»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.
buggable 🎺🎺🎺 It's time for the monthly Accidental /win Lottery 😍😍😍 We have 1 ballots submitted by 1 users! DRUM ROLL PLEASE!... 00:00
And the winning number is 5! Congratulations to Ulti! You win a roll of duck tape!
00:10 patrickb left
holyghost eco: roll of duck tape 00:16
yoleaux 31 Oct 2018 11:15Z <Zoffix> holyghost: you should write an Advent article for your modules. Sign up: github.com/perl6/mu/blob/master/mi...8/schedule
buggable holyghost, Nothing found
yoleaux 31 Oct 2018 11:21Z <Zoffix> holyghost: better link: perl6.party/post/Perl-6-Advent-Cal...or-Authors
jnthn rouking: Oh! It's becuase you're using `rule`, which inserts <.ws> calls for you automatically
rouking: And the default `<.ws>` matches vertical whitespace too
Try token ws { <!ww> \h* } or some such 00:17
00:18 cognominal-p6 left, cognominal-p6 joined 00:19 p6bannerbot sets mode: +v cognominal-p6
jnthn sleep 00:21
'night o/
timotimo gnite o/
00:24 atMaxRink21 joined, atMaxRink21 left 00:31 rindolf left
rouking jnthn: That's progress--it at least tries to match another line, but it fails immediately 00:45
what is <!ww> there, by the way? 00:46
00:48 cognominal-p6 left
leont It's a zero-length assertion 00:49
<ww> checks if you're in the middle of a word, so <!ww> checks the opposite
Not entirely sure what the difference between <wb> and <!ww> would be though 00:50
rouking This is giving me much more grief than I had hoped 00:52
leont I tend to use token or regex instead of rule, that gives less surprises 00:56
(token doesn't do backtracking, so it's faster but there are situations where regex parses things that token can't)
rule tends to be useful when parsing a freeform grammar, and quite a lot of parsing problems aren't that. It just happens to be that most programming languages that aren't python are such freeform grammers. 00:57
timotimo yeah, rule can be very surprising indeed 00:58
rouking Alright, I will just try using token since I know exactly what my format will be 00:59
hmm, still fails on the second line... 01:00
01:09 kaare_ left, kaare_ joined 01:10 p6bannerbot sets mode: +v kaare_ 01:27 fake_space_whale joined 01:28 p6bannerbot sets mode: +v fake_space_whale 01:32 sno left 01:34 sno joined, p6bannerbot sets mode: +v sno 02:05 leont left 02:06 molaf left 02:19 molaf joined 02:20 p6bannerbot sets mode: +v molaf
Xliff \o 02:22
Just managed to bring down a piece of code generating segfaults.
Will generate segfault even with SPESH and JIT disabled. 02:23
02:25 [Sno] joined, donaldh joined, p6bannerbot sets mode: +v donaldh 02:26 p6bannerbot sets mode: +v [Sno] 02:28 sno left
Xliff OK. I can trigger the segfault with what appears to be regularity. 02:29
02:29 [Sno] left 02:33 sno joined 02:34 p6bannerbot sets mode: +v sno
Xliff Segfaulting issue has been submitted as #2455 02:41
Ping me if anyone would like support to help debug this. I will continue my attempts from here. 02:42
AlexDaniel e: chdir ‘sandbox’; run <git clone github.com/Xliff/p6-GtkPlus.git> 02:44
evalable6 Cloning into 'p6-GtkPlus'...
AlexDaniel ahhhhhh that's gtk stuff again
02:46 kurahaupo left
AlexDaniel Xliff: does it require an X server or anything like that? 02:46
I'm wondering if running it bisectable makes sense
02:46 sno left, kurahaupo joined
Xliff AlexDaniel: No. It's a code generator, so it should NOT. 02:47
And yes, AlexDaniel ... the GTK stuff is my current bone! :p
02:47 p6bannerbot sets mode: +v kurahaupo
AlexDaniel ok, then… 02:47
e: chdir ‘sandbox/p6-GtkPlus’; run <git checkout a6d22e9427503ed9274386a2e3b708315ea5403d>
evalable6 Note: checking out 'a6d22e9427503ed9274386a2e3b708315ea5403d'.

