»ö« Welcome to Perl 6! | perl6.org/ | evalbot usage: 'p6: say 3;' or rakudo:, std:, or /msg camelia p6: ... | irclog: irc.perl6.org | UTF-8 is our friend!
Set by masak on 12 May 2015.
dalek ast: 743b620 | TimToady++ | S32-array/shift.t:
test that shift on sparse array produces Nil
 00:00
TimToady BenGoldberg: you're right 00:01
BenGoldberg Add another test, so that the second shift on that array produces the '2' 00:04
m: my @a = 1, 2; @a[0]:delete; say so @a.shift; say so @a.shift; 00:05
camelia rakudo-moar : OUTPUT«False␤False␤»
BenGoldberg ^ Should be False True 00:05
dalek kudo/nom: c016df5 | TimToady++ | src/core/Array.pm:
actually shift the sparse array

  BenGoldberg++ for the eyeballs.		 00:11
ast: 7a2ad38 | TimToady++ | S32-array/shift.t:
test we actually shifted sparse array
BenGoldberg m: say SEQ( say 'ah?'; 42 ); 00:13
camelia rakudo-moar : OUTPUT«ah?␤42␤»
TimToady basically, C-comma semantics 00:26
or a non-block block 00:27
b2gills m: say race map -*, ^100 00:29
camelia rakudo-moar : OUTPUT«HyperSeq.new␤»
TimToady m: say eager race map -*, ^100 00:34
camelia rakudo-moar : OUTPUT«()␤»
TimToady m: say first * %% 7, race map -*, ^100
camelia rakudo-moar : OUTPUT«Nil␤»
TimToady m: .say for race map -*, ^100 00:34
camelia rakudo-moar : OUTPUT«0␤-1␤-2␤-3␤-4␤-5␤-6␤-7␤-64␤-8␤-9␤-65␤-66␤-10␤-67␤-11␤-68␤-12␤-13␤-69␤-14␤-70␤-15␤-71␤-16␤-17␤-72␤-18␤-19␤-73␤-20␤-21␤-74␤-22␤-75␤-23␤-76␤-24␤-25␤-77␤-26␤-78␤-27␤-79␤-28␤-80…»
TimToady seems some things aren't hooked up to HyperSeq quite right, just yet 00:35
BenGoldberg If I'm creating an infinite lazy Iterator, should sink-all throw an exception? 00:50
TimToady what if you want to call it just for the side effects? 00:51
BenGoldberg In this case, it's just generating primes, why would there be side effects? 00:52
Well, unless you count the debugging output as a side effect.
TimToady there's no way to know your intent
m: for 1..* { .say } 00:53
camelia rakudo-moar : OUTPUT«(timeout)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␤31␤32␤33␤34␤35␤36␤37␤38␤39␤40␤41␤42␤43␤44␤45␤46␤47␤48␤49␤50␤51␤52␤53…» 00:53
TimToady should that produce an exception?
diakopter m: .say for 1..* 00:54
BenGoldberg m: (1..*).iterator.say
camelia rakudo-moar : OUTPUT«(timeout)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␤31␤32␤33␤34␤35␤36␤37␤38␤39␤40␤41␤42␤43␤44␤45␤46␤47␤48␤49␤50␤51␤52␤53…»
rakudo-moar : OUTPUT«<anon|304326384>.new␤»
BenGoldberg m: (1..*).iterator.sink-all; say 'ok';
camelia rakudo-moar : OUTPUT«(timeout)»
BenGoldberg p6: gist.github.com/BenGoldberg1/467df...d16fb4af23 01:00
camelia rakudo-moar: OUTPUT«ok 1␤(2 3 5 7 11 13 17 19 23 29 31 37 41 43 53 59 61 71 73 79 83 89 101 103 109 113 131 139 149 151)␤ok 2␤»
diakopter .tell jnthn shaved 8% off CORE compilation by writing the idx after looking it up in interp.c i.imgur.com/WX35mFe.png
yoleaux diakopter: I'll pass your message to jnthn.
BenGoldberg j: gist.github.com/BenGoldberg1/467df...d16fb4af23
camelia rakudo-jvm c016df: OUTPUT«ok 1␤(2 3 5 7 11 13 17 19 23 29 31 37 41 43 47 59 71 77 83 89 101 107 113 119 143 149 161 167 191 203)␤ok 2␤»
BenGoldberg That's not good.
BenGoldberg slaps jakudo. 01:02
The output from moar is correct, and I don't know wtf is wrong with r-j, cause 53 and 61 are prime, and 77 definitely isn't. 01:06
zengargoyle TIL, camelia runs urls... 01:07
BenGoldberg Only github gists. 01:08
zengargoyle neat enough.
Ben_Goldberg m: gist.github.com/BenGoldberg1/7e365...fe49eed7be 02:18
camelia rakudo-moar : OUTPUT«Start␤Produced 1426 primes: 2 3 5 7 11 13 17 19 23 29 31 37 41 43 47 53 59 61 67 71 73 79 83 89 97 101 103 107 109 113 127 131 137 139 149 151 157 163 167 173 179 181 191 193 197 199 211 223 227 229 233 239 241 251 257 263 269 271 277 281 283 293 307 311 313 3…»
Ben_Goldberg j: gist.github.com/BenGoldberg1/7e365...fe49eed7be
camelia rakudo-jvm c016df: OUTPUT«Start␤Produced 67 primes: 2 3 5 7 11 13 17 19 23 29 31 37 41 43 47 53 59 61 67 71 73 79 83 89 97 101 103 107 109 113 127 131 137 139 149 151 157 163 167 173 179 181 191 193 197 199 211 223 227 229 233 239 241 251 257 263 269 271 277 281 283 293 307 311 31…»
flussence j: say ^200 .grep(*.is-prime) # I wonder if it at least gets *this* right? 02:20
camelia rakudo-jvm c016df: OUTPUT«(2 3 5 7 11 13 17 19 23 29 31 37 41 43 47 53 59 61 67 71 73 79 83 89 97 101 103 107 109 113 127 131 137 139 149 151 157 163 167 173 179 181 191 193 197 199)␤»
Ben_Goldberg On the one hand, I'm glad that both moar and rakudo run this code the same. On the other hand, I'm disappointed about how slow jakudo runs. 02:21
Zoffix m: say ^1000 .grep(.is-prime) 02:49
camelia rakudo-moar : OUTPUT«Method 'is-prime' not found for invocant of class 'Any'␤ in block <unit> at /tmp/q1aLJ7nbk6:1␤␤»
Zoffix m: say ^1000 .grep(*.is-prime)
camelia rakudo-moar : OUTPUT«(2 3 5 7 11 13 17 19 23 29 31 37 41 43 47 53 59 61 67 71 73 79 83 89 97 101 103 107 109 113 127 131 137 139 149 151 157 163 167 173 179 181 191 193 197 199 211 223 227 229 233 239 241 251 257 263 269 271 277 281 283 293 307 311 313 317 331 337 347 349 353 359 36…»
Zoffix picks up the blown mind 02:50
Tweet-worthy :P
m: say ^1000000000000 .grep(*.is-prime) 02:52
camelia rakudo-moar : OUTPUT«(2 3 5 7 11 13 17 19 23 29 31 37 41 43 47 53 59 61 67 71 73 79 83 89 97 101 103 107 109 113 127 131 137 139 149 151 157 163 167 173 179 181 191 193 197 199 211 223 227 229 233 239 241 251 257 263 269 271 277 281 283 293 307 311 313 317 331 337 347 349 353 359 36…»
Zoffix 0.o
wait wat: 02:53
m: say ^1000000000000 .grep(*.is-prime).elems
How does it know to return early above, but now it'll timeout?
TimToady .elems is never lazy
camelia rakudo-moar : OUTPUT«(timeout)»
TimToady also, say stops at 100 elements
Zoffix Ah 02:53
gfldex does it stop at 100 all the time or just for camelia? 02:54
TimToady all the time; use put if you want everything 02:56
say is about "gist"
gfldex this cheat should be documented
even if it's clearly documented, it will lead to bugs 02:57
Zoffix It's not really a "cheat". You're asking to print the .gist. 02:59
Zoffix That the doc says about 4 times on one page :P doc.perl6.org/routine/say 03:00
Sure, there'll be bugs, but due to people not reading docs and expecting `say` to be the `print` version of other languages.
gfldex Zoffix: doc.perl6.org/type/List got a generic section for gist and does not mention what happens in github.com/rakudo/rakudo/blob/nom/...st.pm#L591 03:04
I'm trying to make Perl 6 foolproof by being the fool. 03:06
Zoffix gfldex, I opened an Issue: github.com/perl6/doc/issues/206 03:07
dalek href="https://modules.perl6.org/new-db-builder:">modules.perl6.org/new-db-builder: 6dd1ad2 | (Zoffix Znet)++ | web/lib/DbBuilder (4 files):
GitHub Source experimental plugin fully functional

