Perl 6 - the future is here, just unevenly distributed

IRC log for #perl6, 2010-03-04

Perl 6 | Reference Documentation | Rakudo

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

All times shown according to UTC.

Time Nick Message
00:00 TimToady those may not be so simple in p6
00:00 TimToady P5 would consider UTF-16 and UTF-32 to be binary, unless they've changed recently
00:00 TimToady but they're really text
00:01 lue well, the spec lists binary as containing odd characters, "such as control codes or bytes with the high bit set"
00:01 TimToady yes, it's rather latin-1-centric
00:02 TimToady we will not be carrying over the exact behavior
00:02 TimToady would be better to have some kind of file type guesser that has more clue
00:02 lue maybe something like the file command in *nix, or MIME type (I don't trust file extensions)
00:03 athenot joined #perl6
00:03 markjreed How are you going to get MIME type from a file in the filesystem?
00:03 lue I do not know. Just throwing ideas.
00:04 lue the hard-coded way would be a database-of-sorts listing all known 'magic numbers' of binary files and such.
00:04 markjreed Something like the algorithm file(1) uses, but that depends on /etc/magic...
00:07 arnsholt For file type guessing, it might be possible to train some kind of statistical model?
00:07 petdance joined #perl6
00:07 arnsholt Er, encoding, rather than file type
00:07 petdance joined #perl6
00:09 markjreed web browsers have encoding-guessing algorithms..
00:10 markjreed could port Encode::Guess maybe.
00:10 ash_ does icu have something for this?
00:10 ash_ since parrot *can* have icu built into it, if its not there then you'd have to have some sort of fallback
00:11 TimToady replacing icu with a pure P6 library would be a good project at some point
00:11 TimToady could be based on the unicode tables like P5 is
00:11 markjreed according to http://fredeaker.blogspot.com/2007/01/character-encoding-detection.html, ICU does have an encoding detector.
00:14 lue well, let's see how dependant other vital functions (Xorg, KDE, and the like) of the typical system are on icu, and guage our decision on that.
00:14 lue s/guage/gauge/
00:17 lue I'll be back in a little while, then I'll continue musing with you
00:19 ash_ does parrot have this?
00:19 ash_ have you asked the parrot guys about an encoding detection pmc/something?
00:19 cotto_working joined #perl6
00:36 lichtkind PerlJam: ping
00:57 snarkyboojum joined #perl6
01:04 justatheory joined #perl6
01:07 Ivan4 joined #perl6
01:08 Ivan4 left #perl6
01:15 lue wow. Do I have to be here to continue the conversation? :)
01:16 lichtkind lue: hai
01:16 lue hi
01:17 colomon I think "everyone" is getting ready for the big conference / hackathon this weekend, probably.
01:19 lue 7H3R35 G0!/\/G 2 B AY H4K470/\/? \/\/0\/\/!
01:19 lichtkind lue: please calm down
01:20 lue AI D0/\/7 7h!/\/k ai k4n! (I'll stop now)
01:20 colomon yes, there's going to be a hackathon.
01:21 lue I'll find people to bring testimonials. "Call 1-8TIMTOWTDI now, to donate to our cause!"
01:23 * lue .oO(The Larry Wall 45th anual Perl Foundation Hackathon!) #oh, the fun I'll have before I start hacking >:)
01:23 lue s/anual/annual/
01:26 lue .oO(Wikipedia first ran on Perl code (before switching to the PHP MediaWiki))
01:27 colomon No applicable candidates found to dispatch to for '&infix:<=>'
01:27 colomon that's a new one on me....
01:29 lue ACHTUNG!: infix:<=> could not be found. Please try the --things-do-equal-each-other option when you configure next time.
01:34 k23z__ will I get considerable speed boosts if I write some pieces that are now XS in nasm ?
01:35 k23z__ I wrote a module for generating permutations using XS with the best known algorithm called Steinhaus-Johnson-Trotter and it's still 5 times slower than Algorithm::Permute and this disappoints me of my coding skills
01:35 k23z__ I was hoping to beat it
01:35 k23z__ now I am thinking about re-writing parts of it in nasm to make it faster
01:35 TimToady XS is kinda off topic here
01:36 TimToady not that we're strict about that...
01:36 arlinius you should write it in perl 6, then optimize rakudo/parrot to make it fast...
01:36 diakopter .
01:36 lue assembler is the fastest code you could possibly generate (assuming, of course, you're efficient about it)
01:38 yinyin joined #perl6
01:39 k23z__ ^
01:39 k23z__ |
01:39 k23z__ he wrote a dot
01:39 diakopter so I did
01:39 k23z__ why/
01:39 wknight8111 joined #perl6
01:39 diakopter std: .
01:39 p6eval std 29935: OUTPUT«[31m===[0mSORRY![31m===[0m␤Unsupported use of . to concatenate strings; in Perl 6 please use ~ at /tmp/0CIHJKxs7k line 1 (EOF):␤------> [32m.[33m⏏[31m<EOL>[0m␤FAILED 00:01 105m␤»
01:42 diakopter rakudo: say 900.00000000000000000 # masakbot
01:42 p6eval rakudo 4edd19: OUTPUT«PAST::Compiler can't compile node of type BigInt␤current instr.: 'perl6;PCT;HLLCompiler;panic' pc 137 (compilers/pct/src/PCT/HLLCompiler.pir:101)␤»
01:44 diakopter rakudo: say (90^^4).WHAT # masakbot
01:44 p6eval rakudo 4edd19: OUTPUT«Method 'WHAT' not found for invocant of class 'Undef'␤current instr.: '_block14' pc 29 (EVAL_1:0)␤»
01:45 k23z__ can anyone help me understand what the guy did here ? http://cpansearch.perl.org/src/ROBIN/Algorithm-FastPermute-0.999/FastPermute.xs
01:45 k23z__ what's a multicall ?
01:45 k23z__ is this a concept found in other languages ?
01:46 diakopter it appears like he/she wrote an .xs file
01:46 k23z__ or did he invent it ?
01:46 xuzj joined #perl6
01:48 colomon Tene: errr.... you seem to have broken Rakudo somewhat?
01:48 lue .oO(At least I didn't break it *phew*)
01:53 cls_bsd joined #perl6
01:57 lue you guys know of the nazi armbands the soliders wore, right?
01:57 lue the NAZI soliders
01:58 * diakopter squints sidelong
01:59 arlinius soliders?
01:59 diakopter rakudo: (sub {say (*)()})()
01:59 p6eval rakudo 4edd19: OUTPUT«!whatever_dispatch_helper␤»
01:59 diakopter # masakbot ^^ see above; I dunno...
02:00 arlinius hrm
02:01 lue Well, my friends and I came up with the Politically Correct and Copyright Infringment War Machines
02:01 * diakopter lue plz stay closer to topic..
02:01 arlinius rakudo: my $x = *; say $x(0)
02:01 diakopter lue
02:01 p6eval rakudo 4edd19: OUTPUT«Method 'item' not found for invocant of class 'Integer'␤current instr.: '!whatever_dispatch_helper' pc 280 (src/glue/dispatch.pir:25)␤»
02:02 lue diakopter: oh fine, you'll never see the end :) . I just started talking because nothing was happening.
02:02 PZt joined #perl6
02:06 lue .rnd(please be advised the rnd bubbles can be ignored at your will (including this one))
02:07 hicx174 joined #perl6
02:07 * lue ghost hunters &
02:11 TimToady most topics can be related to Perl 6 if you try hard enough, but you *do* have to try hard enough  :)
02:12 TimToady from time to time...
02:12 nihiliad joined #perl6
02:16 k23z__ Rate   tye   SJT dchld  A::P
02:16 k23z__ tye   1.24/s    --  -20%  -52%  -81%
02:16 k23z__ SJT   1.55/s   25%    --  -40%  -76%
02:16 k23z__ dchld 2.58/s  108%   66%    --  -60%
02:16 k23z__ A::P  6.45/s  421%  316%  150%    --
02:16 k23z__ this benchmark
02:17 k23z__ does it say that SJT is slower by 76% than A::P ?
02:17 k23z__ and that in turn A::P is faster by 316% than SJT ?
02:17 k23z__ do I get this right ?
02:17 k23z__ because it doesn't make much sense
02:17 * diakopter tries not very hard
02:17 k23z__ those percents don't make much sense to me
02:17 k23z__ do they make sense to you ?
