Perl 6 - the future is here, just unevenly distributed

IRC log for #perl6, 2012-09-20

Perl 6 | Reference Documentation | Rakudo

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

All times shown according to UTC.

Time Nick Message
00:01 whiteknight joined #perl6
00:17 dayangkun joined #perl6
00:26 jerome joined #perl6
00:26 Circlepuller joined #perl6
00:27 MissingSemicolon joined #perl6
00:36 Circlepuller_ joined #perl6
00:57 benabik joined #perl6
01:05 anuby joined #perl6
01:17 hypolin joined #perl6
01:19 colomon joined #perl6
01:24 jerome joined #perl6
01:31 gootle joined #perl6
01:34 baest joined #perl6
01:45 FROGGS_ joined #perl6
01:48 flightrecorder joined #perl6
02:02 MikeFair_ joined #perl6
02:03 gootle1 joined #perl6
02:11 fgomez joined #perl6
02:42 fgomez joined #perl6
02:42 sftp joined #perl6
02:48 wooden joined #perl6
02:48 Vlavv` joined #perl6
02:48 pmichaud joined #perl6
02:48 [Coke] joined #perl6
02:48 phenny joined #perl6
02:58 orafu joined #perl6
03:02 gootle joined #perl6
03:12 thou joined #perl6
03:40 fgomez joined #perl6
04:15 telex joined #perl6
04:24 satyavvd joined #perl6
04:34 jaldhar joined #perl6
04:36 telex joined #perl6
04:59 am0c joined #perl6
05:06 kaleem joined #perl6
05:11 Targen joined #perl6
05:12 eiro joined #perl6
05:30 wooden joined #perl6
05:30 wooden joined #perl6
05:30 moritz good morning
05:32 sorear good morning moritz
05:37 mucker joined #perl6
05:39 sirrobert joined #perl6
05:51 wooden joined #perl6
05:51 wooden joined #perl6
05:53 gootle1 joined #perl6
05:53 gootle1 joined #perl6
05:58 yarp joined #perl6
06:17 TimToady it seems kinda silly that I'm using a big is_prime routine written in Perl 6 when libtommath implements Miller-Rabin in C..
06:18 TimToady I'm thinking is_prime (or is-prime, or whatever) is something that should be provided as a built-in
06:20 TimToady likewise there's an expmod routine in there that seems to not be visible from Perl
06:24 wooden joined #perl6
06:24 wooden joined #perl6
06:24 eiro hello all
06:25 eiro is there a way to introspect a perl6 grammar to build its AST ? (or something equivalent)
06:27 MikeFair_ eiro: I believe a Grammar has the capabilities you're looking for, but I'm not quite sure exactly what it is you're loking for
06:28 MikeFair_ (nor am I sure I even know how to code it once I do understand ;) )
06:28 gootle joined #perl6
06:29 eiro ok. ingy wrote an awesome piece of perl5 named pegex: it's a PEG parser with multilang emiter so you can write one PEG for python, perl5, javascript, ...
06:30 eiro i wonder if perl6 is not a better tool for that (because of its native PEG support)
06:30 gootle1 joined #perl6
06:31 eiro hmm.. just realized that i can ask on the list: this is a much better place
06:31 wooden joined #perl6
06:31 wooden joined #perl6
06:33 wk joined #perl6
06:33 diakopter TimToady: sounds good to me
06:35 gootle joined #perl6
06:35 yarp_ joined #perl6
06:35 gootle1 joined #perl6
06:36 dalek specs: 5277fef | larry++ | S32-setting-library/Numeric.pod:
06:36 dalek specs: Add expmod and is-prime as built-ins in Int
06:36 dalek specs: review: https://github.com/perl6/specs/commit/5277fefed6
06:37 gootle joined #perl6
06:38 diakopter std: 3.'is-prime?'()
06:38 p6eval std 77327a4: OUTPUT«ok 00:00 41m␤»
06:38 diakopter std: 3.'is-prime?'
06:38 p6eval std 77327a4: OUTPUT«[31m===[0mSORRY![31m===[0m�Unsupported use of . to concatenate strings or to call a quoted method; in Perl 6 please use ~ to concatenate, or if you meant to call a quoted method, please supply the required parentheses at /tmp/z_7V5RQAN6 line 1:�------> [32m3.'is…
06:42 gootle1 joined #perl6
06:42 gootle2 joined #perl6
06:43 MikeFair_ eiro: Sorry, I'm not the right person for that question, but Perl6 Grammars are extraordinary and have a combination of hierarchical nested match rules and automated calling of methods as the matches are found.
06:44 MikeFair_ Parrot uses P6 Grammars specifically for writing interpretters to all kinds of languages.
06:49 eiro MikeFair_, i'm aware of that! that's why i ask :)
06:54 masak morning, #perl6
06:55 sorear morning, masak
06:55 masak eiro: I think PEG has been described as "the closest thing to what Perl 6 has"; i.e. it's not identical, but it's a good first approximation if you want to understand it.
06:59 eiro masak, you mean PEGex ? isn't PEG a general concept ?
07:05 erkan joined #perl6
07:05 erkan joined #perl6
07:10 fhelmberger joined #perl6
07:13 brrt joined #perl6
07:18 cibs joined #perl6
07:18 tadzik good morning #perl6
07:18 moritz \o
07:20 gootle joined #perl6
07:20 flightrecorder joined #perl6
07:22 sorear eiro: https://en.wikipedia.org/wiki/Parsing_expression_grammar
07:24 Woodi morning everyone
07:24 eiro sorear, yep. that's why i don't understand the answer masak gave.
07:25 eiro don't worry: i'll figure out
07:28 Woodi I was thinking about MOP and security a little... MOP is kind of like assembly language, can mess with (near?) everything inside... so imagine app with GUI written in assembler - implementing into it "sandbox" functionality would be for volontiers :)
07:29 diakopter eiro: masak was saying PEG helps to understand P6, not the other way around
07:29 Woodi so maybe maybe security should be written close to GUI/input ?
07:32 Woodi module with "checks" for input, using regex/grammar or even MOP-y based added behaviours for get/lines/recv ? module used with "use", kind of -T from v5
07:32 eiro ohh sure! but that's not the point! i want to write a lib that parses a text and i want it at least in perl5, perl6, python. as i will use a PEG and as perl6 is a target, i thought pegex wasn't optimal for my problem.
07:32 sorear Woodi: implementing sandbox is near impossible for exactly those reasons.  better to implement it at a lower level, restrict what the application can do
07:32 eiro my dream was: write in perl6, use in perl5 and python
07:32 eiro (in perl6 as well)
07:32 Woodi sorear: that's my point
07:33 Su-Shee Woodi: that would be a horrible design. security/grants/permissions - all these decisions should already been made down below before you even display anything remotely interface-ish.
07:33 sorear eiro: talk to ingy, that's what pegex and c'dent is for
07:33 Woodi no sandbox so something close-to-GUI can be used. kind of good practice
07:33 diakopter sorear: I guess you didn't see eiro's original question
07:33 diakopter 1.08 hours ago
07:34 diakopter :)
07:34 Woodi Su-Shee: why horrible before ever designed ?
07:34 eiro sorear, hehe. and that was my first question: why pegex exists when he's aware of perl6
07:34 eiro there must be a technical issue i don't understand
07:34 eiro diakopter++
07:35 Woodi I do not said "do sandbox-in-high-lvl" but "do-filtering-with-module"
07:35 diakopter eiro: well, the p6 compilers don't compile to p5/python
07:35 wooden joined #perl6
07:35 wooden joined #perl6
07:36 * Woodi rereads and what hi sayd could be read that way... sorry
07:36 eiro diakopter, not even just the grammar ? no way to  introspect it?
07:36 diakopter (though STD/viv do of course, but it's just for the parser. So I guess if you wrote viv backends for python/etc you could get what you want)
07:37 Su-Shee Woodi: because this is a well-known question which has been discussed in detail at least since 1978. A GUI is an information editing/displaying engine, nothing more. it doesn't decide on anything security-like. the security on the contrary decides what an interface is allowed to show. and it does just that: "show stuff" AFTER the decision is done in lower levels of your software.
07:37 Su-Shee Woodi: read the orginal smalltalk model-view-controler paper.