Missing docs / tests		 03:21
Zoffix BTW, if we need to support more than just GitHub repos, let me know :P 03:22
At least on modules.perl6.org it'll soon be very trivial to add new sources :) 03:23
(not sure about how easy it is with panda tho)
dalek c: 455f420 | (Wenzel P. P. Peppmeyer)++ | doc/Type/Mu.pod:
List::gist does truncate, others may do so as well.
 03:25
c: 38fc995 | (Wenzel P. P. Peppmeyer)++ | doc/Type/Mu.pod:
Mention .put for machine readable output.
c: 30db8ef | (Zoffix Znet)++ | doc/Type/Mu.pod:
Merge pull request #207 from gfldex/master

  .gist may truncate .put may be better for machine readable output
c: 75af98e | (Brock Wilcox)++ | / (2 files):
Add initial styleguide
c: aaf6572 | (Zoffix Znet)++ | / (2 files):
Merge pull request #202 from perl6/initial-styleguide

Add initial styleguide
flussence iirc there's a few people using gitlab already, might want to add that 03:27
Zoffix k
flussence (I've been thinking of moving over there myself for a while, I just need to start producing actual code to move :) 03:28
dalek c: a188308 | (Zoffix Znet)++ | doc/Type/Mu.pod:
Reword/grammar
Zoffix I use it for private repos (they're free), but I often find the site is slow
herby Hello! 03:56
gfldex m: say 'Hi herby!'; 03:57
camelia rakudo-moar : OUTPUT«Hi herby!␤»
herby m: say 1+1 04:02
camelia rakudo-moar : OUTPUT«2␤»
hankache morning 08:34
yoleaux 21 Nov 2015 20:05Z <TimToady> hankache: re - and ' in names, alpha is only required to the right; to the left can be numeric; think of it as a sequence of identifiers separated by - or '
hankache .tell TimToady re - and ' noted. thank you for going through the doc 08:37
yoleaux hankache: I'll pass your message to TimToady.
FROGGS o/ 09:00
yoleaux 21 Nov 2015 23:49Z <leont> FROGGS: I don't think so, though the interactive output uses \r a lot
FROGGS .tell leont I try to golf it down to a testcase 09:01
yoleaux FROGGS: I'll pass your message to leont.
dalek ast: 3e980b2 | usev6++ | S32-str/ (2 files):
Tests for index/rindex with negative start position
 09:28
nine m: say "hello"; say "world"; 09:33
camelia rakudo-moar : OUTPUT«hello␤world␤»
dalek ast: bae740d | usev6++ | S32-str/ (2 files):
Fix RT number
 09:39
nine Really? Of all the monospace fonts on my system only GNU Unifont and Source Code Pro have the newline symbol. And neither is really pretty :/ 09:42
bartolin sorry, camelia 09:48
[Tux] test 50000 21.222 21.109 09:58
test-t 50000 17.995 17.882
csv-parser 50000 28.856 28.742
dalek ast: 46754e8 | usev6++ | S15-unicode-information/uniname.t:
Use existing RT number
 10:00
arnsholt nine: Inconsolata has it, I think 10:01
[Tux] nine DejaVu Sans Mono, I just tested 10:04
pretty is only in the eye of the beholder :) I personally think *all* serif fonts are ugly
nine [Tux]: Odd. I still get the replacement square with DejaVu Sans Mono. liberation2 AKA Google Croscore may support it though. I've just installed it but to test I'd have to restart konsole which I can't because smart as I am I didn't start my system backup in screen... 10:07
arnsholt: thanks, will try that, too 10:08
[Tux] tux.nl/Files/20151122111014.png 10:10
shown in xterm-319
nine I guess there's something in the konsole -> bash -> ssh -> bash -> screen -> irssi chain screwing things up. 10:12
moritz nine: the nice thing about debugging such a long chain is that you can bisect :-) 10:45
stmuk_ jnthn++ # blog post explanations 10:46
masak today's mini-challenge: write, in Perl 6, a *small* thing which can show Rubik's cube positions (maybe as an ASCII version of commons.wikimedia.org/wiki/File:Ru...colors.svg ), and allow arbitrary moves on the cube. i.e. something a cube solver could be built on top of. 10:48
g'day, #perl6
"small" is in "don't overdo it", not "small" as in "golfed" :P
dr4ngd anyone here? 10:51
masak dr4ngd: hi
in other news, there's a depressing lack of web pages of the form "here's a sane programmatic model for the cube, and here's an algorithm (in Perl/Python/Ruby/whatever) that always solves the cube [assuming it's solvable]" 10:52
I'm *almost* tempted to fill that void :)
dr4ngd I was wondering if you guys knew anything about books that are going to be published on Perl 6
nine Next milestone finished: panda installs itself into a CompUnit::Repository::Installation
masak this is the closest to that ideal that I've found: fulmicoton.com/posts/rubix/ 10:57
dr4ngd is that like some way of saying in heaven you know every possible rubix cube combination 10:59
in multiple dimensions
dr4ngd anyway. I have a site I'm going to be working on in Perl. If you're expert level I'd love to share the project with you if anyones interested. It's a money maker... 11:02
moritz is an expert level procrastinator 11:04
FROGGS nine: do you think it is a good time to join the CUR effort?
nine FROGGS: it is always a good time to join :) 11:06
moritz
.oO( A Frogg is never late! )
nine FROGGS: and considering that it's only a couple of weeks till the big one, we should really merge this sooner rather than later 11:07
dr4ngd when is perl 6 coming out
masak dr4ngd: "God's number" -- the fancy name for "the maximum number of cube moves required to optimally solve any position" -- was shown to be 20 back in 2010.
dr4ngd: Perl 6 is coming out by Christmas. 11:08
moritz this Christmas.
masak dr4ngd: but it's an open source project, so you can download it today and play with it -- get ahead of the curve! :D
masak .oO( o/ this Christmas, I'll give you Perl 6, but the very next day, you'll RT away o/ ) 11:09
stmuk_ hahaha
FROGGS nine: okay, so expect questions :o) 11:10
dr4ngd masak: thanks
nine FROGGS: love them :)
dr4ngd anyone interested in something like this to work on www.kickstarter.com/projects/dante...and-social 11:11
I want to make a social site, for gamers, with their own private guild forums and pretty much everything MMO Players would want in one place... 11:12
dr4ngd which is a rollout project to a more intensive project that you could get involved with too. done in Perl :) 11:12
moritz aren't there two dozen projects like this out there already? 11:15
diakopter gamifying gamer gamification
stmuk_ I think ovid has game plans .. it might be worth contacting him 11:16
masak more than plans
dr4ngd moritz: not even a little near the scope of what mines about
moritz: in content or design
masak can't fault you for lack of ambition :) 11:17
lizmat waves from Echt
nine o/ lizmat 11:20
lizmat nine o/
masak lizmat \o 11:21
stmuk_ you could make a social site about ethics in game journalism :> 11:23
lizmat masak o/
lizmat is slightly jetlagged
dr4ngd why is Perl 6 considered a sister project 11:25
masak dr4ngd: because Perl 5 is still being actively developed, and with a large user base.
dr4ngd: also, Perl 6 is not an incremental update, it's a fresh start in many ways. 11:26
stmuk_ it's intended as a supplement to perl 5 rather than a replacement
masak each of Perl 5 and Perl 6 has its own strengths and specialties.
metasin I'd like to read more about those specialties and strengths
dr4ngd masak: So what happens when the larger user base gets to Perl 6 themselves? they'll both be Perl 6? lol 11:27
metasin dr4ngd: think of how many numbers are in between 0 and 1. 11:28
masak metasin: lots of things. but perhaps most importantly: Perl 5 -- parsing text. Perl 6 -- parsing languages. 11:28
dr4ngd metasin: 0.1 is a number too.
masak metasin: that is, where Perl 5 has excellent regexes, Perl 6 has that plus excellent *grammars*.
metasin masak: yes, which is why I'm having a hard time letter perl6 go 11:29
dr4ngd I'm just concerned with learning Perl 6 without a book lol 11:30
metasin masak: I'd love to put together some kind of comprehensive article on particular semantic differences though
masak: if you know of one... ^_^ 11:31
dr4ngd: don't worry. You have an irc channel right here.
masak dr4ngd: the "Perl 6" name/version number can be seen as an "unfortunate naming circumstance". maybe in retrospect a different name or something would've been more clear. but it's too late to re-brand it, because everyone knows it as "Perl 6".
metasin: there are a couple good resources. moritz++' 5-to-6 articles come to mind.
nine FROGGS: jnthn's design in case you don't have it at hand: gist.github.com/jnthn/47a42b2e86e7e552b2e2 11:32
masak metasin: there's also github.com/perl6/mu/blob/master/do...rences.pod -- but watch out for occasional outdated stuff.
dr4ngd masak: where do I start? 11:32
masak m: say "start here! :)"
camelia rakudo-moar : OUTPUT«start here! :)␤»
metasin masak: oddly, I've already gone through those. More inspiration I suppose. Time to write a lisp->perl6 tutorial or something 11:33
FROGGS nine: yes, I need to re-read it carefully
moritz also doc.perl6.org/language.html has links to various 5to6 documents
FROGGS and then I somehow need the current state
metasin inline::Perl5 is pretty nifty too
^-- (pretty strong about what's different in perl6) 11:34
dr4ngd so learn Perl 5 then move onto 6? cause I know JS, Python, C, some PHP. I decided to make my site in Perl and saw that Perl 6 was coming out, so I guess I was looking for a book to start from scratch 11:35
nine FROGGS: the design is mostly implemented. @*INC still has to be removed and $?REPO has to be added. CompUnit only contains the attributes that are needed for loading ($.dist, $.ver, ... are missing). And I'm very close to having a fully functional panda.
masak dr4ngd: there's also a nice big "Download" button at perl6.org/
metasin dr4ngd: no learn perl6
masak dr4ngd: it's possible to learn first-perl5-then-perl6, but it's also possible to start directly with Perl 6. 11:36
depends what your goals are, I guess.
Perl 5 is a very decent language too.
well worth learning.
metasin masak: is there a Higher Order Perl (6) ?
FROGGS nine: okay, so looking at panda might be a good start
nine: I hope to have some time this evening
nine FROGGS: great! I'll try to be available 11:38
masak metasin: no, but that would be pretty awesome. 11:41
by the way, diakopter++ dug up kociemba.org/cube.htm with what appears to be an optimal C solver.
haven't tested it yet.
dr4ngd so I'm downloading Rakudo 11:43
what database do I use? 11:44
for SQL and shit
diakopter which one are you downloading 11:45
dr4ngd rakudo star 2015.09 jit
moritz dr4ngd: we have support for mysql, postgres and sqlite 11:46
dalek kudo/curli: 681c18b | (Stefan Seifert)++ | src/core/CompUnit/Repository/Installation.pm:
Work towards a fix installation of bin wrapper scripts

