»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.
00:00 vrurg left 00:01 rindolf left 00:03 netrino joined 00:09 vrurg joined 00:25 leont left 00:35 netrino left 00:41 Manifest0 left 00:42 netrino joined 00:44 john_parr joined 00:46 Manifest0 joined 00:59 MasterDuke joined, MasterDuke left, MasterDuke joined 01:03 khisanth_ left 01:05 kurahaupo_ joined 01:06 kurahaupo left, khisanth_ joined 01:15 netrino left 01:19 noisegul_ left
Xliff m: 170408.sqrt.say 01:20
camelia 412.805038728938
01:21 netrino joined 01:28 Manifest0 left, aborazmeh joined, aborazmeh left, aborazmeh joined 01:35 Manifest0 joined 01:49 Manifest0 left 01:52 adu joined, mniip left 01:54 netrino left, Manifest0 joined 01:58 mniip joined 02:00 netrino joined 02:03 Manifest0 left 02:05 Manifest0 joined 02:11 Manifest0 left 02:17 kurahaupo_ left, kurahaupo joined 02:18 Manifest0 joined, MasterDuke left 02:22 kurahaupo left, Manifest0 left
Xliff m: my enum <AAA BBB CCC>; my %n = ( AAA => 'a', BBB => 'b', CCC => 'c' }; say %n{AAA} 02:26
camelia 5===SORRY!5=== Error while compiling <tmp>
Unable to parse expression in parenthesized expression; couldn't find final ')' (corresponding starter was at line 1)
at <tmp>:1
------> 3 = ( AAA => 'a', BBB => 'b', CCC => 'c' 7⏏5}; say %n{AA…
Xliff m: my enum <AAA BBB CCC>; my %n = ( AAA => 'a', BBB => 'b', CCC => 'c' ); say %n{AAA}
camelia a
02:29 Manifest0 joined
elcaro Xliff: I'd say that's because enum values stringyfy as their name 02:32
02:32 netrino left
elcaro and hashes default to string keys unless declared with %h{Any} (or a more specific type) 02:33
m: my enum <AAA BBB CCC>; my %h{Any} = (AAA) => 'a', (BBB) => 'b', (CCC) => 'c'; say %h{AAA}; say %h<AAA>; 02:34
camelia a
(Any)
02:37 eiro left 02:38 eiro joined
Xliff elcaro: Why %h{Any}? 02:38
m: my enum <AAA BBB CCC>; my %n = ( AAA => 'a', BBB => 'b', CCC => 'c' ); say %n{AAA}; say %n<AAA> 02:39
camelia a
a
02:39 netrino joined
Xliff Ah. I think I get it. 02:39
m: rand * 360 02:45
camelia WARNINGS for <tmp>:
Useless use of "*" in expression "rand * 360" in sink context (line 1)
Xliff m: say rand * 360
camelia 179.21518249519428
02:55 aborazmeh left 03:01 kaare__ left 03:02 kaare__ joined 03:04 Manifest0 left 03:09 Manifest0 joined 03:12 netrino left
ChoHag So, sometimes when I get like this I come in here and start ranting at whoever's around for doing this to my favourite language. How're things this time? 03:16
(Also that's a roundabout way of asking what's new in the last, I don't know, 6 months-ish)
Last I heard/really concentrated, perl 6 was "just about done, honestly, any minute now". I think 6d? Something like that. 03:18
03:18 netrino joined
ChoHag By the way if I'm coming across as antagonistic that's just because I'm drunk and trying to be funny. This is an honest question. 03:19
Xliff m: my $s = 0; for ^12 { say $s > 12 ?? 0 !! $s++ } 03:28
camelia 0
1
2
3
4
5
6
7
8
9
10
11
Xliff m: 3.rand.say 03:29
camelia 2.4185029798972404
Xliff m: 3.rand.say
camelia 0.8495419437821555
Xliff m: 3.rand.say
camelia 2.7888226234322904
Xliff m: 3.rand.floor.say for ^10 03:30
camelia 2
1
2
2
0
2
0
0
1
1
03:30 noisegul joined 03:36 Manifest0 left 03:41 Manifest0 joined 03:46 Manifest0 left 03:50 netrino left 03:51 Manifest0 joined 03:57 netrino joined 04:02 Manifest0 left 04:07 Manifest0 joined 04:11 Manifest0 left 04:16 Manifest0 joined
ChoHag What does the ^12 do if there's also a 12-ternary with a ++? 04:23
