Perl 6 - the future is here, just unevenly distributed

IRC log for #perl6, 2012-04-18

Perl 6 | Reference Documentation | Rakudo

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

All times shown according to UTC.

Time Nick Message
00:16 lestrrat joined #perl6
00:20 replore joined #perl6
00:22 sorear good * #perl
00:22 sorear 6
00:23 TimToady happy tax day
00:29 colomon o/
00:32 sorear o/
01:04 szbalint joined #perl6
01:16 wolfman2000 joined #perl6
01:18 zhutingting joined #perl6
01:24 vlixes left #perl6
01:28 scott___ joined #perl6
01:30 [Coke]_ jnthn++ moritz++ doing all the hard work on the release announcements. ;)
01:33 tadzik joined #perl6
01:36 scott_ joined #perl6
01:39 liuchong joined #perl6
01:59 scott___ joined #perl6
02:03 [Coke] was interested to see http://www.warrenellis.com/?p=13931 re: markdown
02:06 scott___ joined #perl6
02:17 vlixes joined #perl6
02:23 samlh joined #perl6
02:28 scott___ joined #perl6
02:29 samlh So, I was wondering whether there is a way for a function to accept a LoL other than a **@foo argument. Actually, I'm not sure how LoL stuff is supposed to work in general.
02:30 * [Coke] sighs, as he has to revert some of his pugs patches from earlier this week.
02:31 [Coke] seen dalek?
02:31 aloha dalek was last seen in #perl6 12 hours 38 mins ago saying "roast: review: https://github.com/perl6/roast/commit/84b3895ce1".
02:37 scott_ joined #perl6
02:38 sorear Can't call method "quote" on an undefined value at modules/irc/seen.pm line 173.
02:38 sorear p6eval@feather3:~/dalek$
02:38 sorear freenode << :sjn!~nn@hagbart.nvg.ntnu.no PRIVMSG #perl6 :!seen pmichaud
02:38 sorear sjn--
02:39 dalek joined #perl6
02:40 uvtc joined #perl6
02:50 scott___ joined #perl6
02:52 cognominal joined #perl6
02:53 uvtc [Coke], Markdown is quite nice. Well, it's spoiled *me* anyway. :)
02:55 rking joined #perl6
02:59 orafu joined #perl6
02:59 scott___ joined #perl6
03:10 scott___ joined #perl6
03:20 scott___ joined #perl6
03:28 cognominal joined #perl6
03:29 scott___ joined #perl6
03:35 zhutingting what [ ] mean in p6 regex? like ( )?
03:35 zhutingting like ( ) but not catch in $\ ?
03:35 benabik I think [] is non-capturing ()
03:36 zhutingting yeah thx~
03:38 sorear [] is the same as (?:) used to be
03:46 cognominal joined #perl6
03:46 derrazo left #perl6
03:46 samlh joined #perl6
03:46 scott_ joined #perl6
03:51 bbkr_ joined #perl6
03:51 att_ joined #perl6
03:54 [Coke] sorear: I don't understand your negative karma there.
03:56 scott___ joined #perl6
03:57 sorear sjn crashed dalek
03:58 diakopter o_O
04:02 thou joined #perl6
04:27 samlh I've made some notes from reading list specs and tinkering with Rakudo. I'd appreciate any comments :). https://gist.github.com/2411012
04:33 TimToady reifying immutable lists can also cache reified values, they're just not writeable; in fact, we rely on this for constant lazy lists
04:34 TimToady esp recursively defined ones (though aspects of that remain unimplemented)
04:37 scott_ joined #perl6
04:46 samlh r: my @a = 1,2,3; @a[3]; say @a.DUMP; say @a.perl
04:46 p6eval rakudo 561682: OUTPUT«Array<712166686>(:items(RPA<705579559>(▶1, ▶2, ▶3)), :nextiter(▶Mu))␤Array.new(1, 2, 3)␤»
04:46 samlh r: say [1,2,3].DUMP; say [1,2,3].perl
04:46 p6eval rakudo 561682: OUTPUT«Array<1143850692>(:items(RPA<1143834356>(▶1, ▶2, ▶3)), :nextiter(▶Mu))␤[1, 2, 3]␤»
04:47 samlh How does it know which is which in .perl?
04:49 samlh Is there a flatten flag?
04:50 TimToady it's probably wrapped in a 'Scalar of' or some such
04:51 samlh TimToady: Ah, thank you.
04:51 TimToady r: my @a = 1,2,3; my $a = @a; say $a.perl
04:51 p6eval rakudo 561682: OUTPUT«[1, 2, 3]␤»
04:56 samlh r: my $a := 1,(2,3); for @$a {.say}; $a.tree.perl.say
04:56 p6eval rakudo 561682: OUTPUT«1␤2␤3␤(1, [2, 3]).list␤»
04:58 samlh TimToady: reifying immutable lists should stay hidden, that is the above should continue to work despite the reification behind the scenes?
05:03 replore_ joined #perl6
05:10 TimToady it would seem to be working right there
05:11 samlh TimToady: yes. So, the cache, if any, must remain invisible?
05:11 Timbus joined #perl6
05:12 TimToady the cache belongs to the @, not to the $a
05:12 TimToady $a is just an immutable parcel
05:13 samlh r: my @a := 1,(2,3); for @a {.say}; @a.tree.perl.say
05:13 p6eval rakudo 561682: OUTPUT«1␤2␤3␤(1, [2, 3]).list␤»
05:13 TimToady if the parcel caches, it shouldn't assume anything about flattening
05:13 samlh r: my @a = 1,(2,3); for @a {.say}; @a.tree.perl.say
05:13 p6eval rakudo 561682: OUTPUT«1␤2␤3␤(1, 2, 3).list␤»
05:14 TimToady list assignment flattens
05:14 samlh r: constant @a = 1,(2,3); for @a {.say}; @a.tree.perl.say
05:14 p6eval rakudo 561682: OUTPUT«1␤2␤3␤(1, [2, 3]).list␤»
05:15 TimToady pseudo-assignment for initialization tends to be more like binding
05:15 TimToady but arguably that should flatten anyway
05:16 samlh Ah, so that is how you are supposed to be able to get an immutable, but cached, list.
05:16 samlh r: constant @a = 1,(2,3); @a.tree.perl.say
05:16 p6eval rakudo 561682: OUTPUT«(1, [2, 3]).list␤»
05:16 samlh What about this? Should this flatten too?
05:16 TimToady I'm inclined to think so
05:17 TimToady especially since one could use := to mean the other thing
05:17 samlh Ok, makes sense. Thanks!
05:17 TimToady and flattening is more useful for a lot of sequence definitions
05:18 samlh r: constant @a = 1,(2..*); @a.tree.perl.say
05:18 p6eval rakudo 561682: OUTPUT«(timeout)»
05:18 samlh r: my @a := 1,(2..*); @a.tree.perl.say
05:18 p6eval rakudo 561682: OUTPUT«(timeout)»
05:19 TimToady I'm still hopeful to get the hamming sequence to work right:
05:19 TimToady constant @hamming = 1, dedup (@hamming X* 2) M (@hamming X* 3) M (@hamming X* 5);
05:20 samlh Ah well, I think I understand the goal a little better now, I'll try updating my notes now. :)
05:20 TimToady if that doesn't flatten, the dedup will return a single parcel that subverts the process
05:21 TimToady at the same time, the = must be understood as definition, so that @hamming refers to the same abstraction on both sides
05:21 TimToady no good replacing a pointer like := does
05:22 samlh So constant @a = does list assignment that guarantees laziness, but @a cannot be written to
05:22 TimToady yes, that's my story, and I'm sticking to it
05:23 samlh Gotcha
05:26 scott_ joined #perl6
05:28 mikemol joined #perl6
05:31 kaleem joined #perl6
05:45 mucker joined #perl6
05:46 uvtc left #perl6
05:53 NamelessTee joined #perl6
05:58 samlh r: constant @a = 1,@a; @a.DUMP.say
05:58 p6eval rakudo 561682: OUTPUT«===SORRY!===␤Variable @a is not declared␤at /tmp/tgU3GkFWsR:1␤»
05:58 samlh r: my @a = 1; @a.push(@a); @a.DUMP.say
05:58 p6eval rakudo 561682: OUTPUT«Array<-666740515>(:items(RPA<-662546955>(▶1, ▶1)), :nextiter(▶Mu))␤»
06:06 wtw joined #perl6
06:08 tarch joined #perl6
06:13 birdwindupbird joined #perl6
06:16 lestrrat joined #perl6
06:20 cognominal joined #perl6
06:29 bbkr joined #perl6
06:49 cognominal joined #perl6
06:50 kaleem joined #perl6
06:54 sjn oh my, I crashed dalek?
06:54 sjn sorry about that :-/
07:05 brrt joined #perl6
07:19 kaleem joined #perl6
07:26 dalek nqp: 1b1106c | moritz++ | tools/build/PARROT_REVISION:
07:26 dalek nqp: bump parrot revision to 4.3.0
07:26 dalek nqp: review: https://github.com/perl6/nqp/commit/1b1106c242
07:30 fglock_ joined #perl6
07:35 stepnem joined #perl6
07:36 araujo joined #perl6
07:36 araujo joined #perl6
07:44 kaleem joined #perl6
07:53 stepnem joined #perl6
08:02 masak o/, #perl6
08:04 daxim joined #perl6
08:04 stepnem joined #perl6
08:04 jferrero joined #perl6
08:05 moritz \o masak
08:13 stepnem joined #perl6
08:19 masak <jnthn> ooh, my French Perl Workshop talk got accepted. Now I've gotta go!