Store resources in the resources directory of the CompUnit::Repository::Installation. Store bin wrappers in the bin directory. Fix lookup of the wrapped bin file.		 11:47
dr4ngd moritz: thank you
nine FROGGS: pushed current state to rakudo/curli and panda/curli
I'm out now for the afternoon for the christmas market with my girlfriend :)
metasin masak: blah.. Guess I'll get started.. Hehe ;) also, awesome!! 11:47
dalek ar: 2dfa7ba | moritz++ | tools/star/release-guide.pod:
Fix off-by-one error in the docs
 11:51
dr4ngd can Perl 6 be used to make applications? 11:54
moritz dr4ngd: as opposed to... what? 11:55
dr4ngd moritz: not in a browser
metasin o_o
moritz dr4ngd: all programs are applications 11:56
dr4ngd moritz: ok I meant nonweb-based
moritz dr4ngd: do you mean something with a graphical user interface?
dr4ngd moritz: yep
moritz dr4ngd: ah. Yes, for example with github.com/perl6/gtk-simple/ 11:57
dr4ngd moritz: thanks 11:58
dr4ngd now something for discussion... what if there were to be an internet loaded operating system that overrides system control of your OS? :) 12:00
moritz dr4ngd: you mean like a configuration management system? 12:01
see puppet, chef, ansible et. al.
dr4ngd moritz: I was thinking something like windows, that literally overrode your windows. as its own operating system.
dr4ngd that was accessible from the internet. from a site with your own account and settings. 12:01
moritz dr4ngd: like a terminal server? 12:02
dr4ngd moritz: um, I was thinking something that had operable ways of taking control of system functions for a large base of computers. but literally ran as the operating system. full GUI. 12:03
dr4ngd sort of like a virtual machine, but no download necessary. 12:03
moritz "taking control of system functions for a large base of computers" -- a botnet? 12:04
dr4ngd moritz: I don't mean a lot of computers. I just mean that I would code it specifically for certain computers. then you pick the computer you're using, and it overrides the OS. 12:05
that way, you could always have your computer wherever you go. a cloud situation
moritz dr4ngd: that's what a terminal server does
dr4ngd: your session is on the server, and you just need a very thin client that forwards your input to the server, and the server's answer to your local monitor 12:06
stmuk_ like VNC
dr4ngd moritz: I honestly don't think we're talking about the same thing lol 12:07
moritz dr4ngd: we are, you just don't want to realize it
stmuk_: or citrix receiver (*shudder*)
dr4ngd moritz: it's a complete OS with its own apps running from the internet streaming your apps with a full GUI 12:08
moritz dr4ngd: exactly
dr4ngd moritz: the only thing I can think of like that is Chrome OS 12:09
moritz: I think you're talking about how things work besides presented and accessed.
moritz dr4ngd: no; I'm just telling you that terminal servers + clients provide exactly the user experience you seem to be looking for 12:10
stmuk_ maybe he means virtualisation
dr4ngd moritz: I'm talking about something more presentable commercially. 12:11
also with an operating system that isn't created by me yet.
moritz doesn't see what's comercially unpresentable about terminal servers
stmuk_ they are in wide use in the "enterprise" 12:12
moritz indeed, at $work we're making lots of money with them
dr4ngd so with a terminal server, I can login to the internet and it will run an operating system that overrides my operating system? 12:15
dr4ngd from a web browser? 12:16
moritz dr4ngd: it makes your local operating system irrelevant
and yes, citrix for example has a web client
dr4ngd moritz: that's fucking weird. why isn't it the status quo
moritz dr4ngd: flakey networks, trust, latency, economics 12:17
dr4ngd: at $work, we have thin clients in all of our meeting rooms, which connect to a terminal server; it works very well 12:18
but it's a local network, and the terminal server is run in-house, so we have full control
I wouldn't want that for my local desktop
stmuk_ my experience of things like thin clients is that the server is under powered and overprovisioned
stmuk_ at least in the businesses I've been unfortunate to have worked in 12:19
dr4ngd moritz: oh. that's just weird to me. I can think of a lot of utilizations of that that would make a lot of money.
dalek ar: a6b6afe | moritz++ | tools/star/Makefile:
Use Rakudo 2015.11
ar: 5dd02c1 | moritz++ | modules/ (21 files):
Update modules to current version
ar: 76e441f | moritz++ | tools/build/panda-state.p6:
Fix HTTP communication in panda-state.p6
moritz dr4ngd: there are lots of use cases where that already makes lots of money 12:20
moritz dr4ngd: but it's more in the enterprise space, where "give everybody a consistent experience" trumps usability, customizability for the user etc. 12:20
stmuk_ and "save money"
moritz and that, yes 12:21
dr4ngd moritz: I have a project that would take that type of connect to a social outlet
moritz dr4ngd: then with all due respect, maybe you should do a thorough review of what's there first :-) 12:22
dr4ngd moritz: yeah that's what I'm doing right now
thanks :)!
moritz I've done my fair share of side projects that ended up with "oh, somebody else has done it much better than me, I'll just use theirs instead" 12:23
dr4ngd I have a whole project that but mines like, oh everyones done everything separately, I think I'll just put it all together
dr4ngd my work is sort of an overhaul with fresh new content. 12:26
the only problem is I'm the only one working on the project. :/ 12:27
metasin oh, merry belated 4:20 to any of you at -8GMT 12:28
dr4ngd I wish I had some acid 12:30
metasin lol... do enough perl and you wont know the difference
masak .oO( what is this, eternal September? )
metasin hehe 12:31
dr4ngd are you telling me perl fries my brain
masak metasin: believing Perl is unreadable is such a 90s notion.
metasin masak: Perl is totally readable, ad-infinitum 12:32
dr4ngd Perl looks intricate not unreadable.
metasin masak: I admit though, I feel attached to the 90s. Help me
dr4ngd yes let us all listen to third eye blind and the smashing pumpkins 12:33
dalek ar: 60378f2 | moritz++ | / (3 files):
Revert "introduce support for sdkroot and sysroot to support cross-builds"

