any'", is that perl6s way to tell me something is undefined? 01:17
01:17 deep-book-gk_ joined, skids1 joined 01:19 skids left, Kyo91 left, deep-book-gk_ left
jeromelanteri yoleaux, really ?!! how you do that ? what was the problem ? 01:19
problem was from Git::Wrapper module or from my code ? and what exactly happened ? 01:21
01:22 skids1 left
jeromelanteri yoleaux, i can not see any pull-request there: github.com/perlpilot/p6-Git-Wrapper/pulls 01:22
TEttinger jeromelanteri: yoleaux is a bot that was referring a message to you from zengargoyle
jeromelanteri TEttinger, OMG !
zengargoyle, i can not see any pull request for Git::Wrapper, but anyway, what was the problem ? 01:23
01:23 skids joined 01:24 espadrine left, skids left
zengargoyle jeromelanteri: the pull-request is on the niqcrocks repo that's used by the ecosystem/zef. 01:26
geekosaur HoboWithShotgun, that is one way, yes
zengargoyle jeromelanteri: github.com/nicqrocks/p6-Git-Wrapper/pull/2 01:28
jeromelanteri: basically one statement needed a 'BEGIN' in front of it. :)
HoboWithShotgun uh, nice. exception handling? just add a catch to a block. any block. how very perlish :) 01:34
do you guys have a favourited paste service? 01:37
01:39 pharv left
zengargoyle i use github gists because they're git repositories and you can edit them that way if you like. 01:40
geekosaur not really. pastebin.com is kinda not liked much though
zengargoyle and the Perl 5 application App::Nopaste provides a nice CLI command to create one.
but App::Nopaste supports various different paste sites. 01:41
01:42 skids joined
jeromelanteri gist github 01:43
zengargoyle, very good ! thank you
zengargoyle, from now, how to profit and update this Git::Wrapper ?
zengargoyle jeromelanteri: jnthn gave a lot of good hints and the final 'this should work' fix. :) 01:44
01:44 someuser left
jeromelanteri zengargoyle, but i mean... from zef, how to tell him to choose perpilot repo or nicqrock one ? 01:45
01:46 ilbot3 left
jeromelanteri and why by default it choose this one or an other ? 01:46
zengargoyle the modules zef finds automagically are listed in a big file on github. when an author has a module they want to publish, they make a pull-request to add the github URL to that list. 01:47
01:48 Swey_Hey left
zengargoyle github.com/perl6/ecosystem/blob/ma.../META.list 01:48
jeromelanteri zengargoyle, i do uninstall the perlpilot one and then install from ur of nicqrock. but it failed. 01:49
from url
zengargoyle zef also searches CPAN, but that's rather new and not used much yet.
nicqrock hasn't accepted my pull-request yet.... 01:50
jeromelanteri zengargoyle, yes. also it doesn't want to install from this url.
zengargoyle, why do you make the choice to pull request this one repo and not the default official one ? 01:51
01:52 ilbot3 joined, ChanServ sets mode: +v ilbot3, nadim left, Swey_Hey joined 01:53 Swey_Hey left
zengargoyle this repo is the one found by zef and listed in that META.list file.... so it's pretty much the official one now. 01:54
jeromelanteri zengargoyle, when i do zef install Git::Wrapper, it does install perlpilot one. 01:55
zengargoyle, anyway... i do find ~/.perl6/ -name "Wrapper*" 01:56
there is no Wrapper.pm6 file find . Where should be source files installed from zef ? 01:57
01:58 Kyo91 joined
jeromelanteri nothing in /usr and nothing in /opt 01:58
i think id do compile things only, doesn't copy sources... correct ?
02:00 coetry left
jeromelanteri la 02:01
zengargoyle i just pushed a fixed version to my github but can't make zef actually load it via URL. 02:02
you can clone github.com/zengargoyle/p6-Git-Wrapper.git
02:02 Kyo91 left, Cabanossi left
zengargoyle cd into p6-Git-Wrapper and do `zef install .` 02:03
02:03 ponbiki left
zengargoyle or just clone the nicqrocks version, add the BEGIN to lib/Git/Wrapper.pm and `zef install .` that way. 02:03
02:04 ponbiki joined, Cabanossi joined
zengargoyle i'm not sure why when i try `zef install github.com/zengargoyle/p6-Git-Wrapper.git` it still installs the old version. 02:05
ah, zef cache.... had to `rm -rf ~/.zef/{tmp,store}/p6-Git-Wrapper.git` before it would actually use my URL. 02:08
if you install and it says :ver('0.0.8') then you have the fixed one. :ver('0.0.7') is the broken version. 02:10
02:14 Kyo91 joined 02:16 mcmillhj joined
jeromelanteri zengargoyle, ho well.. i had the same idea you had first... i just came back from fit clone and vim the Wrapper.pm file from perlpilot. 02:17
02:19 Kyo91 left
zengargoyle jeromelanteri: i doubt perlpilot's version will every get updated... i *think* nicqrocks is the new maintainer. 02:19
jeromelanteri ok then i do remove perlpilot and clone nicqrocks 02:20
zengargoyle .ask perlpilot are you still maintaining Git::Wrapper?
yoleaux zengargoyle: I'll pass your message to perlpilot.
02:20 mcmillhj left
zengargoyle when i looked, nicqrocks repo had more recent changes. like a couple months/weeks ago. perlpilots was like 7 months old. 02:21
jeromelanteri zengargoyle, yeah... but what i need is not the last shiny porsche also... but a strong module who works (i mean... all the time). 02:22
zengargoyle oh, and that .ask is how you get yoleaux to leave a message for somebody. (or .tell does about the same thing) 02:23
jeromelanteri zengargoyle, alos, it looks like there is no pass tests required for post modules for perl6.
zengargoyle, yes, i see. very good to pass message to a robot. 02:24
zengargoyle dpk.io/yoleaux -- more docs for the bot. 02:25
jeromelanteri let's try it... 02:26
.gc perl6 Git::Wrapper 02:27
yoleaux jeromelanteri: Sorry, that command (.gc) crashed.
jeromelanteri (crying)
02:27 deep-book-gk_ joined
jeromelanteri .tr :fr :en mon tailleur est riche 02:28
yoleaux jeromelanteri: Sorry, that command (.tr) crashed.
jeromelanteri (crying again)
02:28 deep-book-gk_ left
zengargoyle mostly it's .ask .tell and .u that i've seen people actually use. 02:28
jeromelanteri ok
zengargoyle .botsnack
synopsebot6 om nom nom
yoleaux :D
zengargoyle fand that one.
jeromelanteri what a wondefull world ! 02:29
wonderfull
zengargoyle, well.. thank you for your fix, it works very well.