08:19 masak <eiro> sure: i did it to be sure you'll join us :)
08:19 * masak .oO( "we didn't care much for the talks themselves..." ) :P
08:20 bonsaikitten masak: conferences are all about beer and food
08:20 bonsaikitten the rest is just bonus features
08:24 masak yes, but eiro is one of the organizers. aren't they expected to keep up appearances? :)
08:27 bonsaikitten ... why? :)
08:34 masak I am unable to think up an answer to that question, even a funny one.
08:34 masak organizers: it's OK! you can drop all pretenses! we're just in it for the beer and food, and you're just in it for the speakers!
08:36 bonsaikitten hey, that's the greatest idea I've heard this week
08:36 bonsaikitten The Ultimate Geek Grill And Beer Festival (without any other distractions)
08:39 masak you sign up as a speaker, but you arrive with a thirst and barbecue cutlery.
08:40 daxim eat the audience!
08:41 daxim bodhidarma advises: when you meet the b33r on the conference, drink it.
08:42 * masak .oO( there's a whole audience at steak here! )
08:43 * daxim *groan*
08:44 zhutingting joined #perl6
08:46 zhutingting r: sub infix:<mean> ($a, $b) { ($a + $b) / 2 } $a mean $b;
08:46 p6eval rakudo 561682: OUTPUT«===SORRY!===␤Confused␤at /tmp/cwiOlFX3Gy:1␤»
08:46 zhutingting perl6: sub infix:<mean> ($a, $b) { ($a + $b) / 2 } $a mean $b;
08:46 p6eval niecza v16-21-g1b64073: OUTPUT«[31m===[0mSORRY![31m===[0m��Strange text after block (missing comma, semicolon, comment marker?) at /tmp/mH2H65Qkqv line 1:�------> [32m infix:<mean> ($a, $b) { ($a + $b) / 2 }[33m�[31m $a mean $b;[0m��Parse failed��»
08:46 p6eval ..pugs: OUTPUT«*** Undeclared variable: ("$a",MkPad (padToList [("$_",PELexical {pe_type = (mkType "Scalar"), pe_proto = <Scalar:0xf6df7289>, pe_flags = MkEntryFlags {ef_isContext = True}, pe_store = <ref:0xf6df8e1d>}),("@_",PELexical {pe_type = (mkType "Array"), pe_proto = <Array:0xf6df7…
08:46 p6eval ..rakudo 561682: OUTPUT«===SORRY!===␤Confused␤at /tmp/ofPWHl96AX:1␤»
08:47 zhutingting how to code infix?
08:47 dakkar joined #perl6
08:49 zhutingting perl6: sub infix:<mean> ($a, $b) { ($a + $b) / 2 } say 2 mean 4;
08:49 p6eval pugs: OUTPUT«3␤»
08:49 p6eval ..niecza v16-21-g1b64073: OUTPUT«[31m===[0mSORRY![31m===[0m��Strange text after block (missing comma, semicolon, comment marker?) at /tmp/GdHnDfpjv6 line 1:�------> [32m infix:<mean> ($a, $b) { ($a + $b) / 2 }[33m�[31m say 2 mean 4;[0m��Parse failed��»
08:49 p6eval ..rakudo 561682: OUTPUT«===SORRY!===␤Confused␤at /tmp/bt_irURYAh:1␤»
08:51 sjn http://code.foo.no/2012/04/18/coding-dojos
08:53 jnthn Hey, this geek beer and grill thing sounds like a good idea...
08:53 jnthn I mean, er, morning #perl6
08:53 jnthn :)
08:55 gfldex geek beer like in http://www.wired.com/wiredscience/2010/08/science-geek-beers/all/1 ?
08:57 jnthn ooh
08:57 bonsaikitten gfldex: that looks like hipster beer
08:58 bonsaikitten ice from the antarctic?
08:58 daxim I participated in a coding dojo.  that's programming by committee.  it was awful
08:58 * bonsaikitten likes a boring Rochefort 10 ... :)
08:58 jnthn Tssk, it's 11am and that page has made me want beer :P
08:59 * jnthn tries to forget about it and get on with $dayjob
08:59 gfldex so you are blaming the internet for wanting beer now?
08:59 bonsaikitten jnthn: and that's a problem why?
09:00 gfldex jnthn: you sound a little like a politican now
09:01 jnthn bonsaikitten: Hmm, good question... :)
09:02 jnthn heh..."Polygamy Porter (Why have just one?)"
09:02 * jnthn notes all of the ones on that page are from the US, and ponders whether Europe has similarly geeky beers
09:03 bonsaikitten jnthn: yes
09:03 bonsaikitten there's even Open Source Beer, where you get the full recipe ...
09:15 noam joined #perl6
09:27 fglock_ good morning #perl6  .oO( I'm up for the grill && beer )
09:29 sjn jnthn: just fyi, I've organized so you and masak and pmichaud can have a meeting room at RL to work in on Friday
09:29 sjn jnthn: it'll be available all day, from 09:00 to 17:00
09:29 sjn jnthn: damian will be giving a class next door too
09:31 d4l3k_ joined #perl6
09:34 jnthn sjn: Oh, wonderful!
09:34 jnthn sjn: Thanks for thinking to do that.
09:39 masak sjn++!
09:40 tadzik nice :)
09:40 phenny tadzik: 17 Apr 22:43Z <uvtc> tell tadzik Oh, I had the "name" field in the META.info file wrong --- it's supposed to be a module name, not the project name. Corrected.
09:40 masak tadzik! \o/
09:48 tadzik \o/
09:48 tadzik we're arriving with fsergot on some crazy hour friday morning
09:53 tadzik I'll have to visit my office, then I have to do some geocaching and I can hack from that time on :)
10:07 * masak decommutes
10:28 samlh gist updated with my understanding of iteration might work :). again, comments welcome https://gist.github.com/2411012
10:29 samlh *of how
10:29 xinming_ joined #perl6
10:37 NamelessTee joined #perl6
10:48 fgomez joined #perl6
10:49 birdwindupbird joined #perl6
10:50 fgomez joined #perl6
11:11 brrt left #perl6
11:22 pernatiy joined #perl6
11:26 masak samlh: interesting.
11:31 lichtkind joined #perl6
11:32 lichtkind moin
11:32 lichtkind moritz: can you tell me please how often the cron runs?
11:33 moritz lichtkind: once per hour
11:34 lichtkind moritz: very good
11:35 lichtkind moritz: after the next run please to the cleanup all files have their supposed final name just have to adapt some links
11:35 lichtkind but the github but doesnt seemed to work yesterday
11:36 moritz lichtkind: I have trouble deciphering what you write. Could you please be more careful?
11:36 moritz what is the "github but"?
11:37 lichtkind i meant bot sorry
11:37 lichtkind i do a commit in 2 min then we wills ee
11:37 lichtkind see
11:40 moritz 2 minutes passed, no commit reported
11:40 * masak .oO( The Github Butt is the ugly part of Github, but fortunately it's facing away from you while you're talking to Github )
11:42 lichtkind moritz: it was just 1 min :)
11:44 dalek tablets: ec38a21 | (Herbert Breunung)++ | docs/index.txt:
11:44 dalek tablets: updated toc(is index now) links to other tablets
11:44 dalek tablets: review: https://github.com/perl6/tablets/commit/ec38a21cfe
11:44 lichtkind oh
11:44 lichtkind what was going on yesterday
11:46 moritz dalek had an outage
11:48 masak better than a dalek having an outrage.
11:51 brrt joined #perl6
11:53 dalek tablets: 11b7eed | (Herbert Breunung)++ | docs/appendix-g-glossary.txt:
11:53 dalek tablets: added dummy for glossary
11:53 dalek tablets: review: https://github.com/perl6/tablets/commit/11b7eed208
11:53 dalek tablets: b599a0b | (Herbert Breunung)++ | docs/styles.css:
11:53 dalek tablets: changed placeholder to a more passive color
11:53 dalek tablets: review: https://github.com/perl6/tablets/commit/b599a0bb1c
11:53 lichtkind joined #perl6
11:58 dalek tablets: 7cb006a | (Herbert Breunung)++ | docs/introduction.txt:
11:58 dalek tablets: updated history a bit
11:58 dalek tablets: review: https://github.com/perl6/tablets/commit/7cb006a3d5
11:58 lichtkind masak: se were funny today?
12:03 facefox3 joined #perl6
12:04 masak yeah. sorry to butt in like that.
12:05 lichtkind i have not the impression you sorry is serious
12:06 masak if you are truly miffed and require an apology, you have it. sorry.
12:06 lichtkind just trying be funny myself
12:07 masak phew :)
12:08 masak when I get challenged like that, I try to be safe and apologise, so as not to make things worse. online communication is narrow-banded and brittle.
