Perl 6 - the future is here, just unevenly distributed

IRC log for #perl6, 2010-08-31

Perl 6 | Reference Documentation | Rakudo

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

All times shown according to UTC.

Time Nick Message
00:02 Psyche^ joined #perl6
00:02 risou_ joined #perl6
00:03 thundergnat rakudo: ("a".."z").perl.say
00:03 p6eval rakudo 6ccdcd: OUTPUT«"a".."z"␤»
00:03 thundergnat hmmm.
00:04 thundergnat Have string ranges been disabled?
00:04 colomon no
00:04 thundergnat string series still work
00:04 colomon that's the correct behavior
00:04 colomon rakudo: ("a".."z").list.perl.say
00:04 p6eval rakudo 6ccdcd: OUTPUT«("a", "b", "c", "d", "e", "f", "g", "h", "i", "j", "k", "l", "m", "n", "o", "p", "q", "r", "s", "t", "u", "v", "w", "x", "y", "z")␤»
00:05 masak thundergnat: minor nit: it's infix:<...> that's usually called "series". infix:<..> is called "range".
00:05 dju left #perl6
00:06 HarryS left #perl6
00:06 dju joined #perl6
00:06 thundergnat masak: I know, I was actually trying to make that distinction.
00:06 Patterner left #perl6
00:06 Psyche^ is now known as Patterner
00:06 thundergnat rakudo: ("a"..."z").perl.say
00:06 p6eval rakudo 6ccdcd: OUTPUT«("a", "b", "c", "d", "e", "f", "g", "h", "i", "j", "k", "l", "m", "n", "o", "p", "q", "r", "s", "t", "u", "v", "w", "x", "y", "z")␤»
00:07 lichtkind left #perl6
00:07 masak thundergnat: pardon me, then. was confused by the surrounding evaluations. :)
00:07 risou_ left #perl6
00:08 fod joined #perl6
00:09 Tene left #perl6
00:09 HarryS joined #perl6
00:10 thundergnat I didn't realize that ranges evaluated in scalar context just return a string.
00:10 thundergnat Amazing what you can learn by reading the docs.
00:11 masak no, that's not so.
00:11 masak rakudo: my $r = 1..3; say $r.WHAT
00:11 p6eval rakudo 6ccdcd: OUTPUT«Range()␤»
00:11 jhuni joined #perl6
00:13 thundergnat Ok, I will attempt to stop saying stupid things now. it may be a struggle....
00:13 shortcircuit joined #perl6
00:14 alester left #perl6
00:15 colomon rakudo: my $r = 1..3; say $r
00:15 p6eval rakudo 6ccdcd: OUTPUT«1 2 3␤»
00:15 HarryS left #perl6
00:19 HarryS joined #perl6
00:23 lucs joined #perl6
00:23 thundergnat left #perl6
00:25 HarryS left #perl6
00:26 shortcircuit TimToady: If the test data for http://rosettacode.org/wiki​/Find_Common_Directory_Path isn't thorough, feel free to add a test case or two and make a note on the talk page.
00:27 shortcircuit Also, thanks a bunch for all the Perl6 code, and I'm pleased to see it help shake bugs out of the language implementation. :)
00:28 lichtkind joined #perl6
00:28 masak thanks for Rosetta Code! :)
00:28 shortcircuit Hey, I just run the servers. It's hackers who add the code. :)
00:29 masak shortcircuit++ # running the servers
00:29 drbean joined #perl6
00:29 HarryS joined #perl6
00:30 lichtkind :)
00:30 lichtkind left #perl6
00:31 TimToady shortcircuit: well, for sure the specced dataset would not have flushed out the problem which assumed nothing of the form /foo/bar/foo existed
00:33 TimToady but yes, it has been very helpful to have a set of problems to work against
00:33 TimToady I've done more Perl 6 programming in the last two weeks than in the previous two years, unless you count the standard grammar
00:34 shortcircuit AFAIK, two other languages (Tcl and Seed7 ) have seen improvements resulting from using RC as a workset, and one other is using it in its design phase.
00:34 TimToady well, since we're still in the design phase, us too :)
00:35 shortcircuit That kind of thing gives me no end of warm opcodes. :)
00:35 HarryS left #perl6
00:36 Sanitoeter joined #perl6
00:37 araujo left #perl6
00:38 [particle] joined #perl6
00:39 HarryS joined #perl6
00:42 * PerlJam wonders what it means to *not* be in the design phase.
00:42 FurnaceBoy PerlJam: that's called Boxing Day!
00:43 colomon TimToady++
00:43 PerlJam FurnaceBoy: There will be no boxing day, only perpetual christmas.
00:43 FurnaceBoy PerlJam: I know :)
00:44 espadrine left #perl6
00:45 HarryS left #perl6
00:48 lue hello o/
00:49 masak hellue \o
00:49 HarryS joined #perl6
00:50 espadrine joined #perl6
00:55 HarryS left #perl6
00:58 lue .oO(how insane is it to think writing a Perl 6 frontend for gcc is fun?)
00:59 masak lue: sounds like the regular level of insane here.
00:59 HarryS joined #perl6
01:00 tylercurtis lue: If you're not picky about your choice of code generator, you could help me work on Bennu (once I start doing so again). :)
01:03 shortcircuit lue: Wouldn't you need to statically-link an interpreter for eval?
01:05 HarryS left #perl6
01:06 lue I read trying to turn a language-dependant AST into something gcc can understand is the best way to go, which is where I got the idea to use STD for the language parsing.
01:06 lue [ Remember: no guarantees I'll actually do it, though :) ]
01:07 masak it's entirely up to you :)
01:07 masak I've had similar thoughts with Yapsi: putting STD as the frontend and C code/gcc as the backend.
01:08 lue It all depends if I can actually get STD running right now :)
01:08 masak there's a version on CPAN that you can install.
01:08 * \xF0 giggles at STD
01:08 TimToady just don't giggle infectiously...
01:09 masak it's a STD pun around here...
01:09 FurnaceBoy sounds more like a sepulchral unhinged evil genius laugh
01:09 HarryS joined #perl6
01:11 lue I would prefer to develop a perl6-gcc frontend locally, on my 12-year old laptop... it's going to take a while :)
01:11 tylercurtis Some categories of typical Lisp macros seem easy to express as block-taking functions in Perl 6. For example, with-open-file, with-accessors, some looping forms.
01:14 lue can the CPAN version be used in Perl 6 code? [or is it just a convenient way to install STD?]
01:16 HarryS left #perl6
01:17 tylercurtis lue: I don't think there are many Perl 6 implementations that can actually run STD. More accurately, I'm not sure if any other than viv can. Niecza is close, I think.
01:17 araujo joined #perl6
01:18 redicaps joined #perl6
01:18 redicaps left #perl6
01:18 masak left #perl6
01:19 HarryS joined #perl6
01:19 lue The second one is able, it'll be a special day.
01:20 araujo left #perl6
01:20 araujo joined #perl6
01:22 tylercurtis phenny: ask sorear Are there any LHF tasks I could work on for Niecza?
01:22 phenny tylercurtis: I'll pass that on when sorear is around.
01:23 araujo left #perl6
01:23 araujo joined #perl6
01:26 HarryS left #perl6
01:28 cjk101010 left #perl6
01:29 HarryS joined #perl6
01:33 whiteknight left #perl6
01:34 araujo left #perl6
01:34 HarryS left #perl6
01:35 ingy what is the proper perl6 executable to run on feather?
01:35 ingy should I build my own?
01:39 HarryS joined #perl6
01:40 cbk It there a special regex metachar that matches numbers that may have a decimals in them?  I need to match numbers like: 1, 5, 5.77, 10.345, etc...
01:41 nimiezko left #perl6
01:41 cbk I was doing /\d+.\d+/ but it's not working too good.
01:42 TimToady esp since . matches anything...
01:42 FurnaceBoy cbk: escape the .
01:42 gfx joined #perl6
01:42 meinwald5 is now known as meinwald
01:42 tylercurtis Even with the escaping, that won't match 1 or 5. Or 1. or .5
01:42 * tylercurtis should have used quotation marks.
01:42 FurnaceBoy tylercurtis: and it goes on from there...
01:42 FurnaceBoy tylercurtis: signs? exponent? ....... :)
01:42 cbk thats what I was haven' problems with
01:44 cbk I want the user of my perl6 program to be able to enter a data set using any number (including numbers with decimals) separated by a space.
01:44 TimToady see token dec_number in STD.pm6
01:45 cbk TimToady, ok
01:45 cbk this is my work in progress so far. (doesn't work yet)
01:45 cbk my @userData = $userInput.comb(/ \d+.\d+/);
01:46 HarryS left #perl6
01:47 TimToady combing for /\d+ ['.' \d+]?/  would be a better start
01:47 cbk yes im using .comb instead of .words
01:48 cbk don't I have to use <> when using []'s?
01:48 TimToady not if it's not a character class
01:48 cbk ie <[1..9]>
01:48 TimToady <[a..z]> is only for character classes
01:48 cbk oh ok.
01:48 TimToady [ ] is just ( ) that don't capture now
01:49 cbk so the [] is optional in the match?
01:49 TimToady noms &
01:49 HarryS joined #perl6
01:50 cbk and the ? is zero or more times
01:51 lue IIRC, ? means 'zero or once', + means 'once or more', * means 'zero or more'
01:53 Quadrescence lue gets a cleany star for good work
01:56 HarryS left #perl6
01:59 HarryS joined #perl6
02:05 HarryS left #perl6
02:08 |CTAPOMAK| joined #perl6
02:09 HarryS joined #perl6
02:14 jferrero left #perl6
02:15 HarryS left #perl6
02:18 Italian_Plumber left #perl6
02:19 HarryS joined #perl6
02:26 HarryS left #perl6
02:29 HarryS joined #perl6
02:31 leprevost joined #perl6
02:34 araujo joined #perl6
02:35 HarryS left #perl6
02:39 HarryS joined #perl6
02:44 Jose___ joined #perl6
02:45 HarryS left #perl6
02:47 Jose___ is now known as YAWPP
02:48 alester joined #perl6
02:48 colomon huh.  the basic Range spec is very off from the current implementation.  :\
02:49 HarryS joined #perl6
02:52 colomon also, current spec seems a bit off.
02:53 FurnaceBoy left #perl6
02:54 HarryS left #perl6
02:54 colomon at least....
02:54 colomon rakudo: say Array ~~ Iterable
02:54 p6eval rakudo 6ccdcd: OUTPUT«1␤»
02:59 HarryS joined #perl6
03:01 colomon How do you reconcile "It is specifically illegal to use anything that does Iterable as implicitly numeric..." with "Hence, Array types in the second argument are assumed to be intended as numeric if the left argument is numeric..."
03:02 colomon I mean, given that Arrays are Iterable...
03:03 lue Good puzzle... I say you can't. The spec did say 0 ... 2,$x was the right way to count every other item up to $x, before TimToady fixed it.
03:04 colomon that's an entirely different section.  but yeah, it's not perfect...
03:05 HarryS left #perl6
03:05 araujo left #perl6
03:06 araujo joined #perl6
03:08 araujo left #perl6
03:09 HarryS joined #perl6
03:10 araujo joined #perl6
03:10 colomon I'm not really sure I see the point in coercing numeric ranges to "the wider type".  And "wider type" isn't well-defined, as far as I can see.
03:11 colomon and I'm tired and repeating myself.
03:12 araujo left #perl6
03:12 araujo joined #perl6
03:13 Holy_Cow joined #perl6
03:14 YAWPP left #perl6
03:15 Holy_Cow left #perl6
03:15 HarryS left #perl6
03:19 HarryS joined #perl6
03:21 sorear Who could I bug about setting up nightly builds for Rakudo?
03:21 phenny sorear: 01:22Z <tylercurtis> ask sorear Are there any LHF tasks I could work on for Niecza?
03:22 sorear I'm looking for someone with 2GB+ of RAM and 24/7 uptime who can compile Rakudo regularly and upload gzipped perl6.pbc somewhere
03:22 sorear perl6.pbc files compress extremely well, don
03:23 sorear 't be scared off by the 15MB...
03:23 sorear a lot of us would appreciate it, I think
03:24 HarryS left #perl6
03:26 cbk left #perl6
03:26 exodist left #perl6
03:27 exodist joined #perl6
03:27 sorear hmm... LHF
03:29 HarryS joined #perl6
03:34 leprevost left #perl6
03:34 espadrine left #perl6
03:36 HarryS left #perl6
03:39 HarryS joined #perl6
03:46 HarryS left #perl6
03:49 HarryS joined #perl6
03:56 HarryS left #perl6
03:59 HarryS joined #perl6
04:01 fod left #perl6
04:06 HarryS left #perl6
04:07 [particle] sorear: get a feather account and set it up!
04:09 HarryS joined #perl6
04:11 cjk101010 joined #perl6
04:14 HarryS left #perl6
04:19 HarryS joined #perl6
04:25 sorear [particle]: I don't think feather has nearly enough memory
04:25 sorear ooo, it's been upgraded
04:26 HarryS left #perl6
04:29 HarryS joined #perl6
04:34 HarryS left #perl6
04:38 HarryS joined #perl6
04:39 sftp left #perl6
04:39 sorear tylercurtis: I can't think of any offhand
04:41 sorear what.
04:41 sorear there appear to be *91* Parrot checkouts on feather1
04:42 sorear ok, only 46; /data was causing double counts
04:42 sorear there are 3 rakudo trees
04:46 HarryS left #perl6
04:47 HarryS joined #perl6
04:47 sorear HarryS: eh?
04:47 hans__ joined #perl6
04:52 HarryS hello
04:53 hans__ Hello, how to connect to a postresql-database when there is no need of a password? my $dbh = MiniDBI.connect( $dsn, $user, $password, RaiseError => 1, PrintError => 1, AutoCommit => 0);
04:57 sorear Juerd: feather1 is currently under painfully heavy load due to googlebot hammering svn.pugscode.org's web-svn-diff-thingy
04:57 sorear Juerd: if I were to add it to robots.txt, would that be good?