02:31 coetry joined
zengargoyle if you're really bored, clone my VimPlugger.git repo... :) i sorta made your thing into more like a regular set of modules while trying to figure out Git::Wrapper thing. 02:32
02:34 SpikeMaster left
zengargoyle mostly just moving files and making Module::Name longer and tests to check if modules compile or not. made it a little easer to try things out and see if tests still passed or not. 02:35
02:35 Kyo91 joined
zengargoyle but in the end... it was just Git::Wrapper that was broken so it didn't really matter... :P 02:36
02:36 Kyo91 left
jeromelanteri i done you fix allready and works. that's ok like that. 02:43
zengargoyle cool.
jeromelanteri yes, i have to merge then and go ahead. my master branch is in front of my Git::Wrapper one. I have to add a comment on my README file. May i talk about your help inside ? you will not become very popular because i am not to... but i have to ask you before to do that. 02:44
02:50 coetry left
jaldhar Hello. Could someone try installing Algorithm::DawkinsWeasel with zef? I'm getting a strange error and I want to know if it is just me. 02:55
02:56 noganex joined 02:58 garpert joined 02:59 noganex_ left 03:02 Cabanossi left 03:04 coetry_ joined, Cabanossi joined
ugexe i sent you a PR already jaldhar 03:05
well, i thought i did 03:06
your source-url is wrong - github.com/jaldhar/Algorithm-DawkinsWeasel needs a .git at the end 03:07
otherwise zef doesn't know to fetch it with git instead of wget
jaldhar ugexe: oh didn't see that. Thanks. 03:10
ugexe zengargoyle: found why that did not seem to update... I had a git fetch where i needed a git pull 03:11
jaldhar ugexe: There's no PR but I presume all I need to do is update META6.json in my repository right?
ugexe jaldhar: yep thats right 03:12
jaldhar ok doing that... 03:13
ugexe it might take some time to show up in the ecosystem, but i verified that `zef install github.com/jaldhar/Algorithm-DawkinsWeasel.git` works
jaldhar ugexe: one thing though, the error message was rather misleading. I thought zef couldn't find unzip or something. 03:14
zengargoyle ugexe: cool, thanks. 03:17
jeromelanteri: you can mention if you want, no problem... or just mention the fine folks on #perl6 :) 03:19
ugexe jaldhar: because zef cant just guess
what that url downloaded might be valid to some other plugin
03:30 Cabanossi left 03:32 Todd_ joined, grondilu joined 03:33 Cabanossi joined
Todd_ Hi. I am trying to turn "123.456.789" into "123456". My match is `$NewRev =~ m/\.(.*?)\.(.*?)\.(*.)/;` which string an unknown "." verb error. What am I doing wrong? 03:33
grondilu m: say grammar { rule TOP { <digit> | <TOP> <digit> }.parse: "01234";
camelia 5===SORRY!5=== Error while compiling <tmp>
Missing block
at <tmp>:1
------> 3digit> | <TOP> <digit> }.parse: "01234";7⏏5<EOL>
grondilu m: say grammar { rule TOP { <digit> | <TOP> <digit> } }.parse: "01234";
camelia (timeout)
grondilu * 03:34
Todd_ I don't understand
ugexe =~ is perl5
Todd_ oh poop. Wrong group. Sorry
03:35 Todd_ left
geekosaur just spotted the actual error 03:35
03:36 coetry_ left 03:38 coetry_ joined
HoboWithShotgun allright, i am shelling out to curl and have a strange bug 03:45
i think there might be a bug in the shell command 03:46
see gist.github.com/holli-holzer/1a042...df2a87184d
03:48 coetry___ joined, coetry_ left
jeromelanteri zengargoyle, done ! thank you dude 03:51
03:52 HoboWithShotgun is now known as HoboWithShotgunA, HoboWithShotgunA is now known as HoboWithShtgnAFK
geekosaur wonders which "system shell" is being used there 03:52
HoboWithShtgnAFK wonders if the D:> isn't clue enough? 03:53
geekosaur that is the thing yiouy are typing to, which may not be what perl 6 is using
unix has $SHELL, windows does not have such convention 03:54
HoboWithShtgnAFK well i have cygwin installed
geekosaur it looks windows specific since it works here 03:56
windows command line parsing is ... an adventure. :/
HoboWithShtgnAFK i changed it to single quotes. the error message then indicates perl6 is indeed using the native shell 03:57
03:59 garpert left, coetry___ left
geekosaur ...the thing about windows is that quote handling is deferred to the application normally (i.e. with cmd, or powershell). if the curl you are using is from cygwin or msys2 then it may be assuming a shell is doing some of the quote processing 03:59
er, a unix-like shell 04:00
04:00 coetry_ joined
geekosaur and some executables for windows try to be "clever" and figure out which to do... which is unreliable at best. 04:00
(that can lead to some real head-scratching errors, especially if you use the other kind of cmd.exe quote (^ which is sort-of like Unix's \ escape but not quite the same) 04:01
HoboWithShtgnAFK apparantly "shell" escaped the double quotes
geekosaur that might be worth reporting as a rakudobug 04:02
buggable, rakudobug 04:03
bleh
huggable, rakudobug
huggable geekosaur, Report bugs by emailing to [email@hidden.address]
geekosaur thought it had more detail than that..
HoboWithShtgnAFK when i say shell qq[echo "foo"];
the output is \"foo"
that is definitly not correct
whatever behaviour it should either be foo or "foo" 04:04
or \"foo\" 04:05
unless my 4am mind is completely off 04:06
geekosaur your analysis sounds correct to me. definitely file a rakudobug. 04:08
(unless your rakudo is several months old; try "perl6 -v"(
HoboWithShtgnAFK 2017.07 04:09
04:09 aborazmeh joined, aborazmeh left, aborazmeh joined
geekosaur yes, file rakudobug. 04:09
HoboWithShtgnAFK allright, will do
tx for your input
04:13 xinming joined 04:15 coetry_ left 04:16 coetry_ joined 04:17 xinming_ left 04:18 pharv joined, aborazmeh left 04:31 Cabanossi left 04:33 Cabanossi joined 04:34 wamba left
pharv is anyone here a core contributor to the web app framework Bailador? 04:36
04:37 xtreak joined
PerlJam pharv: why? 04:43
pharv well, i'm looking for a project to contribute to, but i've heard some people express some concerns about some of the design decisions, so i was just gonna ask about the strategy 04:46
it doesn't have to be a framework, it's just the type of development i'm most familiar with, so i figured i could help out with something like that... i just want to work on something that's well-designed and has some momentum 04:48
PerlJam pharv: the right people to ask about Bailador design are probably ufobat and szabgab. I've contributed some code as I was thinking of using it, but I haven't looked at the design in detail yet.
pharv i've also heard about PW6, but it looks like it's still in design phase 04:49
ok... i haven't used it yet, but it looks like Web::App::MVC got abandoned a few months ago... and PW6 is still in design phase 04:50
if Bailador is the only game in town, i'll help out, i just wanted to explore a few options first 04:51
PerlJam pharv: also, for Bailador stuff, it's probably best to get on the slack team for it and chat there. 04:52
pharv: I don't think it's the only game, but it does have some momentum.
04:52 skids left
PerlJam and szabgab *is* writing a book about it supposedly, so there's some incentive there for him to keep working on it :) 04:52
pharv ahh 04:53
you said you considered using it... what did you end up using instead?
PerlJam I haven't used anything instead. I'm trying to find a way to use Perl6 at work and a small, stand alone web app might just do the trick. 04:54
pharv ahh, cool... what languages do you currently use at work?
PerlJam Mostly Perl 5 and JS 04:55
pharv do you use Moose?
