Perl 6 - the future is here, just unevenly distributed

IRC log for #perl6, 2016-12-13

Perl 6 | Reference Documentation | Rakudo

| Channels | #perl6 index | Today | | Search | Google Search | Plain-Text | summary

All times shown according to UTC.

Time Nick Message
00:02 Xliff CRAP
00:02 Xliff I knew there was a reason I wasn't working on that module.
00:02 Xliff There was a recent change in rakudo (end of summer, maybe) that introduced that.
00:03 Xliff It wasn't there in July, and the XML::LibXML codebase used it extensively.
00:04 MasterDuke btw, don't know if you just typed something wrong, but the line i commented out was "use XML::LibXML::CStructs;" without the ":types"
00:04 MasterDuke (obviously who cares if it's commented out, but just in case you're on a branch or something"
00:05 Xliff OK.
00:05 Xliff in XML::LibXML dir, "git checkout -b 15nodelist"
00:06 Xliff Then pull new changes.
00:07 MasterDuke ===SORRY!=== Expected MAST::Frame, but didn't get one
00:07 Xliff YES!!
00:07 Xliff Er.... Um...
00:07 Xliff Youknowwhuddimean
00:08 Xliff So, there could be something screwy with either the XML::LibXML codebase or XML::XSLT
00:08 Xliff I just don't know.
00:08 Xliff and XML::LibXML needs serious overhaul due to bitrot
00:21 MasterDuke Xliff: well, so far golfed it to this: https://gist.github.com/MasterDuke17/a2c8e38be2580f6d6703025c90220441
00:23 maxp_ is there some kind of problems about capturing groups and ratchet adverbs?
00:24 maxp_ I've this:"<ciao>mamma</ciao>" ~~ /:r '<' (<[\w\W]>+?) '>'/
00:24 maxp_ and I expect the word "ciao" but it returns nil
00:25 maxp_ whitout parentesis it works
00:26 AlexDaniel bisect: "<ciao>mamma</ciao>" ~~ /:r '<' (<[\w\W]>+?) '>'/
00:26 bisectable6 AlexDaniel, On both starting points (old=2015.12 new=29b228c) the exit code is 0 and the output is identical as well
00:26 bisectable6 AlexDaniel, Output on both points:
00:26 AlexDaniel ah, right
00:27 AlexDaniel ok this is the simplified version
00:27 AlexDaniel m: "<ciao>mamma</ciao>" ~~ /:r '<' (.+?) '>'/
00:27 camelia rakudo-moar 29b228: ( no output )
00:27 giraffe joined #perl6
00:27 MasterDuke Xliff: actually, you can even remove "use XML::LibXML::Parser;", but you do still need to include "p6-XML-LibXML/lib/" in your -I
00:27 AlexDaniel m: say "<ciao>mamma</ciao>" ~~ /:r '<' (.+?) '>'/
00:27 camelia rakudo-moar 29b228: OUTPUT«Nil␤»
00:27 AlexDaniel m: say "<ciao>mamma</ciao>" ~~ /:r '<' .+? '>'/
00:27 camelia rakudo-moar 29b228: OUTPUT«「<ciao>」␤»
00:28 AlexDaniel in fact, it does not have to capture
00:28 AlexDaniel m: say "<ciao>mamma</ciao>" ~~ /:r '<' [.+?] '>'/
00:28 camelia rakudo-moar 29b228: OUTPUT«「<ciao>」␤»
00:28 AlexDaniel I think this is part of RT #130117 ?
00:28 synopsebot6 Link:  https://rt.perl.org/rt3//Public/Bug/Display.html?id=130117
00:29 MasterDuke maybe https://rt.perl.org/Ticket/Display.html?id=123934 also
00:29 Xliff Huh!
00:30 AlexDaniel maxp_: while there is probably a bug, I think it is wrong to expect <ciao> here
00:30 Xliff MasterDuke: So it is something in XML::XSLT::Subs
00:30 maxp_ ok thanks
00:30 AlexDaniel m: say "<ciao>mamma</ciao>" ~~ /:r '<' (<-[>]>+) '>'/
00:30 camelia rakudo-moar 29b228: OUTPUT«「<ciao>」␤ 0 => 「ciao」␤»
00:30 AlexDaniel maxp_: ↑ that's how you get it :)
00:30 Xliff Because I have the 2 sub definitions commented out in my current version.
00:30 maxp_ ok thanks again
00:31 Xliff And most of XML::XSLT::Subs is commented out as well.
00:31 Xliff At least here
00:31 samcv more uniprops work in new PR :) will have 'Name' 'Titlecase_Mapping' 'Lowercase_Mapping' 'Uppercase_Mapping' and 'Numeric_Value' working
00:31 samcv i think something has to happen in Moar though for Name1 the unicode version 1 name to work though, and probably a bunch of other ones
00:31 samcv there's a lot of properties
00:32 MasterDuke Xliff: MAST::Frame errors may be above my paygrade, but at least it's been golfed a bit
00:33 Xliff Yes. Thank you so much for looking into it, MasterDuke++
00:33 Xliff So it looks like it is something with XML::LibXML and XML::XSLT
00:33 Xliff But now I am more leaning toward XML::LibXML
00:34 Xliff Which... again... bitrot
00:34 Xliff *sigh*
00:37 Xliff Yup. After that bit of code culling, it looks like it is XML::LibXML and not anything with XML::XSLT at all!
00:37 Xliff wow
00:43 aborazmeh joined #perl6
00:43 aborazmeh joined #perl6
00:57 BenGoldberg joined #perl6
01:07 samcv ok so there are unicode properties to find the matching bracket Bidi_Paired_Bracket is the thing
01:07 samcv and atm it isn't supported by MoarVM or rakudo
01:07 samcv also:
01:07 samcv For legacy reasons, the characters U+FD3E ORNATE LEFT PARENTHESIS and
01:07 samcv # U+FD3F ORNATE RIGHT PARENTHESIS do not mirror in bidirectional display
01:07 samcv # and therefore do not form a bracket pair.
01:08 samcv also idk if that means we should remove it from our list. but i think it was used in 6.c ...
01:08 samcv but i doubt anybody has actually used it
01:08 AlexDaniel u: U+FD3F
01:08 unicodable6 AlexDaniel, U+FD3F ORNATE RIGHT PARENTHESIS [Ps] (﴿)
01:08 samcv and there's a bracket property where you can find out if it's open or close
01:08 phatbot u: U+FD3E
01:08 unicodable6 phatbot, U+FD3E ORNATE LEFT PARENTHESIS [Pe] (﴾)
01:09 samcv ftp://ftp.unicode.org/Public/UNIDATA/BidiBrackets.txt this is the file
01:09 samcv let me check 6.c tests...
01:09 samcv i remember it faliing when i switched the order of the ornate pars in nqp
01:09 AlexDaniel commit: all say Q﴾hello﴿
01:10 committable6 AlexDaniel, https://gist.github.com/ae5fddbcc12cba44510f9a0e14842b7a
01:10 AlexDaniel hehehe
01:10 samcv :(
01:11 samcv i think we should really only support that list i linked to...
01:11 samcv let me check roast though
01:12 canopus joined #perl6
01:12 samcv yeah.. it's in roast
01:13 phatbot Then it'd have to wait until 6.d
01:13 phatbot huggable: 6.d
01:13 huggable phatbot, Proposals for 6.d language: https://github.com/perl6/specs/blob/master/v6d.pod
01:13 phatbot ^ stick it in there, if it's worth it.
01:14 samcv u: “
01:14 unicodable6 samcv, U+201C LEFT DOUBLE QUOTATION MARK [Pi] (“)
01:14 AlexDaniel it's something to think about
01:14 samcv yeah
01:15 samcv the quotation marks aren't listed in that unicode properties but also weird that the closing and opening don't match for those ornate parens either
01:17 SmokeMachine I used my class (that instantiated a singleton) on a module, then I used the same class on my main program and tried to use the singleton... it got a different object... how should I do that?
01:21 phatbot No idea
01:21 * phatbot frequently reaches to .try
01:21 phatbot Anyone else the same?
01:21 phatbot 'fasdfsd'.EVAL.try; ....
01:21 phatbot probably doesn't make any sense :}
01:28 samcv and we will def need the BIDI mirrored glyphs thing if we want to do RTL text well
01:28 samcv i'm looking up what it would take to support the Unicode Bidirectional spec
01:29 * samcv puts on her heavy duty gear for reading this spec
01:29 aborazmeh joined #perl6
01:29 aborazmeh joined #perl6
01:29 samcv there's a BIDI unicode spec test included with unicode sa well... so someday would be good if we can pass it
01:30 phatbot https://www.youtube.com/watch?v=LNKocMiehZA
01:31 samcv i think the spec is most important for DISPLAYING RTL text.. but I will see what we need in all of our levels to support that kind of thing
01:32 samcv but we need to have some way in perl 6 to be able to figure out the order such characters should be displayed in
01:32 samcv However, because these right-to-left scripts use digits that are written from left to right, the text is actually bidirectional
01:32 unicodable6 joined #perl6
01:32 samcv lots to consider
01:34 [Coke] docs - should the perl5-style USAGEs in docs follow the 'space after comma' rule?
01:35 phatbot Yes, but I'm biased since I always write that way :D
01:35 unicodable6 joined #perl6
01:36 AlexDaniel what should whateverables do when gisting something that contains control characters?
01:38 dalek doc: 7dafc5f | coke++ | xt/words.pws:
01:38 dalek doc: learn new function param names
01:38 dalek doc: review: https://github.com/perl6/doc/commit/7dafc5f000
01:39 AlexDaniel MasterDuke: ↑ ?
01:40 pyrimidine joined #perl6
01:45 samcv ok 'Bidi_Mirroring_Glyph', atm nqp::getuniprop_str returns a string, which is the hex codepoint of the mirrored glyph
01:45 samcv i'm thinking probably Moar is storing it as a string not a number?
01:45 samcv how do we want this to work? have nqp::getuniprop_int return the number of the codepoint?
01:46 travis-ci joined #perl6
01:46 travis-ci Doc build failed. Will "Coke" Coleda 'learn new function param names'
01:46 travis-ci https://travis-ci.org/perl6/doc/builds/183459374 https://github.com/perl6/doc/compare/a4292f524145...7dafc5f0000f
01:46 travis-ci left #perl6
01:49 dalek doc: c23878a | coke++ | doc/Language/5to6-perlfunc.pod6:
01:49 dalek doc: prefer space after comma
01:49 dalek doc: review: https://github.com/perl6/doc/commit/c23878a5be
01:49 synopsebot6 Link: http://doc.perl6.org/language/5to6-perlfunc
01:51 skids joined #perl6
01:57 travis-ci joined #perl6
01:57 travis-ci Doc build failed. Will "Coke" Coleda 'prefer space after comma'
01:57 travis-ci https://travis-ci.org/perl6/doc/builds/183461255 https://github.com/perl6/doc/compare/7dafc5f0000f...c23878a5be8d
01:57 travis-ci left #perl6
02:04 dataangel joined #perl6
02:04 kalkin-_ joined #perl6
02:18 rpburkholder joined #perl6
02:19 gfldex joined #perl6
02:29 loveperl joined #perl6
02:30 loveperl How to know the name of a class?
02:30 SmokeMachine there are 2 different modules in memory when I load it at compile time and I load it at run time?
02:30 SmokeMachine m: say 42.^name
02:30 camelia rakudo-moar 29b228: OUTPUT«Int␤»
02:31 AlexDaniel m: say 42.HOW
02:31 camelia rakudo-moar 29b228: OUTPUT«Perl6::Metamodel::ClassHOW.new␤»
02:31 loveperl how?
02:31 AlexDaniel m: say 42.WHAT
02:31 camelia rakudo-moar 29b228: OUTPUT«(Int)␤»
02:31 loveperl ok tenk you
02:31 loveperl tank you
02:31 AlexDaniel m: say 42.WHO
02:31 camelia rakudo-moar 6fc769: OUTPUT«Int␤»
02:33 SmokeMachine when I get a instance of a singleton inside of a trait_mod code, I get a object and when I get it on a regular code I get a different object... is that expected?
02:37 SmokeMachine https://www.irccloud.com/pastebin/3zFvLdlH/
02:37 loveperl What do you think of this syntax?
02:37 loveperl if $saludo.WHO == Humano {  say "hola humano"; }
02:38 SmokeMachine loveperl: try using the smartmatch...
02:38 SmokeMachine m: if 42 ~~ Int { say "Hi Int" }
02:38 camelia rakudo-moar 6fc769: OUTPUT«Hi Int␤»
02:38 AlexDaniel loveperl: what are you trying to do?
02:38 AlexDaniel huggable: xy
02:38 huggable AlexDaniel, nothing found
02:39 AlexDaniel huggable: xy problem
02:39 huggable AlexDaniel, You want to do X, but don't know how. You think you can solve it using Y, but don't know how to do that, either. So, you ask about Y in order to solve X, which doesn't make sense. You should ask about X.
02:41 loveperl I want to know if the name of the class is human?
02:46 loveperl ?
02:46 loveperl what?
02:47 ilbot3 joined #perl6
02:47 Topic for #perl6 is now »ö« Welcome to Perl 6! | https://perl6.org/ | evalbot usage: 'p6: say 3;' or rakudo:,  or /msg camelia p6: ... | irclog: http://irc.perl6.org or http://colabti.org/irclogger/irclogger_logs/perl6 | UTF-8 is our friend!
02:48 dalek doc: 96761b2 | coke++ | xt/space-after-commma.t:
02:48 dalek doc: Add xt test to insure space after comma.
02:48 dalek doc:
02:48 dalek doc: * Only checks non-code.
02:48 dalek doc: * Has several exceptions for text already in docs/
02:48 dalek doc:
02:48 dalek doc: Closes #666
02:48 dalek doc: review: https://github.com/perl6/doc/commit/96761b2209
02:48 [Coke] Still no feedback on https://github.com/perl6/doc/pull/1047 - fuzzy search.
02:49 samcv ok so to implement the Unicode Bidirectional Algorithm, that would look like: given a string, determine:
02:49 samcv 1: whether the paragraph direction is LTR, RTL or autoLTR
02:49 AlexDaniel [Coke]: bring it on! You'll get a bunch of feedback ;)
02:49 samcv 2: the embedding levels of the text (embedded rtl, ltr text etc)
02:49 pyrimidine joined #perl6
02:50 loveperl AlexDaniel tank you
02:50 samcv 3: being able to return a list of the mapping of characters in the string to indexs how they map
02:50 samcv like RTL text obviously the mappings would just be the opposite of from 1->length of string
02:51 samcv or 0 or however we start
02:51 dalek doc/coke/levenshtein: ea1230a | coke++ | / (3 files):
02:51 dalek doc/coke/levenshtein: Allow fuzzy searching
02:51 dalek doc/coke/levenshtein:
02:51 dalek doc/coke/levenshtein: For #999
02:51 dalek doc/coke/levenshtein: review: https://github.com/perl6/doc/commit/ea1230a397
02:51 AlexDaniel SmokeMachine: as always, it is a bit hard to tell if there's no simple example to reproduce the issue
02:51 samcv what is anybodies thoughts on this?