This reverts commit f0def6dc7b4f576668e689cfd44745e506a305ad. See github.com/rakudo/star/pull/53#iss...-158754696 for the reason.
masak metasin: reading good Perl code from the last 15 years helps. 12:39
azawawi hi 12:39
masak hi, azawawi
azawawi how do I access the original text (i.e. including newlines) of a Pod::Block::Named object? 12:40
metasin masak: I think you've missed the joke... >_> may I have some coffee?
sno moritz: to fix an issue with a .git checkout a guide how to prepare a test environment is sane ... 12:41
dalek href="https://perl6.org:">perl6.org: 533502a | (Zoffix Znet)++ | TODO:
Remove severely outdated TODO
 12:43
azawawi moritz: ping 12:52
dalek href="https://perl6.org:">perl6.org: da6e383 | (Zoffix Znet)++ | README.md:
Plack is also needed in deps for plackup
 12:54
azawawi .tell moritz how do I access the original text (i.e. including newlines) of a Pod::Block::Named object? 13:11
yoleaux azawawi: I'll pass your message to moritz.
moritz .tell azawawi if .perl doesn't show it, it's not there anymore 13:25
yoleaux 13:11Z <azawawi> moritz: how do I access the original text (i.e. including newlines) of a Pod::Block::Named object?
moritz: I'll pass your message to azawawi.
dalek href="https://perl6.org:">perl6.org: cf32971 | (Zoffix Znet)++ | source/js.css:
Remove useless file
 13:33
dalek href="https://perl6.org:">perl6.org: 35d9068 | (Zoffix Znet)++ | source/ (4 files):
Add "Recent Blog Posts" section on home page (Closes #42)
 13:49
DrForr breathes deeply while writing a tutorial proposal for OSCON. 13:51
lucasb "brace yourselves, christmas is coming" <-- it would be fun if this sentence gets added to the channel topic :) 13:56
DrForr Not by my hand, unfortunately. 14:01
brrt i think moritz++ can 14:08
DrForr Is anyone interested in reviewing a biography I'm about to submit? 14:09
stmuk_ Zoffix++ # quicker than a JITed quick thing at perl6.org site features 14:13
dalek href="https://modules.perl6.org/new-db-builder:">modules.perl6.org/new-db-builder: 55dd564 | (Zoffix Znet)++ | web/lib/DbBuilder/Dist/Source (2 files):
Make base Dist Source class handle more of readme/tests setting instead of having subclasses do it
dalek href="https://modules.perl6.org/new-db-builder:">modules.perl6.org/new-db-builder: 42fcd2d | (Zoffix Znet)++ | web/lib/DbBuilder/Dist/Source/GitHub.pm:
Be paranoid about data source

If an attacker somehow sneaks arbitrary data into a GitHub commit SHA, they could exploit us. Prevent that.		 14:18
dalek href="https://modules.perl6.org/new-db-builder:">modules.perl6.org/new-db-builder: 78a9552 | (Zoffix Znet)++ | web/lib/DbBuilder/Dist/Source/GitHub.pm:
Use saner code
 14:22
zengargoyle thinks that POD handling of unknown named blocks is brokenish. they shouldn't be processed like POD (all squished up), but left completely intact. 14:24
otherwise you'll never actually be able to use their contents via some other process.
zengargoyle i.e. « =begin Pic␤.....␤=end Pic » will be useless to a Pic processor if the contents have been munged. 14:26
dalek href="https://modules.perl6.org/new-db-builder:">modules.perl6.org/new-db-builder: 3dd2058 | (Zoffix Znet)++ | web/lib/DbBuilder/Dist/Source/GitHub.pm:
Do not die if we don't understand date_updated format
 14:28
dalek kudo-star-daily: c233516 | coke++ | log/ (3 files):
today (automated commit)
 14:29
rl6-roast-data: 95c634b | coke++ | / (9 files):
today (automated commit)
zengargoyle sorta think the whole of POD stuff should be kept intact and not munged until it hits a PodToXXX parser (if that's possible) 14:30
zengargoyle oh, and good * #perl 6 14:44
BenGoldberg Good MornAfternoonEvening. 14:48
hankache hello everyone 14:57
dalek kudo/curli: f967ac2 | (Stefan Seifert)++ | src/core/CompUnit/Repository/Installation.pm:
Fixup: path to resource files has to contain 'resouces/'

This makes the generated panda wrapper work now!		 15:06
jdv79 m: role R[::T = Str] { has T $.r}; R.new.r.say 15:07
camelia rakudo-moar : OUTPUT«Method 'say' not found for invocant of class 'T'␤ in block <unit> at /tmp/xp6lnrXp_I:1␤␤»
jdv79 shouldn't that work as if T was Str and not error like that? 15:08
Skarsnik you can new on a role? that's weird x) 15:10
jdv79 Punning is the term
Skarsnik and hello there
BenGoldberg It's only doable for roles which don't have any unimpelmented required methods. 15:11
m: role R { method cant-pun-me { ... } }; say R.new;
camelia rakudo-moar : OUTPUT«Method 'cant-pun-me' must be implemented by R because it is required by a role␤ in any compose_method_table at gen/moar/m-Metamodel.nqp:2680␤ in any apply at gen/moar/m-Metamodel.nqp:2690␤ in any compose at gen/moar/m-Metamodel.nqp:2826␤ in any make_…»
jdv79 i get the same error with just normal application as well
well, composition i guess its called nowadays 15:12
BenGoldberg m: role R[::T] { has T $.r}; R[Str].new.r.say
camelia rakudo-moar : OUTPUT«Method 'say' not found for invocant of class 'T'␤ in block <unit> at /tmp/ZioBB4oCHI:1␤␤»
BenGoldberg m: role R[::T] { has T $.r}; say R[Str].new.r
camelia rakudo-moar : OUTPUT«Method 'gist' not found for invocant of class 'T'␤ in block <unit> at /tmp/pQ2Rh5b3WS:1␤␤»
BenGoldberg m: role R[::T] { has T $.r}; say R[Str].new.r.WHAT 15:13
camelia rakudo-moar : OUTPUT«Method 'gist' not found for invocant of class 'T'␤ in block <unit> at /tmp/Rdd5V1UeG0:1␤␤»
jdv79 m: role R[::T = Str] { has T $.r}; class A does R {}; A.new.r.say
camelia rakudo-moar : OUTPUT«Method 'say' not found for invocant of class 'T'␤ in block <unit> at /tmp/EP4sdRY8Mj:1␤␤»
BenGoldberg m: role R[::T] { has T $.r}; my $r = R[Str].new; say 'ok';
camelia rakudo-moar : OUTPUT«ok␤»
BenGoldberg m: role R[::T] { has T $.r}; my $r = R[Str].new; my $rr = $r.r; say 'ok'
camelia rakudo-moar : OUTPUT«ok␤»
BenGoldberg m: role R[::T] { has T $.r}; my $r = R[Str].new; my $rr = $r.r; say $rr
camelia rakudo-moar : OUTPUT«Method 'gist' not found for invocant of class 'T'␤ in block <unit> at /tmp/wY6QugAGRZ:1␤␤»
jdv79 it seems to be because type/class T is not real 15:14
which means the bind is failing or something?
nine .tell FROGGS with my last rakudo patch, a panda installed into a Repository::Installation now works with the wrapper generated by rakudo and can install many modules :) The default repo list contains only Repository::Installation (perl6 -e 'say $*REPO.repo-chain') 15:16
yoleaux nine: I'll pass your message to FROGGS.
BenGoldberg m: role R[::T] { method diditbind { T.WHAT } }; my $r = R[Str].new; say $r; say $r.diditbind; 15:19
camelia rakudo-moar : OUTPUT«R[Str].new␤(Str)␤»
dalek href="https://modules.perl6.org/new-db-builder:">modules.perl6.org/new-db-builder: eda8def | (Zoffix Znet)++ | web/ (3 files):
Tests for GitHub dist Source [Repo without a README, tests, or logotype]
 15:19
moritz rakudo star is now in shape that it only needs an announcement and some testing to be releasable 15:21
jdv79 BenGoldberg: huh. 15:22
moritz nine++ # precomp work
El_Che tadzik: nothing but love for panda, but wouldn't the name confuse people coming from python? en.wikipedia.org/wiki/Pandas_%28software%29 15:32
tadzik El_Che: I've never heard of this before 15:34
Begi11115 PandaS =! panda
moritz El_Che: people are goot at putting stuff into context
jdv79 m: role R[::T = Str] { has T $.r =T}; class A does R; A.new.r.say 15:36
camelia rakudo-moar : OUTPUT«5===SORRY!5=== Error while compiling /tmp/lx74dRiHaz␤Too late for unit-scoped class definition;␤Please use the block form.␤at /tmp/lx74dRiHaz:1␤------> 3 = Str] { has T $.r =T}; class A does R;7⏏5 A.new.r.say␤»
jdv79 m: role R[::T = Str] { has T $.r =T}; R.new.r.say
camelia rakudo-moar : OUTPUT«(Str)␤»
El_Che tadzik: pandas is pretty big in the scientific community. Just so you know in case someone asks :)
jdv79 why is it necessary to set the default to the type obj in that case but not in general?
tadzik El_Che: alright, noted :)
jdv79 or is that a bug?
lucasb Is the problem with 'role R[::T] { has T $.x }' the fact that T will only be known at runtime and 'has' needs to know that at compile time? 15:38
timotimo no, it has to be known at composition time 15:39
which is when you apply the role to something or when you pun the role 15:40
not sure why you use ::T there, tbh
m: role R[T] { has T $.r}; say R[Str].new.r.WHAT 15:41
camelia rakudo-moar : OUTPUT«5===SORRY!5=== Error while compiling /tmp/pLd2F20ndY␤Invalid typename 'T' in parameter declaration.␤at /tmp/pLd2F20ndY:1␤------> 3role R[T7⏏5] { has T $.r}; say R[Str].new.r.WHAT␤»
timotimo oh, that's why
m: role R[::T] { has T $.r}; say R[Str].new.r.WHAT
camelia rakudo-moar : OUTPUT«Method 'gist' not found for invocant of class 'T'␤ in block <unit> at /tmp/jxxLkXSbuk:1␤␤»
timotimo it really shouldn't do that, why does it do that?
m: role R[::T] { has T $.r}; class Test does R[Str] { }; Test.new.r.WHAT
camelia ( no output )
timotimo a problem with punning, i suppose?
because it *should* work the exact same way there 15:42
like, punning is designed to be the same as applying the role to an empty class
jdv79 no, composing that onto a class produces the same thing
timotimo huh? 15:43
camelia says no
DrForr How is Bailador coming along?
DrForr (I supposse I could just download it myself...) 15:44
timotimo i don't know if there were many commits to bailador in the recent months? 15:45
DrForr Hrm.
dalek href="https://modules.perl6.org/new-db-builder:">modules.perl6.org/new-db-builder: 4c0fb2e | (Zoffix Znet)++ | / (95 files):
Remove superfluous "web" dir