PerlJam well ... the Ext flavor of JS anyway
yes, we use Moose/Moo
pharv hah, that's funny... perl5/extjs was my first job at verizon wireless :)
PerlJam We've also been using Moops, but we keep running into problems with weird stuff happening that's hard to debug, so whenever that happens the standard response is to use Moo + Function::Paramters + Type::Tiny (you can guess why I want Perl 6 instead :) 04:57
pharv lol, yeah
although when i did perl/extjs, i spent about 90% of my time debugging extjs
hopefully it's made some progress in the last 5 years 04:58
04:59 zengargoyle left
PerlJam luckily we've got some really talented front-end guys who take care keeping us up-to-date wrt extjs and they handle any debugging (they've found probably 3 or 4 extjs bugs in the last year and if extjs hasn't fixed them, they've devised work-arounds) 04:59
pharv oh, that's awesome
yeah, extjs scared me away from front-end development, but i've actually started to enjoy it recently with angular2 05:00
PerlJam what's funny for me is that I first used extjs back in 2007ish and it's definitely come a long way since then ;)
05:00 Ben_Goldberg joined, BenGoldberg left, Ben_Goldberg is now known as BenGoldberg, zengargoyle joined
pharv wow, i can't even imagine what it was like back then 05:01
it probably wasn't that bad, it was just a little overwhelming for me as my first job, and the version we were using wasn't very well documented at the time... i think it was 4.5 or so 05:03
05:05 zengargoyle left 05:13 coetry_ left 05:15 zengargoyle joined
zengargoyle .botsnack 05:16
synopsebot6 om nom nom
yoleaux :D
05:38 Ben_Goldberg joined 05:39 BenGoldberg left, Ben_Goldberg is now known as BenGoldberg 05:41 shmibs left 05:42 shmibs joined, grondilu left 05:43 grondilu joined, khw left 05:44 shmibs left 05:45 BenGoldberg left, shmibs joined 05:47 BenGoldberg joined, HoboWithShtgnAFK left, zoll left 05:56 xtreak left 05:58 xtreak joined 05:59 kyan left 06:03 pilne left 06:11 pharv left 06:14 abraxxa joined 06:25 abraxxa left 06:30 xtreak left 06:38 xtreak joined 06:42 robertle joined 06:47 jaldhar left 06:53 statisfiable6 left 06:54 statisfiable6 joined, ChanServ sets mode: +v statisfiable6 06:58 BenGoldberg left 07:01 Cabanossi left 07:03 Cabanossi joined 07:11 andrzejku joined 07:22 parv joined 07:26 rindolf joined 07:39 mr-foobar joined 07:40 Skarsnik joined 07:45 nadim joined 07:48 darutoko joined 08:00 Cabanossi left
jeromelanteri i get an array, how to show the last entry of the array ? 08:01
my @array = <a b c d e f>; say @array.last; 08:02
08:02 Cabanossi joined
jeromelanteri @array[-1] (not works to) 08:03
08:03 wamba joined
jeromelanteri @array.pop works, but remove the last one... 08:04
Skarsnik .tail(1)
m: my @a = 1..5; say @a.tail(1); 08:05
camelia (5)
jeromelanteri Skarsnik, thank you 08:06
parv m: my @x = (1,3); say @x.tail(0) 08:07
camelia ()
parv m: my @x = (1,3); say @x.tail(2)
camelia (1 3)
parv m: my @x = (1,3); say @x.tail(1)
camelia (3)
parv m: my @x = (1,3,4,5); say @x.tail(2)
camelia (4 5)
parv m: my @x = (1,3,4,5); say @x.tail(3)
camelia (3 4 5)
Skarsnik m: my @a = 1..5; say @a.tail; 08:08
camelia 5
08:08 andrzejku left
jeromelanteri perfect 08:08
parv .tail is not really a substitute for -ve indexing as cannot get only the last i-th element 08:09
zengargoyle m: my @f = < 1 2 3 4 >; say @f[*];
camelia (1 2 3 4)
zengargoyle m: my @f = < 1 2 3 4 >; say @f[*-1];
camelia 4
zengargoyle m: my @f = < 1 2 3 4 >; say @f[2..*]; 08:10
camelia (3 4)
zengargoyle m: my @f = < 1 2 3 4 >; say @f[2..*-1];
camelia (3 4)
jeromelanteri well... which one make working proc the less ?
indexing i think ?
08:11 andrzejku joined
zengargoyle not sure. indexing was first. .tail is more recent and probably just indexing underneath. 08:11
jeromelanteri so maybe the same 08:12
ok one or the other should make the job i need thank you
zengargoyle yeah, with compile wizardry.
08:12 wamba left 08:13 wamba joined
jeromelanteri there is something more i not understand with my code... i do ask an entry from prompt (Prompt:Gruff::Export), and put it in a variable. Well..; that works, but when i ask to change it, it said that it can not assign to an immutable value. 08:14
why immutable value ? 08:15
zengargoyle ouch, complicated. :) Array and List and Seq are all almost same thing... method tail (Array:D: $n) {} is complicated...
jeromelanteri this variable is passed in argument to other sub functions... from there a problem happen. is it because this variable can not be change if pass by argument to an ther function ? do it need to be a referenced variable instead ? 08:16
parv jeromelanteri, probably using binding instead of assignment $x := 4 instead of $x = 4
jeromelanteri parv, i'm going to try
zengargoyle sub whatever (Str $value is copy) { } or $value is rw -- arguments to functions are read-only by default. 08:17
08:18 andrzejku left
zengargoyle is copy makes it so you can write to it. is rw makes it like reference where you change the thing in the calling place. 08:18
jeromelanteri zengargoyle, may i use ref instead of copy ? like \$value ?
is rw then...
zengargoyle don't thing so.
maybe you can use \value (no $) not sure. usually is copy or is rw. 08:19
jeromelanteri := instead works... let's try to define as "is rw"... 08:20
works to ! 08:21
ok, i think define well as "is rw" seems to be more readable code for me. 08:22
zengargoyle jeromelanteri: did you do perl 5 before?
jeromelanteri long time ago one year, then nothing perl for 10 years, and a little bit two weeks ago... prefer perl6 from far now. 08:23
08:24 mr-foobar left, [particle]1 left 08:25 [particle] joined
zengargoyle cool, i did perl5 for many years and can sometimes explain in perl5 but that probably will not help much. 08:25
jeromelanteri i was very happy with perl5, but see python and ruby (ruby very funny). then use C++ and very happy by powerfull and precision of C++. Now perl6 looks like a real powerfull OO language. 08:26
parv jeromelanteri, tried julia? 08:27
jeromelanteri use it for fun and want to make own script
never try julia
want to have a look with haskell and framework yesod soon... python with django (a little bit tired with Rails). See some Go code looks funny to. But never heard about julia. 08:28
zengargoyle i waited patiently a long time for perl6 and still here... it is cool.
jeromelanteri yes, it can be very nice code reading, very clear i think. But need a learning curve first like all langages (much more when need to be precise). 08:29
i look at some module source codes... some of them looks like people coding C lanugage before... not to much use the power of OO. 08:30
parv julia, haskell, clojure in context of fortran: arstechnica.com/science/2014/05/sc...-behemoth/ -- short introduction on page 2 08:31
jeromelanteri i see also that ther eis an OS construct from haskell...
nixOS
strange langages also should be prolog (old)
zengargoyle rosetta code might be a good place to look. same problems in many languages. usually perl6 version is by somebody good. 08:32
jeromelanteri what seems to be very well is the way you can do with perl6 by different options coding. 08:33
R0b0t1 I looked at Perl 6 maybe two years ago
The last two years have made it far more usable
You might want to look at Scala
sorry to interject
jeromelanteri nop
08:34 [particle] left
R0b0t1 Elixir also has a nice web framework 08:34
zengargoyle TIMTOWTDI -- There Is More Than One Way To Do It. almost the meaning of perl 5/6
jeromelanteri scala, julia cclojure... well... lot of information to find. Ok, then i will.
R0b0t1 Erlang, the language Elixir is based on, has some very novel features that found their way into Perl 6. Mainly to do with scheduling I think 08:35
jeromelanteri actually, i use Rails for web framework... i'm not to much happy by Rails. ery slow, eat lot of ressources... i thinking about yesod/haskell because of compiling code ability (then more quick). 08:36
R0b0t1 jeromelanteri, I actually just had a conversation about this elsewhere
zengargoyle thinks perl6 became reasonable with moar replacing parrot. became fast enough to play with.
08:36 mr-foobar joined, [particle] joined
R0b0t1 I think it is very good you are already looking at Haskell 08:36
jeromelanteri What is Erlang need for running ? does it need a kind of engine in background ?
R0b0t1 Knowing what type theory is and how it works makes one capable of reasoning about whether a language is good or not
zengargoyle heh, erlang....