You are in 'detached HEAD'…
AlexDaniel, Full output: gist.github.com/b39762b2547b937cbe...77150ef8bd 02:48
AlexDaniel e: chdir ‘sandbox/p6-GtkPlus’; run <perl6 -Ilib -Iscripts scripts/ui-to-code.pl6>
evalable6 (exit code 1) ===SORRY!===
Could not find Data::Dump::Tree at line 7 in:
AlexDaniel, Full output: gist.github.com/61841e0c3d87cf20c2...702607d1bf
AlexDaniel argh
Xliff Ack! Ditch that line.
Just comment it.
Or would you rather I recommit?
AlexDaniel it's fine 02:49
02:50 MasterDuke left
AlexDaniel e: chdir ‘sandbox/p6-GtkPlus’; run <perl6 -Ilib -Iscripts scripts/ui-to-code.pl6> 02:50
evalable6 (exit code 1) ===SORRY!===
Could not find Pluggable at line 3 in:
AlexDaniel, Full output: gist.github.com/163660444175d5aaaf...f3bc28605c
Xliff I'm horrible at writing bug reports.
Updating. 02:51
AlexDaniel what's Pluggable?
oh so that's a module… ok hmm
Xliff Yeah. Kinda required for that bit.
AlexDaniel: What's the best way to compile rakudo so you can get decent backtrace info out of it? 02:53
AlexDaniel Xliff: maybe run it with perl6-valgrind-m ? You won't get a nice stacktrace unless you Configure it with --moar-option=--debug=3 I think 02:54
Xliff Ahhh... THAT'S what I am looking for. 02:56
02:56 sno joined 02:57 p6bannerbot sets mode: +v sno
AlexDaniel Xliff: I have no idea what that is: gist.github.com/476fa136f6162b8501...065c0e5ec2 03:00
Xliff O fer crissakes. 03:02
It looks like the bot doesn't have permissions to open the lib dir? 03:03
AlexDaniel e: .say for dir ‘/home/bisectable/git/whateverable/sandbox/p6-GtkPlus/sandbox/lib’ 03:04
evalable6 (exit code 1) Failed to get the directory contents of '/home/bisectable/git/whateverable/sandbox/p6-GtkPlus/sandbox/lib': Failed to open dir: 2
in block <unit> at /tmp/teNQ4qNmxf line 1
AlexDaniel huh. 03:05
e: .say for dir ‘sandbox/p6-GtkPlus/sandbox/lib’
evalable6 (exit code 1) Failed to get the directory contents of '/home/bisectable/git/whateverable/sandbox/p6-GtkPlus/sandbox/lib': Failed to open dir: 2
in block <unit> at /tmp/B4mmDOnz8r line 1
AlexDaniel what's 2
no such file or directory?
Xliff: ah, correct! The path is all wrong 03:06
how is it constructing this path?
Xliff Is that Pluggable constructing the path? 03:07
AlexDaniel 🤷 03:09
03:44 cognominal-p6 joined 03:45 p6bannerbot sets mode: +v cognominal-p6 03:48 lizmat left
Xliff OK. I am going to see if I can recompile rakudo and get a backtrace. 03:50
What's the best way to get moar, nqp and rakudo to compile after a recent rakudobrew? 03:51
03:56 cognominal-p6 left, cognominal-p6 joined, moonythevampire is now known as moonythedustpile 03:57 p6bannerbot sets mode: +v cognominal-p6 04:08 sno left 04:09 sno joined
Xliff m: '🤷'.uniname.say 04:09
camelia SHRUG
04:09 cognominal-p6 left, p6bannerbot sets mode: +v sno 04:18 sauvin joined, p6bannerbot sets mode: +v sauvin
donaldh Couple of unassuming PRs if anyone has a moment: github.com/rakudo/rakudo/pull/2454 and github.com/rakudo/rakudo/pull/2457 04:39
Xliff Got a backtrace of the segfault. I've updated github.com/rakudo/rakudo/issues/2455 04:43
AlexDaniel donaldh: thanks! 04:56
Xliff: cool!
04:56 molaf left
Xliff AlexDaniel: Yes. And now I realize how badly some of this was written. :/ 04:57
This is what happens when you code way past your bedtime.