We are not building anything anymore, so we can do without any special dirs to keep the files in.		 15:46
timotimo oh
there was a pull request to bailador 22 days ago
jdv79 m: role R[::T] { has T $.r}; class Test does R[Str] { }; Test.new.r.say
camelia rakudo-moar : OUTPUT«Method 'say' not found for invocant of class 'T'␤ in block <unit> at /tmp/3rGqp2dydl:1␤␤»
timotimo it adds a license file %)
oh, i forgot the "say"!
silly me!
DrForr Well, if it's stable enough that someone can run the Plack equivalent and get a dynamic webpage that's all I'd ask. 15:47
timotimo no clue 15:48
i'm bad at web
DrForr Actually that's another dependency that might scare people off. Backup plan.
timotimo last time i web-app'd i used HTTP::Server::Async 15:49
stmuk_ I think gabor uses a perl6 web system in production for perl6maven
timotimo is it a web server or a generator?
you can easily see if you use siege or ab :P
moritz server
m: put Int 15:50
camelia rakudo-moar : OUTPUT«Use of uninitialized value of type Int in string context␤Any of .^name, .perl, .gist, or .say can stringify undefined things, if needed. in block <unit> at /tmp/6wUq_OA6Gv:1␤␤»
stmuk_ I think its nginx proxying bailidor 15:53
with the perl6 restarted frequently via a cron :) 15:54
timotimo d'oh
Skarsnik damn smoke/smokegen (the tool used by kde to bind qt/kde lib to other language) is not really good with C, but it's close x) pastebin.com/FKZPCAcL 15:57
lucasb m: say '3' / '2' 16:02
camelia rakudo-moar : OUTPUT«1.5␤»
lucasb m: say '3' div '2'
camelia rakudo-moar : OUTPUT«Cannot call infix:<div>(Str, Str); none of these signatures match:␤ (Int:D \a, Int:D \b)␤ (int $a, int $b --> int)␤ in block <unit> at /tmp/Kc_tdIga6M:1␤␤»
lucasb 'div' needs to be stricter for some reason? 16:03
moritz yes.
hankache lucasb div will only accept integers 16:04
dalek ar: e6cb388 | moritz++ | docs/announce/2015.11.md:
Add draft 2015.11 release announcements
hankache m: say 3.0 div 2.0;
camelia rakudo-moar : OUTPUT«Cannot call infix:<div>(Rat, Rat); none of these signatures match:␤ (Int:D \a, Int:D \b)␤ (int $a, int $b --> int)␤ in block <unit> at /tmp/0LNb5Xfj4H:1␤␤»
hankache lucasb not even Rat
moritz div is specificallly *integer* division; use / if you want general division 16:05
hankache m: say 3 / 2; 16:05
camelia rakudo-moar : OUTPUT«1.5␤»
autarch what's the process by which bugs get on the Christmas list? is it too late to add anything?
hankache m: say 3 div 2;
camelia rakudo-moar : OUTPUT«1␤»
dalek href="https://modules.perl6.org/new-db-builder:">modules.perl6.org/new-db-builder: 74a2be4 | (Zoffix Znet)++ | build-project-list.pl:
Add --help usage instructions to builder script
 16:06
lucasb ok, thanks hankache, moritz
moritz autarch: you convince jnthn or TimToady (I think; or maybe me) that it's a Christmas blocker
hankache lucasb plus the result is always rounded down to the closest Int
moritz m: say (-3) div 2 16:07
camelia rakudo-moar : OUTPUT«-2␤»
timotimo rounded down or truncated?
stmuk_ hmmm I'm seeing use.t fail with Inline::Perl5
moritz timotimo: ^^ doesn't look like truncation
hankache timotimo isn't it the same thing?
from a mathematical point of view? 16:08
timotimo well, there's "round towards 0" and "round towards negative inf"
both could be called "round down"
moritz except that zero isn't "down" from a negative number's POV 16:09
timotimo the absolute value goes down, though :P 16:10
from -3.6 to -3
hankache imho it's rounded down because if it was truncated -3 div 2 would have truncated -1.5 which would be -1 16:11
DrForr Is this overkill for a tutorial description? paste.scsys.co.uk/501771
hankache DrForr no 16:13
DrForr Thanks. I tried to hit what I think is going to be a draw, and at least what I'm pretty sure I can teach. 16:14
lucasb DrForr: Don't you think is better to change 'Please come to the tutorial with a laptop running the latest version of Rakudo Star' to 'We will teach you how to install the thing'? :) 16:15
pink_mist maybe have a pre-tutorial workshop where you help people install it 16:16
DrForr Guh. If it's going to be that hard by May then there's no point to a tutorial. 16:17
pink_mist good point :P 16:18
DrForr Rakudobrew went onto my system simply, but I'm guessing by what's being said here that I'm in the vast minority? 16:19
Or am I simply describing the wrong install requirement?
hankache DrForr chekout github.com/hankache/perl6intro 16:20
stmuk_ I suspect most people on the channel install rakudo daily but that the potential audience might be better off installing from docker or something
hankache DrForr or www.perl6intro.com
I tried to compile an introductory tutorial. I would be glad if it can be of any help 16:21
masak metasin: yes, I definitely missed the joke, whatever it was :)
DrForr hankache: It'd be going into an OSCON tutorial link, ust so you're aware. 16:22
*just
OSCON US has a dearth of Perl material and I'm hoping to capitalize on that. 16:23
DrForr And being a bit scared of the prospect of doing a 3-hour tutorial out of the gate. 16:24
hankache DrForr good luck. I am still learning Perl 6 so if you need a beginner point of view i can help 16:27
DrForr Cool, may take you up on that.
Begi11115 That's my case too if you need DrForr 16:28
DrForr scratches notes down.
hankache What is an OSCON tutorial link? Sorry in my part of the world we're way behind those things 16:29
timotimo just means the description of the session will link to your site 16:30
DrForr perl6info.com would be going into my tutorial description, and appear as part of the OSCON schedule, people interested in the tutorial would go there.
hankache DrForr ok. if it can help other people why not 16:31
DrForr Just saying you might get more traffic than you bargained for :)
hankache DrForr its perl6intro.com
timotimo i think it's hosted on github.io
hankache no it's on bluehost but its solid, should be able to handle the traffic 16:32
DrForr Okay, cool. OTOH people interested in this will probably self-select for creativity and desire to do something, so I shouldn't worry *too* much. 16:33
hankache the source is on github though
timotimo ah, ok
DrForr Yah, I'm not expecting much, and this only matters of course if it's accepted, which is a bit of a long shot. 16:34
dalek href="https://perl6.org:">perl6.org: 4271a55 | (Steve Mynott)++ | source/documentation/index.html:
add perl6intro.com
 16:36
hankache DrForr best of luck. ping me if you need any help 16:37
DrForr Thanks.
I'm currently working with 6Gb RAM, do we have any minimal hardware/software requirement docs hanging around? 16:38
DrForr goes looking.
moritz /home/moritz/p6/star/install/bin/perl6-j --target=jar --output=/home/moritz/p6/star/install/share/perl6/lib/JSON/Fast.pm.jar /home/moritz/p6/star/install/share/perl6/lib/JSON/Fast.pm 16:39
===SORRY!===
Cannot unbox a type object
Skarsnik moar spawn like 100Mo at least I think?
moritz for compiling the setting, I think you should have at least 1.5GB RAM 16:40
DrForr That's decent. 16:43
diakopter I just compiled the setting just now; it used max 921 MB real memory and max 3.27 GB virtual memory 16:44
stmuk_ moritz: github.com/rakudo/star/pull/56 16:47
dalek ar: a38ae2b | (Steve Mynott)++ | docs/2015-spw-perl6-course.pdf:
import newer Nov 21 version of jnthn slides
 16:55
ar: 5660f6e | (Steve Mynott)++ | docs/announce/2015.11.md:
ref to updated tutorial in announce
ar: 0548398 | moritz++ | docs/ (2 files):
Merge pull request #56 from stmuk/master