08:37 grumble left
R0b0t1 Yeah, Erlang runs on BEAM which is a VM implemented by Ericsson 08:37
jeromelanteri R0b0t1, yes, i looking at this... but my levele is very poor... need more time. I'm actually very happy with C++, but i'm searching an other one langage to stay with.
R0b0t1 It has a very lightweird process type
08:37 grumble joined
R0b0t1 lightweight* 08:37
jeromelanteri R0b0t1, i thinking that the best is by far to be compiled.
zengargoyle erlang was created by telephone people to run telephone switches. mostly rock-solid and failure proof. prolog heritage. 08:38
jeromelanteri ho well...
what about julia ?
R0b0t1 Meh 08:39
zengargoyle it's good for what it does. i've had erlang jabber (xMPP) servers that run for *years* without anything bad happening.
R0b0t1 The webpage of any of these languages would have you believe they nearly cure cancer
That is why I think it is important to have a wide breadth of field and know about functional languages, logical languages, etc
So you can tell if anything is actually good 08:40
jeromelanteri what is this expression: "cure cancer" mean ? (i'm not english)
R0b0t1 Perl 6 lets you experiment with almost anything
zengargoyle and if you're careful, you can hot-swap code while server is running. and it is built around multiple nodes so is distributed from the start.
jeromelanteri I want to stay use perl6 for playing and kind of little script jobs on computers... nt more. 08:41
R0b0t1 jeromelanteri: A very hard to cure classification of diseases that are usually fatal. I mean that the webpages talk up their respective language a bit much
zengargoyle cancer is disease that kills. the cure is always 10 years in the future.
R0b0t1 zengargoyle: That is the part I have never actually seen duplicated - the hotswapping
zengargoyle cure for cancer is like cure for common cold or fusion reactors. 08:42
R0b0t1 Or a language that actually does what I need it to do
jeromelanteri actually i paid for a little server on Vultr... then i would like to make things going quicker with no more VM need languages (like Rails or some other). Yesod should be the one, but maybe you know some other ? 08:43
zengargoyle R0b0t1: i want slangs and every language at once. this sub does postscript because it returns an image, this sub is in forth because it has a nice stack solution, this sub is in APL because it's math-magic... 08:44
jeromelanteri also, each time you want to update rails is a big big story...
parv perl6 needs VM in form of moar-vm (or a different one) 08:45
08:45 xtreak left
jeromelanteri i see 08:47
i will not use this for a kind of web site because of that. 08:48
zengargoyle thinks forth is the ultimate no-vm. it's basically dynamic super macro assembler.
jeromelanteri forth... added to my searching to do list
zengargoyle jeromelanteri: nah, it's old and mostly used for embedded things. it was just one of my first languages *long* time ago. 08:50
08:51 wamba left
parv zengargoyle, what you (or anybody else) have to say about lua? (i looked at some beginner level code once or twice) 08:51
zengargoyle i don't like lua's "array indexes start at 1 not 0" thing. :) 08:52
jeromelanteri zengargoyle, well... you should like to look at micro-controller books. hobby like PIC micro-controllers... but very powerfull
Skarsnik it's only good to embed in other stuff
jeromelanteri i see lua can works fine, but difficult to read.
have to learn and use this a little for conky application 08:53
i think blender use it also ?
zengargoyle yeah, lua never made me happy.... sorta like javascript. the idea is good, the reality is a bit yuck.
jeromelanteri zengargoyle, you will not like nodes.js then... :) 08:54
zengargoyle it is like different javascript. simple idea of object and slots and simplicity that sounds good like lisp but does not quite hit the right balance for me. 08:55
jeromelanteri: yeah, i don't like javascript or node.js, pretty much avoid them as much as possible. 08:56
jeromelanteri nothing can be bad more than VBA and OpenOffice Basic language
zengargoyle, i use coffeescript up to javascript code (more easy to read/write.. and then... ho no... not to debug... same) 08:57
zengargoyle well.... AMIGA Basic or TRS-80 Basic are arguably worse than VB :)
jeromelanteri i start learn my first langage with Basic on Commodor 64 ! then C. 08:58
zengargoyle or hell, Apple integer BASIC for that matter. no floating point!!!!
jeromelanteri but you know... you can run long time, if you have little legs... you will never run quick. Same for the brain as i see. Never mind, i have not a big brain, then i will learn humility. 08:59
parv speaking of Apple, how is people's experience with Swift?
zengargoyle we could go back to HP calculators but.... 09:00
jeromelanteri HP langage was very funny. You like asm, you like HP
parv zengargoyle, hey I wrote TI-82 program for complex numbers & polar coords. HP was too expensive for me then. 09:01
zengargoyle haven't tried swift, but the talk i watched was intersting enough.
09:01 GhostK joined
jeromelanteri does Swift a kind of GUI ? 09:01
GhostK hi ( tese msg
*test
oops not social ch. sry 09:02
zengargoyle parv: i played with my dads calculator.... there was acutally a lunar-lander game you could type in and try to land on the moon.
jeromelanteri It was the first NASA propaganda maybe ? 09:03
parv zengargoyle, which HP calculator was it, do you remember?
jeromelanteri SX48
zengargoyle it was in the 70's. plastic, wedge like, LED display and slide switch for power.... and had a holster. 09:04
jeromelanteri then older than SX48
zengargoyle around the same time was the TI that had the magnetic strips for programs.... 09:05
eater zengargoyle: are you sure about LEDs?
zengargoyle eater: well i mean the red segmented display that all caculators had at the time.
jeromelanteri 7 segment LED or LCD 16 segment / chars 09:06
09:06 xtreak joined
parv zengargoyle, ok that before TI-{79 .. 83} era. I was thinking of graphical calculators c.1995 09:06
eater yeah, leds were invented physically in '62
web.archive.org/web/20070809062214...Dupuis.pdf 09:07
hehe
jeromelanteri in the 80's, mac was very strong, very different, big/good quality (but can not have one at home... to exepensive). proc was Motorola. 09:08
stmuk the first program I typed in (70s) into a (6502 based) KIM-1 was "Lunar Lander" I remembered it displayed SAFE on its LEDs 09:11
zengargoyle i have a piece of cable that (oral history) is from a Univac, it's about 1 inch diameter and has many bundles shielded wires inside. about 2 feet long. my grandfather kept it in his truck to hit people with. :P
stmuk my dad borrowed the KIM-1 from work 09:12
zengargoyle they hired women to solder the cables to PC boards (-ish) because they had good eyesight from sewing....
eater yea 09:13
the first programmers where females
how did it shift so badly tho 09:14
zengargoyle i'd love to actually identify the cable some day.... i think it's probably some telephone trunk cable. but it is really old...
jeromelanteri telephones cables are mono-brin and they are rotated in the cable. 09:15
zengargoyle eater: do you want to know? i ask this of all my victims. :P
jeromelanteri they are twisted for the background sound level to be down. 09:16
zengargoyle jeromelanteri: don't know what that means... they look like stranded core, insulator, stranded sheild, outer insulator, and there are dozens or more of them in a rubber/plastic outer shell that's about 2mm thick. 09:17
jeromelanteri we actually use some old and new technology mixed for the deep water ROV materials. Due to the problem of data transmissions and energy fr
for make running hrdolic pump, make going camera view, and send commands. 09:18
zengargoyle, they use different normal tecnologies depend of the utility of the cables.. that can be different for analogic sound or for data (depend of frequencies also) and for energy. 09:19
09:19 parv left
zengargoyle it's heavy but flexible and if you smacked somebody with it it would hurt bad. 09:20
jeromelanteri energy can be very specific due to how many current need to go inside...
zengargoyle why not fiber?