04:24 skids joined
ChoHag Does that just ultimately having two iterators, $_ (?) and $s, or something else that I'm not seeing? 04:24
have
04:25 Manifest0 left 04:27 Xliff left 04:30 netrino left, Manifest0 joined 04:36 netrino joined, skids left 04:37 Manifest0 left 04:40 molaf joined 04:43 Xliff joined 04:46 Manifest0 joined 04:48 curan joined 05:08 netrino left, nadim_ joined 05:09 jmerelo joined
elcaro Xliff: sorry was AFK. by default, Hash's are secretly %h{Str}, so all keys get coerced to Str 05:13
05:13 nadim_ left 05:15 netrino joined
elcaro ChoHag: Perl 6 is "done" as in, version 1 was released a few years ago... now it's undergoing constant development like most other languages 05:19
05:19 Manifest0 left
elcaro ChoHag: the '^12' is short-hand for '0 ..^ 12' which means the range "0 up-to-but-not-including 12" 05:19
05:24 Manifest0 joined
Geth doc: 31d7ee314f | (JJ Merelo)++ | doc/Type/Signature.pod6
Addresses first part of #2404
05:30
doc: 9ac0e9e92f | (JJ Merelo)++ | doc/Type/Signature.pod6
Reindexes argument aliases as such, closes #2404
doc: 03274aae4c | (JJ Merelo)++ | doc/Type/Signature.pod6
Minor clarifications
synopsebot Link: doc.perl6.org/type/Signature
05:31 cpan-p6 left, Manifest0 left 05:34 cpan-p6 joined, cpan-p6 left, cpan-p6 joined 05:37 Manifest0 joined 05:41 nadim_ joined 05:44 Manifest0 left 05:47 netrino left 05:49 Manifest0 joined 05:54 netrino joined 06:05 sena_kun joined 06:15 Manifest0 left 06:18 molaf left 06:21 Manifest0 joined 06:22 nadim_ left 06:27 netrino left
Xliff m: role A { submethod BUILD { say 'role' }; }; class B does A { submethod BUILD { say 'class' }; }; B.new; 06:28
camelia class
06:33 netrino joined 06:35 Manifest0 left 06:38 netrino_ joined 06:39 netrino left 06:40 Manifest0 joined 06:51 Manifest0 left, dolmen joined 06:52 noisegul left 06:56 lgtaube joined 06:57 Manifest0 joined 07:02 powerbit joined, jmerelo left 07:12 Manifest0 left 07:16 Manifest0 joined 07:17 zakharyas joined 07:23 kst` joined, Manifest0 left 07:24 kst left 07:30 Manifest0 joined
lizmat clickbaits p6weekly.wordpress.com/2019/05/27/...thats-why/ 07:30
Xliff m: say "\342" 07:39
camelia 5===SORRY!5=== Error while compiling <tmp>
Unrecognized backslash sequence: '\342'. Did you mean $341?
at <tmp>:1
------> 3say "\7⏏05342"
expecting any of:
argument list
double quotes
term
Xliff m: say "\0342"
camelia ␀342
Xliff m: say "\o342"
camelia â
Xliff m: say "\o342\o236\o236" 07:40
camelia â
Xliff m: say "\o236\o236"
camelia
07:40 dakkar joined 07:44 noisegul joined 07:52 sean` joined
noisegul Good morning everyone 07:52
holyghost good morning 07:53
cpan-p6 New module released to CPAN! Soundn::Tina (0.1.2) by 03HOLYGHOST
holyghost ^-- it's a WAV file parser with some sample pattern recognition, it's pretty straightforward code 07:55
07:55 dolmen left
holyghost I had it lying around, it comes from my holly6src scratchpad 07:57
I need to add docs tomorrow
08:01 scimon joined 08:04 netrino_ left
timotimo holyghost: you have a problem in your META6.json 08:07
08:08 patrickb joined
timotimo holyghost: the use statements for Vector32bit are also wrong 08:08
you're publishing code that doesn't even compile. again.
08:08 feldspath joined
timotimo jj already told you in no uncertain terms that that's not cool 08:08
did you perhaps miss that?
Xliff What's the best way to form a unicode character? 08:10
timotimo m: say "\c[snowman]"
camelia
Xliff How about from octal numbers?