02:18 k23z__ do you guys also make benchmarks in Perl6 ? or is it not yet time  to make them ?
02:18 diakopter not really
02:20 k23z__ 316 - (76/100) * (316) = 75.4
02:20 k23z__ wasn't I supposed to get 100 ?
02:21 k23z__ I mean it A::P is 316% of SJT
02:21 k23z__ and SJT is 76 slower
02:21 k23z__ bah nevermind too tired
02:22 colomon rakudo: say 316 - (76/100) * (316)
02:22 p6eval rakudo 4edd19: OUTPUT«75.84␤»
02:23 TimToady this sort of benchmark discussion is usually more appreciated on, say, perlmonks
02:25 k23z__ ok here we discuss Parrot and Rakudo
02:25 diakopter and Perl 6 things
02:25 k23z__ ok
02:25 k23z__ I'll go to bed
02:25 k23z__ gnight
02:25 diakopter 'nite
02:35 eternaleye joined #perl6
02:35 hercynium joined #perl6
02:38 TimToady we also, of course, discuss anything *I'm* interested in.  :)
02:38 TimToady since I'm related to Perl 6 somehow...
02:57 rv2733 joined #perl6
03:00 colomon TimToady: For series, "When no limit is given, the function need not be monotonic:""
03:01 colomon I've already coded up a bunch of non-monotonic series with limits for my tests.
03:02 colomon Is that line a fossil?
03:02 TimToady the monotonic really only applies to intuitied functoins
03:02 TimToady io
03:02 TimToady if you have your own generator, matching of limits must be exact
03:02 TimToady or write your own terminator closure, now that we can have those
03:03 TimToady so it's kinda fossily
03:04 colomon I don't know of any actually useful examples, but the non-exact limit code is smart enough to handle your own generator too.
03:05 colomon Like { (1..2).pick } ... 1.5
03:06 colomon Unless there's some trick I'm missing, NOT supporting that case with the user-generator will actually be more work than just supporting it.
03:06 TimToady okay, most interesting non-monotonics will be to * anyway
03:07 colomon agreed.
03:08 colomon TimToady++ # I'm still really digging the current series spec, even if I'm not quite sure how to efficiently lazily support the arity > 1 cases.  :)
03:08 * colomon to bed.
03:10 TimToady night
03:36 policomb joined #perl6
03:36 policomb hi all
03:36 policomb I've a little question
03:37 policomb how is it possible to change/set an ethernet device mac address in perl (without using ifconfig) ... ?
03:41 spinclad policomb: in perl 6?
03:42 spinclad with the right libraries, and right authorization...
03:42 policomb perl 6 or perl 5, it's the same for me
03:42 policomb i would like to know how to do that
03:42 policomb i'm root
03:44 spinclad in perl 6: you'd have a lot to invent/implement yourself.  in perl 5: dunno, search cpan...
03:46 spinclad or use pack and ioctl, with no portability i expect
03:48 spinclad the replacements for pack in perl 6 are not yet implemented, ioctl likewise
03:50 spinclad (classes managing packed buffers, for pack)
03:50 TimToady my impression is that parrot is still very weak on binary data
03:50 TimToady need compact arrays of ints of various sizes to do it right
03:53 spinclad aye
03:54 petdance joined #perl6
03:58 am0c joined #perl6
04:21 athenot joined #perl6
04:26 lue hello again. Can't wait for the Jerry Lewis Hackathon!
04:26 lue s/Jerry Lewis/Larry Wall/
04:34 lue (methinks the hackathon will be a pain for me on the Rakudo side if I kept on pasting patches. Shouldn't be too much a bother though)
04:38 Trashlord joined #perl6
04:41 * lue --afk &
04:42 LionMadeOfLions joined #perl6
04:54 ned joined #perl6
05:10 yinyin joined #perl6
05:17 meppl joined #perl6
05:46 drasnian joined #perl6
05:46 drasnian left #perl6
05:51 gfx joined #perl6
06:05 yinyin joined #perl6
06:09 snarkyboojum joined #perl6
06:51 kfo joined #perl6
07:02 kaare joined #perl6
07:07 Su-Shee joined #perl6
07:08 Su-Shee good morning
07:08 uniejo joined #perl6
07:28 xabbu42 joined #perl6
08:02 iblechbot joined #perl6
08:16 snarkyboojum joined #perl6
08:19 Toony joined #perl6
08:23 masak joined #perl6
08:23 masak good morning, #perl6.
08:29 flip214 joined #perl6
08:29 flip214 Hi
08:32 mathw hi
08:32 masak hi!
08:33 * Su-Shee sees your hi and raises to "hihi!" ;)
08:34 * masak wonders if that falls under yesterday's ablaut discussion
08:36 sjohnson hi
08:37 masak [haI]
08:37 sjohnson heh
08:38 masak let's do something cool today on #perl6.
08:40 sjohnson that's a good idea
08:40 masak diakopter: I think you've long since filled your quota of 'this number has so many digist/decimals that it causes Rakudo to whimper' rakudobugs :)
08:41 masak rakudo: say (90^^4).WHAT # diakopterbug
08:41 p6eval rakudo 4edd19: OUTPUT«Method 'WHAT' not found for invocant of class 'Undef'␤current instr.: '_block14' pc 29 (EVAL_1:0)␤»
08:41 * masak submits rakudobug
08:41 masak pugs: say (90^^4).WHAT
08:41 p6eval pugs: OUTPUT«Bool␤»
08:47 masak rakudo: say (0 ^^ 0).WHAT # actually, this is wrong too
08:47 p6eval rakudo 4edd19: OUTPUT«Int()␤»
08:47 masak should be Bool().
08:48 mathw why?
08:49 masak mathw: S03:1304.
08:49 masak "Returns the true argument if there is one (and only one).  Returns C<Bool::False> if all arguments are false or if more than one argument is true."
08:50 mathw aaah
08:50 mathw sorry, I was confusing it with infix:<**>
08:52 masak I can see how that would be confusing :)
08:54 masak also from S03: 'Not all types can support the concept of infinity.  Therefore any value of any type may be compared with C<+Inf> or C<-Inf> values, in which case the infinite value stands for "larger/smaller than any possible value of the type."'
08:54 masak I'm confused by the 'Therefore...' phrasing.
08:55 mathw I'm not sure it's 'Therefore'
08:55 masak oh, good.
08:55 mathw 'Therefore' implies that it's obvious
08:55 masak yes.
08:55 mathw which this isn't particularly, it's just rather handy
08:55 masak :)
08:55 masak I'd prefer 'However, "
08:55 mathw but it just needs a slight rephrase
08:55 mathw hmm no, I don't think that'd work
08:56 masak "Nevertheless, " :)
08:56 mathw No, you need to change more than that
08:56 mathw The second sentence has other problems
08:57 eternaleye s/Therefore/So that comparing finite values to infinite ones functions as most people would expect/?
08:57 masak :)
08:57 eternaleye Plus a comma
08:59 masak I feel that Damian's reply to the 'long literals' thread essentially settled the matter.
09:01 masak meeting &
09:04 k23z__ joined #perl6
09:07 k23z__ can Perl5 code be run in Perl6 ?
09:07 am0c joined #perl6
09:07 k23z__ can modules written in Perl5 be called from Perl6?
09:07 k23z__ if not, is this in any future plans of development ?
09:07 araujo joined #perl6
09:16 mathw k23z__: Currently, no and no. jnthn++ has done some work on use of Perl 5 modules. I'm not sure what state that's currently in. It's certainly something highly desirable.
09:16 mathw There is a subset of Perl 5 code which is valid Perl 6, but it's fairly small due to the syntax changes around arrays and hashes and loops and other common things.
09:17 dakkar joined #perl6
09:36 hejki also there is a bigger subset of ill-written perl5 code which is valid in perl6 :)
09:36 hejki e.g. my @ar; @ar[0] = "foo"; :)
09:39 mathw But that's wrong Perl 5
09:40 mathw So it's not ill-written, it's just a syntax error
09:40 hejki err..
09:40 hejki it does work in perl5 tho
09:40 hejki but it's not correct thus making it ill-written imo
09:41 mathw It works??