02:51 samcv what kind of perl 6 routines or whatever would be want to have access to this information
02:52 SmokeMachine There are!
02:52 SmokeMachine https://www.irccloud.com/pastebin/3zFvLdlH/
02:52 SmokeMachine https://www.irccloud.com/pastebin/NK78LBlF/
02:52 AlexDaniel ahhh
02:52 AlexDaniel let's see…
02:52 samcv any thoughts on this AlexDaniel ?
02:53 AlexDaniel I feel like a rubber duck :)
02:53 SmokeMachine AlexDaniel: the first one is the problem, and the second is it working on the "same file"...
02:55 pyrimidine joined #perl6
02:55 AlexDaniel samcv: wait, what problem are you trying to solve?
02:55 * samcv is unsure
02:55 travis-ci joined #perl6
02:55 travis-ci Doc build failed. Will "Coke" Coleda 'Add xt test to insure space after comma.
02:55 travis-ci https://travis-ci.org/perl6/doc/builds/183470235 https://github.com/perl6/doc/compare/c23878a5be8d...96761b22094d
02:55 travis-ci left #perl6
02:55 samcv problem of not implementing the unicode bidirectional algorithm?
02:56 samcv and not being able to pass the test given in the unicode spec
02:56 samcv for said algorithm
02:57 AlexDaniel is it for string operations or for the source code? :)
02:57 samcv well. string operations for sure
02:58 kyclark joined #perl6
02:58 samcv i'm just trying to think about how that would look... are there any perl 6 devs whose native language is RTL?
02:58 travis-ci joined #perl6
02:58 travis-ci Doc build failed. Will "Coke" Coleda 'Allow fuzzy searching
02:58 travis-ci https://travis-ci.org/perl6/doc/builds/183470746 https://github.com/perl6/doc/compare/7ebb0d77b368...ea1230a397f9
02:58 travis-ci left #perl6
02:59 cyphase joined #perl6
03:00 samcv AlexDaniel, also source code would be interesting... hah
03:01 samcv use v6; use RTL; hah
03:01 samcv or even maybe not as something you have to use... hm
03:01 samcv m: say "\x[05D0]\x[05D1]\x[0028]\x[05D2]\x[05D3]\x[005B]\x[0026]\x[0065]\x[0066]\x[005D]\x[002E]\x[0029]\x[0067]\x[0068]"
03:01 camelia rakudo-moar 6fc769: OUTPUT«אב(גד[&ef].)gh␤»
03:01 samcv like this string is one of the unicode spec strings
03:01 AlexDaniel SmokeMachine: the two numbers in the first case…
03:02 AlexDaniel SmokeMachine: I only get one on any subsequent runs
03:02 AlexDaniel SmokeMachine: so one of them is from precompilation, right?
03:02 samcv paragraph direction is supposed to be LTR
03:02 samcv 1 0 2 4 3 5 6 7 8 9 10 11 12 13
03:02 samcv this is the order they are shown in
03:02 [vlad] joined #perl6
03:03 samcv and that parenthesis should prolly be shown so that there's open and closing. at least for me it displays two right parens
03:04 SmokeMachine AlexDaniel: sorry, I didn't get...
03:04 AlexDaniel SmokeMachine: well, your first example with -MFuzz -MSum
03:04 AlexDaniel SmokeMachine: so I run it and get two different numbers
03:05 AlexDaniel SmokeMachine: BUT! If I run it again, I only get one number
03:05 SmokeMachine yes... me too... shouldn't be the same?
03:05 AlexDaniel SmokeMachine: if I delete .precomp, I'd get two numbers again
03:05 SmokeMachine yes, the same with me!
03:05 AlexDaniel that's just how the precompilation works, I think
03:06 SmokeMachine so, I can't use singleton on trait_mod?
03:08 AlexDaniel SmokeMachine: why not?
03:08 AlexDaniel SmokeMachine: you can discard what happens during the precompilation
03:09 kyclark joined #perl6
03:09 SmokeMachine in my case I was pushing the functions with the trait on a array on the singleton to use it in another part of the code...
03:10 SmokeMachine but it returns another object, so I can't get the funds...
03:10 SmokeMachine *funcs
03:10 noganex_ joined #perl6
03:11 SmokeMachine AlexDaniel: My "real" case is this one: https://github.com/FCO/Test-Fuzz
03:11 SmokeMachine https://github.com/FCO/Test-Fuzz/blob/master/lib/Test/Fuzz.pm
03:14 kyclark joined #perl6
03:15 SmokeMachine on https://github.com/FCO/Test-Fuzz/blob/master/lib/Test/Fuzz.pm#L115 and here https://github.com/FCO/Test-Fuzz/blob/master/lib/Test/Fuzz.pm#L104 I save the function on the %!fuzzers and on https://github.com/FCO/Test-Fuzz/blob/master/lib/Test/Fuzz.pm#L157 I use that... but its empty, because its another object...
03:15 SmokeMachine AlexDaniel: ^^
03:16 samcv m: my @o = "\x[05D0]\x[05D1]\x[0028]\x[05D2]\x[05D3]\x[005B]\x[0026]\x[0065]\x[0066]\x[005D]\x[002E]\x[0029]\x[0067]\x[0068]".ords; my @n; my @nums = "1 0 2 4 3 5 6 7 8 9 10 11 12 13".split(' ').».Int; for ^@o -> $i { @n[ @nums[$i] ] = @o[$i] }; say @n.».chr;
03:16 camelia rakudo-moar 6fc769: OUTPUT«[ב א ( ד ג [ & e f ] . ) g h]␤»
03:18 samcv the hebrew characters are the opposite order in this if you look closely
03:19 samcv tho that's more boring cause it's a LTR paragraph
03:19 AlexDaniel SmokeMachine: this is some weird stuff going on there! :)
03:22 SmokeMachine AlexDaniel: So, thats not expected?
03:26 rurban joined #perl6
03:27 AlexDaniel SmokeMachine: I don't think I know enough about traits to answer :(
03:27 SmokeMachine :(
03:27 SmokeMachine AlexDaniel: do you know who could help me?
03:31 AlexDaniel SmokeMachine: I think just asking this question here at a different time of a day is a good idea :)
03:31 pyrimidine joined #perl6
03:31 SmokeMachine AlexDaniel: OK! Ill do that!!! Thank you very much!
03:34 AlexDaniel SmokeMachine: so in other words, you want to push some data into %!fuzzers during the compilation?
03:34 SmokeMachine thats is!
03:34 SmokeMachine thats it!
03:34 SmokeMachine AlexDaniel: ^^
03:40 AlexDaniel yea, how to save some data from a trait… hmm!
03:41 AlexDaniel I don't know! Time to sleep :)
03:47 [vlad] left #perl6
03:55 samcv i want a uniprop-enum ...
03:55 samcv hm
03:55 tailgate how do I match [\w -]+ (alphanumeric + dashes, at least one)
03:55 tailgate can't quite get the syntax
03:56 samcv m: say '-a' ~~ /[\w -]+/
03:56 camelia rakudo-moar 6fc769: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Unrecognized regex metacharacter - (must be quoted to match literally)␤at <tmp>:1␤------> 3say '-a' ~~ /[\w7⏏5 -]+/␤    expecting any of:␤        infix stopper␤»
03:56 samcv m: say '-a' ~~ /[\w '-']+/
03:56 camelia rakudo-moar 6fc769: OUTPUT«Nil␤»
03:56 samcv m: say '-a' ~~ /[<Alpha> '-']+/
03:56 camelia rakudo-moar 6fc769: OUTPUT«No such method 'Alpha' for invocant of type 'Cursor'␤  in block <unit> at <tmp> line 1␤␤»
03:56 samcv oh
03:56 samcv m: say '-a' ~~ /<[\w -]>+/
03:56 camelia rakudo-moar 6fc769: OUTPUT«「-a」␤»
03:56 samcv there
03:56 samcv m: say '-a' ~~ /<[<Alpha> -]>+/
03:56 camelia rakudo-moar 6fc769: OUTPUT«「-a」␤»
03:57 samcv that is what you want
03:57 samcv er alpha num tho
03:57 tailgate thanks
03:57 samcv m: say '-a1' ~~ /<[<Alpha> -]>+/
03:57 camelia rakudo-moar 6fc769: OUTPUT«「-a」␤»
03:57 samcv m: say '-a1' ~~ /<[<AlphaNum> -]>+/
03:57 camelia rakudo-moar 6fc769: OUTPUT«「-a」␤»
03:57 samcv forget which that one is
03:58 samcv m: say '-a1' ~~ /<[<Alpha>\d -]>+/
03:58 camelia rakudo-moar 6fc769: OUTPUT«「-a1」␤»
03:58 samcv that works
04:12 aborazmeh joined #perl6
04:12 aborazmeh joined #perl6
04:15 samcv pythons unicodedata module that comes with it is kinda lame https://docs.python.org/2/library/unicodedata.html#module-unicodedata
04:22 SmokeMachine I just started my "Generative testing" draft. Could someone take a look and give me some feedback please?
04:27 samcv is this the same thing you were talking about earlier with AlexDaniel or somethnig different?
04:28 samcv i'll look at it though, curious not sure how much I can help though
04:29 SmokeMachine samcv: That's my draft for perl6 advent calendar... talking about the module that I was talking with AlexDaniel...
04:29 samcv oh i'll look at it then!
04:29 SmokeMachine samcv: thats on the wordpress...
04:30 samcv oh
04:30 samcv :(
04:32 SmokeMachine samcv: http://pastebin.com/McHWeum7
04:33 samcv nice
04:35 SmokeMachine samcv: did you like it?! :)
04:35 samcv does Test::Fuzz know what the result should be?
04:35 samcv or does it just tell you if it gets an error?
04:36 SmokeMachine by default it just test if it did break...
04:37 samcv ah kk :)
04:37 SmokeMachine but you can set sub function(...) is fuzzed(test => *.is-prime) {}
04:37 SmokeMachine for example...
04:37 SmokeMachine if you want that the function only return prime numbers...
04:38 SmokeMachine like the function blu here: https://github.com/FCO/Test-Fuzz/blob/master/test.p6
04:38 SmokeMachine samcv: ^^
04:39 samcv nice
04:40 SmokeMachine did you find any terrible english error?
04:40 SmokeMachine samcv: ^^
04:40 samcv nope
04:41 SmokeMachine great!
04:41 SmokeMachine thank you very much!
04:41 SmokeMachine samcv: Ill continue tomorrow!
04:41 samcv good start so far :)
04:42 SmokeMachine samcv: :D
04:42 SmokeMachine samcv: thank you very much!
04:45 kyclark joined #perl6
05:06 MasterDuke .tell AlexDaniel it looks like you implemented something already, but i'd say just return the name for them or something like that, i don't see anybody ever wanting them for real
05:06 yoleaux MasterDuke: I'll pass your message to AlexDaniel.
05:12 geraud joined #perl6
05:25 pyrimidine joined #perl6
05:38 telex joined #perl6
05:46 geraud joined #perl6
05:49 Cabanossi joined #perl6
05:55 geraud joined #perl6
06:00 rurban joined #perl6
06:01 brillenfux joined #perl6
06:18 BenGoldberg joined #perl6
06:22 Vynce joined #perl6
06:31 pyrimidine joined #perl6
06:34 darutoko joined #perl6
06:49 dalek perl6.org: cb115bf | samcv++ | source/whatever/index.html:
06:49 dalek perl6.org: Add links for atom-language-perl6 and script
06:49 dalek perl6.org:
06:49 dalek perl6.org: For syntax highlighting and running Perl 6 from Atom
06:49 dalek perl6.org: review: https://github.com/perl6/perl6.org/commit/cb115bf742
06:50 RabidGravy joined #perl6
07:03 jonas1 joined #perl6
07:08 ufobat joined #perl6
07:09 imcsk8_ joined #perl6
07:11 cooper_ joined #perl6
07:11 cooper_ joined #perl6
07:12 RabidGravy morning
07:12 RabidGravy I see I managed to get the scheduled post thing right in the end
07:21 domidumont joined #perl6
07:27 domidumont joined #perl6
07:31 moritz RabidGravy++
07:36 pyrimidine joined #perl6
07:54 kurahaupo joined #perl6
08:03 dataange` joined #perl6
08:03 [particle]1 joined #perl6
08:03 labster joined #perl6
08:07 ilbelkyr joined #perl6
08:07 Actualeyes joined #perl6
08:13 abraxxa joined #perl6
08:14 wamba joined #perl6
08:22 ufobat morning
08:23 ufobat the module Task::Star does not have a travis-ci integration, is that on purpose?
08:23 ufobat i mean, isnt it a pretty good test if Task::Star is installable via zef and/or panda?
08:24 Xliff What are dynvars? Has anything been written up about them?
08:24 zakharyas joined #perl6
08:24 moritz ufobat: sure, go for it!
08:24 ufobat kk :)
08:25 MARTIMM joined #perl6
08:29 arnsholt Xliff: A dynvar (also called contextual in Perl 6 parlance) is a variable that is looked up through *dynamic*, rather than lexical, scope
08:29 arnsholt Where dynamic scope follows the call stack, so if the name is found in the current scope, you look in your caller's scope, and so on
08:30 arnsholt They're somewhat uncommon in day-to-day code, but when you need them you really need them
08:30 moritz you use them when in the bad old days, you'd use a global
08:30 arnsholt They're very useful for global variables for example. Instead of a global STDOUT, Perl 6 has a dynvar $*OUT
08:30 pyrimidine joined #perl6
08:31 moritz but they are more thread safe and generally safer than globals
08:31 arnsholt This way, rather than overriding a global before doing something and making sure it's reset correctly and all that crap when you leave, you create a new binding for it with "my $*OUT = ..."
08:31 arnsholt That way, STDOUT will be altered *only* inside the current function and code called by it
08:32 arnsholt And as moritz says, thread safety
08:32 arnsholt They're also very useful in grammar code
08:32 arnsholt Since Perl 6 grammars are recursive descent, lexical scope in the program being compiled will correspond to the dynamic scope of the parser
08:33 moritz arnsholt: that concept might deserve its own blog post some day
08:33 moritz I've never seen it explained well outside of #perl6
08:33 arnsholt Am I being volunteered? ;)
08:33 arnsholt But more seriously, I agree
08:34 pierre_ joined #perl6
08:34 arnsholt It's a useful concept, and not very complicated either. Just not very common
08:34 arnsholt (Although older Lisps like elisp frequently only have dynamically scoped variables, not lexically scoped)
08:34 arnsholt And apparently you can do some magic with macros and dynamic scope to implement lexical scope
08:38 Xliff arnsholt++ # Brilliant explanation. Thanks!
08:39 samcv what unicode properties are allowed for heredocs? anything?
08:39 samcv except forward slashes or something?