08:10 molaf joined
timotimo m: say "\o777" 08:11
camelia ǿ
timotimo m: say "\o7777"
camelia ࿿
Xliff m: say "\o342\o236\o236"
camelia â
sena_kun github.com/rakudo/rakudo/issues/2169 can be closed
Xliff ^ That's not what I expected, though.
timotimo m: say uninames(0o342, 0o236, 0o236)
camelia 5===SORRY!5=== Error while compiling <tmp>
Calling uninames(Int, Int, Int) will never work with signature of the proto ($, *%)
at <tmp>:1
------> 3say 7⏏5uninames(0o342, 0o236, 0o236)
timotimo m: say .&uniname for (0o342, 0o236, 0o236) 08:12
camelia LATIN SMALL LETTER A WITH CIRCUMFLEX
<control-009E>
<control-009E>
Xliff Interesting.
timotimo m: say .&uniname for (0x342, 0x236, 0x236)
camelia COMBINING GREEK PERISPOMENI
LATIN SMALL LETTER T WITH CURL
LATIN SMALL LETTER T WITH CURL
Xliff When I do "\342\236\236" in C I get an arrow.
timotimo that'd be utf8 probably?
Xliff uniname: →
timotimo m: say Buf8(0o342, 0o236, 0o236).decode("utf8") 08:13
camelia 5===SORRY!5=== Error while compiling <tmp>
Undeclared name:
Buf8 used at line 1. Did you mean 'buf8', 'Buf'?
Xliff m: '→'.uniname.say
camelia RIGHTWARDS ARROW
timotimo m: say buf8.new(0o342, 0o236, 0o236).decode("utf8")
camelia
timotimo m: say buf8.new(0o342, 0o236, 0o236).decode("utf8").ords
camelia (10142)
timotimo m: say "\c10142"
camelia
patrickb sena_kun: Closed.
Xliff m: '→'.unichar.say
camelia No such method 'unichar' for invocant of type 'Str'
in block <unit> at <tmp> line 1
sena_kun patrickb++ vrurg++
patrickb vrurg++ indeed 08:14
HarmtH virt
holyghost I'm working on it timotimo, I upload early 08:15
timotimo no, stop that
cpan is not for early uploads
use github or gitlab or bitbucket or sourceforge or savannah or whatever for early uploads 08:16
stop putting broken code on cpan
holyghost It'll be allright 08:17
08:17 noisegul left 08:19 sean` left
lizmat holyghost: FWIW, at the moment I'm just ignoring *anything* you upload in the Perl 6 Weekly 08:21
holyghost: if you like your uploads to show what you can do as a developer, you need to clean up your act 08:22
08:22 Guest50087 joined
lizmat note to self: I guess we need to work on creating a full-fledged recommendation manager with purging capabilities 08:22
cpan-p6 New module released to CPAN! Soundn::Tina (0.1.3) by 03HOLYGHOST 08:23
08:23 Guest50087 left
holyghost release often, release early, a few bugs won't damage anything 08:24
Xliff holyghost: That works better for git. 08:26
But even then... push carefully.
holyghost I'm doing my best Xliff
Xliff holyghost: Everyone is telling you that you should be trying harder.
Nicely. 08:27
holyghost ok, it used to compile BTW, I need to update it just for the newest perl6
I have it fixed fully by tomorrow 08:28
sena_kun can't imagine this community is the first that encounters such an issue 08:29
holyghost Just don't worry about it, I can manage, I've done it before 08:30
08:31 netrino_ joined
holyghost lizmat, FWIW, you can see at Mathx::Stat and Bayes::Learn if you need to 08:33
those are documented and bugless
then there's Game::AI::Messaging
The rest I need to figure out, I'm probably working on too much modules at the same time 08:34
08:35 Manifest0 left
lizmat what kind of test is this? 08:36
is-deeply $p.Bayes(@indices, @plist, 1), $p.Bayes(@indices, @plist, 1);
holyghost I cannot compare it to a number
It calculates Bayes of course
lizmat feels to me you're testing whether 42 is equal to 42
08:37 rindolf joined
holyghost yes, if ti runs it gives you the conditional probability calculated by the Bayes formula 08:37
I cannot use aq number for it as that's different on e.g. 32 and 64 bit
s/aq/a 08:38
more or less
08:38 netrino__ joined 08:41 netrino_ left
holyghost lizmat : I can help you out if you like but I don't have much time on my hands 08:42
lizmat neither
08:42 Manifest0 joined
ChoHag Perl 6 is done? Does that mean all the synopses? 08:45
Including the one on pod? 08:46
lizmat ChoHag: there's been quite a lot of stuff done on pad in the past year or so, not sure how complete the work is 08:47
ChoHag I was looking forward to a bit more runtime control over the documentation.,
08:47 Manifest0 left
lizmat jmerelo probably knows more about it 08:47
ChoHag It's pad now? Is that a typo or is it short for Plain Annoying?