04:58 sorear Cpu(s):  0.3%us,  0.6%sy,  0.1%ni, 47.7%id, 51.2%wa,  0.0%hi,  0.0%si,  0.1%st
04:58 sorear mmm thrashing
04:58 sorear several 'D' processes
05:11 gfx left #perl6
05:24 diakopter sorear: yeah it'd be ok to add /svn to robots.txt
05:26 alester left #perl6
05:37 alester joined #perl6
05:38 alester left #perl6
05:39 kaare joined #perl6
05:40 kaare is now known as Guest83575
05:42 hans__ left #perl6
05:43 vks joined #perl6
05:44 vks left #perl6
05:57 moritz_ good morning
06:02 sorear hello
06:06 uniejo joined #perl6
06:07 justatheory left #perl6
06:14 wtw joined #perl6
06:16 dream joined #perl6
06:17 crythias left #perl6
06:22 cjk101010 left #perl6
06:26 dual left #perl6
06:36 mberends joined #perl6
06:41 mberends left #perl6
06:42 sorear Why are svn checkouts of Parrot run from feather so SLOW
06:42 sorear they take like 10+ minutes
06:42 sorear rakudo checkout: 4s
06:49 simcop2387 left #perl6
06:50 simcop2387 joined #perl6
07:05 tylercurtis left #perl6
07:06 Ross joined #perl6
07:09 drbean left #perl6
07:14 sorear wtf, I think my pbc_to_exe is broken
07:14 sorear parrot perl6.pbc -> works fine
07:14 sorear ./perl6 -> PackFile_unpack: This Parrot cannot read bytecode files with version 8.2.
07:15 Ross left #perl6
07:18 dju left #perl6
07:18 Trashlord left #perl6
07:19 dju joined #perl6
07:21 Guest83575 is now known as kaare_
07:25 kaare_ Anyone knows who's managing http://planetsix.perl.org - and why it doesn't pick up from rakudo.org?
07:25 kcwu left #perl6
07:25 zag joined #perl6
07:25 kaare_ Should pmichaud change the feed from use.perl.org?
07:27 moritz_ kaare_: most rakudo.org posts are also sent as e-mail to the perl6-announce list, which planetsix picks up
07:27 dju left #perl6
07:28 dju joined #perl6
07:28 kaare_ moritz_: It missed Rakudo Star for august
07:30 kcwu joined #perl6
07:31 moritz_ kaare_: that's unfortunate
07:32 kaare_ That's why I bring it up :-)
07:33 kaare_ I've been busy @ work so I haven't been following here. Just now I looked around and saw it.
07:34 kaare_ I guess that if I miss it, a lot of other people will too.
07:35 * moritz_ not sure what the best solution is
07:36 kcwu left #perl6
07:38 kcwu joined #perl6
07:57 dju left #perl6
07:58 dju joined #perl6
07:58 kcwu left #perl6
07:59 kcwu joined #perl6
08:02 kcwu left #perl6
08:03 kcwu joined #perl6
08:06 Trashlord joined #perl6
08:23 dakkar joined #perl6
08:31 agentzh joined #perl6
08:42 pfusik joined #perl6
08:44 pfusik left #perl6
08:44 amkrankruleuen joined #perl6
08:50 dakkar left #perl6
08:55 cjk101010 joined #perl6
08:56 dakkar joined #perl6
09:01 zulon joined #perl6
09:08 timbunce joined #perl6
09:18 Guest75712 left #perl6
09:20 Guest75712 joined #perl6
09:21 cjk101010 left #perl6
09:22 cjk101010 joined #perl6
09:24 dual joined #perl6
09:27 sjohnson joined #perl6
09:35 dream left #perl6
10:02 bbkr rakudo: / $. / # another one not accepted by STD
10:02 p6eval rakudo 6ccdcd:  ( no output )
10:02 bbkr std: / $. / # another one not accepted by STD
10:02 p6eval std 32120: OUTPUT«[31m===[0mSORRY![31m===[0m␤Unsupported use of $. variable; in Perl 6 please use the filehandle's .line method at /tmp/qiax0mJToC line 1:␤------> [32m/ $.[33m⏏[31m / # another one not accepted by STD[0m␤Parse failed␤FAILED 00:02 116m␤»
10:03 agentzh left #perl6
10:07 azert0x joined #perl6
10:13 moritz_ phenny: tell masak re proto tests, my rakudo was more broken than I thought it was... anyway, fix for tt #1746 pushed, feel free to apply your enum patch
10:13 phenny moritz_: I'll pass that on when masak is around.
10:17 dalek rakudo: f8e959c | nwellnhof++ | src/ops/perl6.ops:
10:17 dalek rakudo: Fix memory handling in rebless_subclass
10:17 dalek rakudo:
10:17 dalek rakudo: Se tt #1746 for details
10:17 dalek rakudo: review: http://github.com/rakudo/rakudo/commit/f​8e959c1f416a4615a72ce1e21ad2dd2535f4f47
10:35 sftp joined #perl6
10:50 espadrine joined #perl6
10:54 daxim joined #perl6
11:08 PZt left #perl6
11:15 jferrero joined #perl6
11:16 Kodi joined #perl6
12:25 _ilbot2 joined #perl6
12:25 Topic for #perl6 is now »ö« Welcome to Perl 6! | http://perl6.org/ | evalbot usage: 'perl6: say 3;' or rakudo:, niecza:, std:, or /msg p6eval perl6: ... | irclog: http://irc.pugscode.org/ | UTF-8 is our friend! | Rakudo Star Released!
12:26 JimmyZ it still is down?
12:26 moritz no
12:26 moritz left #perl6
12:27 JimmyZ I couldn't visit it :(
12:28 moritz_ joined #perl6
12:28 lichtkind joined #perl6
12:32 smash hello everyone
12:33 lichtkind hi smash
12:33 lichtkind lines is lazy as for right?
12:35 colomon lines is lazy.
12:36 lichtkind colomon: thank you
12:36 colomon you're welcome.
12:37 PZt joined #perl6
12:37 moritz_ IRC logs are finally back
12:37 colomon \o/
12:38 moritz_ (and my server, for that matter :-)
12:41 JimmyZ moritz_++
12:43 ash_ left #perl6
12:51 moritz_ http://rosettacode.org/wiki/Bead_Sort the Haskell version looks nice; I don't quite understand it though
12:51 moritz_ if anybody does, porting in to Perl 6 shouldn't be too hard
12:52 Holy_Cow joined #perl6
12:55 M_o_C joined #perl6
12:56 smash moritz_: let me see that, i love haskell
12:57 moritz_ import Data.List
12:57 moritz_ beadSort :: [Int] -> [Int]
12:57 moritz_ beadSort = map sum. transpose. transpose. map (flip replicate 1)
12:59 smash (yeap, loogink at the site, thks)
12:59 smash s/loogink/looking/
12:59 daxim I wrote like 95% of the Perl for that wiki
13:00 moritz_ well, TimToady is working hard to change that :-)
13:02 moritz_ otoh his solutions are often so short that it's probably a small percentage, line-wise :-)
13:04 daxim I'd like to see a data structures cookbook for p6
13:04 moritz_ then write it :-)
13:04 daxim a recursive impossibility
13:05 JimmyZ left #perl6
13:05 daxim I neglected to mention: I'd like to see it to read and learn
13:05 Kodi I wrote like 95% of the Perl 6 (before y'all started contributing).
13:07 smash moritz_: ahh, excelent piece of haskell
13:07 smash very nice indeed
13:08 moritz_ why does it transpose twice? I'd kinda expect it to be the identitity operation
13:09 smash moritz_: let me show you an example, easier than trying to explain i guess
13:09 orafu left #perl6
13:09 orafu joined #perl6
13:12 moritz_ rakudo: say 'a' ~~ /(a) && (.)/; say $0; say $1
13:12 p6eval rakudo f8e959: OUTPUT«a␤␤␤»
13:13 moritz_ anybody want to submit a rakudobug? :-)
13:15 timbunce left #perl6
13:15 meppl joined #perl6
13:17 smash moritz_: http://gist.github.com/558997
13:17 sftp_ joined #perl6
13:17 sftp left #perl6
13:18 tadzik beautiful
13:19 timbunce joined #perl6
13:20 moritz_ so the flip basically corresponds to the p6 R meta operator?
13:21 moritz_ so (flip replicate 1) is   1 Rxx   right?
13:23 Kodi left #perl6
13:24 moritz_ rakudo: say (map {1 xx $_ }, 4, 1, 3, 2)
13:24 p6eval rakudo f8e959: OUTPUT«1111111111␤»
13:24 moritz_ rakudo: say (map {1 xx $_ }, 4, 1, 3, 2).perl
13:24 p6eval rakudo f8e959: OUTPUT«((1, 1, 1, 1), (1), (1, 1, 1), (1, 1))␤»
13:24 moritz_ rakudo: say ([Z] map {1 xx $_ }, 4, 1, 3, 2).perl
13:25 p6eval rakudo f8e959: OUTPUT«maximum recursion depth exceeded␤  in 'Block::count' at line 5773:CORE.setting␤  in 'Block::count' at line 5777:CORE.setting␤  in 'Any::join' at line 1␤  in 'List::perl' at line 2598:CORE.setting␤  in <anon> at line 2598:CORE.setting␤  in 'Any::join' at line 1␤  in
13:25 p6eval ..'List::perl' at …
13:25 moritz_ ew.
13:25 tadzik rakudo: say ([.elems] map {1 xx $_ }, 4, 1, 3, 2).perl
13:25 tadzik oh, noes
13:25 p6eval rakudo f8e959: OUTPUT«===SORRY!===␤Unable to parse postcircumfix:sym<( )>, couldn't find final ')' at line 22␤»
13:25 smash rakudo: say (map {1 xx $_ }, 4, 1, 3, 2).sort.perl
13:25 p6eval rakudo f8e959: OUTPUT«((1), (1, 1), (1, 1, 1), (1, 1, 1, 1))␤»
13:26 tadzik rakudo: say ( map {1 xx $_ }, 4, 1, 3, 2).map( .elems ).perl
13:26 p6eval rakudo f8e959: OUTPUT«No applicable candidates found to dispatch to for 'map'. Available candidates are:␤:(Mu : &block;; *%_)␤␤  in main program body at line 22:/tmp/tzkTwY2Gm6␤»
13:26 tadzik -_-
13:26 moritz_ *.elems
13:26 tadzik rakudo: say ( map {1 xx $_ }, 4, 1, 3, 2).map( *.elems ).perl
13:26 tadzik yeah
13:26 p6eval rakudo f8e959: OUTPUT«(4, 1, 3, 2)␤»
13:26 tadzik duh, too early
13:27 smash you're missing the sort of the inner lists
13:27 tadzik we're missing this transpose-transpose magic I think
13:27 moritz_ right
13:27 smash rakudo: say (map {1 xx $_ }, 4, 1, 3, 2).sort.map( *.elems ).perl
13:27 p6eval rakudo f8e959: OUTPUT«(1, 2, 3, 4)␤»
13:27 moritz_ a "normal" matrix transpose can be written as [Z], if rakudo doesn't suck
13:28 * smash is cheating :P
13:28 moritz_ but that won't work for entries of different length
13:28 moritz_ so we'd have to fill with 0s
13:28 uniejo left #perl6
13:28 tadzik or add a transpose function/method to be as awesome as haskell
13:30 moritz_ or steal one of the other Bead Sort implementations for Perl 6
13:31 smash 'maximum recursion depth exceeded' :)
13:32 espadrine left #perl6
13:33 * [Coke] wonders if perl6 has upped it's recursion depth setting yet.
13:33 [Coke] argh, its
13:33 moritz_ nope
13:33 moritz_ we want to fail fast :-)
13:34 kiffin joined #perl6
13:34 kiffin left #perl6
13:36 kaare_ left #perl6
13:37 colomon rakudo: say [Z] ([[1, 2, 3], [4, 5, 6], [7, 8, 9]]).perl
13:37 colomon no!
13:37 p6eval rakudo f8e959: OUTPUT«No applicable candidates found to dispatch to for 'infix:<Z>'. Available candidates are:␤:()␤:(Any $lhs, Any $rhs)␤␤  in 'reducewith' at line 312:CORE.setting␤  in main program body at line 1␤»
13:37 M_o_C left #perl6
13:37 colomon rakudo: say ([Z] [[1, 2, 3], [4, 5, 6], [7, 8, 9]]).perl
13:37 p6eval rakudo f8e959: OUTPUT«No applicable candidates found to dispatch to for 'infix:<Z>'. Available candidates are:␤:()␤:(Any $lhs, Any $rhs)␤␤  in 'reducewith' at line 312:CORE.setting␤  in main program body at line 1␤»
13:39 colomon oh.  flattens?
13:39 colomon rakudo: say (flat([[1, 2, 3], [4, 5, 6], [7, 8, 9]])).perl
13:40 p6eval rakudo f8e959: OUTPUT«([[1, 2, 3], [4, 5, 6], [7, 8, 9]])␤»
13:40 moritz_ rakudo: say infix:<Z>([1, 2, 3])
13:40 p6eval rakudo f8e959: OUTPUT«No applicable candidates found to dispatch to for 'infix:<Z>'. Available candidates are:␤:()␤:(Any $lhs, Any $rhs)␤␤  in main program body at line 22:/tmp/xetWZ99ZVE␤»
13:40 moritz_ rakudo: say infix:<Z>([1, 2, 3], [4, 5, 6])
13:40 p6eval rakudo f8e959: OUTPUT«142536␤»
13:40 moritz_ rakudo: say infix:<Z>([1, 2, 3], [4, 5, 6], [7, 8, 9])
13:40 p6eval rakudo f8e959: OUTPUT«No applicable candidates found to dispatch to for 'infix:<Z>'. Available candidates are:␤:()␤:(Any $lhs, Any $rhs)␤␤  in main program body at line 22:/tmp/qr6y_7vYSj␤»
13:40 espadrine joined #perl6
13:41 rgrau_ joined #perl6
13:45 colomon rakudo: say [Z]([1, 2, 3], [4, 5, 6], [7, 8, 9])
13:45 p6eval rakudo f8e959: OUTPUT«1 7 4 8 2 9␤»
13:46 espadrine left #perl6
13:46 colomon Is that actually correct?