09:21 parv joined
jeromelanteri fiber can not share electric energy. We use fiber for camera (3 cables with 6 pairs each... need to have spare ready to run without delay 09:21
zengargoyle ah, of course.
jeromelanteri fibers are also very fragil
and repair need between one and two hours to go 09:22
i'm talkign about big ROV (3T5 weight)
zengargoyle the ROV and deep sea stuff i have seen is all about high data transfer and research networking. 09:23
jeromelanteri who go about 2500 meter deep
zengargoyle well maybe not *deep* sea.
eater zengargoyle: are you the reason?
jeromelanteri i do work for inspection in african area sub sea. Inpections of pipes and TLP for oil and gaz industry 09:24
TEttinger TLP?
jeromelanteri Is a kind of plateform
zengargoyle sweet. i worked for .edu in research networking.
jeromelanteri hey... do you guys used allready coding with Gst lib ? 09:25
i'm trying to do something with that... but very difficult for me. 09:26
bitbucket.org/jerome___/gst_video_testeur 09:27
zengargoyle eater: math wise, women are better at calculation then men statistically wise.
jeromelanteri i have to go again on this project.
zengargoyle eater: it's a big-ish discussion in the reasons for lack of diversity in STEM fields. 09:29
09:30 Swey_Hey joined 09:34 wamba joined
Skarsnik jeromelanteri, I worked a bit with gstreamer yes 09:36
eater zengargoyle: I dont follow, because women are better at math, they don't do STEM? or do you not mean that 09:37
TEttinger it's mostly an unwelcoming culture, especially in american startup culture (see: ubeer) 09:38
uber
zengargoyle eater: www.youtube.com/watch?v=9bTKRkmwtGY
TEttinger it should be resolvable, but like other resolvable problems (cough global warming cough) isn't being addressed well. 09:39
jeromelanteri Skarsnik, i'm trying to make dynamics overlay data on video streams... show them together in live, and same time capture video (overlay should be captured inside a file... like it's current to do for overlay sub-title), then later recording result of overlay and video...
but actually, i failed to find any examples for use Gst with GUI Qt5
Skarsnik did you try QtGstreamer ? 09:40
it's a bitch to compile it on windows thou
jeromelanteri i'm just able to capture video and show OR recording... not in the same time.
Skarsnik, yes, that is my bitbucket repo a little code for test it
bitbucket.org/jerome___/gst_video_testeur 09:41
zengargoyle eater: it's long... but decent. it's just different areas... CS is not so much NASA sending moon shots and getting the numbers right than being crazy tech dude drilling tunnels.
jeromelanteri (10 month old.. need to go ahead)
zengargoyle and that's about the end of that digression.
Skarsnik hm, sems like a complicated pipeline 09:42
I only used to play video with my own gstreamer plugin in the mix 09:43
jeromelanteri Skarsnik, for me, i understadn a bit the pipeline Gst prncipe, but it still compicate to make in place. Also, the doc is C oriented, then Gst/Qt use some other way to do for little things i think.
play video ok... but for construct pipeline to make pplaying, then make recording, then push overlay.. all in the same time... OMG !!! it make me crazy 09:44
Qt5 doesn't have something for make this easy to...
and i didn't want to use QML code.
there is also events happen... 09:46
Skarsnik QtGstreamer make just writing the gst code in C++/Qt more c++ friendly (and also better qt integration for widget)
jeromelanteri yes
Skarsnik github.com/Toyunda/toyunda/blob/ma...widget.cpp maybe that can help yu 09:47
it's beem a while I worked with this stuff
jeromelanteri Skarsnik, yes, that should yes... very good and nice code 09:49
09:49 Swey_Hey left
jeromelanteri Skarsnik, , thank you 09:50
09:52 parv left 10:00 Cabanossi left 10:03 Cabanossi joined
Skarsnik moritz, is there a way to have svg-plot to put the value on the graph? 10:19
10:50 TEttinger left 10:54 espadrine joined 11:04 SCHAPiE left 11:06 SCHAPiE joined 11:12 lizmat left, colomon joined 11:20 xtreak left 11:23 xtreak joined 11:29 colomon left 11:31 Cabanossi left 11:33 Cabanossi joined 11:34 vendethiel joined
nadim good day, can someone please remind me what was the shortcut for calling a method only if it exists? 11:37
yoleaux 05:32Z <llfourn> nadim: Thanks for suffering to figure out that problem. Did you manage to make an RT? It still sounds like a bug.
11:38 lizmat joined
Skarsnik ^can ? 11:41
nadim .tell llfourn Hi, I did not managed to make a RT since I could not golf it down, on the other hand I found what the problem was. Maybe this will help nopaste.linux-dev.org/?1160132 11:45
yoleaux nadim: I'll pass your message to llfourn.
nadim Skarsnik: I mean calling a method only if it is possible, not finding out if it is possible
11:45 GhostK left
Skarsnik I think can make a call 11:45
nadim ah! 11:46
ah right caret + can
Skarsnik m: say (0..3).^can("elems);
camelia 5===SORRY!5=== Error while compiling <tmp>
Unable to parse expression in double quotes; couldn't find final '"'
at <tmp>:1
------> 3say (0..3).^can("elems);7⏏5<EOL>
expecting any of:
argument list
double quotes…
Skarsnik m: say (0..3).^can("elems");
camelia (elems Method+{<anon|44581280>}.new)
Skarsnik nvm it's not this one 11:47
nadim could have been :) 11:48
Skarsnik m: say (0..3).+^can("elems"); 11:52
camelia 5===SORRY!5=== Error while compiling <tmp>
Malformed postfix call
at <tmp>:1
------> 3say (0..3).+7⏏5^can("elems");
timotimo there's .* but that'll call multiple candidates if there are more than one 11:54
actually 11:55
.? does what you want
Skarsnik m: say (0..3).?("elems"); 11:56
camelia 5===SORRY!5=== Error while compiling <tmp>
Cannot use .? on a non-identifier method call
at <tmp>:1
------> 3say (0..3).?("elems")7⏏5;
Skarsnik m: say (0..3).?elems;
camelia 4
Skarsnik m: say (0..3).?foo;
camelia Nil
nadim that's the one! 11:58
Skarsnik hu 12:02
that an interesting bug/weird stuff
timotimo you mean that you can't .?()?
12:03 colomon joined
Skarsnik No, I made a script that made profile for a script with multiple rakudo release, but it failed at some point 12:03
I mean the code is basicly for @rakudo-release {run a p6 script with --profile arguments} 12:04
timotimo mh
Skarsnik and for some reason the script fail with undef value toward the end of the releases 12:05
12:05 tzekid joined
Skarsnik and I already run the called script 100 times for each release to make benchmarks 12:05
12:06 itaylor57 left
timotimo that sounds interesting indeed 12:06
Skarsnik my bet is some memory issue since the script call a native lib
timotimo what are you calling a native lib for here? 12:07
Skarsnik the called script use a native lib
timotimo i thought the calling script had the undef errors?