07:37 Su-Shee +l
07:37 eiro Grammar.new( 'rx.p6' ).as_yaml.say
07:37 diakopter eiro: nqp provides a way to supply your own actions to a grammar; I don't know if rakudo does
07:38 eiro argghh... i was affraid of that :) ENOTIME for the moment but i need the result
07:38 eiro so yes: i'll fallback to pegex
07:38 Woodi I was proposing something close to GUI not GUI with sandbox
07:38 marmay_ joined #perl6
07:38 leont joined #perl6
07:38 eiro thanks all for answers
07:38 Su-Shee Woodi: read up on the problem.
07:39 Woodi I know MVC, controler need to do filtering - preferably by "use SomeSecModule"
07:44 FROGGS joined #perl6
08:11 arnsholt diakopter: The NQP actions stuff is the same as in Perl 6
08:13 diakopter k
08:13 wamba joined #perl6
08:19 arnsholt eiro: There's no easy way to introspect grammars (assuming what you want is things like "which subrules does this rule call")
08:20 arnsholt And, given that many of those things can be determined at runtime by things like <$subrule> and code interpolation, it's not really possible in the general case
08:20 Psyche^ joined #perl6
08:22 kresike joined #perl6
08:22 kresike good morning all you happy perl6 people
08:23 masak morning, kresike.
08:23 kresike hello masak o/
08:24 masak eiro: I just meant I don't think it captures the whole truth that Perl 6 implements a PEG. but I might be wrong on that, and it might be subject to interpretation.
08:26 brrt morning kresike
08:26 kresike hello brrt o/
08:30 wooden joined #perl6
08:30 wooden joined #perl6
08:31 brrt for those who care, the fix in parrot 4.8.0 to prevent parrot from dying twice, does not prevent parrot from dying twice, it allows it :-)
08:32 brrt if an embedding application has multiple interpreters, and one of them dies, it used to be the case that the second interpreter dying would not be reported to the embedding application
08:32 brrt this is now fixed
08:32 brrt thus allowing you to do something useful with the resulting exception object, twice
08:33 brrt (or more than once, anyway)
08:33 eiro arnsholt, masak thanks for those answers. that was my guess but perl6 is so astonishing i prefer don't bet before asking
08:47 dakkar joined #perl6
09:01 yves_ joined #perl6
09:09 dalek mu: 53e9d0d | (Konrad Borowski)++ | docs/feather/~/index.html:
09:09 dalek mu: Update http://perlcabal.org/~/
09:09 dalek mu: review: https://github.com/perl6/mu/commit/53e9d0d5c4
09:13 masak I notice today is Release Day and we don't have a release manager.
09:13 masak I volunteer.
09:14 masak (unless someone else wants to do it more than I do, in which case I gladly step down) :)
09:23 moritz masak: go for it
09:23 * masak edits the release guide
09:27 dalek rakudo/nom: 69910dc | masak++ | docs/release_guide.pod:
09:27 dalek rakudo/nom: [docs/release_guide.pod] I'll do September
09:27 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/69910dc761
09:27 masak I'll get to it this evening.
09:27 masak in the meanwhile, please think of an appropriate release name ;) kthx
09:31 dalek rakudo/nom: 94f900b | moritz++ | docs/release_guide.pod:
09:31 dalek rakudo/nom: fix a typo
09:31 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/94f900b3c8
09:31 moritz ah, that's usually the hardest part of doing a release :-)
09:31 moritz the release name
09:31 moritz you could always name it 'Perl', you know :-)
09:33 masak :D
09:34 wk Ooh, i had not noticed, that we had release named Tallinn
09:34 wk thank you, Masak
09:35 masak wk: you're welcome. and thank you for being so awesome, Tallinn. :)
09:37 wk masak: when you were here last time?
09:38 masak after YAPC::Europe last year.
09:38 masak wk: you were just celebrating Estonia's second independence! in a big park in the outskirts of Tallinn.
09:38 masak a huge crowd. thousands of white balloons. awesome.
09:39 wk masak: 21th August this year?
09:40 masak no, 2011.
09:40 wk masak: ok, 20th anniversary, after YAPC in Riga
09:43 masak aye.
09:44 masak jnthn and I went north up to Tallinn, and flew back home from there.
09:44 masak I have relatives in Tallinn. we visited them.
09:46 masak moritz: I'm not happy with the intermediate state unquotes are in in the nom branch. pros and cons of reverting the latest changes and relegating them to a branch?
09:55 moritz masak: what were the latest changes?
09:56 moritz the typed exceptions for splice typecheck failures?
09:56 masak oh, I don't mind those.
09:56 moritz keep it
09:56 masak I'm thinking more about allowing unquotes at all while giving them the wrong scoping.
09:56 moritz nobody uses macros yet, and we know they are experimental
09:56 moritz lunch&
09:57 masak ok, good.
09:57 masak I'll leave them.
09:57 masak I'm starting to have a plan for how to do them right... but $work is sucking up unusually many tuits right now.
10:12 wamba joined #perl6
10:13 [hds] joined #perl6
10:38 kurahaupo joined #perl6
11:05 Su-Shee_ joined #perl6
11:12 mucker joined #perl6
11:15 gootle joined #perl6
11:17 fhelmberger joined #perl6
11:21 jnthn +1 to leave macros in master
11:21 jnthn *no* feature tends to be spot on during its first iteration
11:21 jnthn Well, maybe some of the ones Pm puts in are. But none of mine are. :D
11:22 jnthn s/master/nom, our spiritual master/
11:22 fhelmberger_ joined #perl6
11:27 moritz jnthn: masak and me also tend to find bugs in the features that pm implements :-)
11:29 masak oh, don't worry. we do.
11:29 masak the best Pm bug that I've found has a long ticket on RT. I'm not sure it's resolved yet, actually.
11:30 masak it was about a leaky optimization.
11:30 jnthn nothing beats comet and snowman bug though :P
11:32 fhelmberger joined #perl6
11:33 brrt joined #perl6
11:36 moritz .oO( double bacon bug! )
11:37 tadzik oh yes
11:37 jnthn that was also pretty good :)
11:38 jnthn My students are currently writing a reverse polish evaluator in C#. Who can write one in Perl 6 that fits in a p6eval line? :)
11:41 masak what should it contain? numbers, + - * / ?