13:47 moritz_ don't think so
13:47 smash colomon: depends on what you're trying to do
13:47 colomon it's [1, 4, 2, 6, 3, 6] Z [7, 8, 9]
13:47 moritz_ the spec says that [op] a, b, c  should behave like a op b op c
13:47 moritz_ and since Z is supposed to have list associativity, that's wrong
13:47 colomon reduce meta doesn't even look at list associativity
13:48 espadrine joined #perl6
13:48 colomon and of course, infix:<Z>([1, 2, 3], [4, 5, 6], [7, 8, 9]) doesn't actually work
13:48 moritz_ right; it needs slices for proper implementation
13:49 Mowah joined #perl6
13:50 moritz_ btw I fear I'll miss #cpan6sketch, if there is one
13:50 tadzik I missed it last time and there was no #cpan6sketch :)
13:54 shortcircuit moritz_, daxim: The way Rosetta Code is currently organized, each page is a chrestomathy. Slice it sideways, and you've got three hundred or so cookbooks. I don't see why we couldn't have a few more task pages for data structures. We've got "Compound Data Type" "Enumearation", "Array" and "Associative array" so far.
13:55 shortcircuit We've got enough task pages that I don't try to keep track of them all in my head any more.
13:55 perlygatekeeper joined #perl6
13:57 uniejo joined #perl6
13:58 moritz_ http://perldoc.perl.org/perlfaq4.html might be a nice source of inspiration for more task
14:04 drbean left #perl6
14:05 |CTAPOMAK| left #perl6
14:10 JimmyZ joined #perl6
14:13 fridim left #perl6
14:14 zag left #perl6
14:16 envi^home joined #perl6
14:19 wtw left #perl6
14:20 jferrero left #perl6
14:24 rgrau_ left #perl6
14:25 bbkr rakudo: say "" ~~ /$/ # bug? in p5 it matches
14:25 p6eval rakudo f8e959: OUTPUT«␤»
14:25 bbkr nvm, works fine
14:26 molaf joined #perl6
14:27 molaf left #perl6
14:27 moritz_ rakudo: say so "" ~~ /$/
14:27 p6eval rakudo f8e959: OUTPUT«1␤»
14:27 bbkr rakudo: say "OK" if "0" ~~ /0/ # tricky one..
14:28 p6eval rakudo f8e959: OUTPUT«OK␤»
14:28 bbkr \o/
14:28 JimmyZ left #perl6
14:28 bbkr what is this "say so" ? looks like a nice trick
14:29 moritz_ prefix:<so> is boolean context. Just like 'not', but without the negation
14:29 rgrau_ joined #perl6
14:29 bbkr rakudo: say so 4# expecting 1...
14:29 p6eval rakudo f8e959: OUTPUT«1␤»
14:29 bbkr rakudo++
14:34 patspam joined #perl6
14:40 Guest23195 joined #perl6
14:43 bbkr rakudo: say so "a" ~~ /<<*/ # timeout, but I think already known as /""*/ case
14:43 p6eval rakudo f8e959: OUTPUT«(timeout)»
14:43 moritz_ yes, there's a ticket for quantified zero-width matches
14:44 justatheory joined #perl6
14:46 bbkr matching longest alternative is NYI, right? like in "a bb c" ~~ m/a|bb/
14:47 moritz_ right
14:47 moritz_ but notice that this would match a, even with correct LTM implementation
14:48 moritz_ because it always tries the left-most match, even if shorter
14:48 bbkr what does LTM stands for?
14:48 moritz_ Longest Token Matching
14:48 _jaldhar joined #perl6
14:49 tylercurtis joined #perl6
14:50 * bbkr reads spec...
14:51 [particle] rakudo: "a bb c" ~~ m/(bb|a)/; say $1;
14:51 p6eval rakudo f8e959: OUTPUT«␤»
14:54 slavik rakudo: "a bb c" ~~ m/(bb|a)/; say $0;
14:54 p6eval rakudo f8e959: OUTPUT«a␤»
14:54 slavik rakudo: "a bb c" ~~ m/(bb|a)/; say $/.perl;
14:54 p6eval rakudo f8e959: OUTPUT«Match.new(␤ from => 0,␤ orig => "a bb c",␤ to   => 1,␤ positional => [␤   Match.new(␤    from => 0,␤    orig => "a bb c",␤    to   => 1,␤   ),␤  ],␤)␤»
14:54 bbkr [particle]: first match is $0, not $1 like in P5
14:54 slavik and that, too
14:55 slavik $/ is the smart match object that gets created
14:55 patspam left #perl6
14:59 [particle] bad fingers! bad! [particle]-- needs breakfast
15:00 macroron joined #perl6
15:02 patspam joined #perl6
15:15 bbkr rakudo: my $code = "\$" ~ 1 x 1024; eval $code; # bang!
15:16 p6eval rakudo f8e959: OUTPUT«error:imcc:syntax error, unexpected IDENTIFIER ('Inf')␤    in file 'EVAL_10' line 105593236␤»
15:16 tadzik rakudo: my $code = "\$" ~ 1 x 1024; say $code;
15:16 p6eval rakudo f8e959:
15:16 p6eval ..OUTPUT«$111111111111111111111111111111111111111​1111111111111111111111111111111111111111111111111​1111111111111111111111111111111111111111111111111​1111111111111111111111111111111111111111111111111​1111111111111111111111111111111111111111111111111​1111111111111111111111111111111111111111111111…
15:16 tadzik oh
15:17 cls_bsd left #perl6
15:20 [particle] i need a regex to remove trailing slashes, and following newlines if they exist
15:21 shortcircuit moritz_: Another place we keep lists for inspiration for tasks is here: http://rosettacode.org/wiki/Rosetta_Code​:Village_Pump/Suggest_a_programming_task
15:22 [particle] i suppose it's { \\ $ \n? }{}
15:22 ash_ joined #perl6
15:26 bbkr [particle]: those in rx are trailing BACKslashes :)
15:27 [particle] bbkr: yes, that's what i meant, scratch what i said.  still need breakfast, i'm afraid.
15:27 [particle] let me just fix that now.
15:28 moritz_ \/+\n*$$
15:28 bbkr [particle]: also don't use $ but $$
15:30 tadzik given 'yes' { .say so }
15:31 tadzik rakudo: given 'yes' { .say so }
15:31 p6eval rakudo f8e959: OUTPUT«===SORRY!===␤Unable to parse blockoid, couldn't find final '}' at line 22␤»
15:31 tadzik :(
15:31 tadzik rakudo: given 'yes' { say .so }
15:31 p6eval rakudo f8e959: OUTPUT«1␤»
15:33 ingy [Coke]: that's because Ingy never released a version on Inline::C :)
15:33 bbkr rakudo: given 'no' { say .so } # i don't think 'no' is recognized as False
15:33 p6eval rakudo f8e959: OUTPUT«1␤»
15:33 ingy *of
15:34 tadzik rakudo: given 'yes' { say ~.so }
15:35 p6eval rakudo f8e959: OUTPUT«1␤»
15:36 ash_ rakudo: given 0 { say ~.so }
15:36 p6eval rakudo f8e959: OUTPUT«0␤»
15:37 tadzik ash_: how is try.rakudo going?
15:38 masak joined #perl6
15:38 masak oh hai, #perl6!
15:38 phenny masak: 10:13Z <moritz_> tell masak re proto tests, my rakudo was more broken than I thought it was... anyway, fix for tt #1746 pushed, feel free to apply your enum patch
15:38 ash_ moritz_++ made some changes so use FORBID_PIR works in the REPL, so the security concerns are gone, but i ran into a weird issue with IPC::Run connecting the sub-processes STDIN to the default STDIN, so if you sent the daemon a command, it would do '> ' and wait for you to enter it on the server... which is wrong, that hasn't happened before, so i am trying to find out why, also it only happens in linux, not OS X which confuses me more
15:38 smash masak: hi
15:39 masak will review my own patch a bit, then apply it.
15:39 ash_ tadzik: wall of text was about try.rakudo btw
15:39 tadzik ash_: yeah, I see
15:39 tadzik FORBID_PIR caught my eyes
15:40 tadzik bbl
15:40 tadzik left #perl6
15:40 bbkr rakudo: 1 .. 2 .. 3 # known bug? should not be parsed according to STD
15:40 p6eval rakudo f8e959:  ( no output )
15:40 ash_ the fact that IPC::Run only messes up on linux (with same versions of perl, and IPC::Run) is baffling
15:40 moritz_ :(
15:41 moritz_ would it help if I installed a newer or older IPC::Run on that server?
15:41 Guest23195 is now known as _kaare
15:43 ash_ moritz_: once i get it working on my server thats mimicing feather3 (its just a vm) then i'll see about changing whats on feather3, anyway, i  plan on fixing it today, even if i need to scrap IPC::Run, which would be sad, IPC::Run is the only sub-process managing perl module i have found that has re-set able timers and timeouts, most of them only let you set the timeout once, when a program starts, then never again, which isn't what I want, unless i make th
15:43 ash_ more complex (which i'd like to avoid)
15:44 _jaldhar left #perl6
15:47 ingy http://en.wikipedia.org/wiki/Perl_6_rules - There is a notability threat at the top of the page
15:54 masak ingy: in the long run, trying to acheive happiness through Wikipedia cabal policies and decisions will always fail.
15:54 masak doesn't hurt to try, though. just don't put your soul into it.
15:54 bbkr rakudo: my $x; my $y := $x; say $x =:= $y # known bug? it should return true on comparing variables binding, right?
15:54 p6eval rakudo f8e959: OUTPUT«0␤»
15:55 EvanCarroll left #perl6
15:55 masak bbkr: right. don't think it's known.
15:55 moritz_ yes
15:55 moritz_ it is known, and submitted
15:55 takadonet left #perl6
15:55 masak moritz_++
15:56 bbkr damn, I was hoping for 700th ticket in RT :P
15:56 masak (because 699 are not enough?) :)
15:58 isBEKaml joined #perl6
15:58 tylercurtis phenny: tell sorear that he seems to have misspelled supersede as supercede in Niecza::Actions.
15:58 phenny tylercurtis: I'll pass that on when sorear is around.
15:58 ash_ masak: there is some irony that you would say that :P
15:58 masak ash_: *lol*
15:59 isBEKaml .u 20A8
15:59 phenny U+20A8 RUPEE SIGN (₨)
15:59 isBEKaml \o/ perl6++
15:59 ash_ masak: how many tickets do you have? just curious
15:59 masak ash_: somewhere between 960 and 1e3.
15:59 masak that includes a lot of closed ones, of course.
16:00 isBEKaml hi, #perl6!
16:00 masak isBEKaml! \o/
16:00 ingy masak: too true
16:00 EvanCarroll joined #perl6
16:00 EvanCarroll _sri: what ever happened to your blog entry on dispatchers for dummies
16:00 masak ingy: in this case, they're looking for more sources, which doesn't sound too bad.
16:00 EvanCarroll _sri: that was a really great entry, I can't find it now
16:00 isBEKaml masak: What's up? I'm just so happy to see Rupee sign among the unicode symbols on this channel. :)
16:01 isBEKaml .u 20B9
16:01 phenny U+20B9 (No name found)
16:01 Mowah left #perl6
16:01 isBEKaml :(
16:01 isBEKaml This is the latest position for the actual symbol.
16:01 isBEKaml http://en.wikipedia.org/wiki/Indian_rupee_sign
16:01 masak isBEKaml: I'm doing fine! Yapsi release tomorrow, a few things to do until then. :)
16:02 isBEKaml masak: Yapsi! \o/
16:03 ash_ masak++ # keep up the good work, i am not as good at finding bugs
16:04 isBEKaml BTW, something I noticed this morning, the third Rosetta Code solution for FizzBuzz (I know, :D) seems like a cheat. Can someone clarify?
16:05 isBEKaml More than anything, I was rather surprised to see Fizzbuzz among the tasks. :D
16:07 ingy masak: yeah, and I'm sure we could dig up a few :)
16:07 moritz_ which one do you think is cheating? the Ada solution?
16:07 TimToady Yeah, I wondered about that when I looked, but left it in for the entertainment value.
16:07 TimToady last Perl 6
16:07 TimToady doesn't use %%
16:07 moritz_ oh
16:07 isBEKaml yes.
16:08 ash_ left #perl6
16:08 TimToady but the spec doesn't say you have to use modular arithmetic
16:08 * moritz_ doesn't find that cheating at all
16:08 isBEKaml Btw, it doesn't seem to work in my local build.
16:08 moritz_ spirit of timtowtdi etc.
16:08 TimToady I didn't try that one
16:08 isBEKaml rakudo: for ('' xx 2, 'Fizz') xx 34 Z ('' xx 4, 'Buzz') xx 20 Z 1 .. 100 -> $f, $b, $n { say $f ~ $b || $n; }
16:08 p6eval rakudo f8e959: OUTPUT«No applicable candidates found to dispatch to for 'infix:<Z>'. Available candidates are:␤:()␤:(Any $lhs, Any $rhs)␤␤  in main program body at line 22:/tmp/mLbIHvWcjY␤»
16:08 ash_ joined #perl6
16:08 isBEKaml Z and a space?
16:09 moritz_ TimToady: you could even use Z~ insstead of the first Z, and only two closure params
16:09 isBEKaml can space be used as a op? I think it should be ~ for concat, if that's what the solution is about
16:09 TimToady I didn't write it
16:09 moritz_ oh
16:09 TimToady Z defaults to Z, like X defaults to X,
16:10 Juerd sorear: That would be good, but the whole svn thing is one huge disaster
16:10 TimToady current rakudo doesn't do list associative list infixes
16:10 Juerd sorear: Nobody seems to want to maintain it, and I will probably just kill it off.
16:10 TimToady pugs: for ('' xx 2, 'Fizz') xx 34 Z ('' xx 4, 'Buzz') xx 20 Z 1 .. 100 -> $f, $b, $n { say $f ~ $b || $n; }
16:10 p6eval pugs: OUTPUT«  1  2␤  4 Buzz 5␤  7  8␤ Buzz 10  11␤  13  14␤  16  17␤  19 Buzz 20␤  22  23␤ Buzz 25  26␤  28  29␤  31  32␤  34 Buzz 35␤  37  38␤ Buzz 40  41␤  43  44␤  46  47␤  49 Buzz 50␤  52  53␤ Buzz 55  56␤  58  59␤  61  62␤  64 Buzz 65␤  67  68␤ Buzz 70  71␤  73  74␤  76  77␤  79 Buzz 80␤ …
16:10 Ross joined #perl6
16:11 Juerd sorear: The svn thing in Apache is a source of enormous memory leaks too
16:11 Juerd sorear: And indeed, it's not web spider proof.
16:11 isBEKaml no Fizz?
16:11 isBEKaml :)
16:11 Juerd If nobody comes with a solution, that just means that it'll die.