Skarsnik Yes, the called script end with undef errors for later release
timotimo oh, ok 12:08
12:08 colomon left
timotimo if you call a perl6 script 100 times you'll spend about 10s only doing startup 12:08
12:08 tzekid left
Skarsnik The called script output times that I track internaly, I don't use the running time of the script 12:09
12:09 tzekid joined 12:10 colomon joined
Skarsnik hm 12:10
12:11 Kristjan55555 joined
Skarsnik gist.github.com/Skarsnik/179f3d91f...2d749580e6 12:11
ok this is weird, adding --profile make it goes boom
12:12 Skarsnik_ joined 12:13 jaldhar joined
Kristjan55555 twitter.com/kristjan1125125 imgur.com/a/GZeEy 12:14
timotimo oh 12:15
that
yeah, that's a code-gen issue that the profiler can cause somehow
12:16 Cabanossi left, Skarsnik left, andrzejku joined 12:17 Skarsnik_ is now known as Skarsnik 12:18 Cabanossi joined, kaare_ left, Kristjan55555 left
timotimo yeah, i recommend you not click on that imgur link 12:19
Skarsnik I made the mistake ~~
timotimo moritz: wanna purge that line of spam from the irclogs? 12:20
12:20 itaylor57 joined 12:21 jaldhar_ joined 12:22 jaldhar left 12:25 grondilu_ joined 12:28 grondilu left 12:31 jaldhar_ left, xtreak left, jaldhar_ joined 12:36 rindolf left
rightfold Ugh 12:39
Someone told me not to click it and naturally I clicked it
Skarsnik timotimo, dunno if that usefull gist.github.com/Skarsnik/1c356d1d3...778b6999b4 12:41
12:41 araraloren joined
timotimo rightfold: sometimes i hate the internet 12:42
gotta go! 12:43
rightfold Reminds me of Portal 2 with all the DO NOT ENTER signs 🤡
12:44 colomon left
timotimo except there you get something nice 12:46
12:46 colomon joined
rightfold Yeah I heard you get a cake 12:47
timotimo you get that for following all the orders
12:51 skids joined
Geth ecosystem: 272d253b86 | (Tom Browder)++ | META.list
add new module
 12:56
12:57 rindolf joined
Skarsnik lizmat, I included profiles for each rakudo version gist.github.com/Skarsnik/1c356d1d3...778b6999b4 not sure if that helpful 13:25
13:31 colomon left 13:33 lucasb joined
AlexDaniel Skarsnik: I made the mistake too :| 13:39
next time I'll backlog in the other direction, like I usually do…
13:40 lucasb left
El_Che moritz: if I read your book through safari once it arrives there, do you know how royalties work? Do you get paid? 13:45
13:46 pilne joined 13:53 colomon joined 14:02 colomon left 14:05 BenGoldberg joined 14:16 bronious left, Cabanossi left 14:18 Cabanossi joined 14:38 someuser joined 14:41 Khisanth left 14:42 setty1 left
moritz El_Che: if the publisher gets paid, so do I. But I don't know the specifics for safari 14:45
El_Che: but I'd appreciate a review much more than the roaylties
AlexDaniel do we have a ticket or docs or anything about implicit EVAL in regexes? 14:46
I remember stumbling upon this before, and at the time I was thinking this should not be allowed without MONKEY-SEE-NO-EVAL, but honestly I remember nothing about this now 14:48
14:49 kaare_ joined
AlexDaniel I mean stuff like this: 14:50
m: my $x = ‘hello’; say ‘hello world’ ~~ /<{$x}>/
camelia ｢hello｣
AlexDaniel m: my $x = ‘#hello’; say ‘hello world’ ~~ /<{$x}>/
camelia 5===SORRY!5===
Regex not terminated.
at /home/camelia/EVAL_0:1
------> 3anon regex { #hello}7⏏5<EOL>
Regex not terminated.
at /home/camelia/EVAL_0:1
------> 3anon regex { #hello}7⏏5<EOL>
Malformed regex
at /home/camelia/EV…
14:51 mr-fooba_ joined
AlexDaniel or was it the right behavior with <{}> ? 14:51
14:52 ufobat joined, mr-foobar left 14:54 Khisanth joined
ufobat hello :) 14:55
does anyone have any advice what i can do about this: gist.github.com/ufobat/30a4cd45430...80f694b0a3 it is an installation error of HTTP::Request::Suply on my box. but it works on travis with perl6 nom 14:56
14:59 tzekid left 15:00 colomon joined
Skarsnik hm weird 15:02
AlexDaniel alright I opened this doc issue, feel free to leave your comments if you understand what's going on there: github.com/perl6/doc/issues/1429 15:03
Skarsnik weird 15:04
I know, I am very usefull 15:05
15:05 colomon left
ufobat :D 15:05
AlexDaniel how can I match some generated string from a regex? Like, if I don't want to create a temp variable beforehand
15:06 tzekid joined
AlexDaniel <{/ … /}> ? 15:06
<{my $foo = …; / $foo /}> ?
geekosaur <$var> 15:10
AlexDaniel geekosaur: more information? 15:11
geekosaur oh, I think I was answering ane arlier questiona ctually 15:12
AlexDaniel well you can answer any question :)
<$var> seems to be doing EVAL too
geekosaur btw that <{ }> looks like an LTA internal error stopped things, not intended behavior 15:13
japhb Someone was asking about Terminal::Print and the associated examples? (I updated and/or wrote several of those, maybe I can help?) 15:18
timotimo, nadim: ^^ 15:21
.tell timotimo It looks like from the #perl6 logs that you were asking about creating a TUI with Terminal-Print. I was about to launch into creating high-level widgets for Terminal-Print (e.g. edit boxes, radio buttons, etc.) a few months ago, but got stuck on problems with Uni-level raw input in MoarVM (which blocked e.g. decoding cursor keys and mouse movement). Those problems might have been fixed now after jnthn++'s considerable work, so might be worth pickin 15:24
yoleaux japhb: I'll pass your message to timotimo.
15:25 skids left
andrzejku araraloren 15:27
15:34 kyan joined 15:35 kyan left, kyan joined 15:38 bioexpress joined
bioexpress Hello, can somebody say me how I could describe the change from `function( %options )` to `function( *%options )`? 15:41
AlexDaniel oh, what a great question! 15:42
masak bioexpress: you mean what to call the '*'? or what it means?
bioexpress How to call it
AlexDaniel I wondered exactly the same thing some time ago
masak bioexpress: "slurpy"
m: sub f1(%options) { %options<foo> }; say f1({ :foo<1> })
camelia 1
masak m: sub f1(*%options) { %options<foo> }; say f1(:foo<1>) 15:43
camelia 1
bioexpress Form hash to slurpy?
masak ...and there you see the effects of adding the star
bioexpress: from hash parameter to slurpy hash parameter
bioexpress Thx!
masak phz!
er, *pzh! :) 15:44
(short for "пожалуйста") :P
15:44 BenGoldberg left
AlexDaniel masak: my question is slightly different! :) How can I have a space in a key when doing this? 15:45
you can't just slam a pair into it
m: sub foo(*%z) { dd %z }; foo :5a, :5b, foobar => ‘world’
camelia {:a(5), :b(5), :foobar("world")}
AlexDaniel m: sub foo(*%z) { dd %z }; foo :5a, :5b, ‘foo bar’ => ‘world’
camelia Too many positionals passed; expected 0 arguments but got 1
in sub foo at <tmp> line 1
in block <unit> at <tmp> line 1
15:48 colomon joined
AlexDaniel m: sub foo(*%z) { dd %z }; foo :5a, :5b, |(‘hello world’ => ‘world’) 15:48
camelia {:a(5), :b(5), "hello world" => "world"}
AlexDaniel well, nevermind
15:48 khw joined 15:49 bioexpress left
japhb AlexDaniel: Perl 6 practically makes it a policy not to prevent you from being too clever for your own good. ;-) 15:52
AlexDaniel japhb: which issue are you talking about exactly? :)
japhb AlexDaniel: Oh, just seeing the "But I need a space in the key of my slurpy hash!" unfold. :-) 15:54
Just reminds me of lots of other place where Perl 6 goes for subtle dissuasion rather than outright stopping you. 15:55
15:56 araraloren left 16:00 kaare__ joined
masak japhb: "not to prevent you from being too clever for your own good" -- what a refreshing way of putting it 16:02
16:03 kaare_ left
masak japhb: lately I've been thinking of all the (ES.next, but still) linter rules I want in a project to prevent me from doing silly things. JavaScript is a really neat language when propped up from all sides by just the right linter rules. 16:04
16:04 S007 left
masak but Perl 6 has almost the opposite policy from those linter rules. it allows you to be as clever as you want, and it just goes along with it. 16:04
16:05 S007 joined
masak maybe that's being unfair to Perl and Perl 6, though. in the end what it's doing can be called consistency in all things. not putting up arbitrary limits. 16:07
16:15 Cabanossi left, colomon left
jnthn There are things that Perl 6 outright disallows that you'd catch with a linter in JavaScript though. Such as using undeclared variables. :) 16:16
AlexDaniel m: no strict; $x = ‘but I really want to!’; say $x # :) 16:17
camelia but I really want to!