11:41 jnthn right
11:41 tadzik p6eval: given '22+3* { my@s; for .comb { if $_ ~~ /\d/ { @s.push($_) } else { @s.push(eval "{@s.pop} $_ {@s.pop}}") } } say @s.perl }
11:41 tadzik r: given '22+3* { my@s; for .comb { if $_ ~~ /\d/ { @s.push($_) } else { @s.push(eval "{@s.pop} $_ {@s.pop}}") } } say @s.perl }
11:41 p6eval rakudo 94f900: OUTPUT«===SORRY!===␤Confused␤at /tmp/2Z2RtLYGi7:1␤»
11:41 tadzik dang
11:42 tadzik r: given '22+3*' { my@s; for .comb { if $_ ~~ /\d/ { @s.push($_) } else { @s.push(eval "{@s.pop} $_ {@s.pop}}") } } say @s.perl }
11:42 p6eval rakudo 94f900: OUTPUT«===SORRY!===␤Unable to parse blockoid, couldn't find final '}' at line 2, near "for .comb "␤»
11:43 tadzik yay
11:43 tadzik r: given '22+3*' { my@s; for .comb { if $_ ~~ /\d/ { @s.push($_) } else { @s.push(eval "{@s.pop} $_ {@s.pop}") } } say @s.perl }
11:43 moritz r: $_ = '22+3*'; my @s; for .comb { if m/\d/ { @s.push: $_ } else { @s.push: eval "{@s.pop} $_ {@s.pop}" } }
11:43 p6eval rakudo 94f900: OUTPUT«===SORRY!===␤Unable to parse blockoid, couldn't find final '}' at line 2, near "for .comb "␤»
11:43 p6eval rakudo 94f900:  ( no output )
11:43 tadzik r: given '22+3*' { my@s; for $_.comb { if $_ ~~ /\d/ { @s.push($_) } else { @s.push(eval "{@s.pop} $_ {@s.pop}") } } say @s.perl }
11:43 p6eval rakudo 94f900: OUTPUT«===SORRY!===␤Unable to parse blockoid, couldn't find final '}' at line 2, near "for $_.com"␤»
11:43 moritz r: $_ = '22+3*'; my @s; for .comb { if m/\d/ { @s.push: $_ } else { @s.push: eval "{@s.pop} $_ {@s.pop}" } }; say @s[0]
11:43 p6eval rakudo 94f900: OUTPUT«12␤»
11:44 moritz there you go
11:44 tadzik nice
11:44 moritz though tadzik++ did all the hard work :-)
11:44 tadzik and all the syntax errors :P
11:45 masak hm, I would've thought '22' would mean 22, not two 2s :)
11:46 * masak is definitely not having sour grapes because he didn't get there first
11:46 tadzik masak: \details, \details
11:49 masak 5~:P
11:49 masak :P
11:50 yarp joined #perl6
11:50 tadzik r: given '33 22+3*' { my@s; for $_.comb(/\d+|\D/) { if $_ ~~ /\d/ { @s.push($_) } else { @s.push(eval "{@s.pop} $_ {@s.pop}") } } say @s.perl }
11:50 p6eval rakudo 94f900: OUTPUT«===SORRY!===␤Unable to parse blockoid, couldn't find final '}' at line 2, near "for $_.com"␤»
11:50 tadzik dang
11:50 tadzik r: $_ = '33 22+3*'; my @s; for .comb(/\d+|\D/) { if m/\d/ { @s.push: $_ } else { @s.push: eval "{@s.pop} $_ {@s.pop}" } }; say @s[0]
11:50 p6eval rakudo 94f900: OUTPUT«Element popped from empty list␤  in method Str at src/gen/CORE.setting:9842␤  in method Stringy at src/gen/CORE.setting:753␤  in block  at /tmp/I2DL11wU5I:1␤  in method reify at src/gen/CORE.setting:5217␤  in method reify at src/gen/CORE.setting:5119␤  in method re…
11:50 tadzik I'll just get back to work ;)
11:55 moritz r: $_ = '22+3*'; my @s; for .comb(/\d+ | <[+*/\-]> /) { if m/\d/  { @s.push: $_ } else { @s.push: eval "{@s.pop} $_ {@s.pop}" } }
11:55 p6eval rakudo 94f900: OUTPUT«Element popped from empty list␤  in method Str at src/gen/CORE.setting:9842␤  in method Stringy at src/gen/CORE.setting:753␤  in block  at /tmp/5VHApSl8oW:1␤  in method reify at src/gen/CORE.setting:5217␤  in method reify at src/gen/CORE.setting:5119␤  in method re…
11:55 moritz the problem with tadzik's code is that \D matches the whitespace too
11:55 tadzik ah
11:55 tadzik troo
11:55 tadzik r: $_ = '33 22+3*'; my @s; for .comb(/\d+|\w/) { if m/\d/ { @s.push: $_ } else { @s.push: eval "{@s.pop} $_ {@s.pop}" } }; say @s[0] #pleasepleaseplease
11:55 p6eval rakudo 94f900: OUTPUT«33␤»
11:55 tadzik naah
11:57 JimmyZ joined #perl6
12:00 Timbus r: $_ = '33 22+3*'; my @s; for .comb(/\d+|\S/) { if m/\d/ { @s.push: $_ } else { @s.push: eval "{@s.pop} $_ {@s.pop}" } }; say @s[0]
12:00 p6eval rakudo 94f900: OUTPUT«165␤»
12:03 moritz \S is good thinking
12:09 moritz a cow-orker presented me an internal gitlab installation (open source subset-clone of github). I found a bug in the first 3 minutes :-)
12:10 moritz (I pushed a project without a master branch; gitlab didn't cope with that. It has been fixed upstream yesterday)
12:11 jnthn r: sub rpn($s) { my @s; @s.push(/\d+/ ?? +$_ !! ::("&infix:<$_>")(@s.pop, @s.pop)) for $s.words; @s[0] }; say rpn '5 2 + 3 *'
12:11 p6eval rakudo 94f900: OUTPUT«21␤»
12:11 tadzik nice
12:12 jnthn coffee and cake break! :)
12:14 * colomon finally has a Linux which can run Niecza!  o/
12:14 Timbus ::("&infix:<$_>")
12:15 Timbus well thats just freaking awesome
12:17 * moritz didn't use .words because it requires ws around ops
12:19 masak moritz: maybe .comb can be used to split out ops from runs of digits? / \d+ | <[+-*/]> /
12:22 moritz masak: that's what the \d+ | \S  reges by Timbus++ did
12:25 masak oh!
12:25 masak that's what I get from backlogging partially.
12:25 masak yeah, probably better to \S and then default and error out on something unknown.
12:34 masak which git hooks are connected with the Rakudo repository?
12:34 masak does Niecza have any git hooks?
12:35 moritz both have dalek push hooks
12:36 masak I guess those hooks sit over at Github.
12:38 moritz yes
12:38 moritz masak: any particular reason for asking?
12:41 * [Coke] yawns.
12:47 spider-mario joined #perl6
12:49 masak moritz: yes. the thing at $dayjob that's taking up my time right now is constructing a Git course. I'm looking for real-world examples of everything between heaven and earth (including hooks), that I can use in the course material.
12:51 moritz masak: $work has a git hook that disallows non-fast-forward pushes
12:51 moritz masak: and disallows deleting of several branches (master, dev, bugfix, next)
12:52 * masak likes is-prime, but votes for the $tries parameter to be named, not positional
12:52 masak is-prime( :tries(100) ) reads better than is-prime(100)
12:52 masak moritz: both of those use cases are nice. thank you.
12:57 JimmyZ Is there a usecase for is-prime?
12:58 masak JimmyZ: my guess is that TimToady has encountered use cases for both it and expmod during his adventures in RC-land.
12:58 JimmyZ RC-land?
12:59 moritz rosettacode
12:59 moritz though you need it for project euler too :-)
13:01 JimmyZ r: my $r = 'a'; my $b = 'r'; say $$$$b
13:01 p6eval rakudo 94f900: OUTPUT«r␤»
13:02 masak that's a very rich variable.
13:02 masak maybe even part of the 1%.
13:04 PacoAir joined #perl6
13:04 JimmyZ Is there a way something like $$b to get $r
13:05 * [Coke] wonders if is-prime would be core or modulastic.
13:06 masak [Coke]: well, it's core now.
13:06 masak by spec.
13:07 [Coke] That seems crazy to me.
13:07 JimmyZ r: my $r = 'a'; my $b = 'r'; say eval "\$$b"
13:07 p6eval rakudo 94f900: OUTPUT«a␤»
13:08 masak r: my $r = 'a'; say $::("r")
13:08 p6eval rakudo 94f900: OUTPUT«a␤»
13:08 masak r: my $r = 'a'; my $b = 'r'; say $::($b)
13:08 [Coke] especially if the answer is "it depends".
13:08 p6eval rakudo 94f900: OUTPUT«a␤»
13:08 JimmyZ nice, that's what I want
13:09 masak [Coke]: I don't know enough about tht algorithm to agree or disagree.
13:09 moritz determining primality is one of these things where there's no one good solution yet, and research ongoing
13:10 moritz putting something in core for that seems questionable
13:10 moritz though I understand the motivation behind it
13:10 [Coke] what is the motivation?
13:13 moritz more elegant rosettacode solutions
13:13 masak I always thought a constant list `primes` would go in before `&is-prime`, for some reason.
13:13 moritz well, generating a list of primes is even more difficult than &is-prime
13:14 PerlJam not if it's primes-less-than-1000  or so  :)
13:14 PerlJam (good morning btw)
13:15 moritz good morning PerlJam
13:15 PerlJam even if it's not primes-less-than-1000, it's a one-time cost to generate them and store them.   Each release of Perl 6 could update that list  :)
13:16 * moritz slaps PerlJam
13:17 masak "pull yourself together!"
13:17 * PerlJam hugs moritz *hard*
13:18 masak there aare not many things you get slapped for on #perl6... but apparently joking about lists of primes is one :P
13:18 atrodo joined #perl6
13:18 PacoAir joined #perl6
13:25 gongyiliao joined #perl6
13:25 jnthn .oO( PerlJam's idea shows he sure ain't in his prime on a morning... )
13:25 PerlJam :-P
13:27 sirrobert could just use an Acme::Primes module =)  import a big list of primes if you need them.