16:12 moritz_ rakudo: for ('' xx 2, 'Fizz') xx * Z~ ('' xx 4, 'Buzz') xx * Z 1..100 -> $f, $n { say $f || $n }
16:12 p6eval rakudo f8e959: OUTPUT«No applicable candidates found to dispatch to for 'infix:<Z>'. Available candidates are:␤:()␤:(Any $lhs, Any $rhs)␤␤  in main program body at line 22:/tmp/pVqW08LMCi␤»
16:12 zulon joined #perl6
16:12 TimToady flat champagne, all buzz, no fizz
16:12 moritz_ Juerd: would it make you happy if I migrated the pugs repo to git, and moved it github?
16:12 tylercurtis rakudo: .say for (('' xx 2, 'Fizz') xx 34 Z~ ('' xx 4, 'Buzz') xx 20) Z|| 1 .. 100;
16:12 p6eval rakudo f8e959:
16:12 p6eval ..OUTPUT«1␤2␤Fizz␤4␤Buzz␤Fizz␤7␤8␤Fizz␤Buzz␤11␤Fi​zz␤13␤14␤FizzBuzz␤16␤17␤Fizz␤19␤Buzz␤Fizz␤22␤23␤F​izz␤Buzz␤26␤Fizz␤28␤29␤FizzBuzz␤31␤32␤Fizz␤34␤Buz​z␤Fizz␤37␤38␤Fizz␤Buzz␤41␤Fizz␤43␤44␤FizzBuzz␤46␤​47␤Fizz␤49␤Buzz␤Fizz␤52␤53␤Fizz␤Buzz␤56␤Fizz␤58␤5​9␤FizzBuzz␤61␤62␤Fizz␤64␤Buzz␤Fizz␤67␤68␤Fizz�
16:12 moritz_ then we'll have to come up with a clever solution for the commitbit thing
16:13 Juerd moritz_: Yes. Moving it isn't even necessary. Anything that will tune it down to /normal/ resource usage is good enough
16:13 Juerd moritz_: Also, I believe that something like this belongs on feather2
16:14 TimToady std: .say for (('' xx 2, 'Fizz') xx 34 Z~ ('' xx 4, 'Buzz') xx 20) Z|| 1 .. 100;
16:14 moritz_ Juerd: It does, but I don't dare to migrate it. Too many bad experiences with apache+svn+jifty (for commitbit)
16:14 p6eval std 32120: OUTPUT«ok 00:01 118m␤»
16:14 Juerd moritz_: Currently feather1 can be dossed with just a few http requests, and is dossed regularly by googlebot. That's bad, and hardly googlebot's fault :)
16:14 EvanCarroll left #perl6
16:14 Juerd moritz_: Nobody dares migrate it.
16:14 Juerd +to
16:14 isBEKaml tylercurtis++
16:15 Juerd But without anyone to maintain it, we should ask ourselves if this setup should even be used.
16:18 moritz_ Juerd: currently it seems to me that a migration to git+github is easier than finding a volunteer who maintains svn on feather
16:18 moritz_ so I'll look harder into migration strategies
16:18 Juerd Agreed.
16:18 Juerd I think that my point was that git + gitonfeather would work too
16:18 ingy Juerd: is there a star executable on feather?
16:19 Juerd ingy: If you build one :)
16:19 moritz_ Juerd: yes, but it would also suffer from the maintenance problem
16:19 ingy ok, I just didn't want to waste resources
16:19 Juerd ingy: Currently nobody provides a regurarly updated rakudo
16:19 Juerd moritz_: True.
16:20 ingy Juerd: have /usr/bin/perl6 be the latest star sounds like a good idea...
16:20 * moritz_ thinks any perl6 in $PATH will confuse the crap out of people
16:20 ash_ ingy: which feather are you on? I know on feather3 it won't build due to resource limitations, but if you build it locally, then copy the core.pir file (thats the one that takes to much memory) and modify the makefile, you can get it to build on feather3
16:21 moritz_ rather do a /opt/rakudo-star-2010.08/bin/perl6
16:21 [Coke] harmil?
16:21 moritz_ any maybe an /opt/rakudo/bin/perl6 which symlinks there
16:21 ash_ perl6-on-feather++
16:21 moritz_ and tell people to use that if they want it
16:21 TimToady std: .say for (('' xx 2, 'Fizz') xx * Z~ ('' xx 4, 'Buzz') xx *) Z|| 1 .. 100;
16:21 p6eval std 32120: OUTPUT«ok 00:01 118m␤»
16:21 TimToady rakudo: .say for (('' xx 2, 'Fizz') xx * Z~ ('' xx 4, 'Buzz') xx *) Z|| 1 .. 100;
16:21 p6eval rakudo f8e959:
16:21 p6eval ..OUTPUT«1␤2␤Fizz␤4␤Buzz␤Fizz␤7␤8␤Fizz␤Buzz␤11␤Fi​zz␤13␤14␤FizzBuzz␤16␤17␤Fizz␤19␤Buzz␤Fizz␤22␤23␤F​izz␤Buzz␤26␤Fizz␤28␤29␤FizzBuzz␤31␤32␤Fizz␤34␤Buz​z␤Fizz␤37␤38␤Fizz␤Buzz␤41␤Fizz␤43␤44␤FizzBuzz␤46␤​47␤Fizz␤49␤Buzz␤Fizz␤52␤53␤Fizz␤Buzz␤56␤Fizz␤58␤5​9␤FizzBuzz␤61␤62␤Fizz␤64␤Buzz␤Fizz␤67␤68␤Fizz�
16:22 * moritz_ wants to open an organization account for Perl 6 on github. What name should I use?
16:22 TimToady no need to count the inner zips
16:22 moritz_ 'Mu' and 'perl6' are taken
16:22 ingy ash_: I asked for an account on feather, because I couldn't build on my slicehost. thanks for the tip.
16:22 Juerd ingy: Great idea, even. Will you do that? :)
16:22 ingy Juerd: right away
16:22 ingy :)
16:22 moritz_ 'perlsix' is free
16:22 EvanCarroll joined #perl6
16:22 Juerd ingy: Where shall I symlink it to?
16:23 cdarroch joined #perl6
16:23 cdarroch left #perl6
16:23 cdarroch joined #perl6
16:23 ash_ perl6 is taken as an organization on github? :-(
16:23 moritz_ ash_: no, as a user
16:23 _jaldhar joined #perl6
16:23 ash_ oh, got ya
16:23 ingy Juerd: ~ingy/local/bin/
16:23 ingy Juerd: ~ingy/local/bin/perl6
16:23 moritz_ ash_: currently the 'book' and 'perl6-examples' accounts are in there
16:24 ash_ right right, i understand, i just thought someone had taken the name to be a jerk, but i am wrong it seems
16:25 araujo left #perl6
16:25 moritz_ pmichaud++ took it long before github supported organization accounts
16:26 araujo joined #perl6
16:26 TimToady tylercurtis: are you updating rosettacode or shall I?
16:26 tylercurtis TimToady: I shall.
16:26 lichtkind_ joined #perl6
16:26 TimToady see my *'s above?
16:27 tylercurtis Yes.
16:27 lichtkind left #perl6
16:27 lichtkind_ is now known as lichtkind
16:27 moritz_ http://github.com/perlsix/Mu # the new pugs repo
16:28 Juerd ingy: Done
16:28 moritz_ next step: inform all people with pugs commit bits about it, and add them to the repo if they have a github ID
16:28 araujo left #perl6
16:28 TimToady oh well
16:29 moritz_ TimToady: resistance is futile :-)
16:29 smash moritz_: i give up on that transpose thing (for now at least) :-)
16:29 [particle] moritz_: new bugs repo?
16:29 [particle] HA, silly fingers!
16:29 moritz_ lol
16:29 [particle] pugs? when did that decision happen?
16:29 araujo joined #perl6
16:29 TimToady smash: just do the "Pattern_matching" entry instead (red-black tree matching)
16:30 tylercurtis TimToady: updated it.
16:30 moritz_ [particle]: just now
16:31 moritz_ [particle]: nobody maintains the svn server, so we need to find another solution
16:31 smash TimToady: sorry, which one !?
16:32 [particle] ok, then.  i'd like a commit bit and privs to add new committers, if you wouldn't mind.
16:32 [particle] i r particle
16:32 TimToady smash: http://rosettacode.org/wiki/Pattern_matching
16:32 moritz_ sure
16:33 TimToady smash: I believe signature matching is up to the task
16:33 * szbalint starts following gitpug
16:33 TimToady only one pug?
16:33 isBEKaml oh, that was fast. All the commits are retained! \o/
16:34 smash TimToady: thank you, let me look at it
16:35 moritz_ isBEKaml: nonono, it'll take much longer
16:35 moritz_ all the projects that use pugs checkouts need some time to switch to the new scheme
16:35 szbalint The naughty pug is the one that subverts the rest of the pugs. :)
16:35 alester joined #perl6
16:35 isBEKaml moritz_: I was wondering when I saw the last commit by patrickas++ on 30 Aug.
16:37 shortcircuit Just a general FYI. This page should be useful/interesting: http://rosettacode.org/wiki/Report​s:Tasks_not_implemented_in_Perl_6
16:37 * moritz_ also found that :-)
16:37 TimToady yes, nice page
16:38 molaf joined #perl6
16:39 * TimToady is trying to figure out how we get the color of a screen pixel in Perl 6  :)
16:39 dju left #perl6
16:41 uniejo left #perl6
16:41 dju joined #perl6
16:42 masak moritz_++ # Mu repo
16:42 * isBEKaml goes singing "My heart goes Mu...." {ducks}
16:43 masak that cultural reference is lost to me as a non-Bharatan... :P
16:43 TimToady perhaps the new camel book should be a cow book
16:43 moritz_ now we only need to figure out hwo to checkout out parts of a repo
16:43 TimToady that would be funnier in Deutsch
16:44 TimToady do kuh's say muh?
16:44 TimToady s/"'"//
16:44 isBEKaml *lol* That's from an old Bollywood song "My dil goes hmmm... "
16:44 isBEKaml I just *gasp* transmogrified it.
16:45 isBEKaml "Dil" in hindi means "heart"
16:46 masak ooh
16:47 ash_ moritz_: svn co http://svn.github.com/moritz/Mu.git/t/spec only checks out the spec from Mu
16:47 moritz_ ash_++
16:47 moritz_ with s/moritz/perlsix/
16:48 isBEKaml svn co is enabled from git # I didn't know about git-svn interop.
16:48 ash_ git pulls full repo's as far as i can tell, so i don't see how to do it with git...
16:48 isBEKaml ?
16:48 masak chromatic++ # http://www.perl.com/pub/2010/08/​people-of-perl-6-carl-masak.html
16:48 ash_ github has svn mirrors of git repos
16:48 tylercurtis masak++
16:48 ash_ thats a github special thing
16:48 EvanCarroll left #perl6
16:48 isBEKaml github++
16:48 dakkar left #perl6
16:55 sftp_ left #perl6
16:57 ingy is http://github.com/rakudo/rakudo/tree/2010.08 the star tag for august?
16:59 ruoso joined #perl6
16:59 sftp_ joined #perl6
17:01 EvanCarroll joined #perl6
17:05 sorear bbkr: Did it ever occur to you that std might have have bugs
17:05 phenny sorear: 15:58Z <tylercurtis> tell sorear that he seems to have misspelled supersede as supercede in Niecza::Actions.
17:06 bbkr sorear: yes, and I reported a few STD bugs already. why are you asking?
17:07 ash_ rakudo: sub postcircumfix:<<< >>>($foo) { say $foo }; <"foo">; # just curious
17:07 p6eval rakudo f8e959: OUTPUT«===SORRY!===␤Null PMC access in find_method('lineof')␤»
17:07 ash_ rakudo: sub postcircumfix:«< >»($foo) { say $foo }; <"foo">; # just curious
17:07 p6eval rakudo f8e959: OUTPUT«===SORRY!===␤Null PMC access in find_method('lineof')␤»
17:08 ash_ oh, oops
17:08 ash_ rakudo: sub circumfix:«< >»($foo) { say $foo }; <"foo">; # just curious
17:08 p6eval rakudo f8e959:  ( no output )
17:13 Italian_Plumber left #perl6
17:14 masak sorear: bbkr reported discrepancies between Rakudo and STD, without much consideration for which one is "right". I think that's fine.
17:14 sorear masak: my reading of the tickets was very STD-normative
17:15 masak "Rakudo allows postfix<:> in regexp(?) but STD doesnt"
17:15 masak that just says they differ.
17:15 masak not which one is right.
17:15 bbkr masak: indeed. I'm not skilled enough to distinct Rakudo bug and STD bug. so if something is STD bug then just reject ticked. but discrepancy is something that should be reported anyway and fixed somewhere - in Rakudo or in STD.
17:16 masak bbkr: STD bugs also belong in RT, according to some schools of thought.
17:17 diakopter if it's called perl6 RT... maybe the current perl6 RT needs renamed to rakudo
17:18 bbkr so how should I act when I spot disrepancy if I'm not supposet to report to RT?
17:18 [Coke] you should report it to RT.
17:18 bbkr thanks :)
17:18 [Coke] (unless there is an STD specific bug queue of which I am not aware)
17:19 masak diakopter: people have also been committing doc and book bugs to RT. most are using it as the 'rakudo RT', but not everyone.
17:19 diakopter oh...
17:19 bbkr [Coke]: if I spot something that I know for sure is STD then I report directly to TimToady, otherwise to RT
17:20 Italian_Plumber joined #perl6
17:20 TimToady the current STD bug queue is a file named "todo"
17:20 moritz_ http://github.com/moritz/perlgeek.de/blob/maste​r/source/article/debug-a-perl-6-grammar.html.en # I'm currently working on an article about debugging grammars, and writing robust grammars
17:20 moritz_ any comments/suggestions?
17:21 bbkr moritz_: awesome
17:24 treed fist identifier!
17:25 _kaare left #perl6
17:26 envi^home left #perl6
17:26 moritz_ typos fixed, treed++
17:27 * treed nods
17:27 treed Looks good so far.
17:27 masak moritz_: s/less errors/fewer errors/
17:27 moritz_ rakudo: say 'abc' ~~ /<?DEBUG(1)> .+b/
17:27 p6eval rakudo f8e959: OUTPUT«ab␤»
17:28 moritz_ rakudo: say 'abc' ~~ /<?DEBUG(1)> <alpha>+ b/
17:28 p6eval rakudo f8e959: OUTPUT«ab␤»
17:29 masak moritz_: the '... and test along the way.' paragraph breaks the sometimes-made assumption that ordinary text paragraphs can be read independently of headings.