09:41 mathw Oh wait
09:41 mathw array slice
09:41 huf it's probaby a warning or something
09:41 mathw it probably only works by sheer fluke
09:42 hejki er.. all ill-written perl5 begins with: don't { use warnings; use strict; } :)
09:48 mssm joined #perl6
09:50 jnthn_and_mberen joined #perl6
09:51 jnthn_and_mberen oh hai
09:51 jnthn_and_mberen masak: ping
09:51 jnthn_and_mberen moritz_: ping
09:53 jnthn_and_mberen phenny: tell pmichaud please see http://gist.github.com/321595 if you get chance. :-)
09:53 phenny jnthn_and_mberen: I'll pass that on when pmichaud is around.
09:55 jnthn_and_mberen phenny: tell masak take a look at http://gist.github.com/321596
09:55 phenny jnthn_and_mberen: I'll pass that on when masak is around.
09:56 jnthn_and_mberen phenny: tell moritz_ take a look at http://gist.github.com/321596
09:56 phenny jnthn_and_mberen: I'll pass that on when moritz_ is around.
09:59 jnthn_and_mberen We'll be back. o/ o/
10:16 |newbie| joined #perl6
10:16 |newbie| hi
10:16 |newbie| where can I find gimme5?
10:26 orafu joined #perl6
10:29 masak |newbie|: http://svn.pugscode.org/pugs/src/perl6/
10:29 phenny masak: 09:55Z <jnthn_and_mberen> tell masak take a look at http://gist.github.com/321596
10:29 * masak takes a look
10:30 masak (so, jnthn and mberends have finally merged into a fearsome mechbot?)
10:31 |newbie| hi masak
10:31 masak hi jaffa4
10:32 |newbie| jaffa4?
10:32 masak sorry, am I confusing you with someone else?
10:33 |newbie| no
10:33 masak |newbie|: http://irclog.perlgeek.de/perl6/2010-03-02#i_2055372
10:34 hejki a bit offtopic, but still "related": http://www.patentstorm.us/patents/7028023.html :)
10:35 jaffa4 Could someone tell me how to use gimmi5?
10:36 masak hejki: that's just so depressing.
10:36 hejki almost as depressing as the 'peanut butter sandwich'-patent :P
10:39 sjohnson almost as depressing as having this happen to you: http://images.4chan.org/b/src/1267698521810.png
10:39 * masak sees '4chan', chooses not to be curious
10:39 sjohnson i drew it myself
10:40 sjohnson so it's safe
10:40 sjohnson :)
10:41 masak what does this Japanese twitterer say about Perl 6? http://twitter.com/mr_konn/status/9961701457
10:41 masak something about politics.
10:43 masak jaffa4: I suggest you look at the Makefile in that directory. it uses gimme5.
10:43 jaffa4 never mind, I have figured it out.
10:44 masak goodie.
10:57 jaffa4 masak, how can I downloade the whole http://svn.pugscode.org/pugs/src?
10:58 masak svn checkout http://svn.pugscode.org/pugs
10:58 masak svn checkout http://svn.pugscode.org/pugs/src, even
10:59 hejki s/(heck|ut)//g;
10:59 xinming joined #perl6
11:00 masak both 'checkout' and 'co' should work, according to 'svn help'.
11:01 hejki sure, i just see no reason to waster precious bytes by typing the word as a whole :)
11:02 stepnem joined #perl6
11:02 masak hejki: obbviously you're not using Ctrl+R in bash nearly enough :P
11:06 hejki what? :P
11:07 hejki i'm more of a history | grep-kinda guy :)
11:09 masak Ctrl+R in bash simply rocks.
11:09 hejki it has it's use cases yes
11:09 hejki my favorite bash hotkey is definetly M-_ tho :)
11:11 masak that one's nice as well.
11:11 * masak tends to use Esc . for that, though
11:12 jaffa4 masak, is there an example how to use viv?
11:13 masak jaffa4: there's a USEAGE heredoc in the source.
11:13 hejki masak: well you prolly aren't on a similar keyboard layout either :)
11:14 jaffa4 yes, I know about that.
11:14 jaffa4 I am interested in something else.
11:15 jaffa4 how to catch actions in viv?
11:16 masak hejki: I run a US layout most of the time.
11:16 masak jaffa4: I don't understand the question.
11:16 jaffa4 is there a way to hook actions in viv?
11:17 jaffa4 or walkthrough the datastructure of the syntax tree using a perl program?
11:18 masak the latter is definitely possible.
11:18 masak just pipe it into a perl program.
11:19 jaffa4 without processing
11:19 jaffa4 the output
11:19 masak er.
11:20 masak That's like asking "Is there a way to not be so thirsty? Without drinking."
11:25 jaffa4 look
11:26 jaffa4 here it is
11:26 jaffa4 $r = STD->parse($PROG, actions => 'Actions')->{'_ast'};
11:26 jaffa4 the whole ast will be in $r.
11:26 jaffa4 no need to parse the output!
11:26 masak ok, so there you go then.
11:26 jaffa4 line 91 in viv.pl
11:28 masak jaffa4: good luck, whatever it is you're attempting. it sounds interesting.
11:28 masak lunch &
11:30 hejki masak: ok. i'm on FI layout all the time :)
11:30 am0c joined #perl6
11:38 Gothmog_ joined #perl6
11:40 f00li5h[HireMe] joined #perl6
11:41 stepnem joined #perl6
11:42 k23z__ joined #perl6
11:43 k23z__ does p6 have a way of looping through an array taking each consecutive pair ?
11:43 k23z__ like
11:43 k23z__ suppose the array were @a
11:44 k23z__ and I want   $f->($a[$i],$a[$i+1]) for $i (0..@a-2)
11:44 k23z__ does p6 provide some sugar for that ?
11:44 hejki for @ar -> $x, $y { ... } ?
11:44 k23z__ ah yes forgot
11:44 k23z__ thanks
11:44 hejki wait
11:44 k23z__ I used it a couple of months ago I think
11:44 k23z__ yes ?
11:45 hejki it doesn't work exactly like you want it to
11:45 hejki rakudo: for (1..10) -> $a, $b { say "a: $a b: $b"; }
11:45 p6eval rakudo 4edd19: OUTPUT«a: 1 b: 2␤a: 3 b: 4␤a: 5 b: 6␤a: 7 b: 8␤a: 9 b: 10␤»
11:45 k23z__ hmm ..
11:45 hejki i thought you wanted: 1,2 2,3 3,4 4,5 5,6
11:45 k23z__ yes
11:45 hejki you could use Z or X with 2 arrays
11:46 k23z__ I could shift one of the arrays 1 position forward
11:46 k23z__ how do I shift an array with 1 position forward ?
11:46 hejki rakudo: for (1..10) X (2..11) -> $a, $b { say "a: $a b: $b"; } # might fail :)
11:46 k23z__ I mean having  (1,2,3)  ->   (bla,1,2,3)
11:46 p6eval rakudo 4edd19: OUTPUT«Missing block at line 11, near "X (2..11) "␤current instr.: 'perl6;HLL;Grammar;panic' pc 500 (ext/nqp-rx/src/stage0/HLL-s0.pir:328)␤»
11:46 k23z__ hejki: yeah like that
11:46 k23z__ but ...
11:46 k23z__ it's cartesian product
11:46 k23z__ it's more than I need
11:47 hejki yes
11:47 hejki rakudo: for 1..10 Z (2..11) -> $a, $b { say "a: $a b: $b"; }
11:47 p6eval rakudo 4edd19: OUTPUT«a: 1 b: 2␤a: 2 b: 3␤a: 3 b: 4␤a: 4 b: 5␤a: 5 b: 6␤a: 6 b: 7␤a: 7 b: 8␤a: 8 b: 9␤a: 9 b: 10␤a: 10 b: 11␤»
11:47 hejki voilá
11:49 k23z__ yeah that looks great
11:49 k23z__ hejki: now we just need to generate (2..11) from (1..10)
11:49 betterworld alpha: my @a = 1..10; for @a Z (@a >>+>> 1) -> $a, $b { say "a: $a b: $b"; }
11:49 p6eval alpha 30e0ed: OUTPUT«a: 1 b: 2␤a: 2 b: 3␤a: 3 b: 4␤a: 4 b: 5␤a: 5 b: 6␤a: 6 b: 7␤a: 7 b: 8␤a: 8 b: 9␤a: 9 b: 10␤a: 10 b: 11␤»
11:50 k23z__ betterworld: excellent
11:50 k23z__ thank you
11:50 k23z__ I didn't know p6 had these operators but it's very nice it does
11:50 hejki hmm..