16:17 Cabanossi joined
jnthn It's your foot :P 16:19
AlexDaniel not after I put it into a module :P :D 16:20
16:22 GhostK joined
japhb AlexDaniel: Are you going to call that "FootGun:ver<666>" ? 16:25
16:39 colomon joined 16:44 colomon left 16:49 kyan left 16:55 bvcosta joined
timotimo well done japhb 16:57
16:57 yati joined
yoleaux 15:24Z <japhb> timotimo: It looks like from the #perl6 logs that you were asking about creating a TUI with Terminal-Print. I was about to launch into creating high-level widgets for Terminal-Print (e.g. edit boxes, radio buttons, etc.) a few months ago, but got stuck on problems with Uni-level raw input in MoarVM (which blocked e.g. decoding cursor keys and mouse movement). Those problems might have been fixed now after jnthn++'s considerable 16:57
work, so might be worth pickin
timotimo did you make the rpg tui?
but yeah, should work now 16:58
the inpup stuff
network connection barely exists here
typing blind
yati Hello people, I want to define the infix multiplication operator for my class: what am I doing wrong here? paste.scsys.co.uk/564708 17:00
geekosaur yati, most operators are not methods 17:01
sub infix:<*> (Interval:D $a, Interval:D $b) { ... } 17:02
yati Ah
geekosaur actually, correct that
multi sub infix:<*> (Interval:D $a, Interval:D $b) { ... }
yati What do the :D's do?
geekosaur and it will be selected based on types. (the :D means it won't be selected for undefined values)
17:03 wamba left
geekosaur you could omit them but the error messages tend to be worse because they leak implementation details that a caller generally doesn't/shouldn't care about 17:03
yati geekosaur: that worked thanks! 17:04
geekosaur whereas with the :D-s they get an error pointing at the place where tey used the operator, not one pointing into your implementation
17:04 pmurias joined
yati geekosaur: where do I find info about :D and other such "extended typing operators" or whatever they are called? :) 17:05
geekosaur (this is not just location but also things like the variable names involved, with the type smiley version it will identify the variable the user passed in, otherwise it names one of your internal variables or parameters)
pmurias masak: if Perl 6 forced a linter and coding style on you it wouldn't be a Perl 17:06
17:06 grondilu_ left
geekosaur docs.perl6.org/type/Signature#Cons...ned_Values 17:07
17:07 grondilu joined
yati geekosaur: that is pretty neat, I recall there being a term to describe typing with constraints on values of the types, can't remember it right now. 17:07
Thanks for the link
17:07 pharv joined
Geth ecosystem: fb9e08c1c9 | grondilu++ (committed using GitHub Web editor) | META.list
adding Symbol module
 17:15
R0b0t1 zengargoyle: I suppose that is a solution
grondilu should Travis deal with dependencies defined in META6.json? 17:16
*shouldn't
grondilu realises that's much to ask, probably needs to specifie dependencies in .travis.yml or something 17:18
17:30 Cabanossi left 17:32 Cabanossi joined 17:33 bvcosta left
yati Okay, so thanks for geekosaur, I managed to get this: paste.scsys.co.uk/564709 -- now the question is, if class A extends B, like in my example, how do I wrap B's constructor in A? 17:34
The error I get now concerns the fact that I cannot call method parallel() on Interval values, since the argument is a Resistance value
AlexDaniel greppable6: flatmap 17:35
greppable6 AlexDaniel, gist.github.com/aa5bae6dfcf4096b9e...37b52f4696
geekosaur yati, in general you don't define method new directly. you use the MOP's default new which builds superclasses properly etc. 17:37
in this case you would define submethod BUILD, which gives you a structurally complete object including superclasses. (if you define your own new, you have to build superclasses yourself and even create the object yourself; see the "bless" method. it is *not* the same "bless" as in perl 5!) 17:39
docs.perl6.org/language/classtut#Constructors 17:40
docs.perl6.org/language/objects#Ob...nstruction
geekosaur wonders if there should be an easily-overridden warning about defining method new directly... 17:44
17:45 abraxxa joined 17:48 kaare__ left 17:50 abraxxa left 17:53 kaare__ joined
zengargoyle .botsnack 17:55
synopsebot6 om nom nom
yoleaux :D
17:59 BenGoldberg joined 18:00 statisfiable6 left 18:03 statisfiable6 joined, ChanServ sets mode: +v statisfiable6
yati geekosaur: I see, but I still can't figure out how to do this: Interval exposes an alternate ctor, Interval.from-center-percent(), which I want to delegate to from Resistance.new(). I tried making a submethod BUILD (:$val, :$tol) that then calls Resistance.from-center-percent(), but that does not seem to work. 18:13
So want: Resistance.new(10, 1) to call return a Resistance object that has the underlying Interval built with Interval.from-center-percent(10, 1) 18:14
s/call// 18:15
18:16 kaare__ left 18:18 kaare__ joined
SmokeMachine m: my \undef = 42; say undef 18:26
camelia 5===SORRY!5=== Error while compiling <tmp>
Unsupported use of undef as a value; in Perl 6 please use something more specific:
an undefined type object such as Any or Int,
:!defined as a matcher,
Any:U as a type constraint,
Nil as the ab…
geekosaur yati, for that you would indeed need method new (but you must do more than just delegating, see the second link I sent) or you would need to replace submethod BUILDALL 18:31
18:31 itaipu joined
AlexDaniel m: class Foo { has $.bar = !!! ‘bar is required’ }; Foo.new(:42bar) 18:34
camelia ( no output )
AlexDaniel m: class Foo { has $.bar = !!! ‘bar is required’ }; Foo.new()
camelia bar is required
in method at <tmp> line 1
in block <unit> at <tmp> line 1
AlexDaniel that's pretty cool
I didn't know you could use !!! like this. Although this example would work just fine with die also 18:35
grondilu is there a difference between C<callwith> and C<samewith>?
18:37 miro_ joined
grondilu goes to read perl6.party/post/Perl6-But-Heres-M...with-Maybe 18:38
18:41 HoboWithShotgun joined
HoboWithShotgun good localtime you wonderful people 18:41
normally i use "beautiful" rather than "wonderful", but I was at several YAPCs. I know... I know.