17:30 [Coke] rakudo star has no "issues" tab. any easy way to link back to rt's bug queue from there?
17:30 [Coke] (on github)
17:30 moritz_ masak: I'll fix that
17:30 masak moritz_++
17:31 [Coke] rakudo/rakudo has a link in the README, which is probably sufficient. also, bug in the README -> HTML generation: Rakudo's official web site is <http://rakudo.org/&amp;gt;,
17:31 TiMBuS left #perl6
17:33 masak moritz_: nqp-rx has a debug/trace facility. I haven't tried it much, but it seems to overlap slightly with your suggestion for manually adding debug statements.
17:33 moritz_ masak: I thought <?DEBUG(1)> worked. Right now I don't see any effect.
17:33 moritz_ worked in rakudo, that is
17:34 masak I thought so too. sadly, I don't know any more about it.
17:34 masak pmichaud++ likely will, though.
17:34 TiMBuS joined #perl6
17:36 redicaps left #perl6
17:37 Tene joined #perl6
17:39 moritz_ <?DEBUG(1)> seems to work in grammars
17:40 masak huh. why would there be a difference inside or outside of a grammar?
17:40 plainhao joined #perl6
17:41 moritz_ rakudo: my token foo { <alpha> }; 'abcd' ~~ /<?DEBUG(1)> <foo>+d/
17:41 p6eval rakudo f8e959: OUTPUT«Method 'foo' not found for invocant of class 'Cursor'␤  in <anon> at line 22:/tmp/GwTfXOewyT␤  in 'Cool::match' at line 2420:CORE.setting␤  in 'Regex::ACCEPTS' at line 5823:CORE.setting␤  in main program body at line 10:/tmp/GwTfXOewyT␤»
17:41 moritz_ rakudo: my token foo { <alpha> }; 'abcd' ~~ /<?DEBUG(1)> <&foo>+d/
17:41 p6eval rakudo f8e959: OUTPUT«1283275173.029351 0/0 START    foo␤1283275173.029638 0/0 PASS     foo at pos=1␤1283275173.029860 1/0 START    foo␤1283275173.030061 1/0 PASS     foo at pos=2␤1283275173.030274 2/0 START    foo␤1283275173.030426 2/0 PASS     foo at pos=3␤1283275173.030636 3/0 START
17:41 p6eval ..foo␤1283275173…
17:41 moritz_ seems to work outside too, but not for <alpha>
17:42 masak because <alpha> (along with the other predefined rules) is special-cased
17:42 masak probably should emit debug output, though.
17:42 risou_ joined #perl6
17:43 * masak submits nqp-rxbug
17:45 sorear moritz_: did you head Juerd say he wants to shut down svn.pugscode.org?
17:45 sorear std: /$$x/
17:45 p6eval std 32120: OUTPUT«Can't call method "Str" on unblessed reference at STD.pm line 58885.␤FAILED 00:01 116m␤»
17:45 moritz_ sorear: yes
17:46 Transformer joined #perl6
17:47 risou_ left #perl6
17:48 Transformer left #perl6
17:51 estrabd is now known as jlagron
17:51 jlagron is now known as jlagrone
17:51 dju left #perl6
17:52 jlagrone is now known as estrabd
17:58 dju joined #perl6
17:59 * colomon did not sleep through #phasers
18:00 moritz_ phenny: ask pmichaud in trace output from NQP-rx with <DEBUG(1)>, whas is the second number in the second column? ie the /1 in  1283276995.944721 17/1 START    comment
18:00 phenny moritz_: I'll pass that on when pmichaud is around.
18:01 ingy ingy@feather:~$ perl6 -v
18:01 ingy This is Rakudo Perl 6, version 2010.08 built on parrot 2.7.0
18:01 ingy Copyright 2008-2010, The Perl Foundation
18:01 moritz_ \o/
18:02 masak colomon: ooh, #phasers! thanks for reminding me.
18:04 colomon you're quite welcome.  I didn't remember myself until I woke up from my nap...
18:08 macroron left #perl6
18:08 moritz_ added a section about <?DEBUG(1)>
18:09 sorear std: / x: /
18:09 p6eval std 32120: OUTPUT«ok 00:01 116m␤»
18:09 fglock joined #perl6
18:10 sorear quantifier:<:> requires a space
18:10 fglock hi all
18:10 moritz_ hi fglock
18:12 moritz_ fglock: how are you? how is perlito going?
18:14 fglock I've replaced Pugs::Compiler::Perl6 with Perlito in v6.pm,
18:14 fglock PCP6 was not maintained anymore, and Perlito is getting good enough
18:15 fglock so now v6.pm and http://perlcabal.org/~fglock/perlito.html are the same compiler
18:15 fglock which is kinda cool
18:15 fglock and easier to maintain
18:16 tadzik joined #perl6
18:16 fglock also Perlito got a precedence parser, and lists
18:17 fglock got to update all backends before a release
18:18 dalek niecza: a2ee189 | sorear++ | src/Niecza/Actions.pm:
18:18 dalek niecza: s/supercede/supersede/ (tcurtis++)
18:18 dalek niecza: review: http://github.com/sorear/niecza/commit/a​2ee189793839c69225355ffc79a004bd0b8bea7
18:18 fglock also v6.pm is now pluggable, and v6-mildew could work (thanks to pmurias++)
18:19 tadzik oh hai
18:19 isBEKaml hi, tadzik
18:21 fglock moritz_: a few months ago you've asked about cleaning up mp6/kp6 from the pugs repository, I think it's ok to move those away now
18:21 daxim left #perl6
18:22 tri1 joined #perl6
18:23 moritz_ fglock: do you have a public repository where they live? or are they superseeded by perlito?
18:25 fglock the pugs repository keeps the history after you remove the files, right?
18:25 fglock perlito replaces mp6/kp6 now
18:25 fglock and v6.pm
18:26 zulon left #perl6
18:26 fglock Pugs::Compiler::Rule is used by Pugs, I think
18:28 TimToady rakudo: my @x = ($_ for ^10); say @x.perl  # known bug?
18:28 p6eval rakudo f8e959: OUTPUT«[]␤»
18:28 molaf left #perl6
18:28 sorear fglock: yea, svn can version deleted files
18:28 TimToady rakudo: my @x = (for ^10 { $_ }); say @x.perl
18:28 p6eval rakudo f8e959: OUTPUT«[0, 1, 2, 3, 4, 5, 6, 7, 8, 9]␤»
18:29 fglock rakudo: my $x; say "$x"
18:29 p6eval rakudo f8e959: OUTPUT«Any()␤»
18:29 TimToady that's Str vs Stringy confusion
18:29 fglock I've ran into this when running rakudo on perlito's test
18:30 fglock is this correct?
18:30 TimToady rakudo: my $x; say ~$x;
18:30 p6eval rakudo f8e959: OUTPUT«Any()␤»
18:30 TimToady that's also wrong
18:30 TimToady should produce "" and a warning.  it's the low-level .Str that produced Any(), not .Stringy
18:30 fglock is $x a "Mu"?
18:31 TimToady vars default to Any
18:31 sorear TimToady: Is ::> new?
18:31 TimToady not new in STD, but new in spec
18:31 TimToady syntax is provisional
18:31 TimToady ::> is the old ::
18:31 fglock rakudo: Any()
18:31 sorear std: / ::> / # not in STD yet
18:31 p6eval rakudo f8e959: OUTPUT«Could not find sub &Any␤  in main program body at line 22:/tmp/g1Kl5UMcYk␤»
18:31 p6eval std 32120: OUTPUT«[31m===[0mSORRY![31m===[0m␤Unable to parse regex; couldn't find final '/' at /tmp/7_upg75Mhg line 1:␤------> [32m/ ::[33m⏏[31m> / # not in STD yet[0m␤    expecting any of:␤   quantifier␤       regex atom␤Parse failed␤FAILED 00:01 115m␤»
18:31 TimToady when STD turned :: into LTM always
18:32 sorear ah
18:32 fglock rakudo: $x.perl
18:32 p6eval rakudo f8e959: OUTPUT«===SORRY!===␤Symbol '$x' not predeclared in <anonymous> (/tmp/Et9_LzdSYj:22)␤»
18:32 sorear so what's ::>?  | or ||?
18:32 fglock rakudo: my $x; $x.perl
18:32 p6eval rakudo f8e959:  ( no output )
18:32 TimToady either
18:32 fglock rakudo: my $x; say $x.perl
18:32 p6eval rakudo f8e959: OUTPUT«Any␤»
18:32 TimToady if | it's equiv to ::, if surrounding [] is standalone
18:33 fglock TimToady: thanks
18:34 TimToady hmm, where did I put ::> then...
18:35 moritz_ ::> is new to me
18:35 TimToady maybe it was only in gimme5
18:35 hans__ joined #perl6
18:35 TimToady that were it
18:36 TimToady anyway, attaches to COMMITBRANCH rather than COMMITLTM
18:37 TimToady whatever we decide for the syntax
18:37 TimToady could even be <then>
18:38 TimToady or we steal ::: for that and change ::: to <commit(Rule)> or some such
18:39 TimToady commit should have a smartmatched argument somehow anyway
18:41 TimToady then :: vs ::: would be backwards from a globality standpoint, but right from a huffman viewpoint
18:47 pugssvn r32121 | lwall++ | [STD] add ::> (already handled in viv)
18:53 Lorn left #perl6
19:00 dju left #perl6
19:03 dju joined #perl6
19:08 Lorn joined #perl6
19:11 M_o_C joined #perl6
19:11 _jaldhar left #perl6
19:14 _macdaddy joined #perl6
19:18 mathw o/
19:18 tadzik \o
19:19 masak mathw! \o/
19:19 * mathw is back from holiday
19:19 ash_ moritz_: i now have the server using the rakudo repls working now, i only have 1 last thing to finish up, cleaning up dead sessions, then i'll be ready to update try.rakudo.org
19:19 tadzik ash_: can I try?
19:19 tadzik I won't break anything, promise :)
19:19 Mowah joined #perl6
19:20 ash_ sure, give me a few mins and i'll get you a link to my test server
19:21 ash_ tadzik: break it as best you can
19:21 ash_ so i can fix it before it becomes a problem
19:22 tadzik oh, fun comes
19:22 [Coke] masak: can you try to close a ticket and tell me what happens?
19:23 masak [Coke]: sure thing.
19:23 plainhao left #perl6
19:23 M_o_C left #perl6
19:23 M_o_C joined #perl6
19:23 [Coke] make sure it says "logged in as masak" in the URHS.
19:24 masak [Coke]: ok.
19:24 [Coke] whoops.
19:25 [Coke] ok, I added you as a parrot dev (but not committer)
19:25 [Coke] that should do it.
19:25 masak [Coke]++
19:26 M_o_C left #perl6
19:29 colomon masak, oh master of the bugs: has anyone else reported the problem with .chomp not generating a Str anymore?
19:30 masak colomon: yes.
19:30 masak or something very much like it.
19:30 masak might've been .split
19:30 colomon this one is
19:30 M_o_C joined #perl6
19:31 colomon rakudo: my $a = "this is a test"; say $a.PARROT; say $a.chomp.PARROT
19:31 p6eval rakudo f8e959: OUTPUT«Perl6Scalar->Str␤String␤»
19:31 colomon .trans wouldn't work on a "String", perhaps .split wouldn't either?
19:32 masak colomon: right. here it is. http://rt.perl.org/rt3/Tic​ket/Display.html?id=76564
19:32 masak but that one is resolved, so maybe submit a new one.
19:33 colomon ok, and more important may be learning where the tests go.  masak++
19:34 tadzik ash_: gotta go now, but can test it tomorrow if you paste me a message via phenny
19:34 masak (spec <-> implementation <-> RT <-> spectests)++
19:34 ash_ kk, will do
19:34 tadzik left #perl6
19:36 colomon rakudo: my $a = "this is a test"; say $a.trans("t=>T")
19:36 p6eval rakudo f8e959: OUTPUT«t=>T.perl is not a Pair␤  in <anon> at line 2249:CORE.setting␤  in 'Cool::trans' at line 1␤  in main program body at line 22:/tmp/NkEJ05eA9V␤»
19:36 colomon rakudo: my $a = "this is a test"; say $a.trans("t"=>"T")
19:36 p6eval rakudo f8e959: OUTPUT«This is a TesT␤»
19:36 colomon rakudo: my $a = "this is a test"; say $a.chomp.trans("t"=>"T")
19:36 p6eval rakudo f8e959: OUTPUT«too few positional arguments: 2 passed, 3 (or more) expected␤  in main program body at line 22:/tmp/lhXlaJPKOc␤»
19:36 sorear TimToady: Did you say something about ::> caring about bare []?
19:37 Quadrescence left #perl6
19:37 colomon moritz_: what's the state of the spectest repo at the moment?  if I make changes, should I make them to SVN or a git repo out there somewhere?
19:38 moritz_ colomon: still SVN
19:38 colomon moritz_++
19:40 sorear moritz_: why use perlsix when we already have perl6?
19:41 moritz_ sorear: because afaict perl6 is not an organization account
19:41 timbunce left #perl6
19:41 moritz_ sorear: and because I have no administrative power over Perl 6
19:42 sorear ...hugme?
19:42 sorear pmichaud: do you control 'perl6' on github?
19:43 moritz_ sorear: I have the API token for hugme use
19:43 hans__ left #perl6
19:44 [Coke] Surely someone in this channel has it. ;)
19:44 moritz_ it's pmichaud, yes
19:44 sorear It just bugs me that we now have both perl6 and perlsix
19:44 TimToady sorear: ::> commits to the current meaning of a surrounding [] (or rule) regardless of whether the scope's alternations take part in LTM
19:44 pugssvn r32122 | colomon++ | [t/spec] Add tests to verify that .chomp returns a proper Str (fudged because at the moment, it does not).
19:45 wamba left #perl6
19:47 sorear TimToady: Are [ <foo> ::> ] and <foo>: equivalent?