11:50 hejki yeh
11:55 betterworld or maybe @a Z @a[1..*] so it works with other arrays too
11:57 hejki even better :)
11:57 hejki and looks better as well :>
12:08 korpenkraxar joined #perl6
12:11 lichtkind joined #perl6
12:12 lichtkind moritz_: got my msg?
12:13 lichtkind masak: moin
12:13 rv2733 joined #perl6
12:15 masak lichtkind: hej, hej
12:15 lichtkind :)
12:15 lichtkind masak: currently working on http://www.perlfoundation.org/perl6/index.cgi?timeline
12:16 masak nice.
12:16 masak lichtkind: guess you saw my comments on that page from yesterday?
12:18 masak lichtkind: "lasted until 2006 August 23" -- maybe make this a separate timeline item instead? makes it easier to see where it fits into the history.
12:19 lichtkind yes and yes
12:19 lichtkind i thought about that
12:19 masak lichtkind: "2010 April Rakudo star release" doesn't fit into the list of "what happened so far".
12:19 masak lichtkind: also, you misspelled "happenEd".
12:19 lichtkind soon it will
12:19 masak lichtkind: you're missing the point.
12:20 masak it hasn't happened so far.
12:20 masak thus it doesn't belong on the list.
12:20 masak it belongs on an other list, "things which haven't happened so far".
12:20 daemon joined #perl6
12:20 masak along with a lot of other important things :)
12:21 hejki imagine slacking like that in a corporate job :)
12:21 k23z__ masak: is there a plan for SDL or OpenGL bindings along the way ?
12:21 ruoso joined #perl6
12:22 masak k23z__: don't think so. please go right ahead and start a project.
12:31 athenot joined #perl6
12:32 lichtkind masak: bah http://www.perlfoundation.org/perl6/index.cgi?timeline
12:33 masak lichtkind: looks better and better :)
12:33 lichtkind masak: i added the hague grant too
12:33 masak 'and other stuff' -- questionable phrasing.
12:33 masak it's too informal and not at all informative.
12:34 masak 'for, among other things, Perl 6' would work better, IMHO.
12:42 petdance joined #perl6
12:56 nacho joined #perl6
13:04 takadonet morning
13:09 masak takadonet: \o
13:09 takadonet masak: how are you doing?
13:09 masak takadonet: pretty well. you?
13:10 takadonet masak: busy but good
13:10 payload joined #perl6
13:12 masak takadonet: yes, me too. presentation in two days. :)
13:12 takadonet where and what is your presentation?
13:13 masak http://www.opensourcedays.org/2010/node/267
13:13 takadonet nice
13:14 masak thanks -- really looking forward to the conference, the hackathon, and everyone's talks.
13:15 takadonet will your presentation  be online after the conference?
13:15 masak online, cc-licenced, possibly even with a video (eventually).
13:16 m-locks aye we'd like to see ;)
13:16 masak that makes me happy, too :)
13:22 athenot joined #perl6
13:23 Tene colomon: explain what I broke?
13:23 colomon Tene: 20-30 tests on my machine.
13:23 colomon including the make test   try.t
13:26 colomon reverting 4edd19dbaaee2dd2c61cbc1044ef39651d45f3e4 made everything work again
13:37 riffraff joined #perl6
13:54 Tene colomon: I see you didn't push a revert commit, yes?
13:54 colomon nope.
13:54 colomon just reverted locally
13:55 colomon (well, I think I pushed one in the series branch, but that's a side thing)
13:55 colomon I was hoping it was something easy for you to fix.  :)
13:56 Tene okay, just checking.  The problem is that I'm not returning the value of the try block in cases where the block doesn't fail.
14:06 Tene Looks like it's a bug with the PAST compiler, not setting the return value of a Stmts node properly when there's a handler on it.
14:07 PerlJam good morning #perl6
14:08 Tene oh, and looks like it misbehaves badly in the case where it's handled
14:09 takadonet morning PerlJam
14:09 Tene Let's try this...
14:10 Tene oh, no, that won't work.
14:14 Tene colomon: feel free to push a revert commit.  I'll try to get back to this later today.
14:15 Tene eh, nm, I'll revert it myself.
14:16 Tene pushed
14:16 dalek rakudo: 4055fde | tene++ | src/Perl6/Actions.pm:
14:16 dalek rakudo: Revert "Re-work try {} to ignore control exceptions"
14:16 dalek rakudo: This reverts commit 4edd19dbaaee2dd2c61cbc1044ef39651d45f3e4.
14:16 dalek rakudo: review: http://github.com/rakudo/rakudo/commit/4055fde9dfd07fbff309a2b2ecb6df13a733fd0d
14:17 m-locks a new one?
14:17 Tene what is a new what?
14:19 nacho joined #perl6
14:34 m-locks a new review :)
14:40 nacho joined #perl6
14:43 masak the use.perl.org home page looks depressingly desolate.
14:44 masak the 12th German Perl Workshop in June is announced there, but not this weekend's Hackathon, or NPW in Iceland, or YAPC::EU in Pisa this summer...
14:46 PerlJam masak: submit those stories then!
14:49 masak what, and lose the opportunity to complain? :P
14:49 masak you're right, of course.
14:50 masak I guess this situation occurs, though, because no-one really cares about use.perl any more, and we should all see about giving it a humane final rest.
14:50 masak it used to be that Parrot and Rakudo releases were announced there too, but not anymore.
14:52 Su-Shee I've never used it and rarely even looked at it.
14:53 Jedai joined #perl6
14:57 masak Su-Shee: ...except when you assiduously read pmichaud's, jnthn's and my blog posts, right? :)
14:59 masak std: .say if .so
14:59 p6eval std 29935: OUTPUT«ok 00:01 106m␤»
14:59 Su-Shee masak: I don't read them over use.perl.org.
14:59 masak Su-Shee: oh, good. because frankly, the CSS is terrible over there.
15:00 SmokeMachine joined #perl6
15:00 Su-Shee "what css?" ;)
15:00 masak 'xactly.
15:00 Su-Shee probably, one day a perl (6?) forum is needed.
15:01 masak maybe a combined forum-and-blog-engine?
15:01 Su-Shee no. let blogs be blogs and forums be forums.
15:02 masak fair enuf.
15:02 Su-Shee though most contemporary forums have a feed.
15:02 Su-Shee (and even fvwm (!) has a forum. ;)
15:04 mathw fvwm still exists??
15:05 masak Perl 6 isn't dead??
15:05 mathw I guess I'm not really all that surprised
15:06 mathw I'm just bored
15:06 Tene man, I've never been able to get into forums.
15:07 * masak neither
15:07 Tene IRC and mailing lists are great, but I feel awkward trying to squeeze anything meaningful and interactive into a web browser.
15:07 Tene I don't really blog, either, though... so I'm a bit unusual.
15:07 masak some people are crazy about them, it seems. different tools for different people, I guess.
15:07 Tene nodnod
15:08 mathw I find forums work for some things, and not for others
15:08 mathw I use them, but only for certain things
15:08 mathw A lot of it's down to the people rather than the subject matter
15:09 frettled and the problem that Yet Another Forum means yet another sign-up procedure, another set of credentials, another set of preferences, and another web page to load :)
15:11 masak *sigh*
15:11 masak hopefully Web 3.0 will have some concept of a web identity, so one doesn't have to create new accounts all the time.
15:12 mathw OpenID, then
15:12 masak yeah.
15:14 nihiliad joined #perl6
15:16 mathw You're all intelligent people, so maybe some of you know about good resources for multithreaded system design and programming techniques in C++?
15:17 colomon_ joined #perl6
15:25 vorner mathw: Actually, I say that C++ is extremely unfriendly to multithreaded programming. I use haskell if possible for such things.
15:32 masak joined #perl6
15:35 Su-Shee mathw: of course it does. and it has features very few window managers have. still a bitch to config, though.
15:36 Tene which is this, Su-Shee?
15:36 Su-Shee I HATE web identities. none of them work really properly or fast and hello? I don't want to spread my identity all over everywhere.
15:37 Su-Shee Tene: native svg support for elements of the wm which actually can be scaled for example.