13:28 PerlJam Though, I also question TimToady's judgement in naming ... "is-prime" contains far too much certitude for a probabalistic test.   Perhaps it should be called "maybe-prime" and return a confidence as well as a boolean
13:29 sirrobert sub maybe-prime { return True };
13:30 brrt sub maybe-prime { return True, 0.1 }
13:30 brrt or what is the ratio of prime numbers to non-prime-numbers?
13:30 moritz depends on how far up you go
13:30 sirrobert =)
13:31 hoelzro 1-10 it's 40% =)
13:31 hoelzro 1-7 it's 4/7
13:31 sirrobert http://primes.utm.edu/howmany.shtml
13:32 GlitchMr joined #perl6
13:32 skids joined #perl6
13:32 masak probably-prime
13:32 masak most-likely-prime
13:33 masak i-poked-it-it-is-prime
13:33 hoelzro trust-me-its-prime
13:33 sirrobert odds-of-primacy-are-greater-than-80-percent
13:33 tadzik masak++ # fotc reference
13:33 moritz here's-my-prime-number-mr-crazy
13:33 * masak knew tadzik would like it :)
13:33 jnthn primeval # 'cus it evaluates whether it could be a prime
13:34 tadzik I started my morning with Business Time
13:34 masak tadzik: TMI
13:34 tadzik ...if you know, what I mean :P
13:34 tadzik I just meant the song!
13:34 tadzik you insensitive clod :P
13:34 masak :P
13:35 * masak decommutes
13:48 am0c joined #perl6
13:51 [Coke] it is somewhat depressing to just be getting started on the real work of the day and having people heading out. :)
13:52 jnthn [Coke]: Move west! :P
13:52 jnthn er, dammit
13:52 jnthn Move east!
13:55 moritz or move far enough west :-)
13:57 dalek doc: a6f3177 | (Felix Herrmann)++ | lib/Int.pod:
13:57 dalek doc: [Int] add is-prime and expmod
13:57 dalek doc: review: https://github.com/perl6/doc/commit/a6f3177924
14:00 pmurias joined #perl6
14:01 pmurias jnthn: what does http://pastie.org/4760291
14:01 pmurias mean?
14:01 flightrecorder joined #perl6
14:01 thou joined #perl6
14:02 pmurias jnthn: I'm using classes (in to different files, one used from the other) not knowhows
14:02 stopbit joined #perl6
14:03 pmurias jnthn: putting them in a single file "fixed" the problem
14:08 wamba joined #perl6
14:14 FROGGS joined #perl6
14:19 colomon joined #perl6
14:25 benabik joined #perl6
14:28 leprevost joined #perl6
14:30 wamba joined #perl6
14:36 kaare_ joined #perl6
14:41 thou joined #perl6
14:42 timotimo Cannot add tokens of category 'postcircumfix'   -  but i want my .o( ) postcircumfix operator! for when i want to make my variables think stuff :(
14:44 masak our implementors are currently busy. thank you for waiting.
14:46 timotimo r: multi infix:<.oO>($a, $b) { $a }; my $a = 10; say ($a .oO "I should see if i'm an odd number") mod 2 == 0;
14:46 p6eval rakudo 94f900: OUTPUT«True␤»
14:46 timotimo i think that's cute
14:49 timotimo .o(actually, s/odd/even/ or s/0/1/ ...
14:49 timotimo shame on me!)
14:50 masak r: my $a = 10; say $a %% 2
14:50 p6eval rakudo 94f900: OUTPUT«True␤»
14:50 timotimo of course there's an operator for that :S
14:50 masak timotimo: also, don't use infix:<mod> until you've read up on it in the manual. ;)
14:50 masak it... may not do what you think.
14:52 benabik_ joined #perl6
15:00 benabik_ joined #perl6
15:02 Timbus joined #perl6
15:03 sqirrel joined #perl6
15:08 TimToady masak: I don't think is-prime is terribly misnamed when the default number of tries reduces the probability of error to 4⁻¹⁰⁰
15:09 yarp joined #perl6
15:10 masak yeah. people are a little too wary of probabilistic algorithms sometimes.
15:10 masak TimToady: how fast is it to make 100 trials?
15:10 TimToady not so fast that one isn't tempted to reduce the number :)
15:10 PerlJam masak: "God does not play dice"
15:11 skids r: multi postfix:<ȯO> (\a) { sub { "{ \a } thinks $^t".say } }; my $a = 10; $aȯO("must find a better nonperiod period"); #for timotimo
15:11 p6eval rakudo 94f900: OUTPUT«===SORRY!===␤Variable $aȯO is not declared␤at /tmp/_AyJQWrmTJ:1␤»
15:11 skids whoops.
15:11 TimToady but then if you reduce the number of tries, you supposedly have at least read the documentation for what the argument does
15:11 TimToady sometimes a rough idea of primality is all you need
15:11 masak PerlJam: not a very apt quote, since scientific consensus seems to have landed very much against Einstein on that.
15:11 skids r: multi postfix:<…O> (\a) { sub { "{ \a } thinks $^t".say } }; my $a = 10; $a…O("must find a better nonperiod period"); #for timotimo
15:11 p6eval rakudo 94f900: OUTPUT«10 thinks must find a better nonperiod period␤»
15:12 masak TimToady: how about the argument that $tries should be made :$tries ?
15:12 PerlJam masak: I still can't think of a better person to emulate :)
15:13 TimToady the run time also depends on the number in question, since modular exponentiate is iterative
15:13 masak PerlJam: you should apply for a job at a patent office, then.
15:13 masak TimToady: I suspected that.
15:13 TimToady *nentiation
15:13 PerlJam (at least as far as thinking about things goes ... his personal life wasn't much in the way things to emulate)
15:13 TimToady in http://rosettacode.org/wiki/Find_largest_left_truncatable_prime_in_a_given_base#Perl_6 I'm only using *one* try
15:14 PerlJam masak: I'm not swiss ;)
15:14 TimToady (for the big numbers)
15:14 TimToady because for some reason the actual problem seems to not amplify mistakes
15:15 timotimo when the chance of the algorithm being wrong is lower than the chance of getting hit by a comet, then i say: good enough
15:15 TimToady and just checking a few candidates at the end is pretty certain to find a real solution
15:15 TimToady well, I use 100 tries for the initial seeding and for the final check of all the truncated primes
15:18 TimToady I think the reason it works is because choosing a composite in place of a prime does not confer any advantage over its "children" as its generating longer candidates, so the list of final candidates tends to contain largely unrelated candidates of similar length
15:18 masak timotimo: or, more importantly, lower than a stray cosmic ray hitting the bit representing the boolean result and flipping it.
15:18 TimToady the actual solution is guaranteed to be generated, and the only uncertainty is whether we actually throw it away by not including it in the list of finalists for strict checking
15:19 TimToady anyway, it seems to generate correct results up through base 17
15:20 TimToady the base 18 I was running earlier blew up mono; I'm running it again to see if that was a fluke
15:21 TimToady I'd run it in rakudo, but I'm waiting for is-prime to use the version from libtommath before I attempt that
15:21 spider-mario joined #perl6
15:21 skids masak++ lol.
15:21 TimToady the current Perl-6-based is-prime in rakudo is too slow to be practical for base 18
15:22 TimToady it already takes several days of CPU time on my "supercomputer" in mono, and rakudo runs the algorithm slower
15:23 jnthn decomute &
15:23 TimToady I could run it about 4 or 5 times faster with real parallel hypers...
15:28 flussence joined #perl6
15:29 kresike bye all
15:32 timotimo there's lots and lots of items on the todo list before parallel hypers hit rakudo or niecza?
15:32 * masak .oO( depends if we do things in order on the todo list ) :P
15:32 flussence .oO( oh cool, that's the first time I've noticed 200+ users here. )
15:33 masak flussence: I count 199 nicks in here.
15:33 masak well, irssi does.
15:33 PerlJam maybe flussence sees some invisible users that we don't
15:33 flussence maybe my irssi's broken :)
15:34 * masak .oO( I see dead nicks )
15:35 timotimo masak: well, that's true. but there's a little bit of DAG going on there, too, isn't there?
15:35 TimToady actually, the RC algorithm doesn't even need hyper, race would be good enough, since we don't care about the order of results, since each iteration is just a set of numbers
15:38 colomon TimToady: what brought on is-prime?