AlexDaniel what's bedtime 04:58
05:00 fake_space_whale left
AlexDaniel declares it now 05:09
05:21 kurahaupo left, kurahaupo joined 05:22 p6bannerbot sets mode: +v kurahaupo
SmokeMachine is anyone here used to CQRS? could comment my first attempt to do it with Red? github.com/FCO/Red/blob/master/exa...s/index.p6 05:23
05:31 donaldh left 05:35 fake_space_whale joined, p6bannerbot sets mode: +v fake_space_whale 05:39 fake_space_whale left 05:45 noganex left 05:46 noganex joined, p6bannerbot sets mode: +v noganex 06:21 curan joined, p6bannerbot sets mode: +v curan 06:40 troys left 07:00 kurahaupo left, kurahaupo joined 07:01 p6bannerbot sets mode: +v kurahaupo 07:03 MilkmanDan left 07:05 cognominal-p6 joined 07:06 p6bannerbot sets mode: +v cognominal-p6
holyghost .tell Zoffix my code does not compile yet, if you like I can write something up for it on Advent as it's open source and consistent for the math formulas 07:06
yoleaux holyghost: I'll pass your message to Zoffix.
07:07 Summertime2 is now known as Summertime, cognominal-p6 left, MilkmanDan joined, p6bannerbot sets mode: +v MilkmanDan, cognominal-p6 joined 07:08 p6bannerbot sets mode: +v cognominal-p6
Geth doc/master: 6 commits pushed by (Luis Balderas Ruiz)++, (Juan Julián Merelo Guervós)++ 07:24
07:24 jmerelo joined 07:25 p6bannerbot sets mode: +v jmerelo 07:30 cognominal-p6 left 07:53 dalek left 08:01 cognominal-p6 joined 08:02 p6bannerbot sets mode: +v cognominal-p6 08:08 cognominal-p6 left 08:09 cognominal-p6 joined 08:10 p6bannerbot sets mode: +v cognominal-p6 08:12 kurahaupo_ joined 08:13 p6bannerbot sets mode: +v kurahaupo_ 08:14 kurahaupo left 08:16 Geth left, zakharyas joined 08:17 p6bannerbot sets mode: +v zakharyas
buggable New CPAN upload: Bayes-Learn-0.1.8.tar.gz by HOLYGHOST modules.perl6.org/dist/Bayes::Learn...:HOLYGHOST 08:21
New CPAN upload: Mathx-Stat-0.1.10.tar.gz by HOLYGHOST modules.perl6.org/dist/Mathx::Stat:...:HOLYGHOST
08:30 ufobat joined, p6bannerbot sets mode: +v ufobat
buggable New CPAN upload: Bayes-Learn-0.1.9.tar.gz by HOLYGHOST modules.perl6.org/dist/Bayes::Learn...:HOLYGHOST 08:31
08:32 dakkar joined, p6bannerbot sets mode: +v dakkar
holyghost .tell Zoffix I do not have enough knowledge about the module system except that I read docs.perl6.org/language/modules and set the PERL6LIB path, Mathx::Stat should compile, I can Advent it if you like, AFAIK tests of Mathx::Stat are useless as they are just math formulas and can break an download/install 08:36
yoleaux holyghost: I'll pass your message to Zoffix.
08:39 cognominal-p6 left 08:42 lizmat joined, p6bannerbot sets mode: +v lizmat 08:43 [Sno] joined, sno left 08:44 p6bannerbot sets mode: +v [Sno], [Sno] left
ufobat jnthn++ 08:46
08:46 rindolf joined 08:47 p6bannerbot sets mode: +v rindolf
holyghost My "the legend of kyrandia game" is available at shameen.ddns.net, it's programmed in GNU smalltalk, the site is not reachable from everywhere 08:47
It's a start with a blitter as I said before
08:48 cognominal-p6 joined
holyghost graphics will be licensed from Westwood studios 08:48
08:48 p6bannerbot sets mode: +v cognominal-p6 08:49 robertle joined 08:50 p6bannerbot sets mode: +v robertle
holyghost time for a drink :-) thx ufobat 08:50
ufobat a drink at 10am?