15:38 Psyche^ joined #perl6
15:39 mathw vorner: yes, I know it is but I don't have a choice of language here
15:39 nihiliad joined #perl6
15:40 vorner mathw: I did some small things using pthreads only, usually by trying to separate the things as much as possible
15:42 bronco_creek joined #perl6
15:47 mathw Yeah, we're a bit beyond small things with this
15:47 mathw I've never been able to find anything seriously good on large-scale architecture of multithreaded systems in C++
15:47 mathw maybe nobody knows how to do it well
16:03 m-locks some stack reconstruction approaches have been researched
16:04 bronco_creek Hi.  This is my first time here.  I'm a has-been/wanna-be perl programmer.  I just got around to installing parrot & rakudo and starting to play around.  I may be seeing a bug.  What's my next step?
16:05 m-locks you try telling whats wrong
16:05 Tene bronco_creek: it's pretty likely that you've found a bug.  You'll want to produce as minimal a test case as you can, and then submit it to the bug tracker.
16:05 Tene rakudobug?
16:05 Tene Oh, right, no purl in here.
16:05 bronco_creek Where should I look to see if it is already reported?
16:05 Tene rakudobug@perl.org
16:05 Tene Hmm.
16:06 bronco_creek Thanks.
16:08 masak joined #perl6
16:18 pugssvn r29936 | lwall++ | [file-tests.t] fix parsefail
16:20 k23z__ joined #perl6
16:20 k23z__ joined #perl6
16:22 masak bronco_creek: we have a rakudo evalbot on the channel. reproducing the bug with this often produces some useful discussion which you can use in the bug report.
16:22 masak rakudo: say (1 ^^ 2).WHAT
16:22 p6eval rakudo 4055fd: OUTPUT«Method 'WHAT' not found for invocant of class 'Undef'␤current instr.: '_block14' pc 29 (EVAL_1:0)␤»
16:22 masak like that. :)
16:23 kfo joined #perl6
16:24 bronco_creek What I'm seeing involves instantiating an object.  Can the evalbot handle multiple lines of code?
16:25 m-locks nope
16:26 TimToady yes, it can
16:26 m-locks whoa
16:26 masak it can handle multiple statements.
16:26 masak just separate them by semicolons.
16:26 TimToady you can feed it ␤ characters that translate into newlines
16:26 masak that, too.
16:26 masak but golfing to a minimal program size helps both bots and people :)
16:26 m-locks but it is all single line input
16:27 bronco_creek Thanks, I'll give it a try.
16:27 uniejo joined #perl6
16:28 TimToady to clarify, the evalbot's irc interface is single line, but it will do translation both coming and going of ␤ characters
16:28 TimToady so the program it is running has a multi-line interface
16:29 m-locks ;)
16:29 TimToady rakudo: { say 'I am a bare block'}␤{ say 'and so am I' }
16:29 p6eval rakudo 4055fd: OUTPUT«I am a bare block␤and so am I␤»
16:30 bronco_creek So, how do I type the "HL" characters?
16:30 TimToady it's NL
16:30 TimToady depends on your windowing system, likely
16:30 TimToady but you can always copy/paste from evalbot output
16:31 bronco_creek I'm using the IRC web client on a Windows system.
16:31 masak TimToady: did you see the confusion in the backlog about the 'Therefore' on S03:1340? any opinions?
16:31 masak bronco_creek: copy/paste?
16:31 bronco_creek OK.
16:32 masak that's what I do. I'm on Emacs, so copying is a no-op.
16:32 bronco_creek rakudo: class Test1 { has $.a; has @.b; has %.c; }  my $tester = Test1.new(a => 'test'); say $tester.a;
16:32 p6eval rakudo 4055fd: OUTPUT«Confused at line 11, near "class Test"␤current instr.: 'perl6;HLL;Grammar;panic' pc 500 (ext/nqp-rx/src/stage0/HLL-s0.pir:328)␤»
16:32 TimToady "Most types don't have Inf, therefore we will supply one"
16:32 masak bronco_creek: need a semicolon after the block.
16:33 bronco_creek OK
16:33 masak TimToady: if it really said that, I'd be less confused.
16:33 TimToady or that ␤ character
16:33 TimToady but semicolon is easier to type :)
16:33 bronco_creek rakudo: class Test1 { has $.a; has @.b; has %.c; };  my $tester = Test1.new(a => 'test'); say $tester.a;
16:34 p6eval rakudo 4055fd: OUTPUT«test␤»
16:34 masak but it says 'Therefore any value of any type may be compared...' -- which is not a consequence of not all types having Inf.
16:34 abra joined #perl6
16:35 TimToady it is a pragmatic connector, much like "hopefully" is not saying that the thing you're talking about is happening with hopefulness
16:35 masak ok. it's not vital that it be changed. I just found it unclear when I saw it.
16:36 TimToady Hopefully the technical point is clear.  :)
16:36 bronco_creek Hmm.  Must be something wrong with my installation.  On my system, any class with     has %.x ;   causes errors.
16:36 japhb masak: Someone was asking about SDL and OpenGL bindings earlier?  In what context?  Rakudo?
16:37 masak japhb: Perl 6 in general, I think. but probably implicitly Rakudo, yes.
16:38 masak TimToady: :)
16:39 japhb OpenGL already works in Rakudo (or rather, it did, back in the old branch, I haven't tested it with new master).  SDL bindings for Parrot are under development again (now by kthakore++), so that should be coming soonish too.
16:40 japhb Come to think of it, I guess now is the perfect time to rebuild and try it.  ;-)
16:40 m-locks do eeeet
16:40 masak japhb: cool! that's good news.
16:41 masak a quick Google on Language Log shows that someone has proposed 'hopeably' as the statement-modifying adverb :)
16:43 cotto_work joined #perl6
16:46 bronco_creek Thanks all.  Your reputation for helping newbies is well deserved.  Hope I can contribute something more useful in the future.
16:46 bronco_creek Now, how do I quit the IRC?
16:46 Tene bronco_creek: /quit
16:47 jnthn_mberends_b joined #perl6
16:47 jnthn_mberends_b oh hai
16:47 jnthn_mberends_b the mech bot returns
16:47 jnthn_mberends_b masak: ping
16:48 jnthn_mberends_b (btw, our wifi is HTTP only at the moment, so can't connect properly to IRC :-/)
16:49 masak jnthn_mberends_mechbot: pong
16:49 jnthn_mberends_b masak: hi
16:49 masak \o/ hi!
16:49 jnthn_mberends_b masak: did you read our stuffs?
16:50 masak I did indeed.
16:50 jnthn_mberends_b ok
16:50 masak you stopped right before the good parts! :)
16:50 masak (i.e. the ones I'm interested in)
16:50 jnthn_mberends_b thoughts on the talk?
16:50 jnthn_mberends_b masak: add them to the agenda then! :-)
16:50 masak oh, is it a talk?
16:50 masak I think I'm missing a bit of context here.
16:51 jnthn_mberends_b masak: no, did yuou see the gist about us trying to do something as a group in pm's talk slot?
16:51 masak no, but I've been toying with the same idea myself.
16:51 masak wondering how to pitch it. :)
16:51 jnthn_mberends_b http://gist.github.com/321595
16:51 * masak reads
16:52 masak Plan A sounds like what I thought of doing. Plan B sounds like fun too. I'm up for either. :)
16:52 masak Plan C sounds a bit risky :P
16:53 jnthn_mberends_b Well, we have no input towards Plan A so far.
16:53 jnthn_mberends_b And we're kinda tight on time...
16:53 masak you don't think we can base it on earlier similar pmichaud talks?
16:53 jnthn_mberends_b such as?
16:53 * masak searches
16:54 * jnthn_mberends_b wonders whether there's been more recent Rakudo update talks than YAPC::Europe.
16:54 pmurias joined #perl6
16:54 masak p'haps not.
16:54 Tene let's how you don't end up implementing Plan Omega: Kill All Humans
16:55 masak right. the YAPC::EU slides are here. http://www.pmichaud.com/2009/pres/yapceu-rakudo/slides/start.html
16:56 masak going through them now to see how much is current and salvageable.
16:56 jnthn_mberends_b Welll, they pre-date nqp-rx and ng.
16:57 masak sure. I'm just thinking whether that impacts the message greatly.