timotimo nah, it's pod 08:48
lizmat ChoHag: oops, typo, pod 08:49
08:50 sena_kun left 08:53 Manifest0 joined
timotimo ChoHag: we don't have everything from all synopses, but the synopses haven't been considered "The Language Definition" for a while; that is roast's job now 08:53
Xliff timotimo: Here's one for you. 08:54
m: class A { method Int { 1; } }: multi sub (A $a) { say 'a' }; multi sub (Int() $a) { say 0; }; a(A.new) 08:55
camelia 5===SORRY!5=== Error while compiling <tmp>
Confused
at <tmp>:1
------> 3class A { method Int { 1; } }:7⏏5 multi sub (A $a) { say 'a' }; multi sub
expecting any of:
colon pair
Xliff m: class A { method Int { 1; } }; multi sub (A $a) { say 'a' }; multi sub (Int() $a) { say 0; }; a(A.new)
camelia 5===SORRY!5=== Error while compiling <tmp>
An anonymous routine may not take a multi declarator
at <tmp>:1
------> 3t { 1; } }; multi sub (A $a) { say 'a' }7⏏5; multi sub (Int() $a) { say 0; }; a(A.n
expecting any of:
h…
Xliff m: class A { method Int { 1; } }; multi sub a (A $a) { say 'a' }; multi sub a (Int() $a) { say 0; }; a(A.new)
camelia a
Xliff WTF?
m: class A { method Int { 1; } }; multi sub a (A $a) { say 'a' }; multi sub a (Int() $a) { say 0; }; A.new.Int 08:56
camelia ( no output )
Xliff m: class A { method Int { 1; } }; multi sub a (A $a) { say 'a' }; multi sub a (Int() $a) { say 0; }; A.new.Int.say
camelia 1
08:56 seanrob joined
Xliff m: class A { method Int { 1; } }; multi sub a (A $a) { say 'a' }; multi sub a (Int() $a) { say 0; }; a(A.new) 08:56
camelia a
Xliff Why isn't that giving me the same confusing stuff as my code>?
Hmm...
m: class A { method Int { 1; } }; class B { multi method a (A $a) { say 'a' }; multi method a (Int() $a) { say 0; }; }; B.new.a(A.new) 08:57
camelia a
Xliff m: $*PERL.compiler.version.say
camelia v2019.03.1.385.ga.643.b.8.be.1
timotimo the version is also part of camelia's output 08:58
Xliff timotimo: It's weird. But what I am seeing is that multi methods that have the same number of parameters, if one is coercive, that will BYPASS those that are more restrictive.
timotimo OK, that seems odd
without seeing the actual code, not sure i can help 08:59
09:00 netrino__ left
Xliff /o\ 09:03
No. I figured it out. 09:04
Thanks for waiting.
09:08 farcas_ joined 09:14 netrino__ joined 09:15 reach_satori left 09:18 netrino__ left 09:20 dolmen joined 09:21 Manifest0 left, ravenousmoose joined 09:23 demanuel joined 09:27 dolmen left, netrino__ joined 09:29 Manifest0 joined
lizmat someone needs a reply on twitter: twitter.com/marver/status/1133291661205934080 09:40
:-) 09:43
09:44 seanrob left 09:46 dolmen joined
Xliff lizmat: Replied. 09:53
lizmat Xliff++
09:53 leont joined
timotimo butterfly hit squad assembla 09:56
10:01 netrino__ left 10:03 nadim_ joined 10:04 lizmat_ joined 10:07 lizmat left, netrino__ joined 10:08 leont left, lizmat_ is now known as lizmat 10:15 dolmen left 10:18 reach_satori joined 10:20 ravenousmoose left 10:23 ravenousmoose joined, reach_satori_ joined, reach_satori left 10:26 lgandras joined 10:31 reach_satori_ left 10:34 molaf left 10:39 lgandras_ joined 10:40 Manifest0 left, netrino__ left 10:42 lgandras left 10:44 Manifest0 joined 10:46 netrino__ joined 10:50 lgandras_ left, lgandras joined 10:51 Manifest0 left 10:56 Manifest0 joined 11:00 dolmen joined, Manifest0 left, noisegul_ joined 11:01 sergot joined 11:06 Manifest0 joined 11:09 nadim_ left 11:12 nadim_ joined 11:14 dolmen left 11:19 netrino__ left, dolmen joined 11:23 nadim_ left 11:25 netrino__ joined 11:36 dg_subrata joined 11:37 kurahaupo joined 11:39 robertle_ joined 11:40 zakharyas left
lizmat clickbaits p6weekly.wordpress.com/2019/05/27/...thats-why/ 11:41
11:44 nadim_ joined 11:49 nadim_ left
holyghost I need to adapt @8bitpattern in simplesearch(...) from the PatternChannelLib.pm6 but it should compile now 11:49
It needs to be a scalar
upload is version 0.1.4. from Soundn::Tina 11:50
lizmat have you heard of github, gitlab or a gist ??