15:38 TimToady http://rosettacode.org/wiki/Find_largest_left_truncatable_prime_in_a_given_base#Perl_6
15:38 TimToady and the fact that libtommath already has one :)
15:38 masak timotimo: well, yes. real hyper ops are probably blocking on a more solidly implemented concurrency model.
15:40 estrabd joined #perl6
15:40 timotimo i wonder why there's no syntax highlighting in that code example. i seem to recall seeing syntax highlighting for perl6 on other pages of that wiki
15:40 timotimo or maybe i've hallucinated that
15:41 estrabd joined #perl6
15:41 TimToady no, it seems to be broken there, but all my css from RC seems hosed at the moment
15:43 TimToady I cleared my browser cache, and now it's highlighting correctly
15:45 timotimo indeed it is
15:46 flussence odd, that's the first time I've seen p6 syntax-highlighted better than another language in one place...
15:47 timotimo pow' is a poor choice for that silly syntax highlighter
15:47 timotimo as is n'
16:00 birdwindupbird joined #perl6
16:02 birdwindupbird joined #perl6
16:03 TimToady When all you have is a machine shop, everything starts to look like it wants a custom tool.
16:05 TimToady I'm referring to the leap from BUILD to MOP to grammar mod at http://irclog.perlgeek.de/perl6/2012-09-19#i_6002309
16:05 TimToady when probably all you need is a function that does what the default BUILD does, callable from your BUILD
16:06 pmichaud good morning, #perl6
16:06 masak pmichaud! \o/
16:06 TimToady the extra arguments do show up in %_, after all
16:07 TimToady pmichaud: o/
16:07 masak timotimo: good point. will need to digest that, though.
16:07 pmichaud TimToady: enjoyed your tweets about the 47%  :-)
16:07 markov1 joined #perl6
16:08 TimToady yes, well, I was on a roll...
16:08 stopbit_ joined #perl6
16:08 masak you Usonians sure seem to like your percentages...
16:08 TimToady only 99% of us do
16:08 masak :P
16:09 pmichaud sentiment in my household has been much the same.  It's a little weird for us, too, because nearly all of our extended family are strong Romney backers, and all of them receive some sort of government assistance  :-)
16:09 masak slackers!
16:09 masak :P
16:09 TimToady yes, well, fact is that in the US rich people make more money off of tax breaks than poor people do :)
16:10 perigrin technically they make more money full stop than poor people do
16:10 * TimToady is also mostly embedded in conservative culture, which seems to have moved out from under him, leaving him stranded as a moderate
16:10 perigrin it makes playing the percentages easier.
16:11 pmichaud TimToady: yeah, I'm in the same boat (stranded)
16:11 perigrin a lot of conservatives feel that way too ... from my perspective you're one of the most liberal conservatives I've met :)
16:11 perigrin (both of you actually)
16:11 masak +1
16:12 pmichaud well, the state I live in has a lot to answer for these days :-P
16:12 TimToady I might turn into a conservative liberal if this keeps up. :)
16:13 perigrin there are worse things to be than someone who thinks that history has meaning but we should take care of each other.
16:13 MayDaniel joined #perl6
16:14 * masak .oO( 50 shades of the GOP )
16:14 pmichaud alas, the GOP is more O than G these days.  :-)
16:14 perigrin pmichaud: my state is trying to give yours a run for it's money.
16:15 TimToady what really gripes me is when the pollsters lump me in with all the other "old white males"...
16:16 masak better than being lumped with the dead white males...
16:16 TimToady working on that...
16:16 pmichaud perigrin: perhaps, but we have W, Ron Paul, and "oops".
16:16 flussence .oO( better than being thrown in with the great white sharks... )
16:17 perigrin pmichaud: We have Rubio, Scott, and W's brother ...
16:17 perigrin like I said ... trying to give you a run for your money :)
16:17 pmichaud Jeb actually sounds reasonable to me.
16:17 TimToady because he's *not* running, which is very reasonable :)
16:17 pmichaud lol
16:17 perigrin His first two weeks as Governor he violated our goverment transparency laws.
16:17 TimToady but yes, he seems sane
16:18 TimToady or at least saner, on average
16:18 perigrin Damning with faint praise there. :)
16:19 * perigrin was not impressed with him as Governor ... but also spent most of that time living over-seas.
16:20 TimToady well, this is all rather off-topic, even if rather topical
16:21 huf but how else would i know to like or dislike perl6 if i dont know the political leanings of the core people working on it? :)
16:21 perigrin hyper-politics operators ... you can lean all directions at once and they'll reify at runtime.
16:21 TimToady we tend to be foaming-at-the-mouth moderates, with a large standard deviation :)
16:22 pmichaud huf: we're all folks who believe there are multiple answers to the same question.
16:22 TimToady you can say that again
16:22 TimToady that goes without saying
16:22 huf :)
16:22 huf i'm sure he can say it again, but exactly how many times?
16:22 huf i must know!
16:23 perigrin huf: it's a lazy list.
16:23 huf eh, that way lie the too-easy political jokes, i'll skip those :)
16:31 masak moritz: heh, I also just found the typo you fixed in 94f900b. ;)
16:38 crab2313 joined #perl6
16:54 wk joined #perl6
16:59 skids .oO(Cat should be implemented ASAP just so we (<can>,<haz>).lol.cat.say)
17:02 thou joined #perl6
17:02 * TimToady pictures a lolrat
17:02 skids Iz in yer wallz chewin yer wirez
17:03 skids not as endearing
17:03 TimToady tough job, but someone has to do it
17:04 masak tough... and not without risks. those wires are hot!
17:06 TimToady The wires are only half as hot, which is why Norwegian rats do so well here.
17:09 masak Norwegian rats do well almost anywhere. :)
17:10 masak (and they've emigrated from Norway because the beer prices were too high)
17:11 TimToady I guess rats are subject to notchewall selection...
17:11 masak :P
17:14 xinming_ joined #perl6
17:15 crab2313 joined #perl6
17:15 raiph joined #perl6
17:18 flussence joined #perl6
17:22 * moritz wonders if expmod should be an infix operator
17:23 masak **%
17:23 masak actually, I wonder if it should just be a guaranteed optimization...
17:35 * diakopter wonders what "portable Perl 6 code" will look like
17:36 diakopter obviously least common denominator syntax
17:36 diakopter but style of writing I'm wondering about
17:43 TimToady masak: yes, I have wondered that also
17:44 masak oh, good.
17:44 TimToady though you might want more type inference for that to work out well...
17:47 masak maybe.
17:49 flightrecorder joined #perl6
17:49 FROGGS joined #perl6
17:55 raiph hi all. has anyone reading this used the summary backlog instead of reading the full backlog?
17:55 moritz $ /perl6 -e 'say 3.expmod(2, 8)'
17:55 moritz 1
17:55 moritz $ ./perl6 -e 'say 3.expmod(2, 100)'
17:55 moritz 9
17:56 moritz raiph: I've done that today
17:56 moritz or yesterday, not sure
17:56 * moritz is a bit confused these days
17:57 TimToady moritz: is that tying into the tom code?
17:57 moritz TimToady: yes
17:57 moritz method expmod(Int:D: Int:D \base, Int:D \mod) {
17:57 moritz nqp::expmod_I(self, base, mod, Int);
17:57 moritz }
17:57 TimToady cool
17:57 masak moritz++
17:57 raiph moritz: instead of or as well as?
17:57 TimToady I saw those routines, and figgered it was a simple matter of nqp
17:58 moritz though probably missing some decontainerization
17:58 benabik joined #perl6
17:58 moritz raiph: so far I've settled with reading the summary, and later the full backlog if I have time
17:59 * TimToady feels obligated to read everything always, for some strange reason
17:59 * TimToady would not make a good President
17:59 raiph moritz: \o/ :)
17:59 moritz raiph: my hope was that you (or somebody else) would copy&paste the HTML from the IRC logs and add some headings to it
17:59 moritz and split it up into sections
18:00 moritz but the current state is better than nothing, and i hope it doesn't burn out our summarizers
18:01 sorear good * #perl6
18:02 zby_home_ joined #perl6
18:02 moritz masak: I think I should wait with my expmod and is-prime changes until after the release
18:03 moritz make: *** [src/stage1/QAST.pbc] Segmentation fault
18:03 moritz eeks
18:03 moritz not reproducible though
18:03 benabik Cosmic ray.  No worries.  ;-)
18:04 * TimToady greets sorear with an amish californian greeting...