12:08 bluescreen10 joined #perl6
12:09 ponbiki joined #perl6
12:11 masak also, we all have bad days when we prefer not to be the butt of someone else's wordplay :)
12:12 frettled :)
12:15 lichtkind tables are so much easier in markdown
12:15 lichtkind its almost worth the effort
12:16 tadzik are they?
12:16 tadzik they're awesome in Pod :)
12:16 moritz tadzik: I think lichtkind means "in comparison to the socialtext abomination"
12:17 tadzik oh, maybe :)
12:17 mucker joined #perl6
12:17 dalek tablets: 0866ddf | (Herbert Breunung)++ | docs/ (12 files):
12:17 dalek tablets: simplify and update most headings
12:17 dalek tablets: review: https://github.com/perl6/tablets/commit/0866ddf5b8
12:17 lichtkind yes it was cruel
12:20 pernatiy joined #perl6
12:21 att joined #perl6
12:22 dalek tablets: f9966af | (Herbert Breunung)++ | docs/index.txt:
12:22 dalek tablets: refining TOC
12:22 dalek tablets: review: https://github.com/perl6/tablets/commit/f9966afe4a
12:27 dalek tablets: a414666 | (Herbert Breunung)++ | docs/index.txt:
12:27 dalek tablets: explaining appendix B better
12:27 dalek tablets: review: https://github.com/perl6/tablets/commit/a414666357
12:28 lichtkind allright slowly we getting somewhere sane
12:28 lichtkind later :)
12:28 masak o/
12:29 facefox joined #perl6
12:38 tokuhirom joined #perl6
12:39 dalek tablets: 9e8d9e1 | (Herbert Breunung)++ | docs/ (2 files):
12:39 dalek tablets: fixing broken markdown and placeholder are visible again
12:39 dalek tablets: review: https://github.com/perl6/tablets/commit/9e8d9e15b2
12:46 Woodi allo
12:47 Woodi sjn: ping, you have new mail...
12:50 [Coke] sorear: sjn++ dalek--
12:50 moritz indeed
12:51 jnthn Yeah. :!seen foo should not cause a crash.
12:52 moritz after all it's just a negated :seen :-)
12:53 kaleem joined #perl6
12:55 lestrrat joined #perl6
12:56 masak :)
12:56 masak std: :!seen<pmichaud>
12:56 p6eval std 3d13d53: OUTPUT«[31m===[0mSORRY![31m===[0m�Extra argument not allowed; pair already has False argument at /tmp/CyIrWkjSc0 line 1:�------> [32m:!seen[33m�[31m<pmichaud>[0m�Parse failed�FAILED 00:00 40m�»
12:56 masak dang :)
12:56 masak good error message, though.
12:56 moritz r: :!seen<pmichaud>
12:56 p6eval rakudo 561682: OUTPUT«===SORRY!===␤Argument not allowed on negated pair␤at /tmp/hM_B9k9mzb:1␤»
12:57 moritz r: try eval ':!seen<pmichaud>'; say $!.perl
12:57 p6eval rakudo 561682: OUTPUT«X::Syntax::NegatedPair.new(filename => "eval_0", line => 1, column => Any)␤»
12:57 * moritz can't resist showing off a bit
12:57 answer_42 joined #perl6
12:58 tadzik cool
12:59 brrt joined #perl6
12:59 PacoAir joined #perl6
13:00 masak moritz++
13:00 pmurias joined #perl6
13:00 pmurias fglock: hi
13:01 pmurias $ node perlito5.js  -e 1
13:01 pmurias Warning: Died: Can't find Perlito5/strict.pm in @INC
13:03 sjn Woodi: ook!
13:03 fglock_ pmurias: hi!
13:04 fglock_ strict.pm is now a "real" module, you need to specify the path to the libs (see README-perlito5)
13:05 pmurias fglock_: thanks, that fixes the problem
13:05 fglock_ node perlito5.js -I./src5/lib -e 1
13:05 pmurias fglock_: but why is strict used here? ;)
13:06 fglock_ this will compile "package main; no strict; 1"
13:17 awwaiid joined #perl6
13:20 cognominal_ joined #perl6
13:25 fhelmberger joined #perl6
13:29 cognominal joined #perl6
13:35 preflex_ joined #perl6
13:41 kaleem joined #perl6
13:41 fglock_ joined #perl6
13:42 att joined #perl6
13:46 pmurias fglock_: perl -Ilib5 perlito5.pl -Cjs src5/util/perlito5.pl > perlito5.js
13:46 pmurias Syntax Error in sub 'precedence_parse' at lib5/Perlito5/Grammar.pm line 1302.
13:46 envi_ joined #perl6
13:47 fglock_ looking (you need -I src5/lib before -Cjs)
13:48 fglock_ adding -Isrc5/lib silences the error (still looking)
13:51 fglock_ it may be an error while trying to 'use' a lib, it is failing to find an alternate interpretation and gives a "syntax error" :P
13:53 fglock_ hmm, no - 'use' errors seem to work
13:56 pmurias fglock_: what are the $negate and $depends arguments to proclaim?
13:59 fglock_ I don't know; this file is a direct translation of http://svn.openfoundry.org/pugs/ext/Test/lib/Test.pm
14:00 pmurias I'm considering just copying over the p5 test.pl and adapting it for the missing features
14:01 fglock_ yes, that would be better
14:01 fglock_ it can than be fixed in small steps
14:02 fglock_ I'm getting 426 pass with the tweaked test.pl
14:04 spider-mario joined #perl6
14:05 pmurias fglock_: are you working on copying over the orginal test.pl or should i do that?
14:05 fglock_ please do
14:06 fglock_ pmurias++
14:07 pmurias ok
14:08 pmurias fglock_: none of the tests we ran currently use test.pl?
14:09 fglock_ you mean in t5, no - I'm running the perl test suite in t
14:09 fglock_ and moving to t5 whatever works
14:12 fglock_ (t5 does use Test.pm though)
14:31 colomon joined #perl6
14:31 fgomez joined #perl6
14:47 lichtkind joined #perl6
14:48 benabik joined #perl6
14:50 fsergot joined #perl6
14:50 fsergot Hi o/
14:52 wooden joined #perl6
14:52 colomon \o
14:53 [Coke] phenny: tell au if you do add something for bless or anythign else, be sure to do a spectest first; I found some (to me) surprising) breakage, especially missing multi variants.
14:53 phenny [Coke]: I'll pass that on when au is around.
14:59 kaleem joined #perl6
15:06 kaare_ joined #perl6
15:10 trexy joined #perl6
15:11 NamelessTee joined #perl6
15:12 pmurias fglock_: node perlito5.js -I./src5/lib  -e 1
15:12 pmurias started throwing the can't find strict error again :/
15:14 fglock_ pmurias: maybe you are under t/ (check just in case)
15:15 fglock_ it happened to me
15:17 lichtkind joined #perl6
15:17 pmurias git stashing everything helped
15:18 sjohnson git stash save 'my project' # i use all the time!
15:18 lichtkind moritz: would you please do now a reset?
15:19 moritz lichtkind: it's done on every cron job executiong
15:19 dalek tablets: b07546b | (Herbert Breunung)++ | docs/ (3 files):
15:19 dalek tablets: massive link repair
15:19 dalek tablets: review: https://github.com/perl6/tablets/commit/b07546b828
15:20 lichtkind moritz: you mean deleting all the old stuff?
15:20 moritz yes
15:22 dalek tablets: b97370e | (Herbert Breunung)++ | docs/ (2 files):
15:22 dalek tablets: repaired links to glossary
15:22 dalek tablets: review: https://github.com/perl6/tablets/commit/b97370e3a3
15:29 dalek tablets: e848519 | (Herbert Breunung)++ | docs/index.txt:
15:29 dalek tablets: beautify TOC
15:29 dalek tablets: review: https://github.com/perl6/tablets/commit/e848519df3
15:29 lichtkind dont worry my tiuts will run out soon :)
15:30 lichtkind but current state is a mess
15:30 lichtkind translators are helpless
15:33 lichtkind moritz: did you just changed that in the cron?
15:36 moritz yesterday night
15:36 moritz https://gist.github.com/2409568
15:37 apejens joined #perl6
15:37 ponbiki joined #perl6
15:39 lichtkind moritz: but the old toc is still anschnur
15:40 lichtkind http://tablets.perl6.org/toc.html
15:40 lichtkind should be gone
15:41 moritz wtf is "anschnur"?
15:43 * daxim gesichtspalme
15:45 lichtkind daxim: became an meme for german lang purists :)
15:46 daxim no, it's from ✠chan
15:46 moritz if you equate purity with ignorance, yes
15:47 thou joined #perl6
15:47 lichtkind daxim: the source is defintivly not 4chan but some (fake?) site of a german religous fanatic
15:49 daxim I didn't say 4chan, and that fanatic was fake
15:50 lichtkind i like it anyway :)
15:50 ilogger2 joined #perl6
15:50 lichtkind moritz there was just an refresh and old html is still there
15:51 jnthn .oO( I'm anschnur what he meant... )
15:52 moritz lichtkind: now gone
15:52 lichtkind thank you
15:53 lichtkind jnthn: ++++ that was great
15:54 Psyche^ joined #perl6
16:01 kaleem joined #perl6
16:08 uvtc joined #perl6
16:09 trexy left #perl6
16:10 uvtc tadzik, ping
16:10 tadzik uvtc: pong
16:10 xinming joined #perl6
16:10 uvtc tadzik, Hi, I sent you a pile of emails last night.