11:51 molaf joined
holyghost lizmat : ok 11:52
WAM (wait a minute)
moritz wake me up, before you go go! 11:54
kurahaupo starts humming "wake me up before you go-go"
moritz: *snap*
cpan-p6 New module released to CPAN! Soundn::Tina (0.1.4) by 03HOLYGHOST 11:57
11:58 netrino__ left
holyghost lizmat : It's on github.com/theholyghost2/Soundn-Tina 11:59
It needs to upload the neww code I importedd with the Web GUI, I guess 12:00
anyway, it shouokld compile now from the version on cpan (0.1.4) 12:02
12:04 netrino__ joined
holyghost The docs are included 12:08
lizmat doesn't see any 12:11
Xliff timotimo: p6-Clutter now has working tests in t 12:15
12:16 noisegul_ left 12:17 noisegul joined, dg_subrata left
Geth ¦ problem-solving: AlexDaniel assigned to rba Issue Push access to repositories in perl6 organization github.com/perl6/problem-solving/issues/33 12:20
¦ problem-solving: AlexDaniel assigned to maettu Issue Push access to repositories in perl6 organization github.com/perl6/problem-solving/issues/33
timotimo holyghost: does the test file run now? 12:25
in the META6 there's still Matrix32.pm6 while in the folder it's called Matrix32bit.pm6 12:27
and Vector32bit.pm6 tries to instantiate Vector8bit, which has to be Soundn::Tina::Vector8bit instead, since you renamed that 12:28
and File32Parser.pm6 is the next file that doesn't compile, it doen't find "Parser" 12:29
probably because it's called WAVParser in the file?
12:30 Sgeo__ left
timotimo and there's one } too many in File32Parser.pm6, before the last method. it warns you that you put the method transformtodata outside of a class 12:30
12:30 noisegul left
timotimo oh, you have two WAVParser classes actually, one is in WAVParser and one is in Parser 12:31
why do you keep uploading broken code to cpan when multiple people have asked you to please stop doing that
Xliff holyghost: If you would like, I can try and send you a PR that will allow you to build all of the classes in your project so you can run that BEFORE you release to CPAN.