18:04 masak moritz: agreed.
18:04 masak sorear! \o/
18:13 kurahaupo joined #perl6
18:15 birdwindupbird joined #perl6
18:28 masak ok, all y'all wonder full people.
18:28 masak wonderful*
18:28 masak are you ready to go into Rakudo release mode? :D
18:29 moritz the question is if *you* are ready :-)
18:29 TimToady we're always ready for you to go into release mode, you insensitive clod!
18:29 masak well, then. release mode it is.
18:30 * masak flips the big "RELEASE MODE" switch
18:30 FROGGS the huge red one?
18:30 * TimToady hears a humming sound...
18:30 masak yah. that one.
18:30 FROGGS the "there-is-no-going-back" nob?
18:30 masak also, I'll note that I've been called an insensitive clod twice today.
18:30 moritz $ ./perl6 -e 'say grep *.is-prime, 1..100'
18:30 moritz 1 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
18:30 TimToady masak: yes, I was parodying that
18:30 FROGGS hehe, so it must be true ;o)
18:31 masak TimToady: not sure the first time was well-deserved either :P
18:31 PerlJam masak: you insensitive clod!  (thrice! :-)
18:31 masak FROGGS: no, no. three times means true.
18:31 masak dang.
18:31 PerlJam :-)
18:31 TimToady it's one of those things I wouldn't say if it were true...
18:31 moritz I... didn't expect the libtommath is_prime check to return True on 1
18:31 masak moritz: I *knew* it!
18:31 TimToady hmm
18:31 masak it alwasy *felt* like a prime to me :P
18:32 masak no matter what people said...
18:32 rurban Note that I fixed the broken imcc optimizer for constant encoded strings yesterday, but it's only in a parrot branch. https://github.com/parrot/parrot/issues/837
18:32 TimToady well, it is divisible by itself and 1
18:32 PerlJam looks like a prime, acts like a prime, smells like a prime ...
18:32 pmichaud rurban++
18:32 PerlJam .oO( duck priming? )
18:32 masak TimToady: so are all primes.
18:32 moritz PerlJam: but it doesn't quack like a prime
18:33 * pmichaud detects the awesome rakudo release machine spinning up
18:33 rurban parrot 4.9.0 will cpontain it but I try to fix the internal representaiuon of enconded string in SREG's also. It's 4x times slower currently
18:33 masak let's see if this time I manage to get the nqp release right...
18:33 pmichaud masak: there are a new set of nqp release instructions, so it should be better this time.  (courtesy of mtmh2012
18:33 masak moritz: be prepared to remind me of whatever it is I've forgotten both recent times.
18:33 moritz rurban++ # fixing stuff
18:33 masak pmichaud: oh, good.
18:34 masak All tests successful.
18:34 masak Result: PASS
18:34 masak \o/
18:34 moritz masak: nqp tarball? bumping NQP's VERSION and Rakudo's VERSION?
18:34 masak moritz: I don't remember. :)
18:34 sorear if you make 1 a prime, you lose unique factorization.
18:34 sorear -1
18:34 pmichaud the nqp tarball is now created by doing "make release"
18:34 rurban I think I can optimize string encoding held in registers or constants by factor 4
18:34 * PerlJam looks at NQP's release instructions  (I believe I volunteered for next month's release)
18:34 masak sorear: I wasn't being serious.
18:35 PerlJam pm: nice
18:35 moritz I can't make 1 a prime, and i guess nobody can :-)
18:35 rurban calling conventions not sure yet, but it should be doable also.
18:35 sorear rurban: is this part of the imcc optimization enabled by default?  if not, does rakudo enable it?
18:36 * pmichaud double-checks that nqp release document was indeed updated.
18:36 rurban it's internal in constant folding. it's also that SREG's and const strings cannot hold the encoding information efficiently at all. We need to redesign it.
18:36 masak peeps. release name. seriously.
18:36 masak focus! :P
18:36 sorear masak: Stavanger?
18:36 masak ooh
18:37 rurban But it looks easy to do. See https://github.com/parrot/parrot/issues/837#issuecomment-8717966
18:37 pmichaud is there a Stavanger.pm, though?
18:37 sorear or was that used already
18:37 TimToady have we used Tokyo or Edo?  (YAPC::Asia coming up...)
18:37 rurban p.stolen
18:37 masak pmichaud: doubt it. neither is there a Perl.pm, sadly.
18:37 rurban Preikestolen
18:37 masak that would be wonderfully confusing.
18:37 pmichaud well, I figure we could create a Perl.pm :-)
18:37 PerlJam masak: #56?  :)
18:37 masak PerlJam: you're being very helpful there.
18:37 rurban or Pulpit Rock
18:38 TimToady ooh, I just saw that
18:38 PerlJam Release names are always the hardest part.
18:38 pmichaud I'd be fine with Tokyo.pm
18:38 [Coke] which is why we shouldn't use them. :P
18:38 PerlJam [Coke]: ding!
18:38 pmichaud since jnthn++ will be attending, iirc
18:38 masak going with "Tokyo" for now.
18:38 pmichaud ...is there a reason why the release guide has release #55 labeled as "Frankfurst"?  (note the extra 's')
18:38 masak phew, hardest part is over :)
18:39 masak pmichaud: git pull
18:39 masak pmichaud: as to the reason, I guess moritz was thinking of food at the time :P
18:39 pmichaud lol
18:39 skids r: class A { multi method m (:$f) { "f".say }; multi method m (:$g) { "g".say } }; my A $a .= new(); $a.m(:f);
18:39 p6eval rakudo 94f900: OUTPUT«f␤»
18:39 thou joined #perl6
18:40 skids Believe it or not that's broken in star08
18:40 pmichaud there is a "#Perl.pm", however!  http://www.pm.org/groups/665.html
18:40 skids Probably got fixed along with the constraints?
18:40 masak o.O
18:40 sorear quick, somebody start a PM group in Frankfurt (Oder) to confuse things
18:40 masak tempted to switch over to "Perl" as a release name, then.
18:40 masak we can do Tokyo next month, when jnthn++'s been there.
18:41 pmichaud not sure how '#Perl.pm' is, though -- but it does appear in pm.org's "active groups" list :)
18:41 PerlJam masak: That sounds perfect ... for me  :)
18:42 * pmichaud looks through the other list of groups for more names
18:42 [Coke] star: class A { multi method m (:$f) { "f".say }; multi method m (:$g) { "g".say } }; my A $a .= new(); $a.m(:f);
18:42 p6eval star 2012.07: OUTPUT«Ambiguous call to 'm'; these signatures all match:␤:(A , :f(:$f), Mu *%_)␤:(A , :g(:$g), Mu *%_)␤␤  in method m at src/gen/CORE.setting:323␤  in block <anon> at /tmp/eN7sQqdMEy:1␤␤»
18:42 PerlJam Naming any release of Rakudo "Perl" is tempting fate a little I think.   What confusion there could be!
18:44 skids Save it for the 1.0.0 release :-)
18:44 jnthn o/ from the train
18:44 sorear But we're on 56.0.0 now
18:44 dalek perl6-roast-data: b60f209 | coke++ | / (3 files):
18:44 dalek perl6-roast-data: today (automated commit)
18:44 dalek perl6-roast-data: review: https://github.com/coke/perl6-roast-data/commit/b60f209928
18:45 jnthn this wifi is really greaNO CARRIER
18:45 [Coke] jnthn: back down to 2 rakudo fails.
18:45 PerlJam It would be funny if wifi dropping actually did generate "NO CARRIER" messages
18:45 skids sorear: just chalk that up to endianness.
18:46 pmichaud masak: I'll create the .msi version of the release later tonight or (more likely ) early tomorrow
18:46 sorear pmichaud: and we even have mst contributing :D
18:47 pmichaud sorear: indeed!  mst++
18:47 sorear re. #perl.pm
18:47 jnthn [Coke]: Yay :)
18:48 sorear bleh, the end of the month snuck up on me
18:48 rurban bad performance news from the parrot front. just benchmarked it against a different test and it looks like encoding SREG will not bring 4x times speed advantage. rather none. There are other things slowing it down 4 times. sprintf mostly.
18:54 skids .oO(someday I'll crack a NO CARRIER joke and nobody will get it because they are all too young.  Sigh.)
18:56 jnthn skids: Just hang out with old people like me :P
18:56 TimToady Get off my CARRIER!!!