holyghost yes I work at night mostly 08:51
ufobat what about a gin tonic? :) 08:52
jmerelo Just plotted the contributions per month for the perl6/doc repo raw.githubusercontent.com/JJ/TPF-G...ntropy.png
holyghost I have a terrific phase, I start coding at 4am
gin tonic is declasse :-)
jmerelo Number of authors and contributions has picked up a bit since the Aug-Sep slump, but it's far from peaks. Let's see if this weekend's squashathon helps 08:53
08:54 cognominal-p6 left
jmerelo holyghost: great work, too, on AI stuff. Also, you can test math formulae too. Advent article would be great 08:54
08:54 cognominal-p6 joined
holyghost ok 08:54
08:55 p6bannerbot sets mode: +v cognominal-p6
holyghost the AI-Agent is cool, I know 08:55
I still need to compile, but I might have to refactor 08:56
I have a HashAgent and Agent superclass
They have to go in concert with th rest of the code 08:57
It's just an override on an Actor class with a "method dispatch($message, %optargs)" 09:03
It should be hackable by perlbies
s/perlbies/gnubies, newbies, perl wizards and so on 09:06
09:09 sno joined, p6bannerbot sets mode: +v sno 09:10 sno left 09:11 kdr2 left 09:18 kurahaupo_ left, kurahaupo joined, kurahaupo left 09:19 kurahaupo joined, p6bannerbot sets mode: +v kurahaupo 09:24 kensanata joined, p6bannerbot sets mode: +v kensanata
holyghost .tell Zoffix I've posted a small addon to Advent, AFAIK I can write a HTML page on it, I will do that later 09:31
yoleaux holyghost: I'll pass your message to Zoffix.
09:36 ufobat left 09:40 kdr2 joined 09:41 p6bannerbot sets mode: +v kdr2 09:43 sno joined, p6bannerbot sets mode: +v sno 09:49 sena_kun joined 09:50 p6bannerbot sets mode: +v sena_kun
holyghost Advent is showing up december 2017 in my browser 10:00
jmerelo holyghost: what do you mean?
holyghost I get Bonus Xmas 􏿽x96 Concurrent HTTP Server implementation and the scripter􏿽x92s approach 10:01
December 25, 2017ramiroencinas Leave a comment
jmerelo holyghost: if you want to propose something for the calendar, you have to modify this file github.com/perl6/mu/blob/master/mi...8/schedule
holyghost lol
jmerelo: I did
At the end of the month I'll write something about Mathx::Stat, it's in if you like 10:02
I am going to extend it
jmerelo holyghost: Er. It's not.
holyghost: if you want to have it for the 24th, just add your nick and title behind that day to "claim" it. 10:03
holyghost ok
I can write something about the package as it is now 10:04
I'll figure out the web editor later
jmerelo holyghost: if you mean the Wordpress blog, that comes after. You'll need to obtain privs from Zoffix or one of the other admins (if you don't have them already). Anyway, claiming goes first, to know what we have
holyghost ah now I understand
It's priviledged from the github schedule file ? 10:05
jmerelo holyghost: kinda. It's done by hand. Zoffix or someone will contact you for your Wordpress account after you claim a day.
holyghost ok, I understand
I just need to dump an HTML page with zoffix's tool then 10:06
jmerelo holyghost: yep, when your article is ready.
holyghost ok
I have to talk about a simple statistics package :-) 10:07
jmerelo holyghost: that's good :-)
holyghost It was programmed for speed. done. :-)
I'll talk about games then 10:08
It was made for Game::Markov and Bayes::Learn 10:09
10:09 sno left
holyghost ok, I'll figure it out at the end of the month :-) 10:09
jmerelo holyghost: great idea!
holyghost: you can just claim the day and say "TBA"
holyghost I know
I'll make a second entry later on when I get my chaos theory statistics read and sorted out 10:10
jmerelo holyghost: great!
holyghost upgrades :-)
Now there's a correlation dimension which is a popular chaos theory formula 10:11
It needs refactoring though (with addition of classes I mean, I write everything in modules) 10:12
.pm6 :-)
timotimo jmerelo: DM-ping, but also AFK now 10:16
holyghost jmerelo: basically you just calculate with lists of probabilities until you figure out your variance in a game. Oh! I am posting my article :-)
10:27 ufobat joined 10:28 p6bannerbot sets mode: +v ufobat
jmerelo timotimo: sorry, what? 10:31
10:31 sergot left, SourceBaby left, dalek joined, ChanServ sets mode: +v dalek, Geth joined, ChanServ sets mode: +v Geth, p6lert_ joined, synopsebot_ joined, ChanServ sets mode: +v synopsebot_, p6lert left, synopsebot left 10:32 SourceBaby joined, ChanServ sets mode: +v SourceBaby, niko joined, p6bannerbot sets mode: +v niko, p6bannerbot sets mode: +v dalek, synopsebot_ left, p6bannerbot sets mode: +v Geth, p6bannerbot sets mode: +v p6lert_, p6lert_ left, synopsebot joined, ChanServ sets mode: +v synopsebot, p6lert joined
Geth ecosystem: ccworld1000++ created pull request #413:
CCLog Simple and lightweight cross-platform logs
10:32 p6bannerbot sets mode: +v SourceBaby 10:33 p6bannerbot sets mode: +v synopsebot, p6bannerbot sets mode: +v p6lert 10:35 HaraldJoerg joined 10:36 p6bannerbot sets mode: +v HaraldJoerg
Geth ecosystem: dee60699db | CC++ (committed using GitHub Web editor) | META.list
CCLog Simple and lightweight cross-platform logs

easy-to-use simple learning, and support for multiple languages, such as C, C++, Perl 6, shell, Objective-C .