08:39 dakkar joined #perl6
08:39 samcv m: say q:to/'/;␤test␤'
08:39 camelia rakudo-moar 6fc769: OUTPUT«test␤␤»
08:40 samcv m: my $var = "\0"; say q:to/$var/;␤test␤'
08:40 camelia rakudo-moar 6fc769: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Ending delimiter $var not found␤at <tmp>:3␤------> 3'7⏏5<EOL>␤    expecting any of:␤        whitespace␤»
08:40 samcv any nonwhitespace character?
08:43 psch m: say q:to{/};␤test␤/
08:43 camelia rakudo-moar 6fc769: OUTPUT«test␤␤»
08:44 pierre_ joined #perl6
08:44 psch samcv: <-[ws]> sounds reasonable to me, but you'd probably have to check Perl6::Grammar to be sure
08:45 samcv yeah it looks like <-ws>
08:45 samcv probably
08:45 psch right, not the character class consisting of 'w' and 's' :)
08:47 psch or, actually, 'not "not the character class consisting of 'w' and 's'"' vOv
08:47 psch +plus proper nested quoting parsing :P
08:47 samcv heredocs do quoted parsing? what?
08:48 psch no, just for my statement... :)
08:48 samcv m: say q:to/{/;␤this{}␤}
08:48 camelia rakudo-moar 6fc769: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Ending delimiter { not found␤at <tmp>:3␤------> 3}7⏏5<EOL>␤    expecting any of:␤        whitespace␤»
08:48 samcv oh ok
08:48 psch heredocs parse whatever language you give them
08:48 psch well, from Q and its sisters anyway
08:48 samcv yeah i usually use Q or qq if i need interpolation
08:49 samcv i like Q
08:49 samcv very much
08:49 psch m: say qqw:to/foo/;␤{ "foo", "bar" }␤baz␤foo
08:49 camelia rakudo-moar 6fc769: OUTPUT«(foo bar baz)␤»
08:49 samcv m: say "\r"
08:49 camelia rakudo-moar 6fc769: OUTPUT«
08:49 samcv aww no carriage return chars?
08:49 samcv err
08:49 psch .u carriage return
08:49 yoleaux U+000D CARRIAGE RETURN (CR) [Cc] (␍)
08:49 yoleaux U+240D SYMBOL FOR CARRIAGE RETURN [So] (␍)
08:49 samcv yeah
08:50 samcv lemme see what my bot does
08:50 psch could maybe stuff that into camelia as well, or whateverable maybe
08:50 psch eval: say "\r"
08:50 evalable6 psch, rakudo-moar 6fc7698: OUTPUT«»
08:50 samcv oh it doeqsn't hm
08:50 psch well, that at least doesn't eat the >>
08:50 samcv m: say "\t"
08:50 camelia rakudo-moar 6fc769: OUTPUT«     ␤»
08:50 samcv :)
08:50 samcv i mean :(
08:50 samcv should be ↹
08:51 samcv my bot deos that at least, ↹
08:51 samcv though i guess less important than doing \t
08:51 samcv err \r i'm tired
08:57 andrzeju_ joined #perl6
09:00 g4 joined #perl6
09:01 baest joined #perl6
09:05 rindolf joined #perl6
09:06 dalek ecosystem: 4a13e58 | (Takumi Akiyama)++ | META.list:
09:06 dalek ecosystem: Add JSON::Hjson to ecosystem
09:06 dalek ecosystem:
09:06 dalek ecosystem: See https://github.com/akiym/JSON-Hjson
09:06 dalek ecosystem: review: https://github.com/perl6/ecosystem/commit/4a13e58f81
09:06 Vynce joined #perl6
09:06 dalek ecosystem: e349294 | RabidGravy++ | META.list:
09:06 dalek ecosystem: Merge pull request #273 from akiym/master
09:06 dalek ecosystem:
09:06 dalek ecosystem: Add JSON::Hjson to ecosystem
09:06 dalek ecosystem: review: https://github.com/perl6/ecosystem/commit/e349294d5a
09:08 ccntrq joined #perl6
09:12 thayne joined #perl6
09:12 grondilu joined #perl6
09:14 ccntrq joined #perl6
09:23 bjz joined #perl6
09:31 pyrimidine joined #perl6
09:41 astj joined #perl6
09:42 astj moritz: Hi. Thank you for inviting me to github/perl6! I accepted. But I wonder for which reason I've invited.  Is there any reason?
09:42 ccntrq joined #perl6
09:44 moritz astj: you submitted a pull request to the ecosystem
09:44 moritz astj: in future, you can push directly. And merge other's pull request, should you feel so inclined
09:45 astj moritz: Okay, thanks!
09:45 moritz astj: the ecosystem list was never meant to enable a gate keeper, we just did it because all our stuff is in github
09:45 moritz astj: you're welcome
09:53 rurban joined #perl6
09:53 ccntrq joined #perl6
09:55 pmurias joined #perl6
10:02 bjz joined #perl6
10:05 mniip joined #perl6
10:07 astj joined #perl6
10:10 pierre_ joined #perl6
10:24 sena_kun joined #perl6
10:37 pmurias jnthn: does checking if objects need to be repossesed have a big performance price on MoarVM?
10:38 wamba joined #perl6
10:39 jnthn pmurias: iirc, most of the time once we're running code it's just an integer comparison (because by runtime where's no SCs on the compiling stack)
10:39 jnthn s/where/there/
10:40 jnthn Presumably that stays CPU cache hot, and the branch is predictable, so it's not too awful
10:44 lucasb joined #perl6
10:49 lukaramu joined #perl6
10:51 ufobat moritz, done ;) but i guess tadzik has to add the module to the CI test services https://github.com/tadzik/Task-Star/pull/12
10:52 tadzik oh, will do
10:54 ufobat cool
10:55 El_Che not enough sleep, not enough coffee and reading jvm docs for performance tuning a big jvm service. Bad combination
10:55 ufobat hmm on the other side, does it really make sense
10:56 ufobat the tests are triggered whenever you.. push something?
10:56 ufobat that is probably wrong for Task::Star
10:56 pyrimidine joined #perl6
11:03 bjz_ joined #perl6
11:04 dalek doc: ea1230a | coke++ | / (3 files):
11:04 dalek doc: Allow fuzzy searching
11:04 dalek doc:
11:04 dalek doc: For #999
11:04 dalek doc: review: https://github.com/perl6/doc/commit/ea1230a397
11:04 dalek doc: 2586a29 | (Zoffix Znet)++ | / (3 files):
11:04 dalek doc: Merge pull request #1047 from perl6/coke/levenshtein
11:04 dalek doc:
11:04 dalek doc: Allow fuzzy searching
11:04 dalek doc: review: https://github.com/perl6/doc/commit/2586a29a54
11:07 zakharyas joined #perl6
11:11 cognominal joined #perl6
11:12 dogbert17 joined #perl6
11:12 travis-ci joined #perl6
11:12 travis-ci Doc build failed. Zoffix Znet 'Merge pull request #1047 from perl6/coke/levenshtein
11:12 travis-ci https://travis-ci.org/perl6/doc/builds/183553537 https://github.com/perl6/doc/compare/96761b22094d...2586a29a543b
11:12 travis-ci left #perl6
11:19 pierre_ joined #perl6
11:22 phatbot Same fromspace collectable accesing error "MoarVM panic: Collectable 0x2ae0503e2918 in fromspace accessed"
11:43 phatbot .tell cygx reminder: your post is up next in ~12.5 hours
11:43 yoleaux phatbot: I'll pass your message to cygx.
11:58 lucasb .oO( hmm, a Collectable fromOuterSpace )
12:04 brrt jnthn ^ what phatbot says
12:17 nicky joined #perl6
12:18 nicky ==> Testing File::Find Unhandled exception: Failed to write to filehandle: NULL string given
12:18 brrt /me wonders, just out of interest, how hard it would be to have a 'perl6-syntax-check' daemon process and an emacs mode to use it
12:18 psch brrt: something like "run perl6 -c on the current buffer after n seconds of idle" or what are you imagining?
12:19 psch i suspect that depends more on how emacs works than what perl6 offers :)
12:20 brrt something like that
12:20 brrt well, emacs has flymake-mode, iirc
12:20 brrt so that is just something to hack into
12:20 brrt but the current wisdom is to use a client-server model for that, iirc
12:26 bjz joined #perl6
12:29 CIAvash[m] brrt: https://github.com/hinrik/flycheck-perl6
12:31 tbrowder RabidGravy: nice Advent post! It helps me get a feel for using concurrency in the real world.
12:35 nicky I'm getting a make install failure on rakudo star 2016.11 http://pastebin.com/7PzmcaTS
12:39 sena_kun joined #perl6
12:46 phatbot sena_kun: I don't think building %method_to_role is needed anymore, since we already know method's name by that point.
12:47 brrt ooh that is cool
12:48 brrt thanks CIAvash[m]
12:48 sena_kun phatbot, method name - yes, but role?
12:49 phatbot sena_kun, that's what the added logic needs to do, look for a role with that method and tells us the role name when found. There's no need to store names of all methods for that
12:50 sena_kun phatbot, ah, so that's your point. Yes, you are correct. Another patch is coming.
12:51 phatbot \o/
13:03 arnsholt Good point!
13:04 arnsholt That should make it easier to report all roles at once too
13:06 rurban joined #perl6
13:08 sena_kun I've pushed update. arnsholt, do you think we need reporting of all roles with the unimplemented method after all? It is not hard to write, just asking.
13:08 arnsholt Dunno
13:08 phatbot Wouldn't there be a merge conflict anyway?
13:08 phatbot or whatever it's called
13:08 phatbot m: role A { method foo {} }; role B { method foo {} }; class Z does A does B {}
13:08 camelia rakudo-moar 6fc769: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Method 'foo' must be resolved by class Z because it exists in multiple roles (B, A)␤at <tmp>:1␤»
13:09 phatbot m: role A { method foo {...} }; role B { method foo {} }; class Z does A does B {}
13:09 camelia rakudo-moar 6fc769: ( no output )
13:09 phatbot m: role A { method foo {...} }; role B { method foo {...} }; class Z does A does B {}
13:09 camelia rakudo-moar 6fc769: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Method 'foo' must be implemented by Z because it is required by a role␤at <tmp>:1␤»
13:09 phatbot m: role A { method foo {...} }; role B { method foo {...} }; class Z does A does B { method foo {} }
13:09 camelia rakudo-moar 6fc769: ( no output )
13:09 arnsholt Yeah, that's what I thought. The conflict only crops up if both roles have defaults
13:10 phatbot sena_kun: from the above it does look like reporting all the roles could be useful.
13:10 arnsholt Yeah, I'd add it I think
13:10 psch m: role A { method foo {...} }; role B { method foo {} }; class Z does B does A {}
13:10 camelia rakudo-moar 6fc769: ( no output )
13:10 arnsholt It's relevant knowledge about how to implement the required method, IMO
13:11 phatbot sena_kun: if it's not hard :) I know we've been kinda keeping you on the wheel... pushing all these changes :)
13:12 sena_kun phatbot, no problem, thanks for comments anyway. But all methods commit will be a bit later, since I'm busy right now.
13:13 matiaslina joined #perl6
13:13 canopus joined #perl6
13:14 wamba joined #perl6
13:20 pierre_ joined #perl6
13:21 kurahaupo joined #perl6
13:28 lucasb m: say (my %h{Int}).perl
13:28 camelia rakudo-moar 6fc769: OUTPUT«(my Any %{Int} = )␤»
13:28 dalek doc: 3e88149 | coke++ | xt/space-after-commma.t:
13:28 dalek doc: remove trailing whitespace
13:28 dalek doc: review: https://github.com/perl6/doc/commit/3e88149be2
13:28 dalek doc: 2af9049 | coke++ | xt/trailing-whitespace.t:
13:28 dalek doc: ignore this 3rd party file
13:28 dalek doc: review: https://github.com/perl6/doc/commit/2af9049c11
13:28 lucasb ^^ no need for the equal sign
13:29 phatbot m: say (my %h{Int}).perl.EVAL.perl
13:29 camelia rakudo-moar 6fc769: OUTPUT«5===SORRY!5=== Error while compiling /home/camelia/EVAL_0␤Malformed initializer␤at /home/camelia/EVAL_0:1␤------> 3(my Any %{Int} = 7⏏5)␤    expecting any of:␤        prefix␤        term␤»
13:29 phatbot Yes, .EVAL.perl is meant to roundtrip
13:29 lucasb SourceBaby: help
13:29 SourceBaby lucasb, Use s: trigger with args to give to sourcery sub. e.g. s: Int, 'base'. See http://modules.perl6.org/dist/CoreHackers::Sourcery
13:29 lucasb s: Hash, 'perl'
13:29 SourceBaby lucasb, Sauce is at https://github.com/rakudo/rakudo/blob/6fc7698/src/core/Mu.pm#L527
13:29 lucasb s: {}, 'perl'
13:29 SourceBaby lucasb, Ehhh... I'm too scared to run that code.
13:29 phatbot s: Hash, 'perl', \()
13:29 SourceBaby phatbot, Sauce is at https://github.com/rakudo/rakudo/blob/6fc7698/src/core/Mu.pm#L528
13:30 ZoffixW joined #perl6
13:30 ZoffixW s: (my %h{Int}), 'perl', \()
13:30 SourceBaby ZoffixW, Sauce is at https://github.com/rakudo/rakudo/blob/6fc7698/src/core/Hash.pm#L704
13:30 ZoffixW left #perl6
13:30 phatbot It's not scared of me ;)
13:30 prammer joined #perl6
13:30 phatbot Err... I mean of this Zoffix guy
13:31 lucasb ah, it feels the k/v pairs after the equal sign
13:31 lucasb *it fills :)
13:34 sena_kun I'm going to switch for $must-to-do-it-work now, so the patch will be tomorrow, I suppose. Sigh.
13:35 phatbot sena_kun: I'll merge current version then. And you can submit another PR for multi-role thing when you get a chance
13:36 travis-ci joined #perl6
13:36 travis-ci Doc build failed. Will "Coke" Coleda 'ignore this 3rd party file'
13:36 travis-ci https://travis-ci.org/perl6/doc/builds/183586638 https://github.com/perl6/doc/compare/2586a29a543b...2af9049c11af
13:36 travis-ci left #perl6
13:36 sena_kun phatbot, got it. Thanks for help, everyone.
13:37 pyrimidine joined #perl6
13:38 nicq20 joined #perl6
13:42 kyclark joined #perl6
13:42 sufrostico joined #perl6
13:45 pierre__ joined #perl6
13:50 lucasb m: say Mu.is-lazy, Hash.is-lazy
13:50 camelia rakudo-moar 6fc769: OUTPUT«FalseFalse␤»
13:50 lucasb m: say List.is-lazy
13:50 camelia rakudo-moar 6fc769: OUTPUT«Cannot look up attributes in a List type object␤  in block <unit> at <tmp> line 1␤␤»
13:51 lucasb can List get a method is-lazy(List:U) { False }  ?