18:57 masak tools/contributors.pl finds a contributor named "C". I don't know who that is. I don't find a corresponding author in the git logs. should I delete that entry?
18:57 tadzik C++
18:57 tadzik in the commit logs, I guess
18:59 skids Maybe it's sort of like the C locale.
18:59 sorear rurban: what kinds of things will become 4x faster after you're done?
19:00 fgomez joined #perl6
19:04 rurban I thought that SREG not holding STRING is slowing it down by 4x. Because I have to unesacpe, create a new string for the SREG with "encoding:"\unescaped string\"" as char *ptr. But a new benchmarks shows that its not worthwile fixing this. Rather fix sprintf.
19:04 rurban But I'll research it.
19:05 dalek rakudo/nom: c13ac5e | masak++ | docs/announce/2012.09:
19:05 dalek rakudo/nom: [docs/announce/2012.09] added
19:05 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/c13ac5e535
19:05 masak people are welcome to review the announcement.
19:05 masak spotting typos, errors, etc.
19:07 dalek specs: 1c74b86 | larry++ | S32-setting-library/Numeric.pod:
19:07 dalek specs: Assume Miller-Rabin certainty on composites
19:07 dalek specs:
19:07 dalek specs: Pointed out by Martin D Kealey.
19:07 dalek specs: review: https://github.com/perl6/specs/commit/1c74b86e60
19:09 jnthn masak: is the move to a QAST-based NQP not worth mentioning? Or the new proto support? :)
19:09 rurban #56 "Perl" really? But I like the idea
19:11 masak jnthn: they're in the ChangeLog. I removed them from the announcement as being "a bit internal". happy to be overridden on that point, though.
19:12 * masak decides that "faster compiler" is worth having there
19:13 jnthn yeah, faster is always worth it
19:13 dalek rakudo/nom: 738d1ea | masak++ | docs/announce/2012.09:
19:13 dalek rakudo/nom: [docs/announce/2012.09] mention faster compiler
19:13 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/738d1ea823
19:14 masak phenny: sv en "faster"?
19:14 phenny masak: "aunt" (sv to en, translate.google.com)
19:14 jnthn and the proto thing is actually a nice feature
19:14 GlitchMr "These are some possibly breaking changes:"
19:14 GlitchMr "- tie-breaking with constraints now picks the first matching one rather than demanding they be mutually exclusive"
19:14 GlitchMr Why it would be breaking
19:14 GlitchMr It wasn't working before, now it does
19:15 GlitchMr You could use this same argument for "Str.wordcase" - in 2012.08 Str.Wordcase didn't existed, it could break your 2012.09 if they depend on it to not existy
19:15 GlitchMr exist*
19:16 masak good point.
19:16 masak moving tie-breaking up to "new features".
19:17 GlitchMr Every change in dynamic language could be easily breaking
19:17 GlitchMr your code on 2012.09*
19:18 dalek rakudo/nom: 488521b | masak++ | docs/announce/2012.09:
19:18 dalek rakudo/nom: [docs/announce/2012.09] tie-breaking is a feature
19:18 dalek rakudo/nom:
19:18 dalek rakudo/nom: Not a breaking change. :)
19:18 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/488521be0c
19:19 PacoAir joined #perl6
19:19 skids .oO(breaks ties, not code)
19:19 masak guess it prefers bowties.
19:20 flussence joined #perl6
19:21 PacoAir joined #perl6
19:22 [Coke] bowties are cool now.
19:22 dalek nqp: a318bd5 | masak++ | VERSION:
19:22 dalek nqp: [VERSION] bump to 2012.09
19:22 dalek nqp: review: https://github.com/perl6/nqp/commit/a318bd55e4
19:22 [Coke] so, why is evalbot using a one-back star?
19:22 GlitchMr nqp 2012.09 :-)
19:23 skids [Coke]: At least now that we've all forgotten Tucker Carlson entirely.
19:23 [Coke] skids: ah, I was referencing Doctor Who. Yes, I had forgotten TC.
19:24 * masak wonders if Tucker Carlson is ever going to forgive Jon Stewart
19:32 PacoAir joined #perl6
19:39 PacoAir joined #perl6
19:40 masak NQP 2012.09 uploaded.
19:43 dalek rakudo/nom: bdb50f4 | masak++ | tools/build/NQP_REVISION:
19:43 dalek rakudo/nom: [release] bump NQP revision
19:43 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/bdb50f47a6
19:43 dalek rakudo/nom: 77defe7 | masak++ | VERSION:
19:43 dalek rakudo/nom: [release] bump VERSION
19:43 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/77defe7332
19:43 dalek glitchmr.github.com/perl6-changes: 8179af9 | GlitchMr++ | _posts/2012-09-23-perl-6-changes-2012W38.md:
19:43 dalek glitchmr.github.com/perl6-changes: Add more Perl 6 changes to blog article.
19:43 dalek glitchmr.github.com/perl6-changes:
19:43 dalek glitchmr.github.com/perl6-changes: Currently, the changes aren't too exciting compared to previous week,
19:43 dalek glitchmr.github.com/perl6-changes: but we still have three days - who knows what will happen then...
19:43 dalek glitchmr.github.com/perl6-changes: Besides, isn't Rakudo Star release awesome enough - I mean, it brings
19:43 dalek glitchmr.github.com/perl6-changes: newest changes to useful and usable distribution of Perl 6 - you know,
19:43 dalek glitchmr.github.com/perl6-changes: without installing Panda :-).
19:43 dalek glitchmr.github.com/perl6-changes: review: https://github.com/GlitchMr/glitchmr.github.com/commit/8179af9724
19:45 GlitchMr I know that links to `expmod` and `is-prime` aren't working, but I hope http://doc.perl6.org/ will update before I will merge it with `master` :-)
19:51 dalek nqp: e8a7719 | jonathan++ | docs/qast.markdown:
19:51 dalek nqp: Document more of QAST.
19:51 dalek nqp: review: https://github.com/perl6/nqp/commit/e8a7719544
19:52 sivoais joined #perl6
19:52 dalek glitchmr.github.com: 92a60be | GlitchMr++ | / (2 files):
19:52 dalek glitchmr.github.com: Fix footer on main page.
19:52 dalek glitchmr.github.com: review: https://github.com/GlitchMr/glitchmr.github.com/commit/92a60be42f
19:54 masak all stresstests pass.
19:54 am0c joined #perl6
19:54 dalek glitchmr.github.com: 7111f0d | GlitchMr++ | index.css:
19:54 dalek glitchmr.github.com: Replace 0px with 0, 0px is ugly
19:54 dalek glitchmr.github.com: review: https://github.com/GlitchMr/glitchmr.github.com/commit/7111f0da7f
19:54 masak make release VERSION=2012.09
19:54 GlitchMr I hope I don't spam this channel with commits to my blog
19:54 masak why do I have to pass VERSION here?
19:54 masak if I did everything right up to this point, the correct version is in the file VERSION, no?
19:55 masak when would I ever want to do a release with another version number than the one in the VERSION file?
19:55 jnthn Maybe the Makefile needs it for something
19:57 masak my point is that the Makefile can go look for it in the VERSION file.
19:57 masak oh well. it's not a big deal. it just feels redundant.
19:57 jnthn Yeah but that's probably hassle to (portably) implement :)
19:57 jnthn er, not that make release is portable anyway, I guess...
19:58 GlitchMr Also, my website breaks HTML standards and I seriously don't care
19:58 jnthn GlitchMr: You only break standards if you declare a doctype that says you are following them.
19:59 sorear which ones?
19:59 jnthn If you don't declare a doctype, you can do what you want, 'cus you promised nothing :)
19:59 GlitchMr I declare doctype for Strict Mode
19:59 GlitchMr <!DOCTYPE html>
19:59 GlitchMr Anyways, bye
19:59 jnthn o/
19:59 jnthn bah, my battery is almost out and now the wifi improves
20:00 jnthn 'cus we're back in Skane, of course
20:00 masak Rakudo 2012.09 uploaded to https://github.com/rakudo/rakudo/downloads
20:00 sorear you have wifi on the train but no electricity?
20:01 jnthn There's power but I have something else plugged in :D
20:01 jnthn though, that's probably charged
20:01 jnthn but also I'm almost home :)
20:03 masak p6c announcement sent.