ecosystem: bd5ff55660 | (Elizabeth Mattijsen)++ (committed using GitHub Web editor) | META.list
Merge pull request #413 from ccworld1000/patch-1

CCLog Simple and lightweight cross-platform logs
10:49 yqt joined 10:50 p6bannerbot sets mode: +v yqt
holyghost .tell Zoffix I cannot use your schedule tool, but here's my printed HTML page for Advent if you like (for Mathx::Stat): pastebin.com/x3mENvQE 10:53
yoleaux holyghost: I'll pass your message to Zoffix.
holyghost I hopy you like it, C&C welcome
chsanch japhb: You can install it skipping the tests: zef install --/test croOA
holyghost s/hopy/hope
10:57 isBEKaml joined, p6bannerbot sets mode: +v isBEKaml, sno joined
jmerelo holyghost: the articles for Advent should be a bit longish... Also include a few code examples. Sometimes they are written in a festive-mock-Santa mood, but that's optional. 10:57
10:58 p6bannerbot sets mode: +v sno
jmerelo holyghost: it's a good start anyway. Also, remember to claim a particular day. 10:58
holyghost jmerelo: ok, I'll see what I can do, problem is you cannot say much about core statistics 11:02
11:02 sno left
holyghost I claimed 30 nov 2018 (another login, my theholyghost login has been faulted) 11:03
jmerelo holyghost: Advent starts by Dec 1st 11:04
holyghost 30 dec 2018 I meant then
jmerelo holyghost: it ends by Dec 24th. Sometimes there's a bonus on the 25th, but that's it... 11:05
11:05 kdr21 joined, isBEKaml left
holyghost I've updated the schedule file to 30: 11:05
11:06 p6bannerbot sets mode: +v kdr21
jmerelo holyghost: you can't do that. Please just claim one of the available days. 11:06
holyghost ok 11:07
11:07 Spooktober left, grumble joined, p6bannerbot sets mode: +v grumble, kdr2 left
holyghost I'm not too good at fixing web edits but Zoffix'll have to post the article anyway 11:08
It's just a piece of HTML file
I'll extend anyhow later on
problem is I am flooding .tell 11:09
jmerelo holyghost: that's not how it works. 1. You claim a day 2. You get permission to post in the wordpress blog 3. You schedule the post for that day, and leave it for others to check it out and edit (and you can edit it yourself) 4. It's published on the scheduled day.
holyghost I knew that 11:11
1. I claimed 30 dec 2018 2. I'll post my text/html file if Zoffix concurs 11:12
with my zork9 login
I added further numbers than 25 (dec 2018) 11:13
but I'll add-on sooner :-) 11:14
timotimo there will not be advent posts later than the 25th
11:15 sno joined, p6bannerbot sets mode: +v sno
holyghost timotimo: ok, as I said, I'll have my post ready even for dec 25 11:16
I release often, you know :-) 11:17
timotimo yeah, i see your releases show up on irc every day 11:18
holyghost that's all 11:19
I wrote 1500+ lines for it, but the packages are still not figured out right. I need to factor HLL OOP 11:23
in 20 days
jmerelo holyghost: where's the source?