timotimo you're clearly not even trying to run the code a single time before uploading it, otherwise you would be getting these compiler errors 12:32
Xliff timotimo: When you get some time, please try the latest p6-Clutter. Requires the latest p6-GtkPlus. 12:35
12:35 aborazmeh joined, aborazmeh left, aborazmeh joined
Xliff I will have to update the install docs. It's become a lot easier to get things to run... unfortunately not faster, though. 12:35
12:37 netrino__ left 12:39 Cabanossi left 12:41 Cabanossi joined
Xliff OK. Installation instructions now written and sent to the github project page. 12:42
github.com/Xliff/p6-Clutter/
12:43 netrino__ joined 12:48 Manifest0 left 12:49 lgandras left 12:54 john_parr left 12:55 Manifest0 joined, noisegul_ joined 12:58 john_parr joined 13:11 zakharyas joined 13:16 Manifest0 left, netrino__ left 13:21 Manifest0 joined 13:22 netrino__ joined, molaf left 13:27 Manifest0 left 13:28 skids joined, reach_satori_ joined 13:34 Manifest0 joined
ugexe someone should benchmark how long the 90+ distributions uploaded by holy ghost adds to zef/rakudo parsing and searching :P 13:49
13:50 farcas_ left
lizmat ugexe: perhaps a filter on holyghost would be more effective ? 13:50
13:50 farcas_ joined
ugexe well, its a policy issue and we don't really have any policy yet. additionally it could be argued we need a better api for the ecosytem than "include all meta data for everything" 13:52
13:53 Xliff left 13:54 ravenousmoose_ joined 13:55 netrino__ left
ugexe there are essentially two modes of meta data searching -- the primary one only searches for namespace data, while the other is searching tags/description etc. the former doesn't require all the meta data, while the later does 13:55
13:56 ravenousmoose left
ugexe anywho we *can* dictate policy, but we don't have any prior (perl6) art to work with yet. and such a policy would be that of specific ecosystems. 13:57
what we don't want is various ecosystem generating tools to have stuff hardcoded like `next if $author ~~ /foo/` 13:58
generating *or* consuming
so an ultra naive solution would have some sort of way for members to "vote" for a module to be hidden 13:59
14:01 netrino__ joined 14:02 sena_kun joined 14:05 curan left
lizmat ugexe++ 14:06
14:10 molaf joined 14:11 Manifest0 left
AlexDaniel 90+ distributions? 14:13
ugexe multiple versions of multiple namespaces 14:14
14:16 Manifest0 joined 14:21 Xliff joined
Xliff \o 14:21
I'm getting this when trying to rebuild rakudo: 14:22
Can't locate object method "note" via package "NQP::Config::Rakudo"
And now I can't rakudobrew moar or moar-blead
Using rakudobrew v2 will compile, but I can't get shim or env mode to work... that means I can't run the build executables!
14:23 Manifest0 left
ugexe check last night backlog of #perl6-dev 14:23
Xliff ugexe: OK
patrickb Xliff: WRT rakudobrew v2 not working: Did you update the entry in .profile/.bash_profile/... ? 14:24
AlexDaniel rakudobrew v2 O_o
AlexDaniel is hoping to catch up next week or so
Xliff patrickb: Yes.
patrickb AlexDaniel: It's the normal rakudobrew. He just meant not the v1 branch used in travis builds. 14:25
14:26 MilkmanDan left
patrickb Xliff: I'd be interested in how it fails. 14:27
Xliff: Which branch of rakudobrew you use should have no influence on whether the build works or not. I'd be surprised if it does in your case. 14:28
Xliff After "git submodule update" my entire perl6 install is now not working.
14:28 Manifest0 joined
holyghost timotomo, Xliff : I'll update tomorrow (Matrix32bit META.json issue) 14:29
patrickb vrurg: ^ 14:30
Xliff Back to latest v1. Completely new install. :p 14:31
holyghost also the is ClassX::things 14:32
vrurg Xliff: symptoms? What's broken?
Xliff vrurg: Let me get back to you after the clean install finishes. 14:33
Mine was very long in the tooth.
14:34 netrino__ left
Xliff vrurg: Although if you want the last failure mode, you can find it here: paste.fedoraproject.org/paste/-dsy...tBWiBU~~uQ 14:34
After that bit, I had to force clean, co: rakudobrew: error while loading shared libraries: libmoar.so: cannot open shared object file: No such file or directory 14:35
\o/ -- This is Rakudo version 2019.03.1-492-gca7408e9f built on MoarVM version 2019.05-11-g248e2980a
14:36 dolmen left
holyghost Xliff, timotimo, the upload should show soon 14:37
vrurg Xliff: did you try moar-blead with rakudobrew? I'm just looking for something reproducible.
holyghost Xliff, timotimo, I adpated to my best 14:38
Xliff vrurg: moar-blead has been the only thing I could use from v1 for the last week or so.
ugexe oh if thats the case then rakudobrew will be broke on travis when someone does `perl6: 2019.05` or whatever 14:39
or 2019.03.1 rather 14:40
patrickb I am pretty sure the rakudobrew that is used is irrelevant to these problems.
14:40 netrino__ joined
Guest12727 do we know if the nuke command in rakudobre v2 has been fixed yet? 14:41
14:41 dolmen joined
holyghost I need to add test, still reaching out 14:41
I'm a bit afk now
patrickb Guest12727: What has been broken with nuke in the past? 14:42
14:42 leont joined
Guest12727 patrickb: it used to work but in the new version it's broken 14:42
see github.com/tadzik/rakudobrew/issues/142 14:43
patrickb Just found that. I'll have a look.