20:04 jnthn masak++
20:04 sorear masak++
20:05 jnthn Time to start making the next month's release awesome :)
20:06 masak http://en.wikipedia.org/wiki/Rakudo_Perl_6 updated.
20:06 masak standing down release mode.
20:06 * masak celebrates with an imperial stout :)
20:06 PerlJam masak++
20:06 jnthn \o/
20:06 * jnthn only has unimperial stout at home
20:06 jnthn in fact, I think my fridge only one beer o.O
20:07 jnthn Guess I gotta go shopping after work tomorrow :)
20:08 masak thank you all for participating in this months Rakudo release.
20:08 masak your cooperation is much appreciated.
20:16 dalek rakudo/nom: adc2127 | masak++ | docs/release_guide.pod:
20:16 dalek rakudo/nom: [docs/release_guide.pod] some future release dates
20:16 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/adc21274b3
20:17 Chillance joined #perl6
20:17 * flussence updates the other bit of the wiki page because masak++ forgot :P
20:18 masak oh!
20:18 masak thanks.
20:18 masak let's hope that was my one mistake this time ;)
20:18 dalek rakudo/nom: ee7dfef | duff++ | docs/release_guide.pod:
20:18 dalek rakudo/nom: Move masak's release from the future to the past
20:18 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/ee7dfef791
20:18 masak ah, right. the main text.
20:18 masak PerlJam: heh ;)
20:19 masak flussence++ PerlJam++
20:19 * masak comes to think of Kung Fu Panda
20:19 masak "yesterday is history, the future, a mystery... but today -- is a gift"
20:20 masak "that's why they call it 'the present'"
20:20 tadzik :D
20:20 PerlJam sage words from a cartoon
20:20 tadzik masak++ #release
20:20 masak \o/
20:21 * masak is now having a stout and eating 77% chocolate from the Philippines
20:21 TimToady I suppose the chocolate makes the stout not taste bitter, and vice versa
20:22 sorear Is it blended with 23% Swedish chocolate to reduce importation costs?
20:22 skids Is there a non-fatal way to "use NotExistingPackage;" yet, other than BEGIN{ eval {...} }?
20:22 sorear skids: try { require NotExistingPackage; }
20:22 masak sorear: I don't know what the remaining 23% are. probably disgruntled Americans of some kind. :)
20:23 skids Ah.  Right.  Thanks sorear++.
20:23 sorear masak: eating Americans?
20:23 masak sorear: just referring back to my earlier comment today that Usonians seem unusually keen on percentages right now.
20:24 sorear it's like malaria
20:24 masak 1%, 99%, 47%...
20:24 sorear it flares up every four years like clockwork
20:24 masak :P
20:24 * masak .oO( clockwork flares up every four years? )
20:25 diakopter clockwork * every * *
20:25 PerlJam Is that why it's orange?
20:26 * masak cut the release just in time for the absurdist humor, it seems :)
20:26 [Coke] glitchmr: I am wondering why you post commits to your blog here. isn't it more interesting that your blog post is posted?
20:27 TimToady 14.3% of all colors are orange
20:27 TimToady --Roy G Bif
20:27 TimToady *Biv
20:27 masak r: say (100 / 7).fmt("%4.1f")
20:28 p6eval rakudo 77defe: OUTPUT«14.3␤»
20:28 masak yep.
20:28 * TimToady pictures a fiolet flower
20:28 masak we're in fiolent agreement here.
20:29 sorear what probability measure do you use on colors?
20:29 sorear it seems to me that you're only considering spectral colors
20:29 diakopter sorear: I think he just meant 1/7 of roygbiv
20:30 TimToady npr: sub postfix:<%>($x) { $x * 100 }; say (100% / 7).fmt("%4.1f")
20:30 masak sorear: see above p6eval calculation.
20:30 p6eval pugs: OUTPUT«(timeout)»
20:30 p6eval ..rakudo 77defe, niecza v21-15-gf226bf6: OUTPUT«1428.6␤»
20:30 TimToady er...
20:30 PerlJam clearly "color" must refer to visible light
20:30 masak TimToady: talk about overcompensating ;)
20:31 diakopter PerlJam: *visible to humans
20:31 TimToady 1428.6% of oranges are colored
20:31 flussence spectres normally aren't visible to humans
20:31 masak diakopter: *visible to humans who have sight
20:31 diakopter .. and aren't colorblind
20:32 PerlJam humans are the most important of all beings capable of sight, so all refereneces to color revolve around human perception.
20:32 TimToady I don't see it that way.
20:32 PerlJam It's the most important 300 nm EVAR!
20:33 TimToady Solism is so provincial...
20:34 * sorear mumbles something about imaginary coors
20:34 sorear colors, too
20:35 TimToady didn't know coors made a stout, or chocolate either...
20:35 masak well, the color that flickers in front of my eyes when I rub them or when I'm tired is *exactly* the imaginary purple-green that Pratchett describes in his books.
20:36 masak it's uncanny.
20:36 sorear 'coons?  Moors?
20:37 * TimToady infers that masak rubs his eyes only when he's not tired
20:37 diakopter he didn't say xor
20:37 masak actually, it's when I'm tired that I tend to rub my eyes.
20:37 TimToady this is why TimToady should have his computer do his inferencing for him
20:38 sorear one micro-foot
20:39 thou joined #perl6
20:39 TimToady <- a poem with 7 micro-feet
20:40 * colomon is now thinking of imaginary Ardbeg...
20:42 * TimToady wanders off to take a nap for a millifortnight or two
20:43 masak r: say (14 * 24 * 60 * 60) / 1000
20:43 p6eval rakudo 77defe: OUTPUT«1209.6␤»
20:43 masak r: say (14 * 24 * 60 * 60) / 1000 / 60
20:43 p6eval rakudo 77defe: OUTPUT«20.16␤»
20:43 masak twenty minutes.
20:43 masak or forty :)
20:48 skids I suppose using xor in a natural language conversation is no worse than using iff.
20:51 [Coke] jnthn: my fridge only has a half drunk beer with a wine stopper in the bottle,
20:52 masak beer is so badly behaved when it's half drunk.
20:55 kurahaupo joined #perl6
20:56 spider-mario joined #perl6
20:56 jnthn ah, home :)
20:56 sorear yay!
20:56 jnthn I'll be here for, er, 2 days.
20:57 jnthn Then I think I'm gonna change continent. :P
20:58 sorear YAPC::Tōkyō ?
20:58 masak \o/
20:59 PerlJam jnthn: you need africa, australia, greenland and antarctica for the complete set of major land masses.
21:00 jnthn PerlJam: I went to Morocco, which is Africa. :)
21:00 jnthn Greenland is pretty easy from here. :)
21:00 jnthn Australia is a long flight, and Antarctica is appealing but also distant and very pricey to go to :)
21:01 PerlJam ah, sorry, I wasn't at a good zoom level on your map.   Morocco's placemarks blended into Spain
21:01 jnthn .oO( Why is Greenland icier than Iceland, and Iceland geener than Greenland? )
21:02 * sorear would like to set up shop on Mars or Venus someday
21:02 jnthn sorear: And yes, YAPC :)
21:02 jnthn And vacation.
21:02 jnthn I'm going for 2 weeks to attend a 2.5 day conference. :)
21:06 leprevost joined #perl6
21:09 * [Coke] sees he squeaked onto the release announcement.
21:10 sivoais joined #perl6
21:10 snearch joined #perl6
21:20 stopbit_ joined #perl6
21:25 dalek rakudo/nom: 00533a4 | duff++ | CREDITS:
21:25 dalek rakudo/nom: Add IRC name to CREDITS
21:25 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/00533a4f3d
21:25 dalek rakudo/nom: ef228dd | duff++ | tools/contributors.pl:
21:25 dalek rakudo/nom: Minor improvement to contributors.pl
21:25 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/ef228dd818
22:01 leont joined #perl6
22:14 benabik joined #perl6
22:16 benabik_ joined #perl6
22:22 benabik_ joined #perl6
22:46 masak good night, #perl6
22:46 colomon \o
22:49 fgomez joined #perl6
22:49 thou joined #perl6
23:15 skids joined #perl6
23:18 broquaint joined #perl6
23:22 telex joined #perl6
23:24 Psyche^ joined #perl6
23:49 whiteknight joined #perl6

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

Perl 6 | Reference Documentation | Rakudo