19:48 timbunce joined #perl6
19:48 colomon rakudo: my $a = "this is a test"; say (~($a.chomp)).trans("t"=>"T")
19:48 p6eval rakudo f8e959: OUTPUT«This is a TesT␤»
19:48 colomon rakudo: my $a = "this is a test"; say (($a.chomp)).trans("t"=>"T")
19:48 TimToady sorear: I believe so
19:48 p6eval rakudo f8e959: OUTPUT«too few positional arguments: 2 passed, 3 (or more) expected␤  in main program body at line 22:/tmp/NT27MZaS_Y␤»
19:48 masak colomon++
19:49 colomon does anyone out there atm have an idea how to get Cool.chomp to return a proper Str?  I'm testing just adding a ~ to the output, but that seems like it is probably less than optimal...
19:50 proller_ left #perl6
19:50 moritz_ prefix:<~> is the only way I know about
19:50 moritz_ s/about/of/
19:50 masak same here.
19:50 masak maybe check what prefix:<~> does and do that? :)
19:50 TimToady Stringy() or .Stringy is what it should do
19:51 colomon I think prefix:<~> is deep magic.
19:51 TimToady the functional form is probably more optimizable at compile time
19:51 colomon unless masak++ fixed that during the Buf work?
19:51 moritz_ $P0 = new ['Str']
19:51 moritz_ assign $P0, a
19:51 masak colomon: nope, didn't venture into prefix:<~>
19:51 moritz_ where a is a PIR string
19:52 colomon moritz_++ # will try that next
19:52 moritz_ src/cheats/str-ops.pir
19:52 colomon $P0 = new ['Str']
19:52 colomon assign $P0, $S0
19:52 colomon %r = box $P0
19:53 colomon does that look like the proper translation to Q:PIR?
19:53 [Coke] s/box //
19:53 moritz_ no need for the box
19:53 moritz_ you can use %r instead of $P0 everywhere
19:53 [Coke] or, %r = new [ 'Str'] ; assign %r, $S0
19:53 [Coke] box is for going from basic types to PMCs. if you already have a PMC, it might just explode.
19:54 colomon [Coke]++ moritz_++
19:54 * colomon stopped the compile before the explosion.
19:54 moritz_ my $x = pir::new__pS('Str'); pir::set__PS($x, $x)  maybe
19:55 EvanCarroll left #perl6
19:57 [Coke] I think pir::set($x, ~$x) might be clearer.
19:58 [Coke] also, pir::new('Str') should work these days.
19:58 moritz_ if you use ~$x, there's no need for the other magic
20:00 Italian_Plumber left #perl6
20:00 colomon I'm going with             %r = new ['Str']          assign %r, $S0   for now -- simple, seems like it ought to be reasonably efficient, and it appears to work.
20:01 colomon spectesting...
20:01 colomon afk
20:05 Ross left #perl6
20:07 * moritz_ stll looks for replies to his perl6-user mail
20:09 lamstyle left #perl6
20:10 dju left #perl6
20:10 ash_ man, generating core.pir is intense
20:11 * masak wipes brow
20:11 Italian_Plumber joined #perl6
20:12 sorear moritz_: I haven't gotten it
20:12 sorear ash_: try http://feather.perl6.nl/~sor​ear/rakudo/src/gen/core.pir
20:12 moritz_ sorear: http://www.nntp.perl.org/group/per​l.perl6.users/2010/08/msg1309.html
20:12 lamstyle joined #perl6
20:12 tri1 is it possible to use net::ldap with perl6 ?
20:13 sorear I was trying to set up nightly builds earlier, but pbc_to_exe hates me
20:13 ash_ sorear: which revision is that from? just curious
20:13 moritz_ tri1: might be possible through blizkost
20:13 dju joined #perl6
20:14 fglock left #perl6
20:14 sorear ash_: the build log should say
20:14 sorear moritz_: that sounds like a job for multi-pass parsing
20:15 sorear er, hmm.
20:15 moritz_ sorear: you mean a separate pass that checks the nesting?
20:16 sorear moritz_: yes
20:16 sorear moritz_: both passes being perl 6 grammars, potentially
20:16 moritz_ hm. Possible. But I thought that multi-pass parsing was evil.
20:16 rokoteko ash_: http://cpansearch.perl.org/src/​RURBAN/illguts-0.33/index.html .. Is it insane (the more or the less if you havent took a peek at perl5 sources) ..?
20:16 rokoteko isn't Parrot supposed to be like perl5 xs for perl6?
20:17 sorear no
20:17 sorear not at all
20:17 rokoteko how to make perl6 fast then?
20:17 moritz_ by optimizing it
20:17 ash_ rokoteko: if you need to link to foreign C code, you can do most of that with the parrot NCI system
20:18 sorear parrot is supposed to be like pp_*.c, sv.c, hv.c, cv.c, and Metaconfig for perl 6
20:18 sorear run.c too
20:19 rokoteko Ok, Im speechless. I was (previously) thinking that this is the biggest flaw that perl6 has. everyone knows C, who knows parrot.
20:20 rokoteko how the fuck to *really* optimize eventually.
20:21 ash_ parrot's the vm rakudo is implemented on top of (currently), its kinda like how java can be compiled for the jvm
20:21 sorear RA-KU-DO
20:22 sorear I hate this mindset of "perl6 = parrot"
20:22 rokoteko ash_: are you trying to repeat to me why I think Java is a failier?
20:22 rokoteko sorear: agreed.
20:22 [Coke] sorear: I know. it's really bringing parrot down.
20:22 [Coke] ;)
20:22 rokoteko but Rakudo is the best implementation available.
20:22 ash_ rokoteko: i am not sure i follow
20:22 tylercurtis rokoteko: I'm not too fond of Java, but in what respect do you think that Java is a failure?
20:22 TimToady except for where pugs is still the best implementation :)
20:23 rokoteko tylercurtis: there is only one way to do it.
20:23 TimToady oh that perl 6 would fail as badly as java!  :)
20:23 [Coke] no joke. I /wish/ that many people hated perl 6. ;)
20:23 huf that's impossible, perl6 is far too good a language to be adopted *that* widely
20:24 tylercurtis rokoteko: What do you mean?
20:24 rokoteko TimToady: I wouldnt mind every other kid at University being taught perl6 instead of Java.
20:24 TimToady metoo
20:24 TimToady but we have to get to the university professors while they're still in grade school.  :)
20:25 rokoteko http://mjperry.blogspot.com/2010/06/hig​her-education-bubble-its-about-to.html
20:25 ash_ my professor in the spring did an independent study on vm's where i started on, nq-nqp, which isn't that far off from perl6
20:26 ash_ he teaches the compiler class and thought the idea of a parser that didn't define itself till runtime was awesome btw
20:26 tylercurtis rokoteko: I can understand the complaint of Java not having a TIMTOWDI philosophy, but I don't understand how that relates to the JVM.
20:27 ash_ and, class starts in 5 mins, so got to go
20:27 TimToady htaaof
20:27 sorear The problem with JVMs is that there are too many of them and they all suck.
20:27 ash_ left #perl6
20:27 TimToady some of them don't suck at certain things though
20:28 moritz_ some have good JIT, reliable garbage collection, and good concurrency
20:28 * sorear thinks Perl6 needs a proper ExtUtils::, but nobody else cares enough to make #cpan6sketch happen
20:28 TimToady fsdo "good" and "reliable"
20:28 masak "some of them don't suck at certain things" -- the same could be said of Perl 6 implementations.
20:29 TimToady we care, we're just stretched too thin ourselves
20:29 rokoteko TimToady: I didnt mention pugs. If I only had iq 140+ (I doubt Im even close to that), I wish I could understand it. :)
20:30 rokoteko but Rakudo is good.. But it might take years to have advantages like XS to make it very powerful for the unusual hackers.
20:30 masak sorear: I want you to know that I support your efforts with #cpan6sketch.
20:30 [Coke] rokoteko: what sort of power are you looking for?
20:30 rokoteko [Coke]: performance
20:31 sorear People seem to like joining #perl6 just to complain.  I don't get it
20:31 sorear We all know Perl 6 performance sucks
20:31 [Coke] sorear: be happy they care enough to share.
20:31 TimToady sorear: are you complaining?  :)
20:31 [Coke] hugme: hug sorear
20:31 * hugme hugs sorear
20:31 rokoteko I (before this dicussion) thought that you could write the stuff using parrot (as replacement for perl5 X) to make things faster.
20:31 _sri left #perl6
20:32 rokoteko s/X)/XS)/
20:32 tylercurtis rokoteko: If Perl 6 can't give you good enough performance, write it in C and use a FFI(which will hopefully look like Zavolaj).
20:32 sorear rokoteko: you can, just not using Parrot
20:32 TimToady that might've been the story in, oh, about 2003.
20:32 sorear Parrot is a VM
20:32 sorear the word you're looking for is FFI
20:32 [Coke] there are layers upon layers here.
20:32 moritz_ rokoteko: the problem with rakudo's speed is a mixture of not using parrot optimal, and parrot not providing what rakudo needs
20:32 rokoteko sorear: isnt that an acronym?
20:33 sorear rokoteko: so is XS
20:33 tylercurtis rokoteko: you can actually embed PIR or use PIR ops in Rakudo, but that's a bad idea, because that's not portable across implementations.
20:33 TimToady XS is a pun on "excess"
20:33 rokoteko sorear: ok. you know what XS is? :) mind telling me what FFI is ?
20:33 sorear FFI = foreign function interface
20:34 sorear it lets you run C code from Perl6
20:34 sorear we don't have ExtUtils::CBuilder yet, you'll need to ship a separate .so
20:34 rokoteko yes. lot's of time when I have mentioned XS I get reponses like "Why dont you use Inline::C?"
20:35 \xF0 that's a horrible response
20:35 TimToady they both have their place
20:35 M_o_C left #perl6
20:35 M_o_C joined #perl6
20:36 rokoteko sorear: I just was thinking aloud that it will take a while. :) but I previously thought from this discussion that is .pir is supposed to be the same performance improvement as .xs. please pardon of me being wrong.
20:36 cognominal rokoteko,  the  long term road to performance, with agility and maintainability is done by  having a good high level language runing on a smart VM. That's what Perl 6, Rakudo and Parrot are about even the latter two are not quite there yet.
20:36 cognominal agility is a slogan, Perl 6 will make it a reality.
20:36 rokoteko I really like rapid development. :) so here I agree with you cognominal.
20:37 TimToady ooh, nice slogan. "Making slogans into reality."
20:37 sorear rokoteko: Being wrong is not a sin, so I can't pardon it. ;)
20:37 sorear Learning is even less of a sin
20:37 TimToady .oO(Live and let learn.)
20:37 cognominal TimToady turns every thing into meta
20:38 [Coke] He never meta slogan he didn't leak.
20:38 TimToady Anything you can do, I can do meta / I can do anything meta than you.  No you can't! Yes I can!...
20:39 rokoteko sorear: I only miss the cos of all this. e^(i*pi) = cos(pi) + i*sin(pi) .. ok. I got the sin. where is the cause? and wtf is e?
20:39 sorear TimToady: What are the purposes of the cursor_xact calls on CursorBase.pmc lines 1858, 1872, 1886, 1900 ?
20:39 dh___ joined #perl6
20:40 rokoteko (pun seriously intended)
20:40 dh___ left #perl6
20:40 moritz_ wow, %hash.values.=map: *.elems # works in rakudo
20:40 tylercurtis rokoteko: another thing to consider is that Perl 6 is designed to be optimizable. For example, object attributes are more abstract so they can potentially be stored more efficiently.
20:40 EvanCarroll joined #perl6
20:41 hercynium joined #perl6
20:41 rokoteko tylercurtis: ok. perl6 is designed to be optimizable. Im just sometimes having trouble where the Parrot fits in. but that is probably only because I dont know Parrot. hence the discussion :)
20:42 moritz_ parrot.org tries to describe what parrot is :-)
20:42 patspam left #perl6
20:42 TimToady sorear: they chop the backtracked bits out of the current xact chain so they don't have to be re-traversed later on
20:42 rokoteko moritz_: havent diven in there yet :)
20:42 sorear rokoteko: Parrot is an implementation factor of Rakudo.  It contains no user-servicable parts
20:43 rokoteko moritz_: only scracthed the tip of the mountain (made of ice floating in an ocean)..
20:43 tylercurtis Parrot's a VM that Rakudo happens to run on. And it theoretically makes it easier to integrate different languages together. For example, if you have Blizkost installed, you can do "use Some::Module:from<perl5>;".
20:43 sorear TimToady: ... I don't see how
20:43 tylercurtis Although the interoperability with Perl 5 through Blizkost is a bit limited.
20:44 rokoteko sorear: I was trying to discuss of my (current) views of perl5 XS vs using Parrot underneath.
20:44 sorear rokoteko: perl5 XS leaks a lot of perl5's implementation details
20:44 rokoteko Im not here to dishonour anyone! only to appreciate and learn. dont take me wrong, I love Perl. :)
20:46 rokoteko sorear: yes it does. and I kinda understand that the XS interface has been living (and optimized for years and years) and perl6 is new.
20:46 TimToady sorear: okay, they don't chop it out, but they put anew false floor there to mark the track as "backed"  :)
20:46 rokoteko sorear: but perl6 on parrot.. can it do the same for the Parrot (vs. XS) programmer?
20:46 TimToady the back as tracked?
20:46 TimToady the backtrack as ed?
20:47 rokoteko TimToady++
20:47 rokoteko You're a bit older but I was an edling user when I was younger. :)
20:47 rokoteko -g
20:47 EvanCarroll left #perl6
20:48 sorear I was an ed user when I was younger
20:48 TimToady I watched Mr. Ed when I was young.
20:49 sorear TimToady: Where is the code that checks for the "CL" etc xact entries?
20:49 yrgd joined #perl6
20:50 patspam joined #perl6
20:51 tri1 left #perl6
20:51 FurnaceBoy joined #perl6
20:52 TimToady sorear: not sure there is any at the moment.  it might've been there once, or the marks may have just been info for the debugging log
20:52 sftp_ left #perl6
20:52 TimToady certainly they're all named with C first so that they *could* be recognized if we wanted to
20:53 TimToady and it might well be that we could get some benefit from actually chopping out the backtracked nodes from the new self
20:54 TimToady I might've thought that's what I was doing at the time
20:55 TimToady marking all the xacts as invalid is really only for the other users who still may have them in flight; we could probably chop them from our current list
20:56 masak someone opines 'Rakudo Plus' http://twitter.com/Jinzang/status/22645927237
21:00 sbp double star - exponentiation!