13:51 phatbot s: List, 'is-lazy'
13:51 SourceBaby phatbot, Sauce is at https://github.com/rakudo/rakudo/blob/6fc7698/src/core/List.pm#L1001
13:51 jnthn Should do, yes
13:51 psch m: say Any.is-lazy
13:51 camelia rakudo-moar bff051: OUTPUT«False␤»
13:51 phatbot I think a better approach would be for ^ that one to be a :D and let Mu handle the :U
13:52 lucasb phatbot: right
13:59 phatbot m: use MONKEY; my $r := ^2; nqp::getattr($r, Range, '$!is-int') = 0;
13:59 camelia rakudo-moar bff051: OUTPUT«Cannot modify an immutable int␤  in block <unit> at <tmp> line 1␤␤»
13:59 phatbot how can I change that attribute?
14:00 phatbot m: use MONKEY; my $r := ^2; nqp::bindattr($r, Range, '$!is-int', 0);
14:00 camelia rakudo-moar bff051: OUTPUT«P6opaque: representation mismatch when storing value (of type Int) to attribute (of type int)␤  in block <unit> at <tmp> line 1␤␤»
14:01 phatbot m: use MONKEY; my $r := ^2; nqp::bindattr($r, Range, '$!is-int', nqp::istype(42, Bool));
14:01 camelia rakudo-moar bff051: OUTPUT«P6opaque: representation mismatch when storing value (of type Int) to attribute (of type int)␤  in block <unit> at <tmp> line 1␤␤»
14:01 jnthn _i
14:01 [Coke] phatbot: thanks for the merge. OOC, did you test or just assume we'd get some feedback only after mergeback? ;)
14:02 phatbot jnthn++ thanks
14:02 Ven joined #perl6
14:02 phatbot [Coke]: I didn't test, no.
14:03 phatbot oh shit... perhaps I should have...
14:03 phatbot I get freezes on my box :}
14:04 phatbot If I type "str" it takes it like 2 seconds to react to any further actions
14:04 phatbot Mmm
14:04 phatbot Or maybe it's something else entireely. Seems to be fine on another box in Chrome.
14:05 phatbot oh, but same lag if I search for "a". And after it displays results, moving mouse around results has a delay before it reacts to anything. Is it just me?
14:06 sena_kun phatbot, confirmed on latest firefox.
14:06 kyclark joined #perl6
14:09 cibs joined #perl6
14:11 phatbot [Coke]: so seems like there's a performance issue? Not sure if anything can be improved with it. Perhaps trigger levenshtein only after $x characters have been typed?
14:11 phatbot It's nice that it shows results for in_timezone, for example :)
14:11 [Coke] alexdaniel had suggested a faster levenshtein-like algorithm as well.
14:12 [Coke] I can make it skip a few chars, one sec.
14:12 [Coke] ... er
14:12 cdg joined #perl6
14:13 [Coke] more like 10s of minutes. Have to find what the old default algorithm was, then check term.length before using getEditDistance()
14:15 [Coke] I can do it after $DAYJOB if someone hasn't beaten me to it.
14:16 [Coke] also, looks like someone changed the SASS processing, as external links are now being shown.
14:20 [Coke] https://github.com/perl6/doc/issues/999 was the original ticket, commented about the slowdown and algorithm there.
14:21 phatbot That was me. I didn't change SASS processing, I just followed the proceedure we have for changing styles. You're meant to change assets/sass/style.sass (or whatever it is) and then run the dev app to update html/css/style.css (or whatver it is) and commit both files to repo.
14:21 phatbot And I guess I promised a few months ago to make the doc build do that stuff so we could nix html/css/style.css from repo
14:22 phatbot and still haven't done it :}
14:22 * phatbot adds it onto Christmas Holiday Fixes list
14:26 kurahaupo joined #perl6
14:28 mawkish_ joined #perl6
14:31 mawkish_ joined #perl6
14:37 tbrowder ref p6 publicity: i think i've seen p6 described as a "large" core language, esp. compared to minimalist C.  what's opinion of putting  stats on perl6.org with numbers of built-in routines, classes, modules, dynamic vars, etc.? should, hopefully, be greppable, cacheable, and automatable.
14:38 phatbot So that all the people who compare us to PHP would have actual numbers to point to? :)
14:39 tbrowder roger! but maybe emphasize a subset for most new users.
14:39 phatbot I don't see anything posivite comming out of such an endeavour and only negative. Especially since a lot of routines haven't been moved to Rakudo::Internal yet
14:39 jnthn What do you count as the core of the C language?
14:40 jnthn libc isn't exactly small :)
14:40 prammer joined #perl6
14:40 jnthn And if that doesn't count, then nor does most of CORE.setting. :)
14:40 tbrowder hm, all inside rakudo seen by users?
14:40 phatbot yes
14:41 phatbot m: use Test; is-deeply SEQUENCE(1, 6), (1...6)
14:41 camelia rakudo-moar 1d4600: OUTPUT«ok 1 - ␤»
14:42 phatbot Like that's not a user sub
14:42 jnthn Sure, but my question is more about where do you draw the line between language and standard library.
14:43 phatbot m: CORE::.values.grep({$_ ~~ Sub}).elems.say
14:43 camelia rakudo-moar 1d4600: OUTPUT«121␤»
14:43 jnthn It's for sure that Perl 6 has a decidedly larger grammar than C, Java, etc.
14:43 cdg_ joined #perl6
14:43 tbrowder of course that's a good question, but someone deeply familiar with the lang should be able to draw it
14:43 phatbot And well, you can override core stuff, as opposed to many other languages where you get a conflict
14:44 kyclark joined #perl6
14:44 Actualeyes joined #perl6
14:44 tbrowder part of the super language aspect, for advanced users
14:44 phatbot Draw what?
14:45 tbrowder a line between "core" and library
14:45 phatbot Seems pointless semantics game to me.
14:45 tbrowder there are always folks who like such stats
14:46 tbrowder when, say, comparing languages for suitability for their purposes
14:46 jnthn It could be nice to show the batteries that Perl 6 includes in its standard library.
14:47 tbrowder there is a lot appeal for a lang that can do a lot without hunting for  lots of libraries
14:47 tadzik yep
14:47 tbrowder extra libraries
14:48 phatbot hah
14:48 phatbot You've not seen Zefram's talk then, I guess :)
14:48 tbrowder hm, is he yr brother?
14:49 * lizmat clickbaits https://p6weekly.wordpress.com/2016/12/12/2016-50-highlighting-profilesql/
14:49 lizmat Files=1155, Tests=53893, 200 wallclock secs (11.34 usr  4.53 sys + 1192.98 cusr 126.25 csys = 1335.10 CPU)
14:49 tbrowder no, i don't think so
14:49 jnthn fwiw, it depends what the "a lot" is, to me. Putting Set/Bag/Rat/Complex into the standard library is a non-brainer because math ain't going to suddently come out of date tomorrow. Same for stuff like async sockets/processes.
14:50 jnthn With serialization formats though, it's much more touch and go. The current fashionable ones come and go.
14:51 jnthn And some things are better just having different release cadences than the core language.
14:51 phatbot tbrowder: yes, all people with Z in their name are related.
14:52 phatbot I can't find the URL right now, but his entire talk was about dissing Perl 6 for having too many features, while showing bugs with their interaction and coming to the conclusion that Perl 6 is inherently flawed and must be discarded.
14:52 psch i think one notable point was that assigning IterationEnd to an elemnt in an Array breaks iteration..?
14:52 psch +e
14:52 arnsholt That sounds a lot like the infamous "we found bugs in Bugzilla and CGI.pm, therefore Perl is inherently broken" talk that cropped up a few years back
14:52 phatbot So people like him, along with folks who like to make fun of PHP for having many built ins, would just love your stats, tbrowder
14:53 tbrowder well, one way to handle those types is to ignore them
14:53 phatbot arnsholt: yeah. The premise is inherently flawed. He'd have to reach for library XYZ to do feature X we do in core, giving him with as many "interactions", except now you have completely different projects with completely different styles and authors in play.
14:54 psch i think it was http://act.yapc.eu/lpw2015/talk/6436
14:54 jnthn phatbot: Yes, that's pretty much what I was going to write :)
14:54 phatbot :)
14:54 phatbot psch: aye, title looks familiar
14:55 [Coke] Why do talks like that get approved, I wonder.
14:56 [Coke] at a perl conference, I mean.
14:57 skids joined #perl6
14:59 mawkish_ joined #perl6
14:59 jnthn [Coke]: Perhaps because having a community that won't listen to dissenting ideas isn't really healthy in general. :)
14:59 psch well, the talk has been recorded
14:59 tbrowder i couldn't find the talk itseif, but i think i've seen parts.  he sounds like a large group of. nay-sayers over at perl monks
14:59 psch although i personally find the audio utterly incomprehensible
15:01 zakharyas joined #perl6
15:02 tbrowder listening is fine, i leave it to smarter people to refute facts, and get on to helping advance my own interests with folks who share them--life is too short
15:03 kyclark joined #perl6
15:03 pierre_ joined #perl6
15:04 [Coke] jnthn: It's the question of dissent, but respect.
15:04 [Coke] *sigh*. "it's NOT ..."
15:05 tbrowder i didn't mean refute facts...i meant argue as in debate
15:05 [Coke] I wonder how many of the issues he raised were raised to the community before he felt the need to whine (his word) at a conference. if it's the same zefram, he's also a p5 release manager, among other things.
15:07 mawkish_ joined #perl6
15:07 phatbot Not sure about the timing of the talk, but Zefram did file a whole ton of tickets. Especially the ones with edge case math, like NaN handling, negative zeros, overflows.
15:08 phatbot I'd even say he's my favourite ticket reporter.
15:08 moritz phatbot: don't let masak hear that :-)
15:08 phatbot :D
15:10 jnthn I'm glad of both. :-) masak tends to report things that crop up in sometimes interesting but relatively normal programs. Zefram looks at how stuff can be broken on purpose.
15:10 [Coke] (report) ah, well that's awesome. Zefram++
15:11 lizmat fwiw, zefram liked MoarVM and has said that perl 6 should be implemented on top of that (from scratch)
15:12 lizmat perhaps he will change his mind when we have a fully functional JS backend  :-)
15:13 ilmari I thought he said he liked nqp as well
15:13 mawkish_ joined #perl6
15:13 ilmari (as a language-implementation language)
15:14 khw joined #perl6
15:14 tadzik LPW 2017: Zefram implements Perl 5 on nqp
15:15 wamba joined #perl6
15:16 tadzik oh wow, SQL backend for profiler, that's nice
15:17 lizmat tadzik:  perl 5 on nqp, isn't that "use v5" ?
15:17 tadzik quite :)
15:18 ilmari that still requires rakudo, no?
15:18 lizmat yes
15:18 lizmat eh, well
15:18 lizmat hmmm... technically, not I guess
15:18 jnthn You could feasibly do a Perl 5 built on the NQP compiler toolchain without re-using any part of Rakudo.
15:19 mawkish_ joined #perl6
15:19 * brrt wonders whom he can challenge to do that
15:20 MorayJ joined #perl6
15:20 MasterDuke_ joined #perl6
15:20 perlpilot the next time any of you see Zefram, politely ask him to fix Data::Alias to work with perl 5.24  :-)
15:21 mohae_ joined #perl6
15:23 lizmat perlpilot: I've heard rumours that that is not going to be possible  :-(
15:23 lizmat otoh, I've heard those rumours before
15:24 kyclark joined #perl6
15:25 mawkish_ joined #perl6
15:26 lucasb multi role R[::T] {...}
15:26 lucasb multi role R[::T, ::U] {...}
15:26 lucasb ^^ would such a feature make sense?
15:27 jnthn We already have it, just don't need to write the multi :)
15:27 Khisanth joined #perl6
15:30 lucasb oh, I wasn't aware it already works, nice :)
15:30 arnsholt brrt: The I must confess that "how would I go about implementing Perl 5 on NQP" has been considered by me as an intellectual excercise, but I'm certainly not going to volunteer to actually *do* it!
15:32 arnsholt (Steps one and two being implementing lex and yacc-a-likes for NQP, IMO)
15:33 lucasb m: role R1 {}; role R2[Int] {}; say R1[1,2], R2[Str]
15:33 camelia rakudo-moar 1d4600: OUTPUT«(R1[Int,Int])(R2[Str])␤»
15:34 lucasb ^^ If you try to '.new' these roles, it expectedly dies
15:35 mawkish_ joined #perl6
15:35 lucasb class C does R[invalid args] {} also dies
15:35 lucasb but... couldn't this check happen at compile time for the invalid ones?
15:36 lucasb IOW, should even be possible to refer to them by the names 'R1[1,2]', etc. no?
15:36 lucasb *shouldn't
15:36 lizmat arnsholt: I think the main issue is really: which perl 5 are you going to target?  with all the recent syntactic changes  :-(
15:37 dj_goku joined #perl6
15:38 lizmat I think it's going to be less important now that we have Inline::Perl5
15:38 jnthn lucasb: If you .new a role then we don't calculate the pun until runtime
15:38 jnthn lucasb: For compile-time composition it does die at compile time already
15:38 lizmat and once we can get "supersede" to work and include modules with from<Perl5>, one should be able to upgrade a module to a Perl 6 version transparently without needing to change code
15:39 lucasb jnthn: ok, understood, thanks
15:43 pyrimidine joined #perl6
15:54 kyclark joined #perl6
15:55 lucasb found some strange errors for shaped arrays, will post 3 lines here, ok
15:55 lucasb (my @a[2]).new.perl           # Type check failed in binding to @dims; expected Positional but got Any (Any)
15:55 lucasb (my @a[2] = <a b>).new.Array  # This type (Scalar) does not support elems
15:55 lucasb my @a[<a b c>]                # P6opaque: get_boxed_ref could not unbox for the representation '20' of type Failure
15:56 lucasb also, for 'my @a[...]', idk what method it asks the value inside brackets (if .Int, .elems, or other), but for 'my @a[SomeEnum]', it returns a shaped array
15:57 lucasb m: dd my @a[Bool], my @b[Order]
15:57 camelia rakudo-moar 1d4600: OUTPUT«Array.new(:shape(2,), [Any, Any])␤Array.new(:shape(3,), [Any, Any, Any])␤»
15:57 lucasb I just wondered if that was expected and intended
16:00 phatbot no, crappy errors aren't expected :)
16:00 phatbot The my @a[<a b c>]  one looks like it tries to convert a string to number, gets Failure, and tries to store isbig_I it
16:01 mad__ joined #perl6
16:01 mad__ joined #perl6
16:01 phatbot m: use nqp; nqp::isbig_I(nqp::decont(+'a'));
16:01 camelia rakudo-moar 1d4600: OUTPUT«P6opaque: get_boxed_ref could not unbox for the representation '20' of type Failure␤  in block <unit> at <tmp> line 1␤␤»
16:02 phatbot there are a couple more I have in my notes at home that give exception blah blah produce no messsage
16:04 phatbot Oh, I misread.