16:11 uvtc Do you think that little guide for creating and distributing modules is still sensible after the minor additions I mentioned?
16:12 moritz uvtc: do you have commit access to modules.perl6.org?
16:12 uvtc hi moritz. Y'know, I used to. But no longer.
16:12 moritz uvtc: what's your github ID?
16:12 uvtc uvtc
16:13 uvtc But moritz,
16:13 uvtc Just wanted to briefly discuss something.
16:13 uvtc The topic of the Socialtext Perl 6 wiki has come up here before.
16:13 moritz nothing wrong with discussions, but now you have commit access too :-)
16:13 uvtc Oh, great. Thanks.
16:14 uvtc I've used the SocialText wiki a couple of times, but did not ever get back to it
16:14 tadzik uvtc: yeah, I saw them :) I'm just little fscked over with univesity stuff right now, expect my feedback tomorrow afternoon in the worst case
16:14 uvtc .
16:14 uvtc tadzik, Great, thanks.
16:14 * moritz hates that wiki
16:14 lichtkind haha
16:15 * uvtc is not crazy about it either.
16:15 lichtkind uvtc: klöng
16:15 uvtc Anyhow, I tried out gitit last night, and was pretty blown away.
16:15 lichtkind uvtc: by what?
16:15 lichtkind uvtc: http://tablets.perl6.org/index.html have a look
16:16 lichtkind it gets slowly in shape
16:16 lichtkind ther was still so much broken
16:16 lichtkind but i currently but ass of to fix it
16:16 uvtc lichtkind, (I see. Yes, it's looking good. Just a sec though, I want to finish this thouht.)
16:17 moritz uvtc: I wouldn't mind a migration, under the conditions that 1) I don't have to do it (I'm fine with installing software though), 2) we don't end up with two versions of the same stuff (ie the old wiki must go away) and 3) the URLs continue to work in some way (preferably redirects)
16:17 uvtc So, I just wanted to chime in (an initial chime I suppose), that gitit was pretty amazing. I started making a list of what I liked about it, but the list became probably too long to post here.
16:18 moritz under those conditions I would be very much in favor of such a migration, actually :-)
16:18 moritz uvtc: how does gitit handle access control?
16:19 uvtc moritz, I'm not necessarily suggesting a migration... I don't have an easy way to convert content from one wiki to the other. I converted lichtkind's tablets primarily by-hand, but that was over the weekend, and I don't think I'll have that many tuits available for a while.
16:19 moritz ok
16:21 uvtc moritz, access control is simple --- I like it. For the account creation page, you can set a security question (and answer). If you set the answer to be something unguessable, then all you have to do is have interested parties email you for the unguessable answer. And you can of course change the unguessable answer at any time. :)
16:22 moritz that sounds like it's fine, as long as you don't need to scale
16:23 uvtc I've tried using various wiki's in the past. Gitit is the one. It's easy to run, easy to configure, stores each page as a file on disk ... (gah, I need to write a blog post about it).
16:23 thelazydeveloper joined #perl6
16:24 uvtc That's an interesting supposition. I'm curious how many folks end up contributing to wiki's. My hunch is that, at first, you'd get a flurry of a 10 or 20 people asking to register, but,
16:25 moritz my experience is that there are (nearly) no useful edits from outsiders
16:25 uvtc not sure, but I get the impression that, you end up over time with a core group who contributes to the wiki, and a slow trickle of membership requests after that.
16:26 uvtc moritz, That's what I'm getting it. That's why I think gitit's approach will work.
16:26 moritz uvtc: I don't think it's a problem of the underlying software or of the UI
16:27 uvtc Anyway, the setup looks easy. You just run gitit, and it has its own webserver (based on Haskell's happstack). It comes with instructions on what to add to your apache config to get mod_proxy to pass requests to it.
16:27 moritz uvtc: it's simply that outsiders usually don't have anything interesting to share, or don't care. If they care, they quickly cease to be outsiders
16:28 samlh lichtkind: I have some updates to the tablets
16:29 samlh lichtkind: would you prefer a patch, a pull request, or something else?
16:29 moritz samlh: what's your github ID?
16:29 uvtc moritz, I think that's a good point about outsider ceasing to be outsiders.
16:29 samlh samlh
16:30 moritz samlh: you now have commit access, fwiw
16:30 mikemol joined #perl6
16:30 samlh moritz: ok, thanks!
16:30 uvtc Well, the long and the short of it about the wiki issue, I think, is this:
16:30 moritz to most of the repos in the perl6 organization, because we're too lazy for finer-graded access control :-)
16:32 uvtc * few here like or will use the socialtext wiki, * most here like Markdown pretty well, * gitit is only an apt-get away, and perhaps a few minutes of configuring to make look like a Perl 6 wiki (just replace the logo, change the wiki title in the config, and it's ready to roll),
16:32 uvtc It supports "subpages" nicely by just creating a directory behind the scenes and putting the page file in that directory.
16:32 uvtc Allows file uploads, supports math rendering
16:33 uvtc (moritz, math rendering worked out of the box for me)
16:33 uvtc (moritz, that is, you can write $x^2 + y^2$ and it works)
16:34 uvtc Syntax highlighting for Perl 5 works, but not Perl 6 --- it uses Kate's (the editor) filetype definitions for syntax highlighting. I don't think Perl 6 syntax highlighting would be too far off if someone were to have a look at it.
16:35 uvtc Anyway, sorry, I'll stop gushing. I'm just pretty enamored with it, given the other wiki's I've tried.
16:35 * moritz chuckles briefly (re p6 syntax hilighting)
16:35 moritz yes, sounds like somebody has found True Love
16:36 * uvtc is singing it from the rooftops
16:37 Vlavv` joined #perl6
16:37 uvtc Getting around to a point though, regarding that "create & distribute modules" doc I wrote, if you set up the wiki, I could just add it there.
16:38 moritz or you could simply add the .html file to the modules.perl6.org repo
16:38 moritz then it would be on the same domain
16:38 uvtc Regarding the domain, gitit comes with instructions on how to make wiki.perl6.org proxy to gitit...
16:39 uvtc Is this all on feather? Does feather run apache?
16:39 fglock joined #perl6
16:39 moritz thing is, I'm kinda reluctant to fragment the p6 doc space even more. If there are plans to move other stuff to the new wiki, that's fine. If there's a plan for one page, that's not too convincing
16:39 moritz I should have said "on the same subdomain"
16:39 moritz it's on feather2. And yes, apache2
16:41 moritz though there's nothing to stop us from mod_proxy'ing to an installation on feather1
16:41 uvtc I can understand your reluctance. My hunch is that the opposite of fragmentation will happen. I think people will *like* using gitit a lot, and if there's a choice between, "oh, should I host this myself, or just put it on the Perl 6 gitit wiki", they'll go with the wiki. I know I would.
16:42 uvtc Sorry, I'm unfamiliar with the infrastructure. I don't know how many Perl 6 servers there are, or what runs on which.
16:42 moritz brief summary, because it might interest others too: Juerd++ has donated a box (feather0), on which three VMs (feather1,2,3) run
16:43 moritz what we usually call "feather" around here is feather1. Any p6 hacker can get a shell account on it
16:43 moritz and it runs perlcabal.org and a few other websites
16:44 moritz feather2 is intended to be more secure, hosts *.perl6.org and hugme (because hugme has access tokens to several github accounts)
16:44 moritz feather3 runs p6eval and try.rakudo.org (when it happens not to be down)
16:45 moritz perl.org and perlfoundation.org are run on totally different machines by totally different people
16:45 uvtc (this is exactly something I'd take 2 minutes to put up somewhere on the gitit wiki. Also, I have no interest in editing the Socialtext wiki.)
16:46 uvtc moritz, thanks for that info!
16:48 benabik moritz++ # feather infodump
16:49 vlixes joined #perl6
16:50 uvtc moritz, here's some notes I was keeping, in case I were to write a blog post about gitit. They're just some sloppy notes: http://www.unexpected-vortices.com/doc-notes/benefits-of-gitit.html
16:51 birdwindupbird joined #perl6
16:51 PacoAir joined #perl6
16:51 kaleem joined #perl6
16:52 lichtkind samlh: what you prefer
16:53 lichtkind samlh: but the changes are very rapid currently i cpromise you it will land under a wheel
16:53 uvtc masak, What do you think?
16:54 uvtc masak? masak? Oh no. They've got masak.
16:54 moritz http://perlpunks.de/paste/show/4f8ef1c2.cc6.37c that's what I get
16:54 fgomez joined #perl6
16:55 moritz (installed gitit via apt-get)
16:55 uvtc Is that directory writable? It needs to create files and dirs in the dir where you run it.
16:56 moritz yes, it is
16:56 lichtkind samlh: still around
16:56 uvtc Hm. Let me look on my system for that filestore file...