16:57 masak ooh, pmichaud++ used my show-of-hands gimmick! :)
16:57 masak I had forgotten about that.
16:58 jnthn_mberends_b http://www.pmichaud.com/2009/pres/ seems to not have more recent than the YAPC::EU one on Rakudo.
16:58 masak the cards example is teh awesome.
16:58 masak if we do any of the Plans, we must do that.
17:00 TimToady in the phone call yesterday pmichaud++ suggested a panel discussion
17:00 masak I can see that working, too.
17:00 jnthn_mberends_b ooh, interesting
17:00 masak hm, http://www.pmichaud.com/2009/pres/yapceu-rakudo/slides/slide85.html is the first really out-of-date slide.
17:00 jnthn_mberends_b As in questions from the floor? Or?
17:01 TimToady he didn't give any details
17:01 TimToady just noted that four of the leading lights would be there :)
17:01 TimToady and four is a good number for a panel
17:02 masak seems there's a real danger of questions drying up at minute 14... or a meatspace flamewar breaking out... :)
17:02 masak where do I sign up? :)
17:02 TimToady hopefully among the panel :)
17:02 masak I'd pay to see that!
17:02 masak the real problem we're facing is that there's not really any time to rehearse anything.
17:03 masak from that viewpoint, panel discussion sounds like a good option.
17:03 jnthn_mberends_b yes, it's lwo on prep
17:04 jnthn_mberends_b masak: were you at the nordic perl workshop last year, when the go open panel sessions happened?
17:04 masak aye.
17:04 jnthn_mberends_b (Bruce Perens, Larry, Jon Hall)
17:04 masak sat between half of the jnthn-mberends mech bot.
17:04 jnthn_mberends_b iirc that worked as first there was somebody asking the panel questions to get it seded
17:04 * japhb just tried OpenGL examples under current Rakudo
17:05 jnthn_mberends_b And then it was handed over to the audience
17:05 TimToady Maddog drank ten beers during that
17:05 japhb Looks like 'use OpenGL:from<parrot>;' no workie.
17:05 jnthn_mberends_b *seeded
17:05 japhb Is that syntax changed, or just nyi?
17:05 jnthn_mberends_b TimToady: ooh.... :-D
17:05 Tene japhb: it *used* to work.
17:05 TimToady had to run off to pee in the middle...
17:05 Tene I expect it didn't get put back in.
17:05 Tene I can work on that today.
17:05 uniejo joined #perl6
17:05 masak jnthn_mberends_b: I think I'm onto which half of you is at the keyboard :P
17:05 SmokeMachine joined #perl6
17:06 jnthn_mberends_b how'd you guess? :-P
17:06 japhb Tene, OK, thanks
17:07 * masak reads the last line of http://www.pmichaud.com/2009/pres/yapceu-rakudo/slides/slide93.html and snickers
17:07 jnthn_mberends_b lol
17:08 japhb "Magic unicorns may appear."  ... watches white 4-hooved animal with conspicuous spiral horn trot by ... "Right.  Magic unicorns *will* appear."
17:08 jnthn_mberends_b In light of lack of prep time etc, the panel session seems like a good idea.
17:08 jnthn_mberends_b We're up for it.
17:09 TimToady and you all have different cool accents
17:09 masak vee're yp for it.
17:10 jnthn_mberends_b OK, we haz a plan.
17:10 jnthn_mberends_b \o/
17:10 jnthn_mberends_b And pmichaud++ for suggesting it.
17:10 jnthn_mberends_b And TimToady++ for mentioning.
17:10 masak masak: cool Nordic accent. mberends: cool Dutch accent. moritz: cool German accent. jnthn: cool Lolcat accent.
17:10 TimToady shouldn't that be \o/\o/?
17:11 jnthn_mberends_b the mechbot junction collapsed
17:11 Tene people actually prepare and rehearse talks?  Huh, I've been doing it wrong all this time.
17:12 masak Tene: they've threatened to record me on film! that sure got me working.
17:12 jnthn_mberends_b on your accent?
17:12 masak jeees.
17:12 jnthn_mberends_b lollol
17:13 masak I'm also thinking how I might integrate the "Lifts produced by Schindler" pun into my talk... and then send the video to Tene... :)
17:13 Tene :P
17:14 Tene masak: I'm sure you'll get plenty of other puns in.
17:14 masak some of which are even known to work on real humans!
17:15 masak I pass a bit more of the Turing test each day.
17:15 masak I mean... of course I'm human. *cough*
17:16 Tene I've accidentally failed the turing test on IRC several times before.
17:16 TimToady so how come lollol didn't collapse too?
17:16 jnthn_mberends_b it's a one junction
17:16 masak Do you think that I've accidentally failed the turing test on IRC several times before is the reason you're here, Tene?
17:17 masak jnthn_mberends_b: 1o11o1 <--- one junction
17:18 TimToady 101
17:18 masak The Matrix was clearly before its time.
17:19 TimToady where are those backwards kana in Unicode, I wonder...
17:19 Tene TimToady: they were reversed because they were *outside* the matrix looking *in*.
17:20 Tene the real world is little-endian.
17:20 TimToady oh, I thought maybe they were hebrew kana or some such...
17:20 jnthn_mberends_b OK, if anyone has question suggestions for the panel discussion, prefix them as panelbot: and then we can grep 'em out of the infrared clogs
17:20 * colomon proposes the Higgins Test, where we see if a person can successfully fool another person into thinking he is Eliza.
17:20 masak whoa. Hebrew kana...
17:20 masak colomon++ # :D
17:21 TimToady colomon: don't you mean the 'iggins Test, guv'vor?
17:22 colomon TimToady++
17:26 jnthn_mberends_b panelbot: Is the Rakudo grammar expected to converge with STD? # for example
17:26 jnthn_mberends_b We could also have questions taken from IRC too. ;-)
17:26 masak that'd be cool.
17:27 jnthn_mberends_b Just log someone in on the web irc client as panelbot ;-)
17:28 TimToady if you can can have it up on the screen, pm and I can arrange to be on irc then
17:29 TimToady we mentioned that in the phonecall too
17:29 Tene panelbot: Which line of STD.pm is your favorite?
17:29 jnthn_mberends_b can arrange that :-)
17:34 mberends_jnthn_b joined #perl6
17:35 TimToady oh noes
17:35 mberends_jnthn_b gah, ran out of web minutes
17:35 justatheory joined #perl6
17:36 TimToady you're in a maze of twisty little irc nicks, all different...
17:37 TimToady maybe you should just use each web connection for one IP packet, then run TCP on top of that...
17:38 mberends_jnthn_b We has limited minutes, so we'll log off now...we'll be back around 11 CET.
17:38 TimToady 11 p.m.?
17:38 mberends_jnthn_b TimToady: yes
17:38 mberends_jnthn_b :-)
17:39 mberends_jnthn_b o/ o/
17:39 TimToady \o xx *
17:39 lichtkind :)
17:39 lichtkind TimToady: http://www.perlfoundation.org/perl6/index.cgi?timeline ist so  allright?
17:47 stepnem joined #perl6
17:58 diakopter I keep missing masak
17:59 colomon btw (as I see you were involved) I'm pretty sure 1...^* is getting parsed as 1....(^*)
17:59 colomon rakudo: say (^*).batch(10)
17:59 p6eval rakudo 4055fd: OUTPUT«Method 'batch' not found for invocant of class 'Range'␤current instr.: '_block14' pc 29 (EVAL_1:0)␤»
18:00 colomon rakudo: say (^*).iterator.batch(10)
18:00 p6eval rakudo 4055fd: OUTPUT«␤»
18:00 colomon :(
18:02 colomon rakudo: say (0..^*).iterator.batch(10)
18:02 p6eval rakudo 4055fd: OUTPUT«␤»
18:02 colomon rakudo: say (0..10).iterator.batch(10)
18:02 p6eval rakudo 4055fd: OUTPUT«0123456789␤»
18:02 colomon interesting...
18:03 diakopter colomon: you were talking to me?
18:03 colomon diakopter: more or less.  :)
18:03 diakopter oh :)
18:03 colomon you reported that bug to massk, no?
18:03 diakopter yeah
18:04 diakopter 'cept, as usual, I wasn't certain it was a bug
18:04 colomon well, for sure it's a feature which isn't implemented yet.