16:04 phatbot lucasb: yeah, that's expected. Enums can be used as a shape
16:05 lucasb phatbot: because they numify to their number of elements?
16:05 phatbot m: dd +Bool
16:05 camelia rakudo-moar 1d4600: OUTPUT«Cannot resolve caller Numeric(Bool: ); none of these signatures match:␤    (Bool:D $: *%_)␤  in block <unit> at <tmp> line 1␤␤»
16:05 phatbot They don't. I just recall a commit fly through implementing that feature.
16:06 lucasb hm, interesting, ok
16:07 phatbot It's in 2016.10's changelog: https://github.com/rakudo/rakudo/blob/nom/docs/ChangeLog#L279
16:08 lucasb oh, inserting the commit hash in the changelog only started in 2016.11. phatbot++
16:09 phatbot And that's the commit: https://github.com/rakudo/rakudo/commit/553cedb679f0f191d2f7341e05033505cb20ac7a
16:09 lucasb thanks
16:11 mawkish_ joined #perl6
16:13 lizmat lucasb: I think that was inspired by a blogpost by gfldex, if memory serves
16:16 lizmat https://gfldex.wordpress.com/2016/09/21/are-these-your-keys/
16:17 lucasb ah, you linked other post in you commit https://github.com/rakudo/rakudo/commit/fef3655c0fb43209107b3e8708159280812cbceb
16:17 lucasb but anyway, I wasn't aware the feature was intentional, lizmat++
16:17 dalek doc: a972217 | MasterDuke17++ | doc/Type/Cool.pod6:
16:17 synopsebot6 Link: http://doc.perl6.org/type/Cool
16:17 dalek doc: Add some missing linebreaks to create code blocks
16:17 dalek doc: review: https://github.com/perl6/doc/commit/a9722174d1
16:25 acrussell joined #perl6
16:25 pierre_ joined #perl6
16:25 SmokeMachine I just started my "Generative testing" draft. Could someone take a look and give me some feedback please?
16:26 travis-ci joined #perl6
16:26 travis-ci Doc build failed. MasterDuke17 'Add some missing linebreaks to create code blocks'
16:26 travis-ci https://travis-ci.org/perl6/doc/builds/183640038 https://github.com/perl6/doc/compare/2af9049c11af...a9722174d16d
16:26 travis-ci left #perl6
16:27 SmokeMachine there are 2 different modules in memory when I load it at compile time and I load it at run time?
16:27 SmokeMachine when I get a instance of a singleton inside of a trait_mod code, I get a object and when I get it on a regular code I get a different object... is that expected?
16:27 SmokeMachine 00:37 https://www.irccloud.com/pastebin/3zFvLdlH/
16:27 SmokeMachine but when I do that in the same "file" it gets the same object:
16:27 SmokeMachine 00:45 https://www.irccloud.com/pastebin/NK78LBlF/
16:29 pierre__ joined #perl6
16:32 mr-foobar joined #perl6
16:32 SmokeMachine is that expected?
16:34 pyrimidine joined #perl6
16:35 phatbot SmokeMachine: no idea. Try hunting down nine
16:35 [Coke] fwiw, irccloud is blocked here.
16:37 perlpilot same here.   gist.github.com is fine though :)
16:38 dj_goku joined #perl6
16:38 dj_goku joined #perl6
16:39 mawkish_ joined #perl6
16:40 vike joined #perl6
16:42 wamba joined #perl6
16:43 pyrimidine joined #perl6
16:43 cdg joined #perl6
16:45 mawkish_ joined #perl6
16:48 SmokeMachine [Coke]: what do you mean with "irccloud is blocked here"?
16:48 girafe joined #perl6
16:50 SmokeMachine hum!! I got it!
16:50 perlpilot SmokeMachine: I (and presumably Coke) am at $work where certain URLs are blocked by our corporate rules.
16:50 SmokeMachine Ill paste it on GitHub...
16:53 SmokeMachine [Coke], perlpilot: https://gist.github.com/FCO/65196d4bff0362ef669ec61bafa50ae8
16:55 mawkish_ joined #perl6
17:00 perlpilot SmokeMachine: an object's .WHERE gives a memory location which may change over time due to GC runs.  You want .WHICH for object identity.
17:00 SmokeMachine perlpilot: ok!
17:00 phatbot oh damn...
17:00 * phatbot should've actually read the code :P
17:01 phatbot SmokeMachine: FWIW, if you tell the gist extensionb is .p6 it'll add highlighting
17:01 [Coke] Looks like precomp is involved; run the perl 6 command twice...
17:01 [Coke] (second time you only get one say.)
17:01 rindolf joined #perl6
17:02 SmokeMachine [Coke]: Yes, that happens...
17:02 SmokeMachine [Coke]: but thats not what I want...
17:02 SmokeMachine let me explain with some code...
17:03 [Coke] You want a singleton, that doesn't look like one, yes. :)
17:03 [Coke] Is that not it?
17:04 SmokeMachine on https://github.com/FCO/Test-Fuzz/blob/master/lib/Test/Fuzz.pm#L115 and here https://github.com/FCO/Test-Fuzz/blob/master/lib/Test/Fuzz.pm#L104 I save the function on the %!fuzzers and on https://github.com/FCO/Test-Fuzz/blob/master/lib/Test/Fuzz.pm#L157 I use that... but its empty, because its another object...
17:05 SmokeMachine [Coke]: I want a singleton where I .push() some data at compile time and I can get that data at run time...
17:07 rindolf Hi all.
17:07 phatbot Hi, rindolf
17:08 rindolf TimToady: are you there?
17:08 rindolf phatbot: hi
17:09 phatbot m: class Singleton { my @.stuff; BEGIN @.stuff.push: <some stuff>; method new { self }; }; my $foo = Singleton; my $bar = Singleton; $foo.stuff.push: <moar stuff>; dd $bar.stuff
17:09 camelia rakudo-moar 1d4600: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Variable @.stuff used where no 'self' is available␤at <tmp>:1␤------> 3ss Singleton { my @.stuff; BEGIN @.stuff7⏏5.push: <some stuff>; method new { self }␤    expecting any of:␤        term␤»
17:09 phatbot waat
17:09 phatbot you're trickin' me
17:09 phatbot oh
17:10 phatbot m: class Singleton { my @.stuff; BEGIN self.stuff.push: <some stuff>; method new { self }; }; my $foo = Singleton; my $bar = Singleton; $foo.stuff.push: <moar stuff>; dd $bar.stuff
17:10 camelia rakudo-moar 1d4600: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤'self' used where no object is available␤at <tmp>:1␤------> 3class Singleton { my @.stuff; BEGIN7⏏5 self.stuff.push: <some stuff>; method n␤    expecting any of:␤        term␤»
17:10 phatbot screw you
17:10 perlpilot phatbot: you're exhibiting the first signs of insanity  :)
17:10 phatbot m: class Singleton { my @.stuff; BEGIN ::?CLASS.stuff.push: <some stuff>; method new { self }; }; my $foo = Singleton; my $bar = Singleton; $foo.stuff.push: <moar stuff>; dd $bar.stuff
17:10 camelia rakudo-moar 1d4600: OUTPUT«Array @.stuff = [("some", "stuff"), ("moar", "stuff")]␤»
17:11 phatbot perlpilot: that means I've been hiding it well for all these years
17:11 phatbot Muahahhaahaha
17:12 perlpilot in that case ... good job so far!
17:12 perlpilot ;)
17:14 SmokeMachine phatbot: when I try in the same "file" it works... the problem occurs when I use a different file...
17:15 phatbot SmokeMachine: so you got this singleton in a module... You `use Blagh` it two different files that are being run as a single program and you no longer have a singleton?
17:16 phatbot oh you pasted links
17:16 SmokeMachine phatbot: that's it!
17:18 SmokeMachine phatbot: a small example:
17:18 phatbot SmokeMachine: well, I've no idea if that's even supposed to work, especially with the lexical loading branch we're merging Saturday. From what I understand, you could load two different modules with the same name, for example, and it would work. So I'm thinking the two different uses are two different compilation units and that's why it doesn't work like you expect it to
17:19 phatbot SmokeMachine: but I don't know this stuff, which is why my initial suggestion was to seek out nine :)
17:19 SmokeMachine https://gist.github.com/FCO/1561729d83b6737363ee871d90199184
17:20 phatbot Add .p6 extension to files
17:20 gregf_ phatbot: m: say [*] 1..10
17:20 SmokeMachine phatbot: Im sorry, but what means "seek out nine"?
17:20 SmokeMachine phatbot: thats not a .p6 file...
17:20 phatbot SmokeMachine: a person, nine. He's the expert on all this module stuff
17:21 phatbot gregf_: ?
17:21 Xliff m: say '"' ~~ / <[ '"' ]> /
17:21 camelia rakudo-moar 1d4600: OUTPUT«Potential difficulties:␤    Quotes are not metacharacters in character classes␤    at <tmp>:1␤    ------> 3say '"' ~~ / <7⏏5[ '"' ]> /␤    Repeated character (') unexpectedly found in character class␤    at <tmp>:1␤    ------> 3say '"' ~…»
17:21 Xliff m: say '"' ~~ / <[ " ]> /
17:21 camelia rakudo-moar 1d4600: OUTPUT«「"」␤»
17:21 SmokeMachine hum! nine is a person!!! thanks!
17:23 phatbot gregf_: what was that?
17:23 phatbot m: say '"' ~~ / '"' /
17:23 camelia rakudo-moar 1d4600: OUTPUT«「"」␤»
17:24 cibs joined #perl6
17:26 toolforger joined #perl6
17:26 toolforger g'evening everybody :-)
17:26 phatbot \o
17:26 SmokeMachine .tell nine Hi! could you help me? Could you tell me if it's working as expected? use a module with a singleton in 2 different files (one of then at compile time) should return different object instances? like that: https://gist.github.com/FCO/1561729d83b6737363ee871d90199184
17:26 yoleaux SmokeMachine: I'll pass your message to nine.
17:27 domidumont joined #perl6
17:27 phatbot gregf_: m: say [+] ^9999999999999999999999999999999999999999999999999999999999999999999999999
17:28 phatbot hm, weird. perl6.fail lists tickets if they've been updated after being resolved :/
17:29 toolforger Anybody available for discussing JVM desirables?
17:29 phatbot What are "desirables"?
17:30 toolforger E.g. what package names to use for JVM binaries for a given Perl source file/module
17:31 phatbot vOv
17:31 cyphase joined #perl6
17:31 toolforger I know how Java tools expect things to be, I'm not so sure about what's desirable for Perl
17:31 nicq20 left #perl6
17:31 gregf_ phatbot: sorry, i thought you were a bot *runs*
17:31 toolforger I would like to implement some sensible defaults for that :-)
17:32 phatbot gregf_: so you see a bot and you start feeding him random crap? Pfft... That's racism!
17:33 phatbot m: say [+] ^9999999999999999999999999999999999999999999999999999999999999999999999999
17:33 camelia rakudo-moar 1d4600: OUTPUT«49999999999999999999999999999999999999999999999999999999999999999999999985000000000000000000000000000000000000000000000000000000000000000000000001␤»
17:33 phatbot That one actually works FWIW :)
17:33 jnthn .oO( bigbotry )
17:33 berserk_guts joined #perl6
17:33 gregf_ m: ([*] 1..10).say # phatbot does that look like random crap?
17:33 camelia rakudo-moar 1d4600: OUTPUT«3628800␤»
17:34 gregf_ and racism is too strong a word :/
17:34 berserk_guts left #perl6
17:34 phatbot I guess I think I'm funnier than I really am :(
17:35 gregf_ heh
17:37 samcv good morning perl6
17:37 phatbot \o\
17:37 lizmat samcv o/
17:39 pierre_ joined #perl6
17:39 dj_goku joined #perl6
17:39 dj_goku joined #perl6
17:41 phatbot toolforger: ask psch, bartolin, or arnsholt if you see them around.
17:42 pyrimidine joined #perl6
17:45 toolforger Being a chat noob: Is there a good way to invite two people to a common chat?
17:45 toolforger I'm using Pidgin, FWIW
17:45 * phatbot 's no idea
17:45 ilmari create a new channel, invite them
17:45 geekosaur pretty much just have to create a channel for yourself (by joining it; it's dynamic) and /invite
17:46 toolforger Ah, thanks, that's straightforward enough :-D
17:46 ilmari you can make the channel invite-only, /mode #channel +i
17:46 ilmari that way only people you /invite can join
17:47 geekosaur +s can also be useful (so it doesn't show in a channel list)
17:48 toolforger I think it's okay if I name it #perl6-jvm or something like that
17:49 MasterDuke_ toolforger: are you familiar with ObjectWeb ASM at all?
17:49 toolforger Sort of - I've been using libraries that use it for several years, and been looking at the docs at various times
17:50 toolforger I know my way around the JVM, though I did some brush-up around class file formats and such
17:51 toolforger Plus did a diagonal read of all the ASM manual pages
17:52 MasterDuke_ well, if you're interested in a hard problem, i've been trying to get r-j to support the original file+line annotations that i added to r-m
17:52 MasterDuke_ https://irclog.perlgeek.de/perl6-dev/2016-11-21#i_13602100 has some chat about it (and the previous/next day or two)
17:53 toolforger Mmm... debug info should be easy to add
17:54 toolforger it might be hard to convince tools to pick up Perl sources though
17:54 MasterDuke_ i've got it added, it's reading it back out that's a problem
17:54 toolforger Wrong extension, and most likely wrong subdirectory, too
17:54 * toolforger goes checking the old chat
17:55 toolforger any keywords I should search the chatlogs for?
17:55 phatbot RabidGravy++ good post
17:56 MasterDuke_ i'm adding ASM Attributes to the bytecode with file/line info
17:56 toolforger OK, that would be the way to do it
17:56 MasterDuke_ asm, line directives, attributes
17:56 MasterDuke_ around the timeframe of that link maybe plus/minus a couple days
17:57 [Coke] (#perl6-jvm) (#perl6-dev is probably fine for that, but if you do create another perl6- chat, be sure to mention it to moritz & mst.)
17:57 toolforger Did you try triggering a Java exception? The stack trace should show you what the JVM thinks about the info
17:58 MasterDuke_ well, i got an arrayaccessoutofbounds exception without trying to trigger one
17:58 MasterDuke_ *arrayindex
17:58 toolforger that should be easier to diagnose for mistakes than e.g. Eclipse simply not displaying the right file
17:58 MasterDuke_ but it wasn't very helpful
17:59 toolforger You can simply "throw new RuntimeException()" anywhere
17:59 toolforger then look how the stack trace looks like
17:59 MasterDuke_ i have to head out, but i'll be back on this evening
17:59 toolforger If the stack trace is too short, somebody up the call stack is not rethrowing the exception
18:00 toolforger I saw nqp-j has issues with dealing with exceptions that cross the Java/Perl boundary, maybe that's what you are (not) seeing
18:00 toolforger cu then :-)
18:02 MasterDuke_ before i leave, this is the code/line that was throwing the array exception: https://github.com/MasterDuke17/nqp/blob/source_file_line_numbers_on_jvm/src/vm/jvm/runtime/org/perl6/nqp/runtime/ExceptionHandling.java#L331
18:02 toolforger Oh, that's inside the JDK
18:03 toolforger I'd need to see the stack trace to say anything meaningful about that
18:04 samcv is UNPACK base something that should be documented in the docs?