16:56 moritz and it creates gitit.log and wikidate
16:56 moritz oh
16:56 uvtc wikidata
16:56 moritz I just realized that this is a bad idea
16:56 uvtc oh?
16:56 moritz if you want to access the git repo, you do it via ssh
16:57 moritz so if we run it on feather2, we'd have to hand out shell access to lots of folks
16:57 benabik And we don't want everyone to have ssh access to feather2?
16:57 moritz correct
16:57 benabik You could mirror the git repo out, although that introduces issues syncronizing.
16:57 uvtc I've got a /usr/share/filestore-0.4.0.3 on my system...
16:58 uvtc and extra/post-update is in there.
16:58 moritz uvtc: which package does it come from?
16:58 samlh alright, pull request sent
16:58 uvtc Uhm... (can't recall the command for that... dpkg something something)
16:59 moritz benabik: it would be perfect if it worked with a github repo in some way, because it would be nice not to have another authentication scheme
16:59 moritz uvtc: dpkg -S $path
16:59 uvtc libghc-filestore-data
17:00 benabik moritz: You could have a cron job that does the moral equivalent of `git pull && git push` to a github repo.  But merging and such could become an issue.
17:00 samlh lichtkind: pull request sent https://github.com/perl6/tablets/pull/1
17:03 uvtc samlh, if moritz added you to the team, I think that means you can push directly to perl6/tablets. Just fetch and merge, make sure there's no conflicts, and push.
17:03 lichtkind samlh: sorry to reject that but maybe we have next week a wiki again
17:03 kaleem joined #perl6
17:03 moritz lichtkind: don't
17:04 lichtkind moritz: what?
17:04 moritz lichtkind: change technology so frequently
17:04 moritz as I understood, you were happy with how it's now
17:04 moritz so don't reject contributions just because you might migrate to something else next week
17:05 uvtc lichtkind, I'd assumed you'd prefer editing with your editor (not in the browser, or over ssh). Your pages are rather large and complicated, I don't think you'd want to edit them in a browser.
17:05 lichtkind moritz: we will stay with markdown but maybe use giti as an input means but its unclear now
17:05 moritz lichtkind: yes, but if you continue to use markdown, there's no reason to reject contributions now
17:05 lichtkind in the end yes
17:06 lichtkind yes
17:08 dalek tablets: a3bf94e | samlh++ | docs/ (7 files):
17:08 dalek tablets: Small updates to grammar and wording
17:08 dalek tablets: review: https://github.com/perl6/tablets/commit/a3bf94e8a4
17:08 dalek tablets: e123979 | samlh++ | docs/ (4 files):
17:08 dalek tablets: Updates to meaning.
17:08 dalek tablets: review: https://github.com/perl6/tablets/commit/e1239795ab
17:08 dalek tablets: b6dcc0e | (Herbert Breunung)++ | docs/ (7 files):
17:08 dalek tablets: Merge pull request #1 from samlh/pr1
17:08 dalek tablets:
17:08 dalek tablets: we change format; samlh++
17:08 dalek tablets: review: https://github.com/perl6/tablets/commit/b6dcc0ee57
17:08 samlh lichtkind: thanks! :)
17:09 lichtkind samlh: thank you im just a bit fearful because i currently preparing a huuge patch
17:10 samlh lichtkind: ah, understood
17:12 dalek tablets: 9caeb25 | (Herbert Breunung)++ | docs/ (4 files):
17:12 dalek tablets: still repair markdown conversion mistakes
17:12 dalek tablets: review: https://github.com/perl6/tablets/commit/9caeb25d55
17:12 dalek tablets: c7119eb | (Herbert Breunung)++ | docs/ (7 files):
17:12 dalek tablets: Merge branch 'master' of github.com:perl6/tablets
17:12 dalek tablets: review: https://github.com/perl6/tablets/commit/c7119ebd03
17:12 havenn joined #perl6
17:16 arnsholt joined #perl6
17:19 ilogger2 joined #perl6
17:21 Chillance joined #perl6
17:23 arnsholt o/
17:28 uvtc moritz, sorry, missed some of your comments before. You wrote "I just realized that this is a bad idea ...". I think it would be very rare for users to want direct access to files on the filesyst (ssh in, edit, use git directly).
17:28 uvtc s/filesyst/filesystem/
17:28 uvtc moritz, It's the sort of thing that's there if you need it, but you generally won't ever need it for typical wiki editing.
17:30 thou_ joined #perl6
17:43 mucker joined #perl6
17:44 Moukeddar joined #perl6
17:48 bbkr1 joined #perl6
17:49 fsergot joined #perl6
17:49 fsergot o/
17:55 thou joined #perl6
18:02 moritz uvtc: http://wiki.perl6.org/
18:03 uvtc Oh yeah. :D
18:03 uvtc moritz, a couple of config items:
18:04 uvtc run `gitit --print-default-config > p6-wiki.conf`
18:04 benabik Looks like gitit is blatantly stealing^W^Wborrowing wikipedia's look.
18:05 eiro gitit \o/
18:05 fsergot \o/
18:05 eiro hello everybody
18:05 uvtc moritz, then edit a couple of items: 1, change wiki-title to "Perl 6 Wiki" or somesuch
18:08 uvtc 2. change access-question to something like, "Answer to this security question?", and add a crazy string to access-question-answers.
18:08 uvtc Then quit gitit and restart it as `gitit -f p6-wiki.conf`
18:14 spider-mario joined #perl6
18:14 moritz for now I haven't used a terribly crazy question or answer
18:14 jaffa4 joined #perl6
18:17 spaceships joined #perl6
18:24 uvtc moritz, Next: grab this file http://www.unexpected-vortices.com/temp/logo.png and copy it to static/img
18:24 uvtc I resized it. Should work nicely.
18:24 moritz done
18:24 jaffa4 Hi
18:24 uvtc Nice!
18:24 samlh moritz: s/use use/you use/ ?
18:24 uvtc moritz, I think that, if you don't add an unguessable answer, then it's just going to generate work for you later. My 2 cents.
18:24 uvtc Account created!
18:24 benabik Unguessable isn't required.  Not parsable by spambots is.
18:24 benabik (IMHO)
18:24 benabik Actual people generally don't spend much time adding spam.  It generally only gets bad when a script can do it.
18:24 moritz any idea for a question and answer that's guessable only if you know Perl 6, but still has a pretty unique answer?
18:24 * [Coke] FIRST!
18:24 uvtc Oh, you beat me to it!
18:25 jaffa4 moritz: who are you asking?
18:26 moritz jaffa4: everybody
18:26 jaffa4 Why are you asking?
18:27 benabik TIMTOWTDI makes "unique" a difficult part of that question.  :-D
18:27 [Coke] jaffa4: check the backscroll.
18:27 moritz because I need such a question for protecting a wiki from spam
18:27 moritz benabik: that's why i haven't found anything yet
18:28 moritz maybe "which operator would you use to calculate the product of all items of a list?"
18:28 jaffa4 What is the name of most popular perl compiler?
18:28 benabik It could be a small piece of code with the output as answer.  If the spambots are written in P6 then we're winning anyway.  :-D
18:28 benabik jaffa4: perl
18:28 jaffa4 What is the name of most popular perl 6 compiler?
18:28 jnthn "When will Perl 6 be done?" "Christmas" :)
18:29 [Coke] jnthn++
18:29 jaffa4 You could say, it is already one, we are just extending it
18:29 facefox3 joined #perl6
18:29 jaffa4 one-> done
18:32 facefox3 joined #perl6
18:32 tadzik o/
18:32 uvtc moritz, I don't know much Perl 6 yet, but perhaps a "fill-in-the-blank", such as, "I never ___ I didn't like".
18:34 [Coke] metapunned?
18:34 tadzik :P
18:35 * moritz has something now
18:35 uvtc (metaop)
18:39 masak #perl6! \o/
18:40 havenn_ joined #perl6
18:41 moritz \o masak
18:41 birdwindupbird joined #perl6
18:42 tadzik \o/
18:42 masak jaffa4: the name of the most popular Perl 6 compiler is "Rakudo". but having said that, I think Niecza deserves more popularity than it currently has, and I have thought so for almost a year now.
18:44 icwiener joined #perl6
18:46 uvtc Hi tadzik. I added the create and distribute modules page to [the new wiki](http://wiki.perl6.org).
18:46 tadzik oh, we have a new wiki?
18:46 uvtc :)
18:46 tadzik so cool
18:46 uvtc Ice cold.
18:47 tadzik are you going to Oslo, by any chance?
18:47 uvtc me?
18:48 tadzik aye
18:48 uvtc Added http://wiki.perl6.org/Server%20Infrastructure
18:48 goksie joined #perl6
18:48 uvtc Alas, no. I'm not employed at the moment, and so am not in a position to travel. Would love love to take my family there sometime though. Sounds lovely.
18:50 tadzik I hope you'll be joining us online at least :)
18:50 thou joined #perl6
18:51 uvtc tadzik, aw, thanks. :) Though, I don't yet know enough Perl 6 to be of much help at a hackathon.