18:05 colomon but it may be showing a bug in the existing features, too.  ;)
18:06 colomon What I don't know is if it is being treated as 1...0, 1, 2, 3, etc., or 1...(scalar range 0..^*)
18:06 diakopter ah
18:06 diakopter oh
18:07 diakopter rakudo: say (0..{^*}).iterator.batch(10)
18:07 p6eval rakudo 4055fd: OUTPUT«0123456789␤»
18:07 colomon for sure my examples above indicate some kind of range bug has snuck into the code.
18:07 diakopter rakudo: say (0..^{*}).iterator.batch(10)
18:07 p6eval rakudo 4055fd: OUTPUT«0123456789␤»
18:09 lichtkind http://www.perlfoundation.org/perl6/index.cgi?ofun
18:09 lichtkind rewrote that a bit
18:09 lichtkind any complains?
18:14 ShaneC joined #perl6
18:30 arnsholt lichtkind: "Pugs was allegedly developed this way and grow thatswhy in many directions." looks like a mix of an old and a new sentence
18:30 lichtkind arnsholt: it is, your free to optimze :)
18:31 lichtkind im no native speaker
18:33 _newbie_ joined #perl6
18:33 arnsholt I'll have a look at the edit log and try to figure out what you intended to say =)
18:37 jonasbn joined #perl6
18:37 jonasbn good evening all
18:41 colomon o/
18:42 IllvilJa joined #perl6
18:53 justatheory joined #perl6
18:54 cosimo joined #perl6
18:55 prism joined #perl6
19:04 jaldhar joined #perl6
19:05 jaldhar joined #perl6
19:07 athenot_ joined #perl6
19:08 athenot joined #perl6
19:13 payload joined #perl6
19:13 jmarti joined #perl6
19:13 _jaldhar joined #perl6
19:20 flavioglock joined #perl6
19:20 Chillance joined #perl6
19:21 flavioglock hi
19:23 colomon \o
19:25 sbp left #perl6
19:34 fglock joined #perl6
19:36 diakopter fglock: how goes it
19:36 diakopter or should I say, how "go"es it
19:37 fglock diakopter: all is fine :)
19:37 jaldhar_ joined #perl6
19:38 fglock Perlito now has an interpreter written in plain mp6, so eval() is now possible
19:38 diakopter oo
19:38 fglock doesn't work in Go yet, needs some debugging
19:40 fglock today I've added t/20-eval.t - it passes all tests in the Perl5 backend; I'm now debugging the Javascript backend
19:40 diakopter does it work in JS?
19:40 diakopter oh
19:40 ashleydev joined #perl6
19:40 fglock js is missing "exists"
19:41 fglock (looking up in google for hints how to implement it)
19:41 diakopter I can help
19:42 diakopter wait, which "exists"
19:42 diakopter hash key existence?
19:43 fglock yes
19:43 diakopter jsobj.hasOwnProperty(key) tells you if the object has it defined on itself
19:43 fglock thanks :) - testing...
19:43 diakopter typeof(jsobj[key])!='undefined' tells you if the object has it defined anywhere on its prototype chain
19:45 ashleydev left #perl6
19:45 ashleydev joined #perl6
19:55 fglock hasOwnProperty works, testing eval again...
19:59 ive joined #perl6
20:01 nsh joined #perl6
20:38 payload joined #perl6
20:42 pmichaud good afternoon, #perl6
20:42 phenny pmichaud: 09:53Z <jnthn_and_mberen> tell pmichaud please see http://gist.github.com/321595 if you get chance. :-)
20:43 pmichaud phenny:  tell jnthn I will put together an outline of what I intended for the talk; may either be just an outline or may be slides.
20:43 phenny pmichaud: I'll pass that on when jnthn is around.
20:44 pmichaud phenny: tell jnthn I should have the outline a bit later tonight (about 6 hrs from now, or 03h00 UTC)
20:44 phenny pmichaud: I'll pass that on when jnthn is around.
20:48 frettled pmichaud: by golly, you're right, it's after noon :) (hi)
20:48 frettled pmichaud: how are you doing, recovered from snowy hell yet?
20:49 pmichaud snowy hell?
20:49 frettled Didn't you have some rather bothersome snowy days a while ago?
20:49 pmichaud oh, a couple of weeks ago, yes.
20:50 pmichaud that snow's all gone :)
20:50 frettled \o/
20:51 ashleydev left #perl6
20:59 meppl joined #perl6
21:06 colomon we've still got snow here.  :)
21:08 m-locks :)
21:09 frettled some streets are bare, here
21:09 frettled salt helps a lot :)
21:17 SmokeMachine joined #perl6
21:26 buubot joined #perl6
21:26 k23z__ joined #perl6
21:26 k23z__ joined #perl6
21:27 buu joined #perl6
21:30 justatheory joined #perl6
21:30 colomon frettled: our streets are bare, it's actually pretty warm today.
21:31 colomon but there was a lot of snow of the ground, so where it hasn't been cleared at all, it's still pretty deep.
21:33 frettled it was pretty warm here, too, sun shining for three or four days straight, maybe two or three above freezing
21:35 frettled I'm glad I don't live in inland Canada :)
21:35 colomon for sure I'd prefer Newfoundland to most of it...  :)
21:36 colomon and my winter trips to Toronto always seem to involve blizzards.
21:45 pmurias fglock: hi!
21:58 dalek rakudo: 37e5746 | pmichaud++ | docs/release_guide.pod:
21:58 dalek rakudo: Update release_guide.pod with recent release information.
21:58 dalek rakudo: review: http://github.com/rakudo/rakudo/commit/37e574683ed0c55ca1155bf78a9ace7c2e7a3c6a
22:14 k23z__ joined #perl6
22:15 masak joined #perl6
22:16 masak <diakopter> I keep missing masak
22:16 k23z__ joined #perl6
22:16 k23z__ joined #perl6
22:16 masak diakopter: awww! I missed you too! :)
22:16 masak nom nom nom @
22:17 masak in http://www.perlfoundation.org/perl6/index.cgi?ofun
22:17 masak '-O is the command line option for compiler'
22:18 masak should be '-O is A command line option for compilerS'
22:18 masak (hm, perhaps better written '-O is *a* command line option for compiler*s*')
22:18 masak lichtkind: ^
22:19 lichtkind ive seen it :)
22:19 masak s/we human/us humans/
22:19 masak s/our doing/our actions/
22:19 frettled but that's no fun ;)
22:20 masak frettled: no-one said grammatical correctness was going to be fun :)
22:20 lichtkind frettled: whats no fun?
22:20 lichtkind ah
22:20 dennisroberts joined #perl6
22:20 masak frettled: it's *about* fun, not fun per se. :P
22:20 lichtkind change it later
22:20 lichtkind im officially i my bath
22:21 masak I can see how that might complicate wiki editing.
22:24 lichtkind masak: so feel free to edit yourself :)
22:25 frettled masak: writing about fun should be fun :)  that's why they invented fun-roll-loops
22:25 masak lichtkind: thanks for the offer... I think I'll leave it in your competent hands. still working on my slides over here.
22:25 * masak is less stressed-up about the slides than usual
22:26 masak frettled: I remember the Talk page about Discordianism on Wikipedia at some point.
22:26 frettled masak: that must have been fun :)
22:26 masak frettled: it was all about convincing stubborn people not to write the article in a Discordian tone of voice.
22:26 * masak heads over there to see if that discussion is still there
22:27 * frettled too
22:27 frettled "This article needs additional citations for verification." - bwahahahaha
22:29 diakopter masak: caught!
22:30 masak diakopter: hey there. :)
22:30 diakopter fuzz-stress time
22:31 * masak charges up his bionic arms
22:32 masak I might not be here for long.
22:32 masak forewarned is bionic-armed.
22:32 masak (need to regenerate before tomorrow's big journey)
22:34 masak frettled: it must have been this discussion page I was thinking about: http://en.wikipedia.org/wiki/Talk:Fnord
22:35 masak std: [][]
22:35 p6eval std 29936: OUTPUT«ok 00:01 106m␤»
22:35 masak rakudo: [][]
22:35 p6eval rakudo 37e574: OUTPUT«Confused at line 11, near "[][]"␤current instr.: 'perl6;HLL;Grammar;panic' pc 500 (ext/nqp-rx/src/stage0/HLL-s0.pir:328)␤»
22:35 masak is it just something unimplemented, or worthy of a bug report?