18:04 samcv it's not in roast or anywhere in the docs or in the design docs
18:04 toolforger Or maybe the array access is in NQP's class loaders, though I cannot remember having seen it use arrays
18:05 toolforger The kind of code in line 133 itself cannot throw an ArrayBoundsException, anyway
18:05 pyrimidine joined #perl6
18:06 phatbot samcv: what's UNPACK base?
18:06 phatbot grep -R UNPACK src/ gives me nothing
18:06 samcv err UNBASE
18:06 phatbot samcv: no, it should'nt be documented. It'
18:06 phatbot samcv: no, it should'nt be documented. It's an internal sub that should be moved to Rakudo::Internals. IIRC it's used by grammar to handle rad_number token
18:07 samcv ah ok
18:07 phatbot samcv: we have parse-base() as user-facing version
18:07 samcv gotcha
18:07 phatbot I think the convention is the ALLCAPS subs are all internal and we just haven't had a chance to move them all to Rakudo::Internals yet
18:08 samcv ah ok
18:11 prammer joined #perl6
18:12 phatbot m: CORE::.values.grep(Sub).grep({try .name.uc eq .name})».name.say
18:12 camelia rakudo-moar 1d4600: OUTPUT«WARNINGS for <tmp>:␤Useless use of "eq" in expression ".uc eq .name" in sink context (line 1)␤(RETURN-LIST GATHER GENERATE-ROLE-FROM-VALUE ORDER METAOP_TEST_ASSIGN:<&&> METAOP_HYPER_POSTFIX DUMP VAR DEPRECATED METAOP_HYPER_CALL MD-ARRAY-SLICE-ONE-POSIT…»
18:13 phatbot useless use of useless use
18:13 pyrimidine joined #perl6
18:13 phatbot m: say WHAT 42
18:13 camelia rakudo-moar 1d4600: OUTPUT«(Int)␤»
18:13 phatbot m: say HOW 42
18:13 camelia rakudo-moar 1d4600: OUTPUT«Perl6::Metamodel::ClassHOW.new␤»
18:14 phatbot I wonder why HOW doesn't show up in that list, but WHAT does
18:14 phatbot And these aren't internal despite all caps, it's the sub versions of .WHAT .HOW etc ummm.. what do we call them? pseudo-methods?
18:14 * [Coke] discovers "leider hosen"
18:15 phatbot What's that?
18:15 [Coke] macro-ish]
18:15 [Coke] phatbot: apparently "Unfortunately, pants."
18:15 [Coke] as opposed to lederhosen.
18:15 toolforger "leather pants"
18:15 phatbot And I thought a ladder made of hoses...
18:16 toolforger "leider" = "alas"
18:16 toolforger "ladder" = "Leiter"
18:17 zakharyas joined #perl6
18:17 [Coke] I like "Alas, pants." better. :)
18:18 toolforger Also, "Leiter" = "head" (e.g. of company)
18:18 toolforger So may be "Leiterhosen" might be "pants of the leader"
18:18 ccntrq joined #perl6
18:19 domidumont joined #perl6
18:21 andrzejku joined #perl6
18:21 phatbot m: .say for 42.?dasdasdsa.
18:21 camelia rakudo-moar 1d4600: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Malformed postfix call␤at <tmp>:1␤------> 3.say for 42.?dasdasdsa.7⏏5<EOL>␤»
18:21 phatbot m: .say for 42.?dasdasdsa
18:21 camelia rakudo-moar 1d4600: OUTPUT«Nil␤»
18:22 phatbot Is there a nice-lookin' trick to make it loop only over stuff provided by the method, excluding Nil
18:22 ccntrq joined #perl6
18:32 FROGGS joined #perl6
18:35 ccntrq joined #perl6
18:38 thalunil joined #perl6
18:40 dj_goku joined #perl6
18:40 dj_goku joined #perl6
18:44 ccntrq joined #perl6
18:46 labster joined #perl6
18:47 ccntrq joined #perl6
18:50 phatbot m: Nil.head
18:50 camelia rakudo-moar 1d4600: OUTPUT«Cannot resolve caller head(Nil: ); none of these signatures match:␤    (Any:D $: *%_)␤    (Any:D $: Cool $n, *%_)␤  in block <unit> at <tmp> line 1␤␤»
18:50 * phatbot wonders why that fails
18:53 phatbot s: [], 'head', \()
18:53 SourceBaby phatbot, Sauce is at https://github.com/rakudo/rakudo/blob/1d46004/src/core/Any-iterable-methods.pm#L1880
18:54 b2gills You know whats annoying Ideone has Perl 6, but it is the 2014.07 version. I asked in the middle of June for it to be upgraded, they got back to me in July, and they still have the old version.
18:55 phatbot m: Nil.tail
18:55 camelia rakudo-moar 1d4600: OUTPUT«Cannot resolve caller tail(Nil: ); none of these signatures match:␤    (Any:D $: *%_)␤    (Any:D $: Cool $n, *%_)␤  in block <unit> at <tmp> line 1␤␤»
18:55 phatbot another one :/
18:56 lizmat samcv: undocumented ALLCAPS subs should remain undocumented: they will need to be moved to Rakudo::Internals at some point
18:57 El_Che .
18:58 samcv has rakudo been released yet for the month?
18:58 samcv i'm using Readline and the REPL keeps like. not even coming up sometimes
18:58 samcv so that commit adding Readline history for REPL i think should be reverted before release
18:58 samcv i forget when we release again
18:58 phatbot samcv: no, it's this Saturday. How can I reproduce the issue? Just install Readline?
18:58 phatbot NeuralAnomaly: status
18:58 samcv oh ok
18:58 NeuralAnomaly phatbot, [✘] Next release will be in 3 days and 10 hours. Since last release, there are 50 new still-open tickets (6 unreviewed and 0 blockers) and 43 unreviewed commits. See http://perl6.fail/release/stats for details
18:58 samcv then i will try and fix it then :)
18:58 phatbot samcv: release is 3rd Saturday of the month
18:59 phatbot samcv++ ok cool :)
19:00 ccntrq joined #perl6
19:01 phatbot Ah, doh
19:01 prammer joined #perl6
19:01 phatbot the head/tail are an Any and Nil is an any too
19:02 [Coke] b2gills: sending them a feedback now.
19:02 [Coke] explaining tht a lot's been done and I can help if they need.
19:07 nine ~~
19:07 yoleaux 17:26Z <SmokeMachine> nine: Hi! could you help me? Could you tell me if it's working as expected? use a module with a singleton in 2 different files (one of then at compile time) should return different object instances? like that: https://gist.github.com/FCO/1561729d83b6737363ee871d90199184
19:08 nine SmokeMachine: Precompilation can be tricky. Your singleton module is precompiled first. $instance (an empty scalar container) is stored in the precomp file. Use is precompiled next and causes $instance to be updated. But nothing changes the stored precomp file with the empty container.
19:09 nine SmokeMachine: when you then load ToUse and Use, all you get is an empty container. But how could it be otherwise? E.g. what would happen if two modules both get independently precompiled and both cause a singleton to be created. If your main program loads both these modules, which singleton object would you get?
19:12 phatbot m: say "$2"
19:12 camelia rakudo-moar a980eb: OUTPUT«Use of Nil in string context␤  in block <unit> at <tmp> line 1␤␤»
19:13 phatbot Um, looks like REPL wants fixing for warnings. That gives me "Control flow commands not allowed in toplevel"
19:13 thayne joined #perl6
19:14 kyclark joined #perl6
19:18 phatbot samcv: FWIW, I'm not able to reproduce any hangs in my REPL with Readline installed.
19:18 phatbot I also learned that it's a bad idea to create ~/.inputrc and fill it with random crap
19:18 thalunil joined #perl6
19:20 * phatbot clicks imaginary "Take" on the REPL warnings stuff
19:21 domidumont joined #perl6
19:23 thayne joined #perl6
19:26 phatbot hehe... down a rabbit hole
19:26 phatbot m: CONTROL { .gist.say }; emit; say "haha, no one is deadied";
19:26 camelia rakudo-moar a980eb: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Calling emit() will never work with declared signature (\value --> Nil)␤at <tmp>:1␤------> 3CONTROL { .gist.say }; 7⏏5emit; say "haha, no one is deadied";␤»
19:27 phatbot I mean
19:27 phatbot m: CONTROL { .gist.say }; emit 42; say "haha, no one is deadied";
19:27 camelia rakudo-moar a980eb: OUTPUT«Died with CX::Emit␤  in block <unit> at <tmp> line 1␤␤emit without supply or react␤  in block <unit> at <tmp> line 1␤␤»
19:29 effbiai joined #perl6
19:30 prammer joined #perl6
19:32 asd_ joined #perl6
19:32 asd_ hm...
19:33 phatbot heh
19:37 RabidGravy so I guess I'd better start the next advent post now
19:37 phatbot You're making two?
19:38 phatbot huggable: advent
19:38 huggable phatbot, https://github.com/perl6/mu/blob/master/misc/perl6advent-2016/schedule
19:38 RabidGravy yeah, got a slot for sunday
19:38 phatbot RabidGravy++
19:39 phatbot interesting... now I rebuilt Rakudo and getting the hangs samcv mentioned :/
19:39 samcv :/
19:40 samcv method history-file() returns Str { ... } what
19:40 samcv oh no
19:40 phatbot ?
19:40 phatbot .seen cygx
19:40 yoleaux I saw cygx 9 Dec 2016 14:23Z in #perl6: <cygx> first draft: https://perl6advent.wordpress.com/?p=5746&amp;preview=true
19:41 samcv that's in the REPL since the PR for adding the history file for readline...
19:41 samcv that shouldn't do anything right unless there's some ... magic i don't know about in perl 6
19:41 dj_goku joined #perl6
19:41 dj_goku joined #perl6
19:41 samcv oh linenoise has it set the same way hm
19:41 phatbot it's a stub
19:42 phatbot m: class { method meow { … } }.meow
19:42 camelia rakudo-moar a980eb: OUTPUT«Stub code executed␤  in method meow at <tmp> line 1␤  in block <unit> at <tmp> line 1␤␤Actually thrown at:␤  in block <unit> at <tmp> line 1␤␤»
19:42 samcv yeah i know that
19:42 phatbot Oh ok
19:42 phatbot How do you know so much Perl 6?
19:42 samcv me?
19:42 phatbot Yeah
19:42 samcv idk
19:43 samcv learning it
19:43 RabidGravy how does anyone know so much Perl 6? ;-)
19:44 phatbot No, I mean I don't recall seein samcv here for more than a month, but in some things she knows more than me
19:44 samcv uhm ok it looks like history-file is a method that is defined in a different spot
19:44 phatbot I guess that's more a statement about me than anyone else heh
19:44 samcv i learn quickly ;)
19:44 samcv i only started programming perl 5 5 months ago
19:45 pyrimidine joined #perl6
19:45 samcv my friends i've talked to are super suprised that even tho they've been perling for years i now surpass them haha
19:45 samcv maybe i'm finding what i'm good at.. never been employed as a programmer. but i really enjoy it
19:46 phatbot :)
19:46 RabidGravy don't fancy a job? I'm going to be the only developer in a team that should have five in the new year
19:46 samcv don't fancy? well i am looking for work atm
19:47 samcv maybe i could do something in devops cause my last job was Jr Linux admin
19:47 * samcv really needs to get a job
19:47 RabidGravy where are you based?
19:48 samcv in the bay area, in california
19:48 samcv maybe i can point to contributing to perl 6 and say. pls give me job! i am good! i swear
19:48 dwarring joined #perl6
19:49 nine samcv: there are definitely worse ways to advertise than having public open source contributions
19:49 samcv like? lol
19:49 RabidGravy bit of a commute to London then :)
19:50 samcv RabidGravy, i will drive really fast
19:50 nine samcv: like....anything else *g*
19:50 samcv that's what i was thinking...
19:51 samcv ugh so annoying. the perl language package in atom in built in, but it will go first before the perl 6 one... since it has perl 6 file detection too
19:51 nine If I've ever got to apply for a job again, I'll just point at the list of my conference talks and see who bites ;)
19:51 samcv so if i disable it, it detects the rakudo files... but if i do that i can't look at the perl files
19:51 samcv ha
19:52 samcv really liking it now that i got perl 6 code execution only a keyboard shortcut away :)
19:53 phatbot Yeah, I recall Madcapjake telling me I had to disable the built-in perl highlighter if I wanted perl6fe to work
19:53 samcv just really like automating things at least. i've been doing that a lot
19:53 samcv you don't have to disable it, but
19:54 samcv if it's not named .p6 or .pl6 or .nqp or whatever
19:54 samcv the perl plugin will try and do its bad perl 6 highlighting
19:54 Xliff Is there any way I can block the default constructor of a class?
19:54 phatbot Ah
19:54 Xliff m: class A { has $.n; multi method new($a) { $!n = 1; }; multi method new($a, $b) { $!n = 2 }; }; my $a = A.new(:n(3));
19:54 camelia rakudo-moar a980eb: ( no output )
19:54 RabidGravy which reminds me I need to update the vim perl 6 syntax file at some point
19:54 phatbot Xliff: proto method new (|) {*}
19:54 djbkd_ joined #perl6
19:54 samcv i added some detection to detect if it starts with the perl 6 shebang or 'my class' and detect as perl 6 instead of perl 5, but
19:55 samcv the perl plugin does that too
19:55 Xliff phatbot++ # Thanks!
19:55 samcv and goes first
19:55 samcv dunno if i can get them to disable that, or something..
19:55 samcv RabidGravy, it should just come with it
19:55 samcv unless they updated it since then
19:56 loveperl joined #perl6
19:57 wamba joined #perl6
19:58 RabidGravy samcv it does, it just doesn't match the current state of the language
19:58 samcv yeah
19:58 samcv you should try the atom one ;)
19:59 samcv and if there's _any_ syntax highlighting errors please open an issue and i'll fix it quickly
19:59 samcv unhighlighted methods, routines, etc even tiny things
20:00 Xliff Hmm... phatbot, that works, but the more tricky among us might know that perl6 will allow constructor of the form A.new(1, :n(4)) -- don't want anyone to have the ability to set $.n via new.
20:00 samcv also the Readline plugin still has a problem it doesn't work with readline 7
20:00 samcv it hardcodes version 6 in
20:00 Xliff samcv: I am using atom perl6-fe and loving it.
20:00 samcv nice!
20:00 samcv \o/
20:00 Xliff I also managed to convert the snippets to something usable for Sublime.... although I have not texted that. ;)
20:01 Xliff s/texted/tested/
20:01 phatbot Xliff: it won't get set
20:01 loveperl How did you call a class that is in another file?