Updated jnthn slides
moritz stmuk_: thanks!
stmuk_ np 16:56
moritz sno: I can give you access to hack.p6c.org, a shared linux box where the R* problems can be reproduced
sno: are you interested?
sno moritz: I don't have time this evening for longer remote session 16:57
sno moritz: I have such a machine upstairs, but trouble with family etc. ... 16:57
moritz sno: ok 16:58
sno I can manage that until ~ Tuesday 16:58
and then I do both, fix Darwin tar misusage, too :D 16:59
hankache bbiab 17:00
stmuk_ Usage: panda <action> [options] 17:16
but I have to do panda --notests install 007 17:17
I think it should be Usage: panda [options] <action>?
zengargoyle stmuk_: think it's a limitation of MAIN() argument processing. named things (flags) must be before positionals. 17:18
stmuk_ yes it's caused many problems over the last few months 17:18
also side effects from allowing multiple modules being installed at once 17:19
zengargoyle is that where it sometimes fails to reinstall everything during a rakudobrew build? 17:22
stmuk_ I was thinking more of the commandline arg processing .. not seen that other problem 17:23
zengargoyle seems i'm always double checking and manually reinstalling something.
dalek ar: e33753b | moritz++ | docs/announce/2015.11.md:
Note about beta status, and brokenness of the JVM backend
 17:49
dalek ar: 222cce6 | moritz++ | / (2 files):
Bump version to 2015.11
 18:00
moritz hack.p6c.org/~moritz/rakudo-star-20...rc0.tar.gz # R* 2015.11 Release Candidate 18:01
please test it
xfix perl6: multi a(Int $n) is cached { $n }; say a 20 18:13
camelia rakudo-jvm c016df: OUTPUT«20␤» 18:14
..rakudo-moar: OUTPUT«Cannot invoke object with invocation handler in this context␤ in sub a at /tmp/tmpfile:1␤ in block <unit> at /tmp/tmpfile:1␤␤»
xfix perl6: multi a(Int $n) is cached { $n }; say try a 20; say a 20;
camelia rakudo-moar: OUTPUT«Nil␤20␤»
..rakudo-jvm c016df: OUTPUT«20␤20␤»
xfix Am I not understanding something, or is it a bug?
raiph m: multi a(Int $n) is cached { $n }; say (a 20).WHAT 18:17
camelia rakudo-moar : OUTPUT«Cannot invoke object with invocation handler in this context␤ in sub a at /tmp/F7DYn7By0G:1␤ in block <unit> at /tmp/F7DYn7By0G:1␤␤»
raiph m: sub a(Int $n) is cached { $n }; a 20 18:18
camelia ( no output )
raiph m: sub multi a(Int $n) { $n }; a 20
camelia rakudo-moar : OUTPUT«5===SORRY!5=== Error while compiling /tmp/opEYzLKyNi␤Missing block␤at /tmp/opEYzLKyNi:1␤------> 3sub multi7⏏5 a(Int $n) { $n }; a 20␤ expecting any of:␤ new name to be defined␤»
raiph m: multi a(Int $n) { $n }; a 20
camelia ( no output )
raiph m: multi a(Int $n) is cached { $n }; a 20
camelia rakudo-moar : OUTPUT«Cannot invoke object with invocation handler in this context␤ in sub a at /tmp/3lx3Ni259v:1␤ in block <unit> at /tmp/3lx3Ni259v:1␤␤»
BenGoldberg j: multi a(Int $n) is cached { $n }; a(20).WHAT.say 18:19
camelia rakudo-jvm c016df: OUTPUT«(Int)␤»
BenGoldberg m: multi a(Int $n) is cached { $n }; a(20).WHAT.say
camelia rakudo-moar : OUTPUT«Cannot invoke object with invocation handler in this context␤ in sub a at /tmp/ozhOVoIt_6:1␤ in block <unit> at /tmp/ozhOVoIt_6:1␤␤»
BenGoldberg raiph, Maybe you should file a bug report? 18:20
diakopter xfix: it's plainly a bug
xfix Calling a function once while catching exceptions is a workaround, but I feel it's not supposed to work that way. 18:21
raiph currently exploring if it's related to rt.perl.org/Ticket/Display.html?id=124016 18:23
hankache raiph thank you for the typo correction 18:25
psch RT #125278 exists, fwiw 18:26
synbot6 Link: rt.perl.org/rt3/Public/Bug/Display...?id=125278
psch no idea if "you need to proto that" would be the fix or if it should be more along the lines of "oh, it's cached, lets gen a cached proto" 18:28
lucasb m: proto f(|) is cached {}; multi f($n) { $n }; say f 20 18:29
camelia rakudo-moar : OUTPUT«Nil␤»
lucasb ^^ even a proto doesn't fix that
psch well, it fixes the LTA error :S 18:30
m: proto f(|) is cached {*}; multi f($n) { $n }; say f 20
camelia rakudo-moar : OUTPUT«20␤»
psch and an empty proto just does nothing
m: proto f(|) is cached {*}; multi f($n) { say "uncached yet"; $n }; say f 20; say f 20 18:31
camelia rakudo-moar : OUTPUT«uncached yet␤20␤20␤»
lucasb ^^ I made a mistake, forgot the {*} thing. nevermind my eval :) 18:32
psch i don't know about use-cases for is cached, and whether caching on a proto-level (i.e. for all multis) or not is something a user might want to decide themselves... 18:33
psch as in, i can imagine a "give me a fresh result here" use-case, but i'm not sure MMD is what this should be delegated to 18:34
although considering only the first non-cached proto multi invocation throws that error it probably should work on a candidate level
raiph hankache: yw 18:37
psch: now *that* looks like the right rt :)
psch: "first non-cached proto multi invocation throws that error" <-- please repeat that with camelia so i understand :) 18:38
psch m: multi a(Int $n) is cached { $n }; say try a 20; say a 20; # this one 18:39
camelia rakudo-moar : OUTPUT«Nil␤20␤»
psch m: multi a(Int $n) is cached { say "uncached yet"; $n }; say try a 20; say a 20; # this one
camelia rakudo-moar : OUTPUT«Nil␤uncached yet␤20␤»
psch m: multi a(Int $n) is cached { say "uncached yet"; $n }; say try a 20; say a 20; say a 20# this one
camelia rakudo-moar : OUTPUT«Nil␤uncached yet␤20␤20␤»
psch note it still caches afterwards
(alternative let me rephrase: "[...] considering only the first invocation of a multi that is cached without a proto that is cached [...]") 18:40
xfix perl6: say 9 ** 9 ** 90 18:44
camelia rakudo-jvm c016df: OUTPUT«-Inf␤»
..rakudo-moar: OUTPUT«0␤»
xfix 0?
xfix I see, Rakudo cannot calculate cannot Int ** Int where an exponent doesn't fit in a native integer. 18:46
pink_mist try a FatRat? 18:47
psch m: say 9 ** FatRat.new(9 ** 90, 1) 18:50
camelia rakudo-moar : OUTPUT«Inf␤»
psch m: say FatRat.new(9 ** 90, 1)
camelia rakudo-moar : OUTPUT«76177348045866392339289727720615561750424801402395196724001565744957137343033038019601␤»
psch m: say FatRat.new(9, 1) ** FatRat.new(9 ** 90, 1) 18:51
camelia rakudo-moar : OUTPUT«Inf␤»
skids m: grammar foo { rule TOP { . } }; foo.parse("foo",:subclass_arg_to_ignore) 18:52
camelia rakudo-moar : OUTPUT«Unexpected named parameter 'subclass_arg_to_ignore' passed␤ in block <unit> at /tmp/cUwyy7Hppz:1␤␤» 18:53
skids star: grammar foo { rule TOP { . } }; foo.parse("foo",:subclass_arg_to_ignore)
camelia ( no output )
skids That was breaking Grammar::BNF but I worked around it.
stmuk_ moritz: github.com/rakudo/star/pull/57 18:54
raiph psch: thanks 18:55
skids files as RT#126708 18:59
psch m: say f(*%opts) { say %opts.perl }; f :subclass_arg_to_ignore 19:00
camelia rakudo-moar : OUTPUT«5===SORRY!5=== Error while compiling /tmp/ZrVtxYeq93␤Unexpected block in infix position (missing statement control word before the expression?)␤at /tmp/ZrVtxYeq93:1␤------> 3say f(*%opts)7⏏5 { say %opts.perl }; f :subclass_arg_to_␤ expecting…»
psch m: sub f(*%opts) { say %opts.perl }; f :subclass_arg_to_ignore
camelia rakudo-moar : OUTPUT«{:subclass_arg_to_ignore}␤»
psch m: class A { method m(*%opts) { say %opts.perl } }; A.new.m :subclass_arg_to_ignore
camelia rakudo-moar : OUTPUT«{:subclass_arg_to_ignore}␤»
psch m: say Grammar.^can('parse')>>.candidates>>.signature
camelia rakudo-moar : OUTPUT«(((Grammar $: $target, :$rule = { ... }, :$args = { ... }, Mu :$actions = { ... }, *%opt)))␤»
skids Maybe it is actually the call to the .cursor method complaining. 19:01
timotimo if it's a call to a method, a named parameter wouldn't complain without some extra code to do the complaining 19:03
skids Do nqp methods follow that convention as well? 19:04
timotimo i believe so, yeah
if you really do mean methods and not nqp:: things 19:05
skids Well if it has to be caled as self."!cursor_init" who knows how it was cobbled together. 19:06
raiph m: my $*foo = 1; say :$*foo # searched rakudo src and clogs for "unexpected named parameter"; I think this one's coming from `say`
camelia rakudo-moar : OUTPUT«Unexpected named parameter 'foo' passed␤ in block <unit> at /tmp/t_aKAYrx6z:1␤␤» 19:07
psch nqp-m: class A { method foo() { } }; A.new.foo :bar 19:08
camelia nqp-moarvm: OUTPUT«Confused at line 2, near "A.new.foo "␤ at gen/moar/stage2/NQPHLL.nqp:521 (/home/camelia/rakudo-m-inst-1/share/nqp/lib/NQPHLL.moarvm:panic:105)␤ from gen/moar/stage2/NQP.nqp:921 (/home/camelia/rakudo-m-inst-1/share/nqp/lib/nqp.moarvm:comp_unit:872)␤ from …»
psch nqp-m: class A { method foo() { } }; A.new.foo(:bar)
camelia nqp-moarvm: OUTPUT«Unexpected named parameter 'bar' passed␤ at /tmp/K857Ys5RPJ:1 (<ephemeral file>:foo:3)␤ from /tmp/K857Ys5RPJ:1 (<ephemeral file>:<mainline>:37)␤ from gen/moar/stage2/NQPHLL.nqp:1303 (/home/camelia/rakudo-m-inst-1/share/nqp/lib/NQPHLL.moarvm:eval:190)␤…»
psch no interface consistency in nqp it seems 19:09
dalek ast: 7e1ef3e | usev6++ | S16-io/lines.t:
Fudge test for :nl-in and :chomp on JVM