11:23 kurahaupo left
holyghost eco: Mathx::Stat, Bayes::Learn, Game::Markov, AI::Agent 11:24
buggable holyghost, Nothing found
holyghost eco: Mathx::Stat
buggable holyghost, Mathx::Stat 'Statistics package unto Correlation Dimension.': modules.perl6.org/dist/Mathx::Stat:...:HOLYGHOST
11:24 kurahaupo joined
holyghost eco: Bayes::Learn Game::Markov AI::Agent 11:24
buggable holyghost, Nothing found
holyghost eco: Bayes::Learn
buggable holyghost, Bayes::Learn 'Bayesian Inference based upon ditributions. Minimax': modules.perl6.org/dist/Bayes::Learn...:HOLYGHOST
holyghost eco: Game::Markov
buggable holyghost, Game::Markov 'Markov Strategy kit for games': modules.perl6.org/dist/Game::Markov...:HOLYGHOST
holyghost eco: AI::Agent
buggable holyghost, AI::Agent 'AI network Agent system, example: Music Link Agent': modules.perl6.org/dist/AI::Agent:cpan:HOLYGHOST
holyghost there you go
11:24 p6bannerbot sets mode: +v kurahaupo
jmerelo holyghost: right, but where's the repo so that people can create issues, pull requests and things like that? 11:25
lizmat holyghost: have you thought about adding tests ???
11:26 noganex_ joined
holyghost there's no github repo, I thought CPAN was ok 11:27
11:27 p6bannerbot sets mode: +v noganex_
holyghost github doesn;t wuantify everything with just a web interface 11:27
s/does;t wuantify/doesn't quantify
lizmat: tests for a statistics package are uneasy 11:28
jmerelo holyghost: it's OK for release, but if you want people to contribute, create issues, all that, there needs to be a repo. Need not be GitHub, could be Gitlab or Bitbucket
lizmat holyghost: at least write some usage examples in the documentation (and make sure they run) 11:29
holyghost jmerelo: I'm on my own for now
lizmat: sure, but it's open source code, anyone knows their perl inclusion path 11:30
11:30 noganex left
lizmat having a way to report issues, allows people to do pull requests for you, e.g. for s/probabiltiy/probability/ 11:30
holyghost lizmat: ok but I can manage 11:31
timotimo i mean, you can always do it the classical way and do it via mail, but that won't let others see if there's already a PR for what they found 11:33
unless you have a publically archived mailing list
holyghost you probably cannot pull older versions but my package stays final and so is archived on CPAN for everyone who knows 11:34
That was the perl5 way
I don't need webeditors for that to be frank 11:35
lizmat well, zef install Mathx::Stat does not work
jmerelo lizmat: it's not released in CPAN, but in backpan: metacpan.org/author/HOLYGHOST
lizmat 'Mathx::Stat::Mathx::Stat::DistributionPopulation' cannot inherit from 'Population' because it is unknown.
jmerelo lizmat: for some reason, it was picked up by modules. 11:36
lizmat: that's because the class is declared as an unit module, and then as a class.
lizmat jmerelo: I don't care what the problem is at this stage
jmerelo lizmat: right
lizmat It's for this reason that I specced the concept of a recommendation manager 11:37
jmerelo lizmat: sorry.
11:37 kurahaupo left, kurahaupo joined
jmerelo lizmat: point is, that code does not compile. 11:37
lizmat not having tests in your distribution could be a reason to not recommend a distribution 11:38
11:38 kurahaupo left
holyghost lizmat: I know my perl 5 classes, I just need to remove the non-intuitive way 11:38
lizmat well, if the distribution had had tests, then the author could have known that it doesn't compile
11:38 kurahaupo joined
jmerelo lizmat: you are right. 11:38
11:39 p6bannerbot sets mode: +v kurahaupo
lizmat if you use App::Mi6, you always get a single test added 11:39
"use Module"; pass "it lives" !
11:39 pmurias joined, p6bannerbot sets mode: +v pmurias
lizmat that's already a good start 11:39
but holyghost's distributions don't even have that :-(
pmurias holyghost: re perl5 way, modern modules have a repository up 11:40
jmerelo lizmat: :-( Right so. And there's no repo where we can report that as an issue. Or provide a pull request. 11:41
lizmat FWIW, I have removed mention of holyghost's modules from all Perl 6 Weeklies 11:42
jmerelo lizmat: makes sense
lizmat I'll be glad to mention them if they have tests and install
holyghost lizmat: some modules do not install with tests, you can hack them as they are open source 11:46
lizmat: you do know perl -M -CPAN ?
lizmat holyghost: that's fine, but that doesn't mean we need to recommend them
"Missing argument to -M" 11:47
holyghost I don't mind but they do stay in the ecosystem
perl -MCPAN -e shell
lizmat *
yeah, I know about that, so ? 11:48
do you know "cpan"
holyghost I'd rather write an interface to CPAN than use a simple github server
lizmat that's a lot shorter then perl -MCPAN -e shell 11:49
jmerelo holyghost: your code is simply broken. Just try to "use" it from an example and run it using perl6. It will not run. Saying that people can hack it to make it run does not make it any better
lizmat it's a bit like graffiti: anybody can do a tag and add a tag to that 11:50
jmerelo holyghost: the whole point of running tests is to force authors to actually create something that, at least, compiles.
