Perl 6 - the future is here, just unevenly distributed

IRC log for #perl6, 2013-05-23

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 cognominal__ joined #perl6
00:03 japhb_ Interesting, if I make install for nqp-jvm, ./nqp in the top level directory works, but install/nqp (or cd install; ./nqp) does not
00:03 japhb_ $ install/nqp --version
00:03 japhb_ Unhandled exception: Method 'options' not found
00:03 japhb_ in command_line
00:03 japhb_ in MAIN
00:03 japhb_ in <anon>
00:03 japhb_ in <anon>
00:03 japhb_ (Same error with cd install; ./nqp)
00:19 PacoAir joined #perl6
00:38 dalek perl6-bench: 8ea7c51 | (Geoffrey Broadwell)++ | analyze:
00:38 dalek perl6-bench: Change width attribute of chart divs to max-width
00:38 dalek perl6-bench: review: https://github.com/japhb/perl6-bench/commit/8ea7c517df
00:38 dalek perl6-bench: 0a119a6 | (Geoffrey Broadwell)++ | TODO:
00:38 dalek perl6-bench: Add a feature TODO
00:38 dalek perl6-bench: review: https://github.com/japhb/perl6-bench/commit/0a119a6a8e
00:38 dalek perl6-bench: 9f640ae | (Geoffrey Broadwell)++ | / (2 files):
00:38 dalek perl6-bench: Add nqp-jvm component, build instructions, and compiler info
00:38 dalek perl6-bench: review: https://github.com/japhb/perl6-bench/commit/9f640ae8c4
00:41 japhb_ nqp-jvm REPL seems to silently not work on master -- is this known
00:41 japhb_ ?
00:47 timotimo i knew about it
00:52 sftp joined #perl6
00:54 japhb_ What's the correct syntax for eval'ing a string in NQP?
00:57 japhb_ (In the benchmark code, I need to eval "pir::getinterp__P().recursion_limit(100000);" only when running on Parrot.  If it's in the [unquoted] source, nqp-jvm will choke just trying to compile it, since pir:: isn't recognized)
01:05 xinming joined #perl6
01:07 anuby joined #perl6
01:48 FROGGS_ joined #perl6
02:04 jaldhar joined #perl6
02:08 lustlife joined #perl6
02:09 estrabd joined #perl6
02:10 dayangkun joined #perl6
02:19 prevost joined #perl6
02:23 jaldhar joined #perl6
02:36 TimToady http://rosettacode.org/wiki/Variable_size/Set#Perl_6 to go with the Get entry
02:37 TimToady 3 behind D, and 70 behind Python :)
02:47 Timbus|Away joined #perl6
02:47 labster TimToady++
02:51 colomon TimToady: I'd love to work on some of the bezier curve stuff (assuming someone else hasn't already done it) but it is like #30 on my to-do list of projects, and I'm having a heck of a time getting anywhere with the much more important things at the top of the list.  :\
03:03 dukelet0 colomon: mmmmm, bezier curves :)
03:05 pecastro joined #perl6
03:15 TimToady I promise not to work on them any time soon.
03:30 cognominal__ joined #perl6
03:33 odoacre joined #perl6
03:38 fildon_ joined #perl6
03:57 gdey joined #perl6
03:57 preflex_ joined #perl6
04:25 Psyche^ joined #perl