22:35 masak alpha: [][]
22:35 p6eval alpha 30e0ed:  ( no output )
22:36 diakopter berg
22:36 masak it regressed, so I'm reporting it.
22:36 * masak submits rakudobug
22:36 masak diakopter: berg? as in salty mountain?
22:36 diakopter rakudoberg
22:36 masak oh, it's a distortion of 'bug'. I see.
22:37 masak well, there's been a berg of bugs lately, that's for sure.
22:43 masak alpha: my $a = 0; $a++++; say $a
22:43 p6eval alpha 30e0ed: OUTPUT«1␤»
22:43 masak rakudo: my $a = 0; $a++++
22:43 p6eval rakudo 37e574: OUTPUT«Cannot assign to readonly value␤current instr.: '&die' pc 16799 (src/builtins/Junction.pir:347)␤»
22:43 diakopter that seems more proper
22:43 masak does it? got spec to back that up?
22:44 masak I always thought postfix:<++> produced lvalues.
22:44 diakopter seeing as how $a++ should return a copy of the value
22:44 diakopter oh
22:44 masak or is that only prefix:<++>?
22:44 masak what does Perl 5 do?
22:44 masak "Can't modify postincrement (++) in postincrement (++)"
22:45 diakopter Can't modify postincrement (++) in postincrement (++) at -e line 1, near "++++"
22:45 diakopter yes, that
22:45 masak excellent error message. :) funny, and to the point.
22:45 wknight8111 joined #perl6
22:45 masak (incidentally funny, that is. I'm sure it's a specific case of a more general error message)
22:45 masak ok, so I won't report Rakudo for that, then.
22:45 jnthn o/
22:45 phenny jnthn: 20:43Z <pmichaud> tell jnthn I will put together an outline of what I intended for the talk; may either be just an outline or may be slides.
22:45 phenny jnthn: 20:44Z <pmichaud> tell jnthn I should have the outline a bit later tonight (about 6 hrs from now, or 03h00 UTC)
22:46 masak rakudo: my $a = 0; ++$a++; say $a
22:46 p6eval rakudo 37e574: OUTPUT«2␤»
22:46 masak \o/
22:47 diakopter rakudo: say 1%%1
22:47 p6eval rakudo 37e574: OUTPUT«Method 'postcircumfix:<[ ]>' not found for invocant of class ''␤current instr.: '!postcircumfix:<[ ]>' pc 10568 (src/builtins/Role.pir:90)␤»
22:47 masak ooh!
22:47 * masak submits rakudobug
22:47 * diakopter wins
22:47 jnthn phenny: tell pmichaud I think we'll go with pannel session rather than a replacement talk.
22:47 phenny jnthn: I'll pass that on when pmichaud is around.
22:47 masak diakopter: not so. it's 1-1 so far.
22:47 diakopter oh
22:47 * diakopter catches up
22:48 diakopter std: say 1%^^1
22:48 p6eval std 29936: OUTPUT«[31m===[0mSORRY![31m===[0m␤Preceding context expects a term, but found infix ^^ instead at /tmp/Yminm8Zt2F line 1:␤------> [32msay 1%^^[33m⏏[31m1[0m␤FAILED 00:01 105m␤»
22:48 diakopter rakudo: say 1%^^1
22:48 p6eval rakudo 37e574: OUTPUT«1␤»
22:49 * masak submits [WEIRD] rakudobug
22:49 diakopter heh
22:49 diakopter that counts as 0.5
22:49 masak TimToady still hasn't acknowledged the 'my @stuff[ %100_000 ]' STD.pm-bug I found a few days back.
22:50 diakopter I didn't see it
22:50 masak std: my @stuff[ %100_000 ]
22:50 diakopter maybe you were netsplit
22:50 p6eval std 29936: OUTPUT«[31m===[0mSORRY![31m===[0m␤Whitespace is required between alphanumeric tokens at /tmp/mTBTXUFHnX line 1:␤------> [32mmy @stuff[ %100[33m⏏[31m_000 ][0m␤FAILED 00:01 105m␤»
22:50 diakopter isn't that just a LTA error?
22:51 k23z__ joined #perl6
22:52 diakopter rakudo: say @*_ ~ 'hi'
22:52 masak it should be kosher, but due to how STD.pm doesn't special-case % there, it isn't.
22:52 masak no, it shouldn't be an error.
22:52 masak by my reading of the spec.
22:52 masak '%' here means 'wrap the edges'/
22:52 masak s[/][.]
22:52 masak according to S09.
22:52 p6eval rakudo 37e574: OUTPUT«Contextual @*_ not found␤current instr.: 'perl6;Perl6Exception;throw' pc 14482 (src/builtins/Seq.pir:52)␤»
22:52 diakopter rakudo: say @*_
22:52 p6eval rakudo 37e574:  ( no output )
22:53 diakopter rakudo: say @*_; say 'hi'
22:53 p6eval rakudo 37e574: OUTPUT«hi␤»
22:53 diakopter sometimes say fails silently
22:53 masak yes. I've had that too.
22:54 masak time for me to sleep.
22:55 masak diakopter: I'll collect any further bugs you find when I wake up :)
22:55 diakopter rakudo: @_.=[3]
22:55 p6eval rakudo 37e574: OUTPUT«Method '!postcircumfix:<[ ]>' not found for invocant of class 'ResizablePMCArray'␤current instr.: '!dispatch_.=' pc 440 (src/glue/dispatch.pir:118)␤»
22:55 masak that one, for example :)
22:55 masak o/
22:55 diakopter anytime I see ResizablePMCArray in an error msg, I consider it a bug
22:55 diakopter l8r
22:57 diakopter rakudo: say @_.perl
22:57 p6eval rakudo 37e574: OUTPUT«Method 'perl' not found for invocant of class 'ResizablePMCArray'␤current instr.: '_block14' pc 29 (EVAL_1:0)␤»
22:57 * jnthn wonders why it's init'd to @_.
23:12 rv2733 joined #perl6
23:18 lichtkind i saved http://www.perlfoundation.org/perl6/index.cgi?poniecode_org
23:20 lichtkind jnthn: o/
23:21 quietfanatic joined #perl6
23:39 pmichaud jnthn: ping
23:39 phenny pmichaud: 22:47Z <jnthn> tell pmichaud I think we'll go with pannel session rather than a replacement talk.
23:41 pmichaud phenny:  tell jnthn  panel session sounds awesome.  Between today and tomorrow I'll put together a list of bullet points of things I wanted to get across/accomplish with my talk; I suspect you all can work those in (or use them as starting items for the talk) on Saturday
23:41 phenny pmichaud: I'll pass that on when jnthn is around.
23:43 colomon pmichaud: Did you see the gather / take bug the other day?  (I'm somewhat out of it at the moment, but...)
23:45 colomon rakudo: ((1...*) Z (3...*)).batch(20).perl.say
23:45 p6eval rakudo 37e574: OUTPUT«(3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22)␤»
23:47 colomon That's a little vague, but I'm 99% sure the issue is that you can only have one instantiation of a given gather block running at a time.
23:48 pmichaud colomon: okay, I can look into a bit.
23:48 pmichaud *into it
23:49 pmichaud the code is supposed to be set up to handle multiple given/take blocks
23:49 colomon Thanks.
23:49 colomon original (more complicated) report was:
23:49 colomon http://lastofthecarelessmen.blogspot.com/2010/03/lazy-sieve-of-eratosthenes-sidetracked.html
23:50 Tene pmichaud: the problem with gather/take is the sub cloning, unless that's been fixed.
23:51 pmichaud that makes sense
23:52 pmichaud I did some work on the sub cloning, but my "fix" caused other things to break so I abandoned it
23:52 pmichaud I'll look into it again.
23:53 colomon just to make sure I was clear -- I'm not seeing a problem using different gathers at the same time.  It's when the same gather code is called twice and you then try to run both iterators at the same time.
23:53 colomon which makes it sound like cloning could indeed be the issue.
23:55 snarkyboojum joined #perl6
23:58 lue hello! (looking at the backlogs, it seems we could define "day" and "night" by the most prominent people around at certain times)
23:58 lue (ie. not day, but masak. Not the nightshift, but the jnthn-shift)

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

Perl 6 | Reference Documentation | Rakudo