18:51 tadzik that
18:51 tadzik 'd be a good occasion :)
18:51 tadzik anyway, if you are eager to hack on modules stuff, we need every pair of hands :)
18:51 masak Perl 6 knowledge not a strict requirement! :D
18:51 tadzik I'd really like to bring emmentaller back to life
18:51 tadzik we _do_ need it, especially those days
18:51 tadzik and it was mostly perl 5 (back in the days when it worked)
18:52 jaffa4 How to iterate through a perl6 hash?
18:52 dalek tablets: b0b04c8 | (Herbert Breunung)++ | docs/appendix-b-grouped.txt:
18:52 dalek tablets: repair mardown tables
18:52 dalek tablets: review: https://github.com/perl6/tablets/commit/b0b04c8b7c
18:53 lichtkind jaffa4: you meant through all pairs or keys?
18:53 jaffa4 all pars.
18:53 lichtkind for (@(%h)) { }
18:54 lichtkind for pairs %h {}
18:54 lichtkind i think thats the best
18:54 lichtkind and if you want a named var insted of $_
18:54 lichtkind for pairs %h -> $p  {}
18:54 benabik r: .say for pairs { a => 1, b => 2 }
18:54 p6eval rakudo 561682: OUTPUT«"a" => 1␤"b" => 2␤»
18:54 spider-mario what about kv ?
18:55 spider-mario r: .say for kv {a => 1, b => 2}
18:55 p6eval rakudo 561682: OUTPUT«a␤1␤b␤2␤»
18:55 spider-mario yeah, right
18:55 spider-mario r: .say @_ for kv {a => 1, b => 2}
18:55 p6eval rakudo 561682: OUTPUT«===SORRY!===␤Confused␤at /tmp/rA1WBC_fKj:1␤»
18:55 spider-mario oops
18:55 lichtkind it gives you a key value list
18:55 spider-mario r: say @_ for kv {a => 1, b => 2}
18:55 p6eval rakudo 561682: OUTPUT«===SORRY!===␤Cannot use placeholder parameter @_ in the mainline␤at /tmp/HaWjKdRjht:1␤»
18:55 spider-mario r: say list $_ for kv {a => 1, b => 2}
18:55 p6eval rakudo 561682: OUTPUT«a␤1␤b␤2␤»
18:55 jaffa4 r: for %*ENV.kv -> $a,$b { say $a,$b}
18:56 p6eval rakudo 561682: OUTPUT«PERLBREW_VERSION0.42␤MAIL/var/mail/p6eval␤USERp6eval␤SHLVL1␤HOME/home/p6eval␤OLDPWD/home/p6eval␤PERLBREW_ROOT/home/p6eval/perl5/perlbrew␤PERLBREW_MANPATH/home/p6eval/perl5/perlbrew/perls/perl-5.14.1/man␤LOGNAMEp6eval␤_./run.sh␤TERMscreen␤PATH/home/p6eval/perl5/perl…
18:56 jnthn A nice way is
18:56 jnthn yeah, that one :)
18:56 samlh left #perl6
18:56 spider-mario that’s what I was about to write
18:56 TimToady r: -> $k, $v { say "$k $v" } for %*ENV.kv
18:56 p6eval rakudo 561682: OUTPUT«PERLBREW_VERSION 0.42␤MAIL /var/mail/p6eval␤USER p6eval␤SHLVL 1␤HOME /home/p6eval␤OLDPWD /home/p6eval␤PERLBREW_ROOT /home/p6eval/perl5/perlbrew␤PERLBREW_MANPATH /home/p6eval/perl5/perlbrew/perls/perl-5.14.1/man␤LOGNAME p6eval␤_ ./run.sh␤TERM screen␤PATH /home/p6eva…
18:57 jaffa4 Rotated version?
18:57 TimToady r: { say "$^k $^v" } for %*ENV.kv
18:57 p6eval rakudo 561682: OUTPUT«PERLBREW_VERSION 0.42␤MAIL /var/mail/p6eval␤USER p6eval␤SHLVL 1␤HOME /home/p6eval␤OLDPWD /home/p6eval␤PERLBREW_ROOT /home/p6eval/perl5/perlbrew␤PERLBREW_MANPATH /home/p6eval/perl5/perlbrew/perls/perl-5.14.1/man␤LOGNAME p6eval␤_ ./run.sh␤TERM screen␤PATH /home/p6eva…
18:59 dalek modules.perl6.org: 920735f | tadzik++ | web/index.tmpl:
18:59 dalek modules.perl6.org: Link to a more appropriate module writing guide
18:59 dalek modules.perl6.org: review: https://github.com/perl6/modules.perl6.org/commit/920735f6af
18:59 tadzik uvtc++
19:00 uvtc And Bingo was his name-o.
19:00 uvtc tadzik, thanks.
19:00 uvtc p6eval, nice to see you back.
19:01 uvtc moritz++ for making the wiki happen!
19:03 Alano joined #perl6
19:07 kaare__ joined #perl6
19:08 jaffa4 How to define an infix operator within a grammar if I could not find a decent example....
19:10 moritz token myinfix { '+' | '-' }; token term { <termish> [ <infix> <termish> ]* }
19:10 moritz s:2nd/infix/myinfix/
19:11 jaffa4 I thought there is a build in operator parser...
19:11 jaffa4 and one can specify precedence levels..
19:14 mdupont joined #perl6
19:15 masak yeah, that bothers me too.
19:15 masak there's too little reach into the built-in optable parser.
19:15 jaffa4 Not implemented, or not in the spec?
19:15 masak a little as if it were only to be used by Perl 6 itself.
19:19 s1n joined #perl6
19:28 dalek tablets: 9a22b66 | (Herbert Breunung)++ | docs/appendix- (2 files):
19:28 dalek tablets: fixed all special var
19:28 dalek tablets: review: https://github.com/perl6/tablets/commit/9a22b66c38
19:32 noam_ joined #perl6
19:34 kaare_ joined #perl6
19:42 masak jaffa4: it's not in the spec.
19:43 masak jaffa4: the only way to create operators is to put them into the language grammar currently being parsed. I know of no mechanism to put them into a user-defined grammar for later use.
19:43 masak jaffa4: clearly there should *be* a way, but... there isn't.
19:53 raiph joined #perl6
19:56 havenn joined #perl6
19:59 TimToady the P6 dependencies in STD's EXPR method are commented with "# XXX P6", if anyone cares to try figuring out how to make a generic version
19:59 pmurias joined #perl6
20:04 noam_ joined #perl6
20:06 masak sounds like a nice project for a rainy day.
20:07 sjohnson masak.project
20:08 pmurias fglock_: what do you intent to use Perlito5 for? or is mainly -Ofun?
20:08 moritz masak.schedule: Project.new('generalize STDs precedence parser')
20:08 icwiener szabgab: Did you see that your media links are broken? http://szabgab.com/download/media/perl6_01.avi
20:09 moritz .oO( 'schedule' -- word that can be both verb and noun. Not good. )
20:09 szabgab icwiener: I thought I removed those links  :(
20:09 icwiener szabgab: http://szabgab.com/perl6.html#screencast
20:09 icwiener There they are.
20:10 szabgab icwiener: thanks for telling me, I'll remove the links for now
20:10 szabgab till I find the files :)
20:10 icwiener You lost the files? :D
20:12 cognominal_ joined #perl6
20:12 brrt joined #perl6
20:13 szabgab icwiener: probably not, I jus t moved server and have not copid the files so I think they are somewhere on my desktop computer :)
20:13 fsergot joined #perl6
20:14 arnsholt star: use NativeCall; my $arr = CArray[int].new(); $arr[0] = 1; say $arr[0];
20:14 p6eval star 2012.02: OUTPUT«(signal SEGV)»
20:14 noam_ joined #perl6
20:14 jnthn arnsholt: Wow...
20:15 jnthn arnsholt: I thougth CArray had passing tests...
20:15 arnsholt Yeah. Didn't really expect that
20:16 arnsholt There are some tests, but none with int, now that I look
20:16 arnsholt star: use NativeCall; my $arr = CArray[num].new(); $arr[0] = 1.2; say $arr[0];
20:16 p6eval star 2012.02: OUTPUT«Cannot unbox argument to '$v' as a native num␤  in method <anon> at /home/p6eval/star/lib/parrot/4.1.0/languages/perl6/lib/NativeCall.pm6:159␤  in <anon> at src/gen/Metamodel.pm:2891␤  in block <anon> at /tmp/1V9eTk6lWB:1␤␤»
20:16 arnsholt Oh, right
20:16 arnsholt star: use NativeCall; my $arr = CArray[num].new(); $arr[0] = 1.2e0; say $arr[0];
20:17 p6eval star 2012.02: OUTPUT«(signal SEGV)»
20:17 arnsholt None of the tests try just creating an array and reading back from it, so that might be it
20:17 icwiener szabgab: Ah, ok. :)
20:17 s1n joined #perl6
20:18 jnthn arnsholt: Hm, they only pass it?
20:18 jnthn Aren't there tests for returning an array too? :)
20:18 jnthn Very curious, anyway
20:18 jnthn Do you know why it SEGVs?
20:18 arnsholt Null pointer
20:19 arnsholt This is a smaller version of the problem in my CStruct test, I'm reasonably sure
20:19 arnsholt But yeah, there are tests for returning an array, and for passing an array
20:20 bruges joined #perl6
20:20 arnsholt But no test simply creating and then reading from the array
20:20 jnthn heh, but not reading on you made yourself...