lizmat generally, people like to see a complete painting
buggable New CPAN upload: Mathx-Stat-0.1.11.tar.gz by HOLYGHOST modules.perl6.org/dist/Mathx::Stat:...:HOLYGHOST 11:51
holyghost SFY
lizmat like twitter.com/streetartmagic/status/...8465891329
holyghost Mathx::Stat needs to resolve, but I do not think there are any faults left 11:52
pmurias holyghost: the point of including a repo isn't that people download the code to use from their it's that they can have the version control history?
holyghost: do you use version control?
jmerelo holyghost: if there's no source repo, if there's no Changelog, we don't even know what's changed. There's still no test, we still don't know if it runs or not.
lizmat just plonked
afk& 11:53
holyghost jmerelo: I just put in the minial LICENSE and README.md 11:54
You need a patch or fidd to see the changes 11:55
11:55 Zoffix joined, p6bannerbot sets mode: +v Zoffix
jmerelo holyghost: no, *you* need to upload that to a repo to show changes. 11:55
Zoffix holyghost: but didn't you say your modules don't even compile?
yoleaux 10:53Z <holyghost> Zoffix: I cannot use your schedule tool, but here's my printed HTML page for Advent if you like (for Mathx::Stat): pastebin.com/x3mENvQE
holyghost Zoffix: Mathx::Stat should compile, it needs to resolve 11:56
The rest I am doing tomorrow
Zoffix holyghost: advents are about showing off technology and building cool stuff with it. That pastebin reads like someone's TODO list rather than showcase of tech
jmerelo holyghost: "should compile" as in "I haven't compiled it, but it looks OK"? 11:57
Zoffix "The package was meant to support Markov strategies for calculating several variances" great. What's the code to do that?
holyghost that's in the Game::Markov package
I hope nobody frames me for releasing early and often ? 11:58
Zoffix holyghost: even if Mathx::Stat compiles, there's no way it works correctly, because this isn't the syntax for calling methods on `self`: modules.perl6.org/dist/Mathx::Stat...on.pm6#L13
holyghost you just said I should write a handout, I did, so I tried to compile
jmerelo holyghost: "tried to" as in "I did it, but it didn't work"? 11:59
Zoffix holyghost: you can release as often as you want, but if you're gonna write an Advent Post about modules that don't work when people try them, I imagine a lot of people will be pissed off with you.
holyghost: well yeah, I suggested you should write something. Don't mean you should, if you're not ready :)
holyghost I concurred to support the community, the system still holds
Zoffix You can always write a non-Advent blog later on when the modules work. 12:00
holyghost I only wanted to help
Zoffix It's apreciated.
holyghost Mathx::Stat is easy to debug
as I said I'll do the rest of the packages later on (to compile I mean)
s/compile/compile and test 12:01
pmurias was once slightly pissed off after going to a super boring YAPC talk that was going through minute internals of a module that didn't compile
jmerelo holyghost: debugging supposes they compile. If there are no tests, it's impossible even to know what they're supposed to produce.
holyghost jmerelo: problem is it's instance.Covariance for example 12:02
you cannot calculate a covariance/correlation without knowing the result, thing is it can differ on x-bit systems
pmurias: I understand but released early, I've put them in the eco to support open source 12:03
pmurias holyghost: what's an x-bit system?
holyghost 8,16,32,64
jmerelo holyghost: problem is the whole thing. First upload it to a repo. Then write a basic "use Module" test, as lizmat has said. Then start writing tests, one by one, for all functions in your class. Then start to work on the code until it passes tests.
holyghost jmerelo: ok but I'm still working on the codebase 12:04
jmerelo holyghost: the first step to support open source is to upload it to a repo so that people can engage the code at the repo level, comment, all that.