Guest12727 ++patrickb
14:44 MilkmanDan joined
cpan-p6 New module released to CPAN! Soundn::Tina (0.1.5) by 03HOLYGHOST 14:44
vrurg rakudobrew doesn't specify --prefix for Configure.pl. This is what breaks the build (nqp/moar are picked from PATH). 14:45
patrickb .tell tadzik Gentle reminder about github.com/tadzik/rakudobrew/pull/136
yoleaux patrickb: I'll pass your message to tadzik.
patrickb .tell tadzik I'd like to request a commit bit for rakudobrew. 14:46
yoleaux patrickb: I'll pass your message to tadzik.
tadzik heh, this chain of ifs reminds me of some of my JS code
yoleaux 14:45Z <patrickb> tadzik: Gentle reminder about github.com/tadzik/rakudobrew/pull/136
14:46Z <patrickb> tadzik: I'd like to request a commit bit for rakudobrew.
tadzik patrickb: request granted :)
use it responsibly ;)
Xliff tadzik: Heyy! If it's that easy to get a commit bit... .oO( I hope he realizes I'm kidding! ) 14:47
holyghost .tell Xliff I'm going to do an SDL2 game later on, I'd like you to debug it
yoleaux holyghost: I'll pass your message to Xliff.
Xliff o_O
yoleaux 14:47Z <holyghost> Xliff: I'm going to do an SDL2 game later on, I'd like you to debug it
Xliff holyghost: I'm quite busy these days. When you are ready, let me know. Give me a couple of days, however. 14:48
And I'm utterly serious about getting you to use those scripts. They may need a bit of modification, but they will help you improve the code you submit.
holyghost Xliff, we'll manage 14:49
14:49 cpan-p6 left
holyghost Xliff, I'm glad you're on game software, even if they're not Gtk based :-) 14:51
14:51 cpan-p6 joined, cpan-p6 left, cpan-p6 joined 14:52 aborazmeh left
holyghost barf *lol* 14:52
vrurg tadzik: may I ask you to have always specify --prefix for Configure.pl in rakudobrew? 14:53
patrickb vrurg: What did the rakudo build do without --prefix before the build-revamp? 14:54
vrurg patrickb: it was using ./install by default.
That's what it does now if no nqp found in PATH. 14:55
ugexe if i have an nqp from 2019.03.1 installed from rakudobrew, and do rakudobrew build moar-blead, couldn't the old nqp in the PATH get chosen by mistake? 14:57
patrickb I guess that's what happens now. 14:58
14:58 noisegul_ left
vrurg ugexe: it currently will. That's why I'm asking for explicit --prefix. Or I'd have to pull PATH seeking from Configure.pl. 14:58
ugexe i dont think the PATH thing is compatible with multiple perl6 installs then
patrickb We should either fix rakudobrew by adding --prefix to the build, or remove the logic to check for nqp in PATH. 14:59
I'm unsure whether I like that behavior...
ugexe ignore rakudobrew -- what if i manually want to install two perl6s?
vrurg ugexe: --prefix or --with-nqp turns that logic off.
ugexe rakudobrew doesn't use --with-nqp then eh? 15:00
patrickb Hm. It's using --gen-nqp IIRC
In that case it should obviously also use the generated NQP...
15:01 noisegul_ joined
patrickb it uses `Configure.pl --backends=moar --gen-moar --git-reference=\"$git_reference\" --make-install` 15:04
The `triple` command uses `Configure.pl --backend=moar` after installing moar and nqp into ./install 15:05
vrurg patrickb: I copied the logic of handling --gen-nqp from the previous configure. It was pretty relaxed about this parameter. nqp takes --gen-moar more literally.
ugexe you copied the logic, but the problem is you said --gen-nqp itself disables the PATH thing while also removing --gen-nqp from the command
(im assuming it used to be there but can be wrong) 15:06
vrurg ugexe: no, --gen-nqp doesn't disable it. --with-nqp does. That's because --gen-nqp was always kinda optional things saying 'gen if not found' 15:07
ugexe oops
vrurg I'm not sure yet which way is better: make it always rebuild nqp with --gen or leave as it is now. 15:08
patrickb I'm leaning towards disabling the search in PATH thing. There is quite some potential for messing up and breaks backwards compatibility. On the other hand it's easy to print a message notifying the user that no NQP could be found and she needs to specify one. 15:10
vrurg I'd like to keep PATH in place for beginners who would have nqp package pre-installed. Because what a typical beginner does – ./Configure, no params.