20:20 jnthn *one
20:21 * jnthn blames the guy who wrote the tests.
20:21 arnsholt ^_^
20:22 arnsholt But I'm not entirely sure what could be going on. I suppose it could be some kind of spooky action-at-a-distance making the get return garbage, but not sure why the NULL pointer shows up in P6int, not CArray
20:28 arnsholt Hmm. One possible explanation: bug is related to CArray reading from an array CArray created
20:29 jnthn I suspect it is that
20:29 arnsholt Return array test passes, so reading of at least some arrays does indeed work
20:29 jnthn There's the two modes, one where the array is managing its own memory and another where it isn't.
20:29 arnsholt Similarly, passing array test passes, so layout appears to be correct as well
20:29 jnthn I'm wondering if it's a confusion between the two.
20:31 arnsholt I have a CArray[int], and stepping in the getter
20:31 arnsholt elem_size is 8
20:31 jnthn That's fine on 64 bit
20:31 arnsholt Right. Not that then
20:31 arnsholt Kind is 1
20:31 jnthn But it's what the computed offset gets added to taht must be null
20:34 jnthn oh, wait, is it in at_pos_boxed when it explodes, or at_pos_ref?
20:34 arnsholt The explosion is in P6int.c
20:34 jnthn yeah, I see it now
20:35 jnthn Or, partially
20:35 arnsholt The get is at_pos_ref
20:35 jnthn oh
20:35 jnthn That's the correct one then
20:35 jnthn So is body->storage NULL?
20:36 arnsholt Nope
20:37 jnthn star: use NativeCall; my $arr = CArray[int].new(); $arr[0] = 1; $arr[1] = 2; say $arr[0];
20:37 p6eval star 2012.02: OUTPUT«1␤»
20:37 jnthn OK, two problems
20:37 jnthn if (body->allocated && index >= body->elems)
20:37 jnthn return NULL;
20:37 jnthn That's an off-by-one
20:37 jnthn Should be >
20:37 jnthn star: use NativeCall; my $arr = CArray[int].new(); say $arr[0];
20:37 p6eval star 2012.02: OUTPUT«(signal SEGV)»
20:38 jnthn Second, at_pos_ref can return NULL if there's no element at that reference.
20:38 jnthn I'm guessing that this case is not being handled.
20:38 arnsholt Yeah
20:39 jnthn It doesn't crash in the tests because we can't bounds-check a return value.
20:39 arnsholt Yup
20:39 jnthn (Because C arrays don't communicate their size.)
20:40 arnsholt But >= should be right, no?
20:40 arnsholt It's elems that's wrong I think
20:41 arnsholt In my session with a one-element array, elems is 0
20:42 jnthn oh
20:42 jnthn yeah, you're right.
20:42 * masak .oO( "...and off-by-one errors" )
20:43 arnsholt Oh, I see it
20:44 arnsholt if (index >= body->elems) body->elems = index;
20:44 arnsholt There's the off-by-one
20:44 arnsholt It should be = index+1
20:47 jnthn aha!
20:47 jnthn good find :)
20:47 arnsholt Thanks =)
20:47 jnthn arnsholt++ # oneing in on the bug
20:48 arnsholt I'll leave the accessor-returns-NULL issue to you =)
20:49 arnsholt I'll add a test and fix in master and pull that back into my branch
20:49 arnsholt With a bit of luck my structs will work then
20:53 arlinius joined #perl6
20:54 alester joined #perl6
20:56 uvtc moritz, new wiki down?
20:56 dalek joined #perl6
20:57 uvtc moritz, oh, it's back.
20:58 lichtkind raiph: hai
20:58 p6eval joined #perl6
20:59 raiph lichtkind: hi
20:59 dalek perl6.org: 97cecea | (John Gabriele)++ | source/community/index.html:
20:59 dalek perl6.org: Added link to new Perl 6 wiki
20:59 dalek perl6.org: review: https://github.com/perl6/perl6.org/commit/97cecea272
20:59 lichtkind very much changed :)
21:00 dalek nqp: 769b277 | (Arne Skjærholt)++ | src/6model/reprs/CArray.c:
21:00 dalek nqp: Fix off-by-one bug in CArray.c.
21:00 dalek nqp: review: https://github.com/perl6/nqp/commit/769b2773e8
21:00 ggoebel joined #perl6
21:00 lichtkind who is john gabriele?
21:00 Juerd joined #perl6
21:00 uvtc hi
21:00 uvtc that's me
21:01 tadzik joined #perl6
21:02 [Coke] joined #perl6
21:02 Util joined #perl6
21:02 hugme joined #perl6
21:02 PerlJam joined #perl6
21:02 arnsholt There. Fix pushed
21:02 jnthn yay
21:02 masak joined #perl6
21:02 tadzik heh. Everyone loses connection and gets back to the channel. arnsholt: "There. Fix pushed"
21:03 tadzik I feel updated
21:03 Chillance joined #perl6
21:04 arnsholt ^_^
21:04 arnsholt And even better, my CStruct with CArray test case now works properly
21:04 * arnsholt celebrates a bit
21:05 Juerd Just a bit? For arrays and structs I'd expect at least a few bytes.
21:05 arnsholt =D
21:05 tadzik heh, where do I have this..
21:05 jnthn Maybe grab a nibble to eat...
21:05 tadzik http://goo.gl/VBMbr
21:05 jnthn Anyway, word up!
21:05 tadzik aye, here
21:06 dalek perl6.org: 9ac8a61 | (John Gabriele)++ | source/documentation/index.html:
21:06 dalek perl6.org: Updated link to Perl 6 Tablets
21:06 dalek perl6.org: review: https://github.com/perl6/perl6.org/commit/9ac8a6145c
21:06 jnthn tadzik: :P
21:07 tadzik rseen tadzik?
21:07 tadzik heh, rseen would be nice. Then I could write "tadzik. Providing not-really-related images from the internets since 200x"
21:07 [Coke] please see the rseen(1) and rseen.conf(5) man pages for full documentation.
21:08 tadzik so there is hope :)
21:09 dalek nqp/cstruct-work: 3a23406 | jnthn++ | src/6model/reprs/CStruct.c:
21:09 dalek nqp/cstruct-work: Don't try to allocate 0 bytes in CStruct REPR.
21:09 dalek nqp/cstruct-work: review: https://github.com/perl6/nqp/commit/3a23406005
21:09 dalek nqp/cstruct-work: 1b1106c | moritz++ | tools/build/PARROT_REVISION:
21:09 dalek nqp/cstruct-work: bump parrot revision to 4.3.0
21:09 dalek nqp/cstruct-work: review: https://github.com/perl6/nqp/commit/1b1106c242
21:09 dalek nqp/cstruct-work: 769b277 | (Arne Skjærholt)++ | src/6model/reprs/CArray.c:
21:09 dalek nqp/cstruct-work: Fix off-by-one bug in CArray.c.
21:09 dalek nqp/cstruct-work: review: https://github.com/perl6/nqp/commit/769b2773e8
21:09 dalek nqp/cstruct-work: b85e904 | (Arne Skjærholt)++ | / (3 files):
21:09 dalek nqp/cstruct-work: Merge branch 'master' into cstruct-work
21:09 dalek nqp/cstruct-work: review: https://github.com/perl6/nqp/commit/b85e9046e2
21:09 dalek nqp/cstruct-work: 682e311 | (Arne Skjærholt)++ | src/6model/reprs/CStruct.c:
21:09 dalek nqp/cstruct-work: Implement setting of CArray members in CStructs.
21:09 dalek nqp/cstruct-work:
21:09 dalek nqp/cstruct-work: Confirmed working in Perl 6-land, C-land working or breakage to be determined.
21:09 dalek nqp/cstruct-work: review: https://github.com/perl6/nqp/commit/682e311bdc
21:10 arnsholt Proper testcases in Zavolaj can wait 'till tomorrow
21:10 tadzik writing something that generates Perl 6 modules from .h files could be a nice thing for the hackathon
21:11 havenn joined #perl6
21:11 arnsholt Quite. We should discuss that on friday =)
21:11 tadzik certainly :)
21:11 arnsholt See you folks in cyberspace tomorrow, and some even in meatspace on friday =D
21:11 tadzik o/
21:11 * arnsholt &
21:12 * jnthn looks forward to it
21:12 jnthn o/, arnsholt
21:17 uvtc Who handles updates to rakudo.org?
21:18 libertyprime joined #perl6
21:18 jnthn Any of the folks who have a login to it and time.
21:18 jnthn I know myself, pmichaud and moritz have access.
21:19 uvtc I was going to suggest that I think it should have a separate heading for Rakudo Star.
21:19 nif joined #perl6
21:20 jnthn ah, not a bad idea.
21:20 uvtc Would be useful to link new users directly to a R* page there.
21:20 uvtc Currently, the "How to get Rakudo Perl 6" is mostly about building the compiler release.
21:21 uvtc There's also still a "How to get Rakudo nom" heading there.