21:00 TimToady well, *+1 is kind of an inside joke
21:01 dalek yapsi: 21732ee | svatsan++ | / (4 files):
21:01 dalek yapsi: Added 'unless' keyword.
21:01 dalek yapsi: review: http://github.com/masak/yapsi/commit/21​732eee9c6567fe84acecb6ceb0b35647272d45
21:01 dalek yapsi: 515246f | svatsan++ | / (4 files):
21:01 dalek yapsi: added 'until' keyword.
21:01 dalek yapsi: review: http://github.com/masak/yapsi/commit/51​5246f5cefed43578ad32cbc93c8ec01f52b8fc
21:01 tylercurtis I think we should wait until HyperWhatever is implemented to even consider using it as a Rakudo Star release name.
21:01 whiteknight joined #perl6
21:01 masak isBEKaml++ \o/
21:01 TimToady going from Star to Plus has the same kind of issue as going from a one-humped camel to a two-humped camel; what comes after that?
21:01 masak TimToady: times. oh wait.
21:02 sorear easy, add tetration syntax
21:03 TimToady eventually we'll get to Rakudo @#$!&?#@
21:03 sbp the subname should be valid perl6, for bonus points
21:03 masak TimToady: that's how I feel already sometimes. :)
21:04 sbp and should print JAP6H...
21:09 masak can an action method have any say in whether a rule fails?
21:10 [Coke] it's too late by then, isn't it?
21:10 moritz_ it's too late indeed
21:10 masak it's never too late to fail :)
21:10 Quadrescence joined #perl6
21:10 FurnaceBoy TimToady: nothing genetic engineering can't now fix.
21:10 [Coke] you can fail in the action, but that's not the same thing.
21:10 masak what's the closest thing to failing a grammar match from within an action method?
21:11 moritz_ die()
21:11 moritz_ $/.CURSOR.panic('OH NOEZ')
21:11 masak I like that.
21:11 masak I don't like die(). that's the same problem as I had with ~
21:11 TimToady FurnaceBoy: I know that; see http://rosettacode.org/wiki/E​volutionary_algorithm#Perl_6
21:12 FurnaceBoy I suppose exceptions are called OHNOEZ in LOLCAT.
21:12 masak everything else in a regex returns false, but ~ dies.
21:12 moritz_ unless they are called FEJL
21:12 rokoteko Rakudo log_e(z)
21:12 FurnaceBoy TimToady: excellent
21:12 huf that's a bit too yiddish to be lolcat, moritz_ ;)
21:12 sorear in viv, that's spelled $/.panic
21:12 FurnaceBoy moritz_: I certainly had to look twice. :)
21:12 sorear you can also say $/.sorry to fail the match but keep going
21:13 sorear niecza: augment class Any { }; say $foo;
21:13 TimToady .oO(You'll be sorry!!!)
21:13 p6eval niecza 1d65d67: OUTPUT«Can't locate STD.pm in @INC (@INC contains: /home/p6eval/niecza/src /opt/perl-5.12.1/lib/site_perl/5.12.1/x86_64-linux /opt/perl-5.12.1/lib/site_perl/5.12.1 /opt/perl-5.12.1/lib/5.12.1/x86_64-linux /opt/perl-5.12.1/lib/5.12.1 .) at /home/p6eval/niecza/src/Niecza/Grammar.pm
21:13 p6eval ..line 3.␤…
21:13 * masak suspected he'd find the METHINKS IT IS LIKE A WEASEL sentence at http://rosettacode.org/wiki/E​volutionary_algorithm#Perl_6
21:13 masak that's from one of Dawkins' books.
21:14 masak "Blind Watchmaker", prob'bly.
21:14 huf isnt it shakespeare?
21:14 FurnaceBoy masak: seems likely
21:14 patspam left #perl6
21:14 FurnaceBoy huf: the example may be from Dawkins
21:14 masak aye. http://en.wikipedia.org/wiki/W​easel_program#Weasel_algorithm
21:15 FurnaceBoy huf: not the text per se
21:15 masak huf: the phrase is from Hamlet, yes.
21:15 patspam joined #perl6
21:15 TimToady it's about monkey typing :)
21:15 huf :)
21:15 patrickas joined #perl6
21:16 TimToady to be or not to be, that is the fobnitz  "Come over here Fred, I think we got something..."
21:16 dju left #perl6
21:16 FurnaceBoy TimToady: unfortunately the ionosphere was too small to contain it.
21:16 dju joined #perl6
21:16 TimToady perhaps they named their fitness function "Fred".
21:18 masak I think this might be an urge for Perl 6 to hurry up replacing Perl 5: http://twitter.com/Hannibal7878/status/22635870630
21:18 timbunce left #perl6
21:18 masak right before Hamlet and Polonius talk about weasels, they're talking about a camel.
21:19 sbp and then ruby on whales
21:19 sorear masak: ?
21:19 M_o_C Google says: "Dispatch by the number of arguments in Perl5 do not ... soon Perl6 Koi Koi ~" So he's missing some sort of polymorphism based on signature with perl 5?
21:19 _jaldhar joined #perl6
21:20 masak sorear: ¿
21:20 masak M_o_C: yeah. methinks so.
21:20 sorear koi koi...
21:20 patrickas masak: ؟
21:21 masak koi koi, #perl6!
21:22 masak I don't know Japanese, but to me it sounded like KUAI4 KUAI4 快快
21:23 dalek niecza: 8f9cfd5 | sorear++ | lib/Cursor.cs:
21:23 dalek niecza: [nrx] Simplify xact/backtrack correspondence
21:23 dalek niecza: review: http://github.com/sorear/niecza/commit/8​f9cfd5cc9d20064b6a59587343322ebbc9830da
21:23 dalek niecza: 33f66ad | sorear++ | / (2 files):
21:23 dalek niecza: [nrx] Implement CommitAll, CommitTo(name), CommitRule
21:23 dalek niecza: review: http://github.com/sorear/niecza/commit/3​3f66ad0ae008751fedae268563d3644d860048b
21:23 stepnem left #perl6
21:24 isBEKaml whoa! http://en.wikipedia.org/wiki/Koi_Koi_Seven
21:24 timbunce joined #perl6
21:25 isBEKaml Google gave me this as the second link.
21:26 isBEKaml hanufada?
21:26 stepnem joined #perl6
21:27 isBEKaml http://www.gamedesign.jp/f​lash/hanafuda/rule_e.html
21:29 M_o_C left #perl6
21:29 rurban joined #perl6
21:32 sorear or keep going (referred to as 'koi-koi,'
21:37 isBEKaml g'night, #perl6!
21:38 masak sleep well, isBEKaml!
21:38 patrickas bonne nuit, isBEKaml!
21:38 isBEKaml :)
21:39 * isBEKaml zzz
21:39 isBEKaml left #perl6
21:39 Mowah left #perl6
21:42 drbean joined #perl6
21:42 Italian_Plumber left #perl6
21:43 treed is now known as treed-rlpowell
21:44 treed-rlpowell is now known as rlpowell
21:46 araujo left #perl6
21:46 araujo joined #perl6
21:49 araujo left #perl6
21:49 rlpowell is now known as urmom
21:49 urmom is now known as tswett
21:50 masak it happens quite often that I accidentally name variables $a or $b in shorter Perl 5 scripts. then I spend time, sometimes several minutes, trying to figure out why they behave so strangely.
21:50 araujo joined #perl6
21:50 tswett is now known as treed
21:51 patrickas I do the same in perl6 but name them $^a and $^b and then things get messy too! (just being silly :-)
21:52 masak if things get messy in that case, then it's more a matter of LTA error messages. :)
21:52 masak for example, it makes no sense to use those hats outside of a block.
21:52 masak std: say $^a
21:52 p6eval std 32122: OUTPUT«[31m===[0mSORRY![31m===[0m␤Placeholder variable $^a may not be used outside of a block at /tmp/4FPBUW6jjF line 1:␤------> [32msay $^a[33m⏏[31m<EOL>[0m␤Check failed␤FAILED 00:01 116m␤»
21:52 patrickas hence the "being silly" part
21:52 masak I know, I know :)
21:53 frettled masak: as long as you don't use the hats indoors.
21:53 frettled Perl 6: outdoor use only?
21:53 masak patrickas: just taking the opportunity to show that your statement not only happened to be silly, but must have been :)
21:54 masak yapsi: my $a = 0; unless $a { say 42 }
21:54 p6eval yapsi:  ( no output )
21:54 masak :/
21:54 patrickas BTW I find it interesting that in http://rosettacode.org/wiki/Evolutionary_algorithm the perl5 solution uses @target and @valid_chars but in perl6 it uses $target and $range...
21:54 masak yapsi: something that definitely doesn't parse
21:54 p6eval yapsi:  ( no output )
21:54 masak yapsi: say 42
21:54 p6eval yapsi:  ( no output )
21:54 masak moritz_: p6eval/yapsi iz teh broke
21:57 patrickas rakudo: my $a = 5; say $^a;
21:57 p6eval rakudo f8e959: OUTPUT«5␤»
21:57 patrickas is that reported ? ^^
21:58 masak yes.
21:58 patrickas I guess it's too obvious to not have been masaked a long time a go :-)
21:58 timbunce left #perl6
21:59 rschuster joined #perl6
22:01 masak never hurts to check.
22:01 masak rakudo: my %h = 1 => "foo", 2 => "bar", 3 => "foo"; %h = grep { .value ne "bar" }, %h.pairs; say %h.perl
22:01 p6eval rakudo f8e959: OUTPUT«{}␤»
22:01 masak anyone but me thinks that the result should contain two pairs?
22:01 Holy_Cow left #perl6
22:03 ruoso left #perl6
22:04 patrickas rakudo: my %h = 1 => "foo", 2 => "bar", 3 => "foo"; my %n = grep { .value ne "bar" }, %h.pairs; say %n.perl
22:04 p6eval rakudo f8e959: OUTPUT«{"1" => "foo", "3" => "foo"}␤»
22:04 * masak submits rakudobug
22:05 masak patrickas++ # workaround
22:05 timbunce joined #perl6
22:08 s_mosher left #perl6
22:09 masak rakudo: my %h = 1 => "foo"; %h = %h; say %h.perl
22:09 p6eval rakudo f8e959: OUTPUT«{}␤»
22:09 masak rakudo: my %h = 1 => "foo"; my %n = %h; say %n.perl
22:09 p6eval rakudo f8e959: OUTPUT«{"1" => "foo"}␤»
22:10 patrickas duh!
22:10 masak doubleplusungood.
22:11 patrickas rakudo: my %h = 1 => "foo"; %h = %h.clone; say %h.perl;
22:11 p6eval rakudo f8e959: OUTPUT«{"1" => "foo"}␤»
22:13 patrickas might be slightly nicer than creating a names temporary variable as a workaround
22:13 patrickas s/names/named/
22:13 masak indeed.
22:15 japhb left #perl6
22:17 * masak .oO( :$value.method as sugar for :value($value.method) -- tempting, but perhaps a job for a module )
22:22 amkrankruleuen left #perl6
22:25 patspam left #perl6
22:26 patspam joined #perl6
22:29 drbean left #perl6
22:32 cdarroch left #perl6
22:32 leprevost joined #perl6
22:33 leprevost hi, i have a question about rakudo star installation
22:33 masak excellent.
22:33 masak let's hear it.
22:33 leprevost I have a previous version of rakudo installed and now I whant to update it to the recent version, what is the best way to do it ?
22:34 masak just install the new one, I'd suppose.
22:34 rschuster left #perl6
22:35 leprevost i thought that, I just need it to confirm
22:35 * masak stamps 'confirmed' on a paper and gives to leprevost :)
22:35 leprevost hehe, thank you Sir.
22:36 masak best of luck with your new Rakudo.
22:38 FurnaceBoy oil regularly
22:42 masak rakudo: say 'oil regularly'.comb.sort.join
22:42 p6eval rakudo f8e959: OUTPUT« aegilllorruy␤»
22:42 masak rakudo: say 'oil regularly'.comb.sort.trim
22:43 p6eval rakudo f8e959: OUTPUT«a e g i l l l o r r u y␤»
22:43 masak rakudo: say 'oil regularly'.comb.sort.join.trim
22:43 p6eval rakudo f8e959: OUTPUT«aegilllorruy␤»
22:43 masak rakudo: say 'oil regularly'.comb.sort.uniq.join.trim
22:43 p6eval rakudo f8e959: OUTPUT«aegiloruy␤»
22:44 masak rakudo: say 'oil regularly'.comb.sort.uniq.join.trim.subst('ae', 'æ')
22:44 p6eval rakudo f8e959: OUTPUT«ægiloruy␤»
22:44 leprevost left #perl6
22:44 masak :)
22:44 jhuni joined #perl6
22:45 masak jhuni: aloha!
22:46 dalek rakudo: 2ac5feb | colomon++ | src/core/Cool-str.pm:
22:46 dalek rakudo: Have Cool.chomp return a proper Str.
22:47 dalek rakudo: review: http://github.com/rakudo/rakudo/commit/2​ac5feb2b9e6b8791a4e5e2e71a86536df24ad59
22:50 jhuni masak: mahalo
22:53 alksentrs joined #perl6
22:54 s_mosher joined #perl6
23:02 timbunce Having a rand_char function in http://rosettacode.org/wiki/E​volutionary_algorithm#Perl_6 just to do $range.pic seems like a needless distraction in an otherwise clear and concise implementation. (Also, if $range were renamed $chars then $chars.pick would make the in-place use even more natural.)
23:03 * FurnaceBoy sees many eyeballs at work
23:04 sahadev1 left #perl6
23:05 saaki left #perl6
23:07 colomon does that even work?  (the classic colomon question)
23:07 rgrau__ joined #perl6
23:07 masak colomon: what, Range.pick?
23:07 colomon no, [+] $str Zeq $target
23:08 rgrau_ left #perl6
23:08 colomon where $str is a string, and $target is an array of letters.
23:08 masak the function of the [+] eludes me.
23:08 colomon sums the number of matches, I presume.
23:08 masak ooh!
23:09 masak of course, because the result from Zeq will be 1 0 0 1 1 0...
23:09 colomon ah, maybe $str isn't actually a string
23:10 alksentrs Hello #perl6
23:10 colomon o/
23:10 arnsholt Huh. Changing version control systems seems to have changed my workflow a bit for the better too ^^
23:10 masak alksentrs: hi!
23:10 alksentrs masak: Hello
23:10 masak hello.