mst sulks 18:42
18:44 wamba joined
HoboWithShotgun that's good. as long as you do that, you don't talk 18:44
;-)
18:44 AlexDaniel is now known as AlexWithFootgun
zengargoyle mst: do you have any code up that shows your "shipping a CUR" like thing? 18:44
mst zengargoyle: code? 18:47
I mean the point here is basically you can deploy via tar or rsync (or a fat .deb) 18:48
zengargoyle an example that i can look at... my mind goes to fatpack and Object::Remote and tak and the like. 18:49
18:50 andrzejku left
zengargoyle i.e. do you have a sh wrapper that does perl6 -Imycur -e 'Module::run' or something to get that almost fully contained like a fatpack. 18:50
mst in perl5, all I mean is "including the local::lib in the tarball"
HoboWithShotgun rakudobot: say "äöü".encode("UTF-8").decode("latin1"); 18:51
meh, what was the bots' name again? 18:52
AlexWithFootgun m:
evalable6
mst zengargoyle: but, yeah, you can do that, or stick a 'use lib' in the script, or whatever
AlexWithFootgun m: say "äöü".encode("UTF-8").decode("latin1"); 18:53
camelia Can not decode a utf-8 buffer as if it were latin1
in block <unit> at <tmp> line 1
18:53 eroux joined
HoboWithShotgun yeah, that's what i get too 18:53
what am i doing wrong? 18:54
18:55 zakharyas joined
mst trying to do something that doesn't make sense? 18:56
zengargoyle i was just wondering if you had an automated process for the "all deps but core installed to this cur" and wrap this 'module/script' up like fatpacker. or if it's still a sorta manual process. 18:57
18:57 zakharyas left 18:58 zakharyas joined 18:59 shu joined
mst I've not particularly automated it 19:00
honestly though "use a user that can't write to the global CUR to install everything, hey presto now ~/.perl6 contains what you wanted" is not a terrible approach
HoboWithShotgun possibly, mst. as i understand it string literals are utf-8, right. the windows shell talks latin1. and so my output is borked
what do i do to fix this. 19:01
zengargoyle with fresh perl6 that doesn't have any needed modules already installed....
or removing the site CUR somehow...
19:02 Cabanossi left
geekosaur HoboWithShotgun, chcp 65001 19:02
although I thought it was supposed to check that and translate (possibly to -c8)
zengargoyle and zef can install to a specific CUR.... i'm just not sure there's a core-only sort of option at the moment to ensure every dep needed goes into CUR...
19:02 shu left 19:03 zakharyas left, Cabanossi joined
mst zengargoyle: for perl5 I use a fresh build a lot of the time too 19:04
"I could be clever, or I could use a perl with an empty site directory that the build can't write to"
19:04 darutoko left
mst this being systems, clever is the wrong answer 19:04
zengargoyle ah, *nods*
19:05 zakharyas joined
HoboWithShotgun geekosaur: now it's broken twice 19:05
geekosaur huh?
HoboWithShotgun with chcp65001: Negasphäre, without: Negasphäre, should be: Negasphäre 19:06
zengargoyle i've just used the cpanm flags for *everything not core* goes into this ./lib
19:07 committable6 left, greppable6 left
mst the cpanm -L flag uses lib::core::only which was extracted from local::lib because I don't entirely trust it 19:07
HoboWithShotgun interesting. with and without actually is different but comes out identical in my client 19:08
geekosaur normalized and non-normalized?
zengargoyle spinning up a frest perl or whatever is probably better...
19:08 greppable6 joined, committable6 joined, ChanServ sets mode: +v greppable6, ChanServ sets mode: +v committable6
mst -L is useful for a quick check 19:08
for build work I like a fresh perl 19:09
geekosaur (ä can be \x00e5 or \x0065\x0308; the former is the normalized one per Unicode spec.) 19:10
(and feeding a non-normalized one in might be just the thing to trigger odd issues, given that windows is not great at locales) 19:11
zengargoyle yeah, makes sense... just a little too much effort for the things i was doing where just trying it once to see if it works was pretty much "it works, i'm done." :) 19:12
19:13 itaipu left
geekosaur er \x0061 19:13
no 0065
zengargoyle hehe, because the whole reason i was fatpacking was because they wouldn't change/upgrade/add to the perl that was there... :P 19:14
and it was shared over NFS, so same perl everywhere. :) 19:15
19:16 andrzejku joined 19:19 itaipu joined 19:28 itaipu left 19:35 itaipu joined 19:37 nhywyll joined 19:52 TEttinger joined 20:01 Cabanossi left
japhb .tell timotimo Yes, I did the RPG TUI demo. And I have a present for you: github.com/ab5tract/Terminal-Print...es/tris.p6 (Note the branch.) 20:02
yoleaux japhb: I'll pass your message to timotimo.
20:03 Cabanossi joined 20:05 colomon joined, andrzejku left 20:06 skids joined 20:09 colomon left 20:11 skids left
ufobat when i install a module with zef, it will install the dependencies of the module as well. (if the tests of the module fail, zef wont install nothing at all) but that leads me to the question, what if the tests of the module run another perl6, are the depenencies of the module available to this perl6 invokation. i think they aren't. and thing this has recently been changed, could that be? 20:16
20:18 rindolf left
ufobat s/and thing/and i think/ 20:18
20:24 pharv left 20:37 clarkema joined
masak pmurias: re "if Perl 6 forced a linter on you" -- yes, my point exactly. Perl culture is anti-linter-rules, not because linters are a bad idea, but because it's against the Liberté ideals of Perl. 20:40
20:41 andrzejku joined
nadim . tell japhb Hi, I wrote this github.com/nkh/P6-Data-Dump-Tree/b...olding.pl. It's a begining a of NCurses appliction to display data. The idea was to make a data viewer that could be used by scripts. The folding is handled by the dumper, it just needs ome nice UI. if you'd like to make something or even just translate this example I'll put it in the distro. 20:41
20:47 Skarsnik_ joined
nadim masak: I want linter forced on me, looking forward to a Perl::Critic like for P6 20:49
forced by me ;)
20:49 Skarsnik left 20:50 zakharyas left 20:53 ufobat left 21:07 mr-foobar joined 21:08 skids joined 21:09 mr-fooba_ left 21:10 kaare_ joined 21:11 cog_ joined 21:12 itaipu left, kaare__ left, BenGoldberg left 21:13 cog__ left 21:15 Skarsnik_ left, Geth left 21:16 Geth joined 21:19 itaipu joined 21:20 andrzejku left 21:21 llfourn left 21:22 skids left
El_Che moritz: I'll keep you posted when I see it arriving in Safari 21:31
21:34 itaipu left 21:36 itaipu joined 21:37 nhywyll left 21:43 eroux left 21:56 itaipu left 22:00 yati left 22:01 Cabanossi left 22:03 Cabanossi joined 22:11 jaldhar_ left 22:12 jaldhar_ joined, itaipu joined 22:14 Rawriful joined 22:30 Cabanossi left 22:32 espadrine left, Cabanossi joined 22:43 miro_ left, clarkema left 22:46 vendethiel left 22:47 HoboWithShotgun left 23:01 benjikun joined 23:02 rubio_terra joined
benjikun Why is the build failing? github.com/perl6/gtk-simple 23:02
I've installed it successfully on the latest version
23:05 BenGoldberg joined 23:06 pmurias left
geekosaur looks like a bug in zef when it was last updated which, note, was 6 months ago) 23:07
benjikun hm 23:10
23:19 Rawriful left 23:26 deep-book-gk_ joined
timotimo i just restarted the gtk-simple build on travis-ci 23:26
yoleaux 20:02Z <japhb> timotimo: Yes, I did the RPG TUI demo. And I have a present for you: github.com/ab5tract/Terminal-Print...es/tris.p6 (Note the branch.)
timotimo i don't have access to the appveyor because azawawi hosts it
though i have my own appveyor of that project 23:27
geekosaur the error there looks no worthwhile (strawberry perl install complaining about unsigned package)
timotimo bleh, that thing still? 23:28
i feel like i fixed that like five times already
ci.appveyor.com/project/timo/gtk-s...sxq0876dxu - works here it looks like
23:29 deep-book-gk_ left 23:36 nadim left 23:46 Cabanossi left
benjikun is Larry Wall in this irc? lol 23:46
23:47 Cabanossi joined
geekosaur sometimes 23:49
23:53 wamba left, pharv joined
timotimo it's quite possible that we have to bump the lowest version for gtk-simple 23:58
since zef just doesn't work on such an old version .. or something 23:59
alternatively, put lost of backwards-compatibility hacks into zef :D

Powered by Cro and the Raku® Programming Language.

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

Thank you!