patrickb vrurg: But such users would probably rather have a fitting NQP built for them than using some unkown NQP a distro provided. 15:11
15:12 robertle_ left 15:13 netrino__ left
vrurg patrickb: I'm not certain about that. What's probably more relevant is compatibility matter. The package would most likely be outdated comparing to what one clones from git. 15:14
patrickb True. 15:15
vrurg I have to go now. Would think of it over the day. But most likely will remove the feature tonight.
lizmat fwiw, the current behaviour was a bit of a WAT to me
yoleaux 14:29Z <vrurg> lizmat: picking up of moar/nqp from PATH is aimed at beginners who might have pre-installed packages. It is only used if there is no --prefix or --with-nqp/with-moar are used.
patrickb o/ 15:16
vrurg lizmat: sorry for that. :(
o/
lizmat vrurg: no worries 15:17
15:19 patrickb left, netrino__ joined 15:20 jmerelo joined
jmerelo o 15:21
15:27 nadim_ joined
Geth ¦ doc: JJ assigned to antoniogamiz Issue Cache generated documentation github.com/perl6/doc/issues/717 15:30
15:32 aindilis left 15:34 aindilis joined 15:35 feldspath left
Xliff timotimo: You might need to bump Cairo. zef isn't picking up the latest changes. 15:47
holyghost lo JJ, I'm into games with Xliff later on this week
Xliff : indeed 15:48
timotimo, writing interfaces really is for a true-hacker
15:49 Manifest0 left
Xliff
.oO( wtf? )
15:49
Geth doc/master: 5 commits pushed by (JJ Merelo)++ 15:51
15:52 netrino__ left 15:53 Manifest0 joined
jmerelo holyghost: I'd be very grateful if you read this I wrote a few days ago: colabti.org/irclogger/irclogger_lo...05-23#l222 15:54
holyghost: today you have released a new module without tests or documentation. You need to stop doing that, for the sake of the whole perl6 community 15:55
holyghost jmerelo, there's docs
I make the tests tomrrow
I will read it 15:56
I will
jmerelo holyghost: no, you won't. You've said that repeatedly, yet you keep uploading stuff to CPAN that does not meet the minimum requirements for modules.
holyghost: the TL;DR of those messages is: stop uploading modules to CPAN if they are untested, undocumented, developed outside a repository where people can submit patches. 15:57
15:57 abraxxa left
jmerelo holyghost: I would also kindly ask you to remove all modules in CPAN that do not meet those standards. 15:57
15:58 netrino__ joined
jmerelo holyghost: and to stop uploading them until someone or a group of someones has reviewed them to check that they meet the community standards. You can upload them to some repo like GitHub or GitLab for that purpose, and share the link here. 15:58
holyghost ok 15:59
Did you see the docs of Soundn::Tina, they're in however
jmerelo holyghost: do you understand that this means that if you upload any other module, or any other version of existing modules, we might have to take some measure about it? 16:00
holyghost sure
jmerelo holyghost: I saw the docs in the one previous to that. They're not docs, they're comments. It does not have a README.md file. No one will be able to find out what it does. That's the problem with all your modules I've checked.
holyghost: OK, thanks for understanding this. 16:01
holyghost I'll try to troll les on cpan 16:03
16:04 nadim_ left
holyghost I'll try to add tests and docs to my modules 16:05
I understand that cpan is good code for everyone
JJ, is Mathx::Stat and Bayes::Learn all right for you ? 16:06
and Game::AI::Messaging ?
they have docs and tests 16:07
16:07 nadim_ joined
jmerelo holyghost: this is one of the tests in Mathx::Stat. is-deeply $p.CalculatedCondP(3,0), $p.CalculatedCondP(3,0); 16:07
holyghost: do you think that's OK?
holyghost as it's not a number, I explained lizmat before
jmerelo holyghost: that's not the point. Something is going to always be equal to itself. Saying that's a test is cheating in Solitaire. 16:08
holyghost yes but it runs ...
lizmat m: use Test; is 42,42, 'success!'
camelia ok 1 - success!
jmerelo holyghost: you might need to check out some literature about unitary tests. You can start here: docs.perl6.org/language/testing and continue here: perl6advent.wordpress.com/2017/12/...he-things/ 16:09
timotimo jmerelo: but that calls the method twice, so if CalculateCondP accidentally modifies the $p, this could perhaps catch the mistake
holyghost no it won't