12:04 Zoffix left
jmerelo holyghost: that's great, but it would be much better if you had waited to release it until it compiles and passes tests. You have produced several versions already, and we have no guarantee even one of them actually works. 12:05
holyghost jmerelo: ok, I just mirror my code
jmerelo holyghost: we really appreciate your enthusiasm, but it would really help if you waited until you have some code that compiles to upload the next version. 12:06
holyghost ok, I didn't know that 12:07
I wrote the code out of my head, I am still working on it, non-versioned except cpan, but people sometimes like an upload to debug/compile themselves if they know what it is all about 12:09
It's called hacking
jmerelo holyghost: hacking is twice as good if you use a source control system like git locally and git(lab|hub) offsite and check that everything works by compiling and testing it. 12:10
holyghost ok, but CPAN doesn't have that system 12:12
although somebody might have written that for cpan back in the day
W3C HTML is 23 years old at the least
jmerelo holyghost: that's not the point. CPAN is for releasing modules that work. It just reads metadata from your module and publishes it. You still have to use a source control system for managing yours and others' contributions. 12:13
holyghost ok 12:14
I can do that, but I want to work to a package that works as you concurred 12:15
to be honest, my updates are multiple such as Zoffix said
jmerelo holyghost: please do that. I would be very grateful if you did. 12:16
holyghost thx
I know git from savannah, same thing, of course
jmerelo: could you (once) see that Mathx::Stat compiles, I told Zoffix, I am a bit confused with the module system. Back in the day we hacked that way by logging into someone else 12:18
jmerelo holyghost: git is the client, savannah is the host. You can use savannah to host your code if you want, although there might be better choices, also free
holyghost: OK, I'll try it out 12:19
holyghost It should upload first to the ecosystem
0.1.11 I believe
jmerelo holyghost: here you go: gist.github.com/JJ/750c8d8c75239a1...63d0aa265a 12:23
holyghost: you'd do well, after you upload it to a repository, to include that test and repeat it for every module in your distribution. When it compiles, take it from there and write tests for what they are actually supposed to do. 12:25
holyghost jmerelo: it's probably on line 11 : .population = DistributionPopulation.new; ? 12:31
I program smalltalk, Scheme and other languages so I get confused 12:32
jmerelo holyghost: that's exactly what it says in the error. Perl6 compiler is usually great at errors.
holyghost I know :-) I love perl for that
I'll reupload, I comment out such things to compile 12:33
jmerelo holyghost: First, you should try to fix them, not comment them out. That's in the BUILD submethod, it's doubtful that anything will work if you comment that out. Second, I hope that by "reupload" you mean "upload it to a source control hosting site". And before you do that, include that test (and tests for all the other modules). When they pass, we at least know it's syntactically correct code. 12:35
12:35 mniip left
holyghost jmerelo: do try to understand perl6 doesn't in this case check a type of a ctor 12:36
12:37 kensanata left, leont joined
jmerelo holyghost: do try to understand that it's a syntax error and that you should try to fix that. Checking type does not have to do anything with that. 12:38
12:38 p6bannerbot sets mode: +v leont
holyghost jmerelo: to uncomment I mean 12:38
s/uncomment/comment 12:39
jmerelo holyghost: you can't comment it out. You need to fix it.
holyghost jmerelo: the new syntax, yes
I've got the point now :-) 12:40
jmerelo holyghost: great. Good luck. 12:41
jmerelo goes AFK
holyghost thx
s/thx/thx for the help
12:46 yqt left
jmerelo comes back to keyboard 12:46
holyghost: You have just uploaded a new version. Does it compile now? If it still does not, why did you do that? 12:47
12:49 daxim left
buggable New CPAN upload: Mathx-Stat-0.1.12.tar.gz by HOLYGHOST modules.perl6.org/dist/Mathx::Stat:...:HOLYGHOST 12:51
holyghost jmerelo: I thought you were AFK, it should compile now 12:52
jmerelo holyghost: it "should"? You haven't tried to compile it either? 12:55
holyghost I commented out the .new and it compiles 12:56
jmerelo holyghost: it compiles but it does not do a thing 12:57
holyghost: and the rest does not compile either. 12:58
holyghost I need to fix the rest
jmerelo holyghost: ... before you upload anything else to CPAN. 12:59
jmerelo *really* goes AFK 13:00
13:01 GDPR is now known as emerson 13:05 jmerelo left 13:09 zakharyas left 13:13 ufobat left 13:14 cjkinni left 13:15 cjkinni joined 13:16 p6bannerbot sets mode: +v cjkinni 13:47 isBEKaml joined, p6bannerbot sets mode: +v isBEKaml
Geth doc: 456118e16e | Coke++ | doc/Language/modules.pod6
run more tests with :solo

change skip-test to :lang for clarity on one.
synopsebot Link: doc.perl6.org/language/modules