20:01 Xliff Also: https://github.com/MadcapJake/language-perl6fe/issues/53
20:01 phatbot Xliff: and well, reading your multi method new()... it's wrong. It needs to self.bless
20:02 Xliff phatbot: LOL! Yeah. Fixed that already in a camelia window.
20:02 Xliff m: class A { has $.n; proto method new (|) {*}; multi method new($a) { self.bless(:n(1)); }; multi method new($a, $b) { self.bless(:n(2)) }; }; my $a = A.new(1, :n(4));
20:02 camelia rakudo-moar a980eb: ( no output )
20:02 phatbot Xliff: right, so that's what will only be set. The extra args would be ignored. Any methods can take any number of named args
20:02 Xliff So how can I block the constructors from allowing $.n to be set like that?
20:03 Xliff Ooh!
20:03 phatbot m: say 'lulz'.uc: :42a :72meows
20:03 samcv phatbot, just made a PR to fix readline
20:03 camelia rakudo-moar a980eb: OUTPUT«LULZ␤»
20:03 samcv err the REPL
20:03 Xliff m: class A { has $.n; proto method new (|) {*}; multi method new($a) { self.bless(:n(1)); }; multi method new($a, $b) { self.bless(:n(2)) }; }; my $a = A.new(1, :n(4)); say $a.n;
20:03 camelia rakudo-moar a980eb: OUTPUT«1␤»
20:03 Xliff Right.
20:03 cygx joined #perl6
20:03 samcv Xliff, see the current readme for Perl6FE for how to get perl 6 scripts running from atom
20:03 * cygx has scheduled his advent post
20:03 yoleaux 11:43Z <phatbot> cygx: reminder: your post is up next in ~12.5 hours
20:03 Xliff REPL... REPR... what's the difference?
20:03 phatbot Xliff: but if you want it to actually whine about the extra args, just add %_ and die "no args!"
20:03 phatbot cygx: \o/
20:04 cygx I'm not entirely happy with it, but it is what it is...
20:04 phatbot That's what my mom said about me!
20:04 samcv Xliff, https://github.com/perl6/atom-language-perl6#integration
20:05 Xliff samcv++ # \o/
20:05 samcv :)
20:05 Xliff If only Atom was working under my Linux VM.
20:05 samcv you don't even have to save the file too
20:05 samcv :(
20:05 samcv what is the host?
20:05 Xliff Currently I am using it with Windows via SMB export to my host.
20:05 Xliff Windows
20:05 samcv atom is for windows too
20:05 Xliff I know. That's the only way I can use atom right now.
20:05 samcv ohhh
20:05 samcv oh i see the problem i guess
20:05 Xliff I can't get it to run under Linux for some reason. The GUI will not start.
20:05 samcv perl 6 not working under windows properly?
20:06 Xliff No. My perl6 install is on the linux vm.
20:06 samcv kk
20:06 Xliff So I export home dir to Windows so I can use Windows Atom for editing.
20:06 samcv ah
20:06 cygx our wordpress theme looks sub-optimal, though: it's too narrow, so code blocks end up with horizontal scroll bars :(
20:06 Xliff Sublime is working under Linux VM, so I will try getting the TextMate rules over there to see if they work better than ModernPerl.
20:06 AlexDaniel joined #perl6
20:07 samcv Xliff, maybe hack something together?
20:07 Xliff samcv: And I do have to say, Atom+language-perl6fe > Sublime+ModernPerl
20:07 samcv idk can you make a perl6 names executable which gets the vm to run it
20:07 Xliff samcv: Actually, I already have the .sublime_package. I just have been too lazy to try it!
20:07 samcv and returns the result ha
20:07 samcv what .sublime_package?
20:07 Xliff language-perl6fe
20:08 samcv how did you create it
20:08 Xliff I converted it.
20:08 samcv how? pls tell
20:08 Xliff cson => json => plist => zip
20:08 samcv you're using the updated plugin too right? cause the dl package moved
20:08 Xliff There are scripts.
20:08 samcv hmmmmm
20:08 samcv links
20:08 Xliff But I haven't tested it under Sublime, yet.
20:08 Xliff Yeh. I am using the repo.
20:08 samcv kk
20:10 tbrowder Anyone using Komodo for p6?
20:11 Xliff samcv: Actually, it is more convoluted.
20:12 Xliff cson => json => yaml => plist => zip
20:12 samcv well send it to me
20:12 samcv i will test it
20:12 samcv nice
20:12 samcv haha
20:12 Xliff Uh. OK. I will create issue.
20:12 Xliff And attach package.
20:12 Xliff Or can I email you?
20:12 samcv you can email me
20:12 Xliff kk
20:12 samcv pm'd you
20:13 phatbot Xliff: if you make sublime highlight perl6 decently, I'll promise not to assimilate you :)
20:13 phatbot sublime > atom
20:17 samcv kk
20:17 samcv yeah i need perl 6 highlighting to be happy tbh
20:18 AlexDaniel phatbot: source
20:18 yoleaux 05:06Z <MasterDuke> AlexDaniel: it looks like you implemented something already, but i'd say just return the name for them or something like that, i don't see anybody ever wanting them for real
20:19 samcv https://github.com/drforr/perl6-readline/pull/15 hmm if this can be pulled would we want to include Readline in Star?
20:19 samcv because linenoise doesn't support utf-8
20:19 phatbot AlexDaniel: https://github.com/zoffixznet/perl6-phatbot
20:19 samcv oh AlexDaniel
20:19 samcv you need to add something to unicodable6
20:19 AlexDaniel MasterDuke: the problem is, if you do 「say ‘hello’ ~ 0.chr ~ ‘world’」 or something like that, what should committable/evalable gist?
20:20 Xliff phatbot: LOL! You'll try to do that anyways. I wonder what else I can get out of you. :D
20:20 RabidGravy Hmm, now do I either split some of the utility functions of Test::META into a separate module or make Test::META a dependency the thing I want to use them in
20:20 AlexDaniel phatbot: bug tracker
20:20 AlexDaniel the link is broken
20:20 AlexDaniel I wanted to submit a PR that will change the nickname…
20:20 AlexDaniel samcv: alright, what?
20:20 phatbot AlexDaniel: https://github.com/zoffixznet/perl6-phatbot/issues
20:21 AlexDaniel samcv: dumping of all unicode properties
20:21 AlexDaniel ?
20:21 samcv no
20:21 samcv well sorta
20:22 samcv m: for 0..0x1FFFF -> $i { if $i.unimatch('LF') { say $i.uniname } }
20:22 camelia rakudo-moar a980eb: OUTPUT«LINE FEED (LF)␤»
20:23 samcv the LF is a unicode property called the 'Line_Break' property
20:23 Tonik joined #perl6
20:23 samcv unimatch will check if the codepoint has the specified property
20:23 AlexDaniel u: LF
20:23 unicodable6 AlexDaniel, U+000A LINE FEED (LF) [Cc] (control character)
20:23 unicodable6 AlexDaniel, U+00BD VULGAR FRACTION ONE HALF [No] (½)
20:23 unicodable6 AlexDaniel, U+02BE MODIFIER LETTER RIGHT HALF RING [Lm] (ʾ)
20:23 unicodable6 AlexDaniel, https://gist.github.com/dc202f8aaf0aa2ec27e4242457309795
20:23 samcv LF is a bad example tbh
20:23 AlexDaniel u: LF Cc
20:23 unicodable6 AlexDaniel, U+000A LINE FEED (LF) [Cc] (control character)
20:24 phatbot AlexDaniel: what's the new nickname?
20:24 AlexDaniel samcv: alright. OK. What could be the syntax?
20:24 AlexDaniel phatbot: Zoffix
20:24 samcv m: for 0..0x1FFFF -> $i { if $i.unimatch('Latin') { say $i.uniname } }
20:24 camelia rakudo-moar a980eb: OUTPUT«LATIN CAPITAL LETTER A␤LATIN CAPITAL LETTER B␤LATIN CAPITAL LETTER C␤LATIN CAPITAL LETTER D␤LATIN CAPITAL LETTER E␤LATIN CAPITAL LETTER F␤LATIN CAPITAL LETTER G␤LATIN CAPITAL LETTER H␤LATIN CAPITAL LETTER I␤LATIN CAPITAL LETTER J␤LATIN …»
20:24 samcv uh just search it when searching other things
20:24 phatbot AlexDaniel: I can't use that one. I have to stay undercover
20:24 AlexDaniel u: { .unimatch(‘Latin’) }
20:24 samcv or maybe not uh
20:24 unicodable6 AlexDaniel, U+0041 LATIN CAPITAL LETTER A [Lu] (A)
20:24 unicodable6 AlexDaniel, U+0042 LATIN CAPITAL LETTER B [Lu] (B)
20:24 unicodable6 AlexDaniel, U+0043 LATIN CAPITAL LETTER C [Lu] (C)
20:24 unicodable6 AlexDaniel, https://gist.github.com/27443d7c8b9213125e42a90950abb8e7
20:25 phatbot u: { .unimatch("{use NativeCall; sub system (Str) is native {}; system 'ls'}") }
20:25 RabidGravy tadzik++ # mergey goodness
20:25 samcv but it will find all properties no matter what they are as long as moarvm has it i guess
20:25 AlexDaniel phatbot: :|
20:26 phatbot >:}
20:26 AlexDaniel it should say something like ‘meh, fuck off’
20:26 samcv AlexDaniel, you could just search it when searching others
20:26 samcv it won't slow it down really
20:26 samcv tho you could do a special case too
20:26 samcv m: for 0..0x1FFFF -> $i { if $i.unimatch('9.0') { say $i.uniname } }
20:26 camelia rakudo-moar a980eb: OUTPUT«<reserved>␤<reserved>␤<reserved>␤<reserved>␤<reserved>␤<reserved>␤<reserved>␤<reserved>␤<reserved>␤<reserved>␤<reserved>␤<reserved>␤<reserved>␤<reserved>␤<reserved>␤<reserved>␤<reserved>␤<reserved>␤<reserved>␤<reserved…»
20:27 AlexDaniel samcv: sure, but it's already possible now, and unless you can propose some syntax I'm not sure I will be able to add it :)
20:27 AlexDaniel phatbot: aren't you supposed to bring them back on when you crash them?
20:27 phatbot OK will do
20:27 AlexDaniel no need
20:27 unicodable6 joined #perl6
20:28 samcv u: match 'thing'
20:28 unicodable6 samcv, U+006D LATIN SMALL LETTER M [Ll] (m)
20:28 unicodable6 samcv, U+0061 LATIN SMALL LETTER A [Ll] (a)
20:28 unicodable6 samcv, U+0074 LATIN SMALL LETTER T [Ll] (t)
20:28 unicodable6 samcv, https://gist.github.com/a6a297d998af3ae1141bb9d7d96ece98
20:28 samcv maybe like that?
20:28 samcv idk something with the match word in it would be easiest to remember
20:29 AlexDaniel phatbot: I'm still waiting for a day I get at least some small bit of time to work on containerizing them… until that happens, can you please not advertise the holes in our RESTRICTED setting pseudofeature? :)
20:30 AlexDaniel phatbot: or… perhaps consider disabling nativecall if it is on
20:30 phatbot You mean the NativeCall one or the nqp one? :D
20:30 AlexDaniel … both
20:30 phatbot ehehe
20:34 loveperl Could not find form at line 1 in:
20:35 loveperl use form
20:35 mst AlexDaniel: perhaps talk to simcop2387 about how he's locked down perlbot's eval system
20:35 loveperl ?
20:35 mst AlexDaniel: I mean, 'stick it in a container' won't help
20:35 mst loveperl: you tried to load a module that doesn't exist
20:35 AlexDaniel mst: wait… why not?
20:35 simcop2387 rlimits, chroot, seccomp.  I can give a better overview if you'd like
20:35 phatbot :o
20:35 mst AlexDaniel: you want security. containers are not designed for security.
20:35 phatbot simcop2387!
20:36 simcop2387 the one and only
20:36 phatbot \o/
20:36 mst AlexDaniel: every time somebody goes "I'll use docker for security", apeiron eats a kitten
20:36 AlexDaniel mst: sooo, are you saying that LXC is insecure or something?
20:36 toolforger left #perl6
20:36 simcop2387 AlexDaniel: LXC and docker won't protect you from kernel exploits.
20:37 loveperl Do I want to call a class that is in a file that is called form.p6?
20:37 mst AlexDaniel: ... no, I'm saying it's not designed for the sorts of things you'd want to do to secure an eval server
20:37 loveperl Do I want to call a class that is in a file that is called form.p6?
20:37 loveperl form.pm6
20:37 simcop2387 seccomp can got a LONG LONG way to making it mostly impossible to do most exploits
20:37 mst loveperl: http://trout.me.uk/vaguecat.jpg
20:38 loveperl lol
20:38 mst AlexDaniel: I'm not sure if you're intentionally being a dick or just completely clueless here, but responding to "this is not designed for securing what you're securing" does not mean it's "Insecure or something", it just means you don't know what you're doing yet. maybe ask the sysadmin of the machine they run on for help?
20:38 AlexDaniel simcop2387: alright. alright… good that you're here
20:39 mst loveperl: show us all the code you currently have and explain what you're trying to do?
20:39 AlexDaniel mst: well, that's kinda my machine…
20:39 simcop2387 AlexDaniel: here's the code i do in perlbot's eval to secure things with seccomp, https://github.com/simcop2387/perlbuut/blob/master/lib/eval.pl#L31..L106 , the idea is that it doesn't let the user supplied code open anything for writing, open any sockets etc.  they're only allowed to do a few things.
20:40 mst AlexDaniel: then I recommend listening to simcop2387 and not being deliberately obtuse to him like you were to me, since he's a nicer person than I am :P
20:40 AlexDaniel mst: stop accusing me of shit, I'm listening already
20:40 simcop2387 i won't abuse anyone (unless they ask for it nicely) but i will start ignoring people.
20:42 ccntrq joined #perl6
20:42 loveperl call.pl : code=( use form; say form) form.pm6 is: code=( class form extends common{})
20:42 AlexDaniel simcop2387: I'd have to read about seccomp to get the idea about this stuff. So I hope you don't mind if I ping you a bit later
20:42 phatbot loveperl: the current directory is not included by default. you need use lib <.>; or supply -I. on command line
20:42 simcop2387 AlexDaniel: oh no problem at all.
20:43 dj_goku joined #perl6
20:44 simcop2387 AlexDaniel: This one might or might not affect you, but i gave the perl interpreter CAP_SYS_CHROOT with `setcap cap_sys_chroot+ep $PERL`, that one let me do everything running from nobody:nogroup without ever touching the root user.  probably not THAT big of a deal but it helps against the possibility of some module i'm using having code sitting around specifically for me to hit.
20:44 loveperl I do not understand?