21:21 uvtc Gah. My time for now just ran out. Gotta go. o/
21:21 raiph \o
21:24 Chillance joined #perl6
21:24 libertyp1ime joined #perl6
21:25 brrt joined #perl6
21:37 dalek tablets: 23d1466 | (Herbert Breunung)++ | docs/introduction.txt:
21:37 dalek tablets: update content of into to recent order
21:37 dalek tablets: review: https://github.com/perl6/tablets/commit/23d1466be4
21:37 dalek tablets: 4b4f95c | (Herbert Breunung)++ | docs/ (3 files):
21:37 dalek tablets: moved it terms to glossary
21:37 dalek tablets: review: https://github.com/perl6/tablets/commit/4b4f95cbcf
21:39 lichtkind raiph: ping?
21:41 lichtkind moritz: is it to every full hour?
21:42 thou joined #perl6
21:46 fridim_ joined #perl6
21:56 tokuhirom joined #perl6
22:02 fsergot good night o/
22:03 rking left #perl6
22:03 tokuhirom joined #perl6
22:03 brrt left #perl6
22:10 libertyprime joined #perl6
22:23 mdupont joined #perl6
22:24 diakopter phenny: tell masak Whether you believe you can autopun or not, you are right.
22:24 phenny diakopter: I'll pass that on when masak is around.
22:25 localhost joined #perl6
22:27 raiph phenny, ask phenny (and lichtkind) is this pong an autopun?
22:27 phenny Hey, I'm not as stupid as Monty you know!
22:28 lichtkind raiph: no just wanta know if your still around
22:28 masak diakopter: belief doesn't come into it! I think you're veering too close to self-reference right now ;)
22:28 phenny masak: 22:24Z <diakopter> tell masak Whether you believe you can autopun or not, you are right.
22:28 masak diakopter: the key is use/mention collapsing.
22:29 masak like "Whr dd ll m vwls g?"
22:29 masak (though that particular one is quite tired)
22:29 diakopter Whether you believe this is an autopun or not, you are right.
22:30 TimToady speaking of which, have you seen http://imgur.com/r/funny/4r8uO ?
22:30 masak yeah. typical self-reference.
22:31 diakopter I skimmed an explanation somewhere; I don't remember it.
22:31 jnthn Ouch!
22:31 masak that one is nice because you can actually start down the problem, do a few iterations on it, and then go "oh, I see what you did there..."
22:31 jnthn ...it was instantly funny then I started thinking about it :)
22:32 masak but srsly. if you do a Monte Carlo on that one, you're bound to get a probability out, aren't you?
22:32 masak and since it's a computer program, you have to make assumptions, and it's those assumptions that determine that answer.
22:33 diakopter I don't see how all puns about autopunning aren't autopuns
22:33 masak the one assumption that matters here seems to be "to what degree are (A) and (D) 'the same'?"
22:33 TimToady .oO(This car no autopun.)
22:33 masak TimToady: :P
22:34 masak diakopter: well, the trick isn't to get "autopun" in there somewhere. it's to collapse a mention into a use and catching the listener slightly unawares.
22:34 masak "We can repair anything. (Please knock hard on the door, the doorbell doesn't work.)"
22:35 TimToady well, but that one depends on assumptions too
22:35 TimToady "can" == "always do"
22:35 diakopter Teach a man to autopun, you'll feed him for a lifetime.
22:36 * diakopter is hungry
22:36 masak TimToady: well, they don't have to make perfect sense. just tickle the imagination enough so that it connects the dots.
22:36 masak TimToady: and not repairing the doorbell in this case has a certain sense of "heal thyself" justice to it that I think makes people smile.
22:37 masak diakopter: aww :) I love to see you try, though.
22:37 diakopter AUTOFAIL
22:39 TimToady I think any autopunnyness of the multiple choice question hinges on use/mention of the concept of "correct"
22:40 masak oh, you consider the multi-choice question an autopun? I just see it as an Epimeaniness.
22:41 diakopter still I struggle with the definition
22:41 masak yeah. me too.
22:41 diakopter use/mention collapsing
22:41 masak ok, a clear example.
22:41 TimToady self-reference always involves some degree of use/mention, but I can see where it doesn't tickle you enough to call it a pun
22:42 TimToady it certainly tickles me :)
22:42 TimToady and part of the is the garden-pathology of it, when you see the second 25%
22:43 uvtc joined #perl6
22:44 masak "...you think 'I know, I'll use multithreading! Nothhw tpe yawrve ..."
22:44 masak er
22:44 masak "...you think 'I know, I'll use multithreading!' Nothhw tpe yawrve ..."
22:44 uvtc haha
22:44 masak see? :)
22:44 masak diakopter: the 'multithreading' is the mention.
22:44 masak diakopter: the garbled characters are the use.
22:45 masak diakopter: the fact that the garbled characters *enter the reality of the joke* is the collapse.
22:45 jnthn "I reviewed your code and found two errors. It is an off-by one."
22:46 huf :))))))
22:46 TimToady btw, off-by-one is just a naming problem
22:46 diakopter "masak reviewed your autopun and didn't get it. He laughed."
22:47 diakopter s/your/this/
22:47 diakopter argh.
22:47 diakopter still I mix too much self-reference
22:48 diakopter also, the fact it's not funny makes it use the mention.
22:51 whiteknight joined #perl6
22:53 diakopter There are only 10 kinds of people - those who understand autopuns and those who don't.
22:54 masak :D
22:55 masak just by using fixed snowclone-ish joke templates like that, you can approximate funny pretty well :)
22:56 diakopter "To autopun or not to autopun"
22:57 masak the "...and 9 other kinds!" joke is funny because it (a) deviates from the pattern at the last moment when the listener has already committed down the garden path, and (b) in retrospect puts the speaker squarely in the category of "doesn't understand binary", closing the autopun loop.
22:59 diakopter Self-reference alone or even mentioning autopuns as well does not make an autopun.
22:59 sjohnson diakopter: hi
23:00 diakopter sjohnson: hello from the punfailer
23:00 sjohnson punfailer eh
23:00 sjohnson i haven't told any bad puns today tho!
23:01 diakopter I meant myself
23:02 sjohnson hmm, i wonder if i can whip up a purposely corny pun
23:02 * masak .oO( just put enough maize in it )
23:02 diakopter just assume everything you read intends to self-refer
23:02 * masak .oO( you're in a corny maize of twisty little autopuns, all alike )
23:03 sjohnson a maze of maize! HAHAHAHAHA
23:03 sjohnson bbl
23:03 cognominal joined #perl6
23:03 * masak .oO( he got a sudden urge to make popcorn )
23:04 silug joined #perl6
23:05 diakopter as soon as masak teaches me to autopun, I'll make one.
23:06 diakopter YAFAIL
23:06 diakopter pretty soon here TimToady will correctly label me as perseverating
23:09 javascriptnewbie joined #perl6
23:09 masak javascriptnewbie: oh hai!
23:10 javascriptnewbie hi!
23:11 flussence joined #perl6
23:11 masak welcome to our humble channel.
23:11 TimToady diakopter: I keep doing that.
23:11 diakopter :)
23:12 masak there! see!?
23:12 masak TimToady is like, built out of autopuns.
23:12 javascriptnewbie i'm shinobicl... i'm just asking stuff on the #javascript channel :)  .... with no luck :(
23:12 masak oh.
23:13 masak carry on, then.
23:13 javascriptnewbie i'm just here to ask... i finished a module (WorkdayCalendar) and i want to "publish" it
23:13 * flussence was slightly confused there until checking the ir clogs
23:13 javascriptnewbie what should i do? is there a tool for that?
23:14 uvtc javascriptnewbie, http://wiki.perl6.org/Create%20and%20Distribute%20Modules
23:14 javascriptnewbie thanks :D uvtc
23:15 uvtc javascriptnewbie, sure, y/w
23:16 javascriptnewbie is that a wiki written in haskell?
23:17 uvtc Yes.
23:17 javascriptnewbie oh... nice... well got to go now... i'll upload my module this week :) good night all!
23:18 uvtc g'night
23:19 diakopter I don't see why the use/mention of autopuns can't be an autopun. :| too directly self-referential?
23:20 diakopter no pun intended.
23:20 masak ;)
23:20 diakopter I think.
23:20 masak I guess it could be an autopun, it's more like... it isn't automatically one.
23:20 masak it doesn't become an autopun because it's a self-reference, or because you throw in the word "autopun" in there somewhere.
23:22 daniel-s joined #perl6
23:22 mspaulding joined #perl6
23:23 awwaiid joined #perl6
23:26 sorear good * #perl6
23:26 diakopter howdy
23:27 lichtkind uvtc: i have trouble with the anchors
23:27 uvtc What's the trouble?
23:28 sftp joined #perl6
23:30 uvtc sorry, have to go
23:30 uvtc left #perl6
23:30 sorear diakopter: heya
23:32 lichtkind good bye
23:32 lichtkind they dont work :)
23:35 jnthn_ joined #perl6
23:39 libertyprime joined #perl6
23:47 sftp joined #perl6
23:50 libertyprime joined #perl6
23:51 alvis joined #perl6
23:51 pochi joined #perl6
23:51 yeltzooo7 joined #perl6
23:52 flussence joined #perl6
23:57 thou joined #perl6
23:59 awwaiid joined #perl6

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

Perl 6 | Reference Documentation | Rakudo