Major duplication of code was necessary to be able to fudge failing tests. As elegant as nested 'for' loops are for testing different parameters, as hard it becomes to fudge specific tests.
Please revert if you find a better way to do the fudging.
mj41 .tell jnthn github.com/rakudo/star/blob/master...course.pdf slide 23 s/# (Int)/# (Str)/ 19:29
yoleaux mj41: I'll pass your message to jnthn.
timotimo hm, the rakudo star announcement didn't go out yet, right? 19:35
stmuk_ its in RC 19:38
stmuk_ I can't "panda install XML" with star rc0 I see "./prove: 1: ./prove: Cannot fork 19:47
The spawned process exited unsuccessfully (exit code: 2) in method throw at /home/steve/othersrc/star/rakudo-star-2015.11-rc0/install/share/perl6/runtime/CORE.setting.moarvm:1
I've seen this both on linux and os x 19:48
flussence came across this interesting article today about the PRNG in V8 being subtly, horribly broken: medium.com/@betable/tifu-by-using-...c308c4fd9d
gfldex m: for 1..5 { say ([+] 1.rand xx 100000) / 100000 } 19:50
camelia rakudo-moar : OUTPUT«0.499856562498002␤0.498229974484659␤0.498956127554163␤0.498720718972541␤0.499470305202736␤»
psch j: for 1..5 { say ([+] 1.rand xx 10000) / 10000 } # less precision, 'cause timeout 19:54
camelia rakudo-jvm c016df: OUTPUT«0.500889111480991␤0.49810811090140594␤0.4977089177080055␤0.49838340840390394␤0.5054720095566693␤»
psch yay noise :) 19:55
lucasb This is a friendly reminder that the advent calendar schedule is empty. Hope you people are thinking about your post ideas :)
El_Che In case no one linked here to the latest Larry Wall interview: twit.tv/shows/triangulation/episodes/225 19:56
metasin El_Che: thanks
grondilu there will be an advent calendar even this year? Isn't that a bit of a waste of time/energy before Xmas? 19:58
stmuk_ "panda install XML" works fine for me with current rakudobrew
lucasb grondilu: yep, I understand the developers will be busy with the release. Don't know what others think. Having advent calendars in all the past years and the release year doesn't have one would look strange. 20:00
ilmari it would be nice (and better for search engin findability) if one of docs?.perl6.org redirected to the other 20:01
jdv79 surely advent is better than no advent 20:03
masak +1 to advent calendar 20:05
as usual, I'm ready to chip in with posts. can't do it alone, though :)
ilmari m: say 2³ 20:07
camelia rakudo-moar : OUTPUT«5===SORRY!5=== Error while compiling /tmp/H2qRJC5PlK␤Bogus postfix␤at /tmp/H2qRJC5PlK:1␤------> 3say 27⏏5³␤ expecting any of:␤ infix␤ infix stopper␤ postfix␤ statement end␤ statement modifier␤ …»
ilmari m: say 2²
camelia rakudo-moar : OUTPUT«5===SORRY!5=== Error while compiling /tmp/phUHQ9bZgB␤Bogus postfix␤at /tmp/phUHQ9bZgB:1␤------> 3say 27⏏5²␤ expecting any of:␤ infix␤ infix stopper␤ postfix␤ statement end␤ statement modifier␤ …»
ilmari m: my \x = 2; say x³ 20:08
camelia rakudo-moar : OUTPUT«5===SORRY!5=== Error while compiling /tmp/ibSEnL61t7␤Bogus postfix␤at /tmp/ibSEnL61t7:1␤------> 3my \x = 2; say x7⏏5³␤ expecting any of:␤ infix␤ infix stopper␤ postfix␤ statement end␤ statement m…»
ilmari m: sub postfix:<²> ($x) { $x ** 2 }; say 3² 20:09
camelia rakudo-moar : OUTPUT«9␤»
ilmari docs.perl6.org/language/unicode_ent...subscripts implies they should work OOTB
masak ilmari: that's not how I read it. 20:10
ilmari oh, no, that's not actually code
masak it's talking about which characters might be useful when defining stuff like operators.
ilmari yeah, I realised 20:11
what I was actually looking for was the equivalent of perl5's \N{...} 20:14
dalek ar: 59c0536 | (Steve Mynott)++ | docs/announce/2015.11.md:
correct two minor typos
ar: 349fb98 | moritz++ | docs/announce/2015.11.md:
Merge pull request #57 from stmuk/master