20:45 simcop2387 AlexDaniel: also, if you're not familiar with strace, it's a big help in understanding what you can do with seccomp.  basically you can filter system calls that show up in strace based on the arguments to them.
20:47 phatbot loveperl: perl6 -I. call.pl
20:48 phatbot loveperl: you need to tell it where to find form.pm6 and common.pm6
20:49 loveperl How do I say that?
20:49 thalunil joined #perl6
20:50 phatbot loveperl: Poner lib <.> En su archivo para decir perl6 para buscar los módulos en el directorio actual
20:50 * phatbot relocates
20:52 AlexDaniel mst: thanks
20:56 loveperl asi
20:56 loveperl lib <.>;   2 use form;   3 say form;
20:59 AlexDaniel mst: by the way, any news about the button?
20:59 loveperl =begin ⏏comment?
21:00 loveperl I get error
21:01 loveperl How to comment en bloc?
21:01 loveperl =begin comment;     brief utilities for the creation of HTML forms.     with special datatypes not found in HTML like date, etc     todo fieldset+legend, label for field     todo BLOCKs (field.block?)     todo block show/hide  =end comment;
21:01 mst loveperl: please use http://fpaste.scsys.co.uk/perl6 to show complete code
21:01 nowan_ joined #perl6
21:01 mst AlexDaniel: I'm probably going to have to email somebody, will try and remember to do that this week
21:01 bartolin joined #perl6
21:02 AlexDaniel mst: that would be nice
21:02 mst AlexDaniel: at this point I suspect "attempt to invoke a higher power" is going to be the right answer
21:02 shadowpaste "loveperl" at 217.168.150.38 pasted "como comentar en perl" (7 lines) at http://fpaste.scsys.co.uk/540217
21:02 mst key question: are you ok if it's "registered users only" and are you ok if we have to ask on a per-user basis?
21:02 mst (trying to pre-anticipate compromises that might be easier to win)
21:03 loveperl how to comment eb bloc?
21:03 loveperl How to comment in block?
21:03 loveperl http://fpaste.scsys.co.uk/540217?tx=on&amp;wr=on&amp;ln=on&amp;submit=Format+it%21
21:04 mst loveperl: what happens if you remove the ';' and just have '=begin comment' ?
21:04 mst loveperl: ... wait, there's one space in front of the = in your code. no spaces!
21:05 phatbot mst, IIRC it was already registerer users only so that's fine. The other option isn't ideal but better than nothing
21:05 loveperl tank you
21:05 AlexDaniel mst: “registered users only” is how it has always been
21:05 mst AlexDaniel: right, so maybe it needs some sort of additional whitelist
21:06 xinming_ joined #perl6
21:06 AlexDaniel mst: not sure if that would be easy, but perhaps checking if a user has at least one open or resolved ticket is good enough
21:07 loveperl class form extends common{
21:07 AlexDaniel or a script that builds a whitelist based on this, that could work too
21:07 mst loveperl: please always use the paste site to show code.
21:07 mst loveperl: random lines in channel do not help
21:08 loveperl ok
21:08 SmokeMachine nine: so, I cannot have a trait that saves data on a singleton because I'll never get that singleton again?
21:09 jnthn loveperl: If you are trying to do inheritance, you may find https://docs.perl6.org/language/classtut#Inheritance helpful
21:10 bartolin AlexDaniel: you pinged me to rebase rakudo pr 565. I can do that, but I wonder what git commands are best to do that. should I merge current nom into my old branch (resolving the conflict while doing that) and push the updated branch to my repo?
21:10 RabidGravy SmokeMachine, sure you can
21:11 SmokeMachine RabidGravy: not inter files...
21:11 RabidGravy eh?
21:13 SmokeMachine nine: when the second file uses the module it's gotten by precompiled file? Not from the preloaded module in memory with the $instance?
21:13 bartolin AlexDaniel: or is it acceptable to just force push something without a conflict?
21:14 SmokeMachine RabidGravy: this: 17:07 <yoleaux> 17:26Z <SmokeMachine> nine: Hi! could you help me? Could you tell me if it's working as expected? use a module with a singleton in 2 different files (one of then at compile time) should return different object instances? like that: https://gist.github.com/FCO/1561729d83b6737363ee871d90199184
21:14 AlexDaniel bartolin: you can force push into your own branch, yes
21:14 phatbot bartolin, you can forcepush into your own branch the PR is from
21:14 AlexDaniel in your own fork rather
21:15 loveperl constructor?
21:15 bartolin thanks! I'll look into it (but not today)
21:16 loveperl Constructor of a class
21:18 nine SmokeMachine: maybe you want to initialize your singleton in an INIT block?
21:20 SmokeMachine nine: in my real case of use I do that on a trait... or do you mean initialize the $instance on a INIT block?
21:22 SmokeMachine nine: this is the case: https://github.com/FCO/Test-Fuzz/blob/master/lib/Test/Fuzz.pm
21:22 shadowpaste "loveperl" at 217.168.150.38 pasted "constructor" (26 lines) at http://fpaste.scsys.co.uk/540218
21:23 loveperl_ joined #perl6
21:23 loveperl_ How do I build a builder?
21:23 loveperl_ http://fpaste.scsys.co.uk/540218?tx=on&amp;wr=on&amp;ln=on&amp;submit=Format+it%21
21:24 SmokeMachine nine: I'm sorry the leg.. I'm on my doughter's ballet presentation
21:24 RabidGravy joined #perl6
21:35 dwarring m: role R {}; my $a = 42 but R; say 42.WHICH
21:35 camelia rakudo-moar 7b84ad: OUTPUT«Int|42␤»
21:35 dwarring m: role R {}; my $a = 42 does R; say 42.WHICH
21:35 camelia rakudo-moar 7b84ad: OUTPUT«Int+{R}|42␤»
21:36 dwarring ^^ shot myself in the foot the other day
21:37 dwarring used 'does' on simple scalars
21:37 dwarring has a very global affect
21:39 cygx there's an integer cache, so stuff like that can indeed happen
21:39 dwarring yeah, a but of a trap
21:41 lukiramu joined #perl6
21:42 cygx I believe early Fortran version even had the ability to inadvertently redefine numbers dues to by-reference argument semantics
21:43 dwarring heh, like redefining 0 in fortran
21:44 masak in Perl 6, all we can do is redefine 2 + 2
21:44 masak but in a very safe, lexically scoped way :)
21:44 masak m: sub infix:<+>(2, 2) { 5 }; say 2 + 2
21:44 camelia rakudo-moar 7b84ad: OUTPUT«5␤»
21:44 masak oh, I guess that should be spelled `multi` for clarity.
21:44 geekosaur sounds like stupid haskell let tricks :p
21:45 masak m: multi infix:<+>(2, 2) { 5 }; say 2 + 2
21:45 camelia rakudo-moar 7b84ad: OUTPUT«5␤»
21:46 cygx m: 2 does role { method gist { "42" } }; say 2
21:46 camelia rakudo-moar 7b84ad: OUTPUT«42␤»
21:46 pyrimidine joined #perl6
21:47 kurahaupo_ joined #perl6
21:49 rindolf joined #perl6
21:49 dj_goku joined #perl6
21:55 cdg_ joined #perl6
21:59 Ven joined #perl6
21:59 rurban joined #perl6
22:00 cdg joined #perl6
22:02 bjz joined #perl6
22:15 kyclark joined #perl6
22:25 loveperl_ exit in perl
22:25 loveperl_ ?
22:37 kyclark joined #perl6
22:46 pyrimidi_ joined #perl6
22:50 dj_goku joined #perl6
22:50 dj_goku joined #perl6
22:54 masak these random comments three years later are sometimes amusing, but rarely very deep: https://perl6advent.wordpress.com/2013/12/22/day-22-a-catalogue-of-operator-types/#comment-16238
22:56 rburkholder joined #perl6
22:56 kyclark left #perl6
22:58 loveperl_ my $s = any; here null?
22:59 masak m: my $s; say $s
22:59 camelia rakudo-moar 7b84ad: OUTPUT«(Any)␤»
23:01 cygx masak: that comment is hilarious
23:01 cygx "the factorial function is worth putting in your product"
23:09 bpmedley joined #perl6
23:11 cdg joined #perl6
23:11 phatbot hehe
23:11 phatbot masak: nice one :)
23:11 * phatbot is glad "infix_prefix_meta_operator:<[ ]>" didn't stick :}
23:13 phatbot loveperl_: exit is exit
23:13 phatbot m: say "hi"; exit; say "this doesn't print"
23:13 camelia rakudo-moar 7b84ad: OUTPUT«hi␤»
23:13 loveperl_ tank you
23:14 phatbot loveperl_: no, `any` is not null; it's a Junction http://perl6.party/post/Perl-6-Schrodinger-Certified-Junctions
23:15 loveperl_ what is null?
23:16 phatbot loveperl_: depends on what you're trying to do. Perl 6 has infinite number of "nulls". There's Empty, Nil, and all the type objects (Int is undefined; but 42 is, for example)
23:19 b2gills m: say #`(There is a factorial function, its just called) 6.&permutations.Int
23:19 camelia rakudo-moar 7b84ad: OUTPUT«720␤»
23:21 phatbot m: say [*] ^6+1; #`(There is a factorial function, its just called)
23:21 camelia rakudo-moar 7b84ad: OUTPUT«720␤»
23:22 AlexDaniel joined #perl6
23:25 samcv omg fuzzy search
23:25 samcv finding wayyyy too many things
23:26 samcv it needs to list the fuzzyly found things at the end of the list
23:26 samcv (re the perl 6 docs)
23:26 samcv also it seems to be realllyyy slow sometimes
23:27 samcv and maxing out the cpu on my computer
23:28 samcv try typing 'run' into it. it justs... kinda
23:28 samcv doesn't do anything and uses all my cpu
23:29 samcv oh hey it finally showed up... but has a billion pages all returned in that scrolling menu..
23:30 samcv i think it should search how it normally would and then have a link to click to go to a fuzzily searched list of things if nothing shows up
23:31 AlexDaniel samcv: fix it then!
23:32 AlexDaniel this problem has been there for quite a while
23:32 AlexDaniel it just got a little bit worse
23:32 samcv you mean since it was changed to fuzzy search?
23:32 AlexDaniel no, previously if you typed ‘e’ it would have given you waaaaay too many results
23:32 AlexDaniel so now there are just a bit more ways to get too many results
23:32 AlexDaniel but same problem
23:33 samcv ah
23:35 dalek doc: d5c7777 | samcv++ | template/search_template.js:
23:35 dalek doc: Change levenshtein distance from 2 to 1
23:35 dalek doc:
23:35 dalek doc: Try to get it for now to not return way too many things
23:35 dalek doc: for short or common words
23:35 dalek doc: review: https://github.com/perl6/doc/commit/d5c77772c7
23:38 kurahaupo joined #perl6
23:39 samcv ok that change made it like 10x better
23:39 samcv ok maybe not 10x better but usable
23:40 AlexDaniel OK: https://github.com/perl6/doc/issues/1050
23:43 travis-ci joined #perl6
23:43 travis-ci Doc build failed. Samantha McVey 'Change levenshtein distance from 2 to 1
23:43 travis-ci https://travis-ci.org/perl6/doc/builds/183766447 https://github.com/perl6/doc/compare/a9722174d16d...d5c77772c72b
23:43 travis-ci left #perl6
23:44 samcv MoarVM panic: Collectable 0x2b3668c81980 in fromspace accessed :(
23:44 samcv was wondering why that would have broken the docs
23:45 skids joined #perl6
23:48 AlexDaniel mmmmm collectable
23:51 dj_goku joined #perl6
23:51 dj_goku joined #perl6
23:53 AlexDaniel u: Lo
23:53 unicodable6 AlexDaniel, U+00AA FEMININE ORDINAL INDICATOR [Lo] (ª)
23:53 unicodable6 AlexDaniel, U+00BA MASCULINE ORDINAL INDICATOR [Lo] (º)
23:53 unicodable6 AlexDaniel, U+01BB LATIN LETTER TWO WITH STROKE [Lo] (ƻ)
23:53 AlexDaniel … that's it???
23:53 AlexDaniel there are only 3 LO characters???
23:54 AlexDaniel Lo*
23:54 AlexDaniel unicodable6: Lo
23:54 AlexDaniel no, we have a problem!
23:54 unicodable6 AlexDaniel, U+00AA FEMININE ORDINAL INDICATOR [Lo] (ª)
23:54 unicodable6 AlexDaniel, U+00BA MASCULINE ORDINAL INDICATOR [Lo] (º)
23:54 unicodable6 AlexDaniel, U+01BB LATIN LETTER TWO WITH STROKE [Lo] (ƻ)
23:54 AlexDaniel or maybe not
23:55 samcv u: { .unimatch('Lo') }
23:55 samcv did i do that right?
23:55 AlexDaniel :| it ate too much memory and died
23:55 phatbot There are 112804
23:55 samcv oh
23:55 samcv that's not enough
23:55 phatbot m: ^0x110000 .grep(*.uniprop eq "Lo").elems.say
23:55 camelia rakudo-moar 7b84ad: OUTPUT«112804␤»
23:55 unicodable6 joined #perl6
23:56 AlexDaniel alright, so it died trying to gist it, I guess
23:56 AlexDaniel or… I don't know
23:56 phatbot What does it use to gist?
23:56 samcv or get the remaining results?
23:56 phatbot Zofware?
23:56 AlexDaniel phatbot: your module I think
23:56 samcv ugh i got this error just now building docs MoarVM panic: Collectable 0x7f8879020cd0 in fromspace accessed
23:56 samcv travis got it too
23:57 phatbot samcv: https://rt.perl.org/Ticket/Display.html?id=130327#ticket-history
23:57 AlexDaniel yea, Pastebin::Gist
23:57 AlexDaniel phatbot: by the way, that module does not do any sort of escaping, and it probably should
23:58 dj_goku joined #perl6
23:58 dj_goku joined #perl6
23:58 samcv escaping?
23:58 AlexDaniel e.g. if you put some control characters in there it will not work
23:58 samcv oh
23:58 samcv why can't it just gist the control characters
23:58 AlexDaniel dunno, let's try something
23:58 samcv does it write it to a file before gisting it?
23:59 phatbot AlexDaniel: I never updated it since I wrote it and it was my second piece of code I ever wrote in Perl 6....
23:59 AlexDaniel evalable6: say ‘hello’ ~ 0.chr ~ ‘world’; .say for ^999
23:59 evalable6 AlexDaniel, rakudo-moar 7b84adc: OUTPUT«hello␀world␤0␤1␤2␤3␤4␤5␤6␤7␤8␤9␤10␤11␤12␤13␤14␤15␤16␤17␤18␤19␤20␤21␤22␤23␤24␤25␤26␤27␤28␤29␤30…»
23:59 evalable6 AlexDaniel, Full output: https://gist.github.com/
23:59 AlexDaniel samcv: no
23:59 samcv it probably should

| Channels | #perl6 index | Today | | Search | Google Search | Plain-Text | summary

Perl 6 | Reference Documentation | Rakudo