23:10 arnsholt Oh, and hi =)
23:11 alksentrs I've been following #perl6 a while
23:11 saaki joined #perl6
23:11 alksentrs This is the first time I've logged on
23:11 masak hugme: hug alksentrs
23:11 * hugme hugs alksentrs and blushes
23:11 masak :)
23:11 alksentrs and is the first time I've used IRC
23:11 alksentrs masak: Thanks
23:11 masak I'd say it's going well so far!
23:11 alksentrs masak: Who created the hugbot?
23:12 masak moritz_++
23:12 alksentrs Why?
23:12 masak its chief function isn't even hugging.
23:12 masak it's handing out commit bits to diverse repos.
23:12 patspam left #perl6
23:12 masak hugme: show yapsi
23:12 hugme masak: the following people have power over 'yapsi': T⁣imToady, [⁣particle], c⁣olomon, j⁣nthn, m⁣asak, m⁣oritz_, p⁣michaud, s⁣narkyboojum. URL: http://github.com/masak/yapsi/
23:12 masak hugme: add alksents to yapsi
23:12 hugme masak: ERROR: Can't add alksents to yapsi:  404 Not Found
23:12 masak just an example.
23:12 \xF0 alksentrs: welcome to irc. you are now a man
23:13 alksentrs masak: Thanks.
23:13 * FurnaceBoy titters girlishly
23:13 alksentrs masak: Sorry, didn't see type
23:13 masak \xF0: it's interesting how you assume alksentrs is male :)
23:13 alksentrs *typo
23:13 alksentrs I am male
23:13 masak so I assumed. :)
23:13 alksentrs alksentrs = corruption of Alexander
23:13 \xF0 :D
23:13 masak alksentrs: are you from .ru?
23:13 \xF0 masak: or, alternatively, irc makes him into a man!
23:13 alksentrs No
23:14 alksentrs England
23:14 masak oh!
23:14 alksentrs I got it from the Code Book by Simon Sigh
23:14 masak :)
23:14 alksentrs *Singh
23:14 alksentrs Aha, Rakudo has finished compiling
23:15 alksentrs Sometimes I get weird error about mismatched versions
23:15 alksentrs What causes that?
23:16 TimToady dunno, usually deleting one or the other of parrot/ or /parrot_install fixes it though
23:16 masak I think we'd better show you another bot.
23:16 masak rakudo: say "Alexander".lc.subst(/e/, "").subst("x", "ks").subst(/a/, "e", :nd(2)).subst("der", "trs")
23:16 alksentrs rakudo: 3
23:16 TimToady .u smiling
23:16 phenny U+263A WHITE SMILING FACE (☺)
23:16 p6eval rakudo f8e959: OUTPUT«alksentrs␤»
23:16 p6eval rakudo f8e959:  ( no output )
23:17 alksentrs rakudo: () .= ()
23:17 p6eval rakudo f8e959:  ( no output )
23:17 TimToady it doesn't autosay
23:17 alksentrs segfaults
23:17 smash left #perl6
23:17 * masak checks locally
23:17 TimToady () isn't mutalbe
23:17 TimToady *bl
23:17 masak alksentrs: so it does. good catch.
23:17 TimToady though it shouldn't dump
23:17 * masak submits rakudobug
23:17 masak alksentrs++
23:18 TimToady buubot: karma alksentrs
23:18 buubot TimToady: alksentrs has karma of 1
23:18 masak alksentrs: now you're officially part of the Perl 6 effort.
23:18 alksentrs rakudo: say +&[+]   # number of multi matches
23:18 p6eval rakudo f8e959: OUTPUT«13␤»
23:18 masak yeah, that's probably not spec...
23:18 masak more like a Parrot see-through.
23:18 TimToady definitely not spec
23:19 masak :)
23:19 masak better plug it before someone uses it for something.
23:19 alksentrs I tested Rakudo a few weeks ago, and found a few oddities
23:19 masak we love oddities.
23:19 alksentrs I have a list...
23:19 masak we tend to frame oddities and put them up on the wall.
23:19 TimToady and TheDamian suggests that &[+] is not as useful anymore, now that we have *+*
23:20 masak nobody seems to prefer &[+] except me... :)
23:20 rgrau__ left #perl6
23:20 TimToady he suggests that &[+] should be put up against the wall...
23:20 rgrau__ joined #perl6
23:20 masak let's wait until the revolution comes.
23:21 alksentrs rakudo: say 1 % 0
23:21 colomon how does &[+] work?  Is it always infix:<+>?
23:21 p6eval rakudo f8e959: OUTPUT«1␤»
23:21 masak colomon: aye.
23:21 masak alksentrs: o.O
23:21 masak alksentrs: that should be a 'div by zorro' error, if you ask me.
23:21 FurnaceBoy masak: what, the revolution isn't *here*?
23:21 TimToady shh
23:21 alksentrs rakudo: say 0 % 0
23:22 p6eval rakudo f8e959: OUTPUT«0␤»
23:22 masak FurnaceBoy: it was here while you left the room. said it'd be back again tomorrow, same time.
23:22 colomon $x % $y == $x - floor($x / $y) * $y
23:22 alksentrs rakudo: say 0 / 0
23:22 p6eval rakudo f8e959: OUTPUT«Divide by zero␤  in 'infix:<div>' at line 3565:CORE.setting␤  in 'Rat::new' at line 3866:CORE.setting␤  in 'infix:</>' at line 3955:CORE.setting␤  in main program body at line 22:/tmp/wGT2CbE7AM␤»
23:22 colomon rakudo: say floor(1/0).perl
23:22 p6eval rakudo f8e959: OUTPUT«Inf␤»
23:22 TimToady rakudo: say (0/0).WHAT
23:22 p6eval rakudo f8e959: OUTPUT«Divide by zero␤  in 'infix:<div>' at line 3565:CORE.setting␤  in 'Rat::new' at line 3866:CORE.setting␤  in 'infix:</>' at line 3955:CORE.setting␤  in main program body at line 22:/tmp/UVbMWk9Ytg␤»
23:22 colomon rakudo: say floor(1/0) * 0
23:23 p6eval rakudo f8e959: OUTPUT«NaN␤»
23:23 alksentrs rakudo: say (0 := 1)
23:23 p6eval rakudo f8e959: OUTPUT«1␤»
23:23 colomon okay, so that's not how it's being calculated...
23:23 colomon ah, just a call to pir::mod
23:24 * masak submits $x % 0 rakudobug
23:24 alksentrs rakudo: { use v5; }
23:24 p6eval rakudo f8e959: OUTPUT«Null PMC access in setprop()␤  in main program body at line 22:/tmp/RUrRrsffwz␤»
23:24 TimToady alksentrs++
23:24 * masak submits binding-of-literal rakudobug
23:24 * masak submits 'use v5' rakudobug
23:24 masak alksentrs++
23:25 masak rakudo: use v5
23:25 p6eval rakudo f8e959:  ( no output )
23:25 * colomon hopes masak doesn't expect us to trap LHS of 0 on infix:<%> before calling into Parrot...
23:25 masak rakudo: use v5; say "alive"
23:25 p6eval rakudo f8e959: OUTPUT«alive␤»
23:25 masak colomon: I expect $x % 0 to die just as much as $x / 0 does.
23:26 alksentrs rakudo: say ~('1' .. '43')   # shouldn't this be similar to (1 .. 43)?
23:26 p6eval rakudo f8e959: OUTPUT«1 2 3 4␤»
23:26 lichtkind colomon: what is LHS?
23:26 alksentrs left hand side
23:27 masak alksentrs: no.
23:27 masak alksentrs: because ranges of strings follow different increment/halting rules than ranges of numbers.
23:28 TimToady which will probably be revised, but still may not work the way you expect :)
23:28 alksentrs rakudo: say ~('α' .. 'ω')
23:28 p6eval rakudo f8e959: OUTPUT«α β γ δ ε ζ η θ ι κ λ μ ν ξ ο π ρ σ τ υ φ χ ψ ω␤»
23:29 TimToady single chars are a special case currently
23:30 TimToady not for ranges, but for the ... that the .. turns into
23:30 masak I think I'd prefer the string ranges semantics to err on simplicity.
23:30 alksentrs rakudo: '' ~~ / { $0.perl } /
23:30 p6eval rakudo f8e959: OUTPUT«Null PMC access in find_method('perl')␤  in <anon> at line 1192:CORE.setting␤  in 'Any::join' at line 1␤  in 'Mu::attribs' at line 1193:CORE.setting␤  in 'Mu::perl' at line 1197:CORE.setting␤  in <anon> at line 22:/tmp/QKdlrTtTtd␤  in 'Cool::match' at line 2420:CORE.setting␤
23:30 p6eval ..in 'R…
23:30 masak preferably it should have a wide common base with string increment.
23:30 masak which should also preferably be extremely simple.
23:31 alksentrs string increment seems very... odd to me
23:31 masak alksentrs: that one's known. but thanks.
23:31 TimToady there is another consideration, however, which is that string increment can't know the goal, while a series can
23:31 jferrero joined #perl6
23:31 timbunce left #perl6
23:31 alksentrs only Perl 5 and Perl 6 have it?
23:31 TimToady and they have it differently
23:32 TimToady which explains some of the problem right there :)
23:32 alksentrs What is it used for?
23:32 TimToady things like generating filenames
23:32 TimToady or other symbols that want to be unique
23:33 alksentrs Would it be safer to use interpolation of numbers, or something?
23:33 huf c has string increment for very short strings :)
23:33 * TimToady snorts
23:33 huf well... sortof ;)
23:34 TimToady alksentrs: it depends on how you define "safer"
23:34 TimToady certainly it could be just as well defined
23:34 alksentrs In terms of Unicode?
23:34 TimToady in the sense of any algorithm being repeatable
23:34 alksentrs Locale? Codepoints vs Graphemes vs etc
23:34 masak TimToady: the uses cases with generating filenames and other unique symbols doesn't really motivate the big section in S03 on exotic alphamets and Unicode number-y things.
23:35 rurban left #perl6
23:35 TimToady Graphemes don't have well-ordered codepoints
23:35 TimToady there are many reasons to want sequences of single chars
23:36 TimToady the carry is a bit less well motivated, but is there to keep the tendency toward uniqueness
23:36 TimToady rakudo: ('⚀' .. '⚅').pick.say
23:37 p6eval rakudo f8e959: OUTPUT«⚁␤»
23:37 alksentrs How do other languages solve similar problems?
23:37 TimToady by forcing the complexity onto the programmar and code reader, usually
23:38 masak for the filenames/unique use case, you could keep a number in a separate variable, and build the strings yourself just-in-time.
23:38 masak Yapsi does that in a couple places.
23:38 masak rather than using the magical string increment behavior.
23:39 jhuni left #perl6
23:39 alksentrs masak: Yapsi uses bytecode, doesn't it? How does it differ from LLVM, JVM, .NET, ...?
23:39 alksentrs ...Parrot...
23:39 masak no, right now Yapsi only compiles down to a text opcode format called SIC.
23:40 TimToady strange intermediate code?
23:40 alksentrs Isn't that the name of the Intercal compiler?
23:40 masak TimToady: :)
23:40 alksentrs No wait: ick
23:40 jhuni joined #perl6
23:40 masak Yapsi has a small runtime which probably doesn't deserve to be called a VM.
23:40 alksentrs sick was the CLC-INTERCAL compiler
23:40 masak all of it in ~500 lines of Perl 6 code.
23:40 TimToady maybe it "came from" there
23:41 masak \o/ all tests successful in Yapsi!
23:41 masak for the first time since the refactor.
23:41 masak making 'variable not declared' errors work right was ridiculously hard to do during the parse phase.
23:42 TimToady tell me about it
23:42 masak but it was the first time I tried, so there might be better ways than the one I settled on.
23:42 alester left #perl6
23:43 alksentrs I gave a half-hour talk on Perl 6 a few days ago
23:43 masak the last hump I stumbled on was just hilarious: it found the <variable> inside <declaration> before it found the <declaration> itself (because actions work bottom-up), and so it always gave the error, even for declarations.
23:43 patrickas left #perl6
23:43 masak alksentrs: wow! cool!
23:43 TimToady alksentrs: did you survive?
23:43 masak alksentrs: are there slides somewhere?
23:43 alksentrs No slides
23:44 alksentrs I just wrote a couple of Perl 5 programs on the whiteboard, and went from there
23:44 alksentrs I have handwritten notes
23:44 alksentrs *had
23:44 alksentrs It was for an interview, no less
23:45 masak Perl 6 in an interview?
23:45 alksentrs I was supposed to do a talk on the topic of my choice
23:45 * masak would have loved to hear that
23:47 alksentrs I showed them the "idomatic" Fibonacci generator thing
23:47 alksentrs *idiomatic
23:47 alksentrs I'm not sure if they were awed or apalled
23:47 * masak beams
23:48 TimToady sometimes s/pal/w/ over time
23:48 alksentrs One asked if Perl 6 was optimized for toy examples, not production software
23:48 TimToady it's a fair question
23:49 masak basically the same objection mst had on Twitter.
23:49 TimToady to which the correct answer is, the two are not mutually exclusive
23:49 masak http://twitter.com/shadowc​at_mst/status/22112066276
23:50 TimToady toys are how we learn Real Life when we're young
23:50 alksentrs Another question that was asked: Why should people use Perl6 instead of Perl5? (c.f. People used C++ instead of C because of its OOP support.)
23:50 drbean joined #perl6
23:51 TimToady same with Perl 6 :)
23:51 TimToady only in our case, it's true  :P
23:53 alksentrs rakudo: ''>>.()
23:53 p6eval rakudo f8e959: OUTPUT«Tried to find null name␤  in main program body at line 22:/tmp/LpPTTKSQcE␤»
23:54 TimToady the more serious answer is that Perl 5 programmers often don't notice when they're getting frog-boiled, and Perl 6 is designed to scale better
23:54 alksentrs MONKEY_PATCHING doesn't scale...
23:54 alksentrs does it?
23:54 TimToady indeed, which is why we discourage it
23:55 TimToady that's a strange one
23:57 alksentrs What does unable to parse blockoid mean?
23:57 TimToady usually it means it got a more specific error later, but didn't know how to describe it, and backtracked till the start of block
23:57 TimToady and often in that case, STD will give you a better error
23:58 TimToady if it is an error...
23:59 alksentrs Is there a low-precedence method call operator?

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

Perl 6 | Reference Documentation | Rakudo