correct two minor typos
masak m: say "\c[WHITE SMILING FACE]" # <-- ilmari 20:15
camelia rakudo-moar : OUTPUT«☺␤»
ilmari where is this documented? 20:16
masak ilmari: perhaps more importantly, where did you expect but fail to find it? :) 20:16
stmuk_ examples.perl6.org/categories/cookb...icode.html 20:18
lucasb Just noticed that these 2 cheatsheets have minor differences: mu/docs/Perl6/Cheatsheet/cheatsheet.txt and star/docs/cheatsheet.txt. One should be the official and the other just a copy. 20:20
stmuk_ I suspect the mu one is outdated 20:21
FROGGS o/ 20:22
yoleaux 15:16Z <nine> FROGGS: with my last rakudo patch, a panda installed into a Repository::Installation now works with the wrapper generated by rakudo and can install many modules :) The default repo list contains only Repository::Installation (perl6 -e 'say $*REPO.repo-chain')
FROGGS nine: and that's with the panda/curli branch as well? 20:24
ilmari masak: docs.perl6.org/language/quoting was the first place I looked 20:27
lucasb stmuk_: I saw you updated the star one. But it is a little misleading since the mu one says 'cheat sheet v4' and the star one says 'cheat sheet v2'
ilmari masak: since it's part of interpolating string syntax
I also looked under design.perl6.org/S15.html 20:28
and design.perl6.org/S02.html#Strings%2...e_Str_Type
masak I agree those are good places to look. 20:29
ilmari ah, design.perl6.org/S02.html#Unicode_codepoints
I didn't look far enough down
masak yes, S02 would be the first place I'd look :)
stmuk_ lucasb: maybe vimdiff the two? 20:30
or ediff or *
FROGGS nine: I'm unable to build the restricted setting: Too few positionals passed; expected 3 arguments but got 0
masak but I have the advantage of knowing roughly how the synopses are structured...
diakopter m: my int8 @a[-1;-1]; say @a[0;0] 20:34
camelia rakudo-moar : OUTPUT«Index 0 for dimension 2 out of range (must be 0..-2)␤ in block <unit> at /tmp/Hf5peDxaf0:1␤␤»
diakopter O_O
FROGGS hehe
diakopter masakbug: lookie
FROGGS diakopter: good catch :o)
diakopter m: my int8 @a[;]; 20:35
camelia rakudo-moar : OUTPUT«===SORRY!===␤Cannot find method 'returns': no method cache and no .^find_method␤»
diakopter m: my int8 @a[0;0]; say @a[;] 20:36
camelia rakudo-moar : OUTPUT«===SORRY!===␤Cannot find method 'flat': no method cache and no .^find_method␤»
diakopter masak: ^^ three buggoes
timotimo hehe. 20:38
dinner time \o/ 20:39
diakopter m: my int8 @a[1;1]; @a[0;0] = 256; say @a[0;0]
camelia rakudo-moar : OUTPUT«0␤»
diakopter m: my int8 @a[0;0]; @a[0;0] = 256; say @a[0;0] 20:40
camelia rakudo-moar : OUTPUT«Index 0 for dimension 2 out of range (must be 0..-1)␤ in block <unit> at /tmp/lRPSfpcJBs:1␤␤»
diakopter # imho it shouldn't allow zero as a dimension
FROGGS aye
Skarsnik m: int4 @a[1;1]; @a[0;0;] = 256; say @a[0,0];
camelia rakudo-moar : OUTPUT«5===SORRY!5=== Error while compiling /tmp/L7o7TkaOQX␤Two terms in a row␤at /tmp/L7o7TkaOQX:1␤------> 3int47⏏5 @a[1;1]; @a[0;0;] = 256; say @a[0,0];␤ expecting any of:␤ infix␤ infix stopper␤ statement end␤ …»
Skarsnik m: int4 @a[1;1]; @a[0;0] = 256; say @a[0,0];
camelia rakudo-moar : OUTPUT«5===SORRY!5=== Error while compiling /tmp/TYr6XRowHF␤Two terms in a row␤at /tmp/TYr6XRowHF:1␤------> 3int47⏏5 @a[1;1]; @a[0;0] = 256; say @a[0,0];␤ expecting any of:␤ infix␤ infix stopper␤ statement end␤ s…»
Skarsnik m: my int4 @a[1;1]; @a[0;0] = 256; say @a[0;0]; 20:41
camelia rakudo-moar : OUTPUT«MultiDimArray: Unhandled slot type␤ in block <unit> at /tmp/DslJ3nq3Jt:1␤␤»
diakopter it doesn't get below int8 yet
psch m: my int4 $x; 20:42
camelia rakudo-moar : OUTPUT«===SORRY!===␤Invalid int size for local/lexical␤»
Skarsnik m:use NativeCall; say nativesizeof(int8); say nativesizeof(int4);
m: use NativeCall; say nativesizeof(int8); say nativesizeof(int4);
camelia rakudo-moar : OUTPUT«1␤0␤»
FROGGS nine: I have no idea what's wrong: gist.github.com/FROGGS/8f38f2ca2887d82ff0bb 20:43
dalek kudo/nom: bc52faa | usev6++ | lib/Test.pm:
Report long name for exceptions in 'throws-like'
 20:44
kudo/nom: ae8216b | FROGGS++ | lib/Test.pm:
Merge pull request #597 from usev6/throws-like_longname

Report long name for exceptions in 'throws-like'
masak this might be of interest to the package/module people on this channel: blog.ezyang.com/2014/08/the-fundame...anagement/ 20:53
just discovered it now, though it's more than a year old, as you can see by the URL. 20:54
moritz if it's as fundamental as the URL makes us believe, a year of age won't change much :-) 20:55
masak ;) 20:57
Skarsnik it's quite empty in the end x) 21:01
diakopter dependencies vs requirements; until we have holistic proof-oriented programming [languages, tools, AND entire OSes], we'll be in this quagmire 21:03
dalek kudo-star-daily: ba06d10 | coke++ | log/ (10 files):
today (automated commit)
rl6-roast-data: 8a49548 | coke++ | / (9 files):
today (automated commit)
masak 'night, #perl6 21:13
nine FROGGS: that's...odd 21:21
FROGGS: it build NQP revision 8ffafe6?
FROGGS: have you nuked your install dir? 21:22
FROGGS no, have not nuked my install... that might be it 21:24
nine: nqp is latest
nine I haven't merged nom in a while. So the branch might no longer be compatible with current nqp.
FROGGS hmm, let's see
FROGGS okay, looks like I want an older nqp 21:28
nine I'll have a look at merging nom
nine It's just two simple conflicts. Building now. 21:31
FROGGS All positional args must appear before named args 21:32
maybe I'd need an older moar too :o)
(or I'll just wait)
dalek Heuristic branch merge: pushed 104 commits to rakudo/curli by niner 21:34
nine Built successfully :)
FROGGS cool :o)
FROGGS ./perl6-m --target=mbc -Ilib --output=lib/NativeCall/Compiler/GNU.pm.moarvm lib/NativeCall/Compiler/GNU.pm 21:39
D05D5F27E9A6C63503C21B4A647CCBA1E706405A
FROGGS nine: it built fine 21:39
\o/
nine yeah!
jdv79 what just happened? 21:43
FROGGS nine: ohh, panda is installing to home rather than site? 21:44
nine It's installing to the first repo that allows installing. 21:44
FROGGS k 21:45
then the search order is wrong I guess :o)
but it's nice that it works anyway :o)
nine That's one of the points that could use some discussion. Search order for loading modules should probably be HOME, site, vendor but for installation just site, HOME
moritz doesn't R* for example want to install into vendor? 21:46
FROGGS there probably should be a way to tell panda to a specific repo anyway
nine There already is
jdv79 what's wrong with installing to home in some cases? 21:47
nine As root I usually want to install system wide
panda install --prefix /home/nine/install/rakudo/install/share/perl6/site Foo will install into the site repo 21:49
jdv79 is that env var able? 21:51
or config or whatever 21:52
nine There's PANDA_DEFAULT_OPTS
FROGGS nine: one missing thing is precomp, right? 21:53
nine: or is there something else that I could look into in the upcoming days? 21:54
nine FROGGS: CompUnit::Repository::Installation precompiles on installation (line 194), Repository::FileSystem in need (line 57). 21:59
FROGGS ohh, cool
nine I'm not sure if we invalidate precomp files in all situations. The one thing that's sorely missing is tests :/ 22:00
FROGGS hmmm, maybe then I do that
jdv79 where is this stuff that needs tests? 22:01
nine jdv79: rakudo's curli branch
jdv79 ok
nine niner.name/curli_test.tar is the only synthetic test I've used. The rest was just building rakudo (which installs files), running the spectests (which do use lib quite a lot) and running panda 22:03
nine I also think that we should install the wrappers for scripts into /usr/local/bin and ~/bin directly instead of the repo. The wrappers are looking through repos and distributions to find the correct scripts anyway. And it'd be much more user friendly. 22:08
FROGGS nine: I'm going to bed now but tomorrow I'll add tests to roast 22:12
gnight 22:13
mj41 The Christmas is comming. Nice momentum in number of commits in October github.com/mj41/Perl6-Analytics-re...ll-commits 22:16
gfldex is there any way to force binding for a value returned by a sub, from inside the sub? 22:22
jdv79 wow, the git tree is a bit crazy unless i'm failing to follow this 22:24
moritz gfldex: this is a bit like asking "I want to sell those apples, but force everybody who buys them not to put them in a box" :-) 22:28
gfldex moritz: i believe to be aware of some of the consequences but still would like to have that answer answered. 22:29
bartolin .tell [Coke] the spectest for JVM should be nearly clean by now. the three failures in S05-mass/properties-general.t (596-598) seem to happen because we test unicode characters not supported by Java 7 (the tests pass with JDK 1.8) 22:52
yoleaux bartolin: I'll pass your message to [Coke].
dalek p: 0944ec5 | peschwa++ | src/vm/jvm/QAST/JASTNodes.nqp:
Make --target=jast work and somewhat comprehensible.
 22:59
psch ...and with that i can actually *see* where those args disappear (i hope :/ ) 23:00
psch ah, duh, of course not /o\ 23:05
'cause compilation breaks before reaching there
dalek ast: f1df147 | usev6++ | S32- (2 files):
Do not rely on order of @.unexpected in X::Adverb

Fixes RT #126671		 23:09
synbot6 Link: rt.perl.org/rt3/Public/Bug/Display...?id=126671
lizmat good *, #perl6! 23:49
lizmat wonders how you would do blogs.perl.org/users/ingy_dot_net/2...eness.html in Perl 6
lizmat and whether that shouldn't be a rosettacode entry :-) 23:58
psch m: my @a = <you got mail man mailman manners>; say "yougotmailmanners" ~~ m:ex/@a/ # am i misunderstanding :ex..?
camelia rakudo-moar : OUTPUT«(｢you｣ ｢got｣ ｢mailman｣ ｢manners｣)␤»
psch i thought :ex gets all possible matches..?
m: my @a = <you got mail man mailman manners>; say "yougotmailmanners" ~~ m:g:ov/@a/ # ...and i don't know if :ov and :ex should do the same here 23:59
camelia rakudo-moar : OUTPUT«(｢you｣ ｢got｣ ｢mailman｣ ｢manners｣)␤»

Powered by Cro and the Raku® Programming Language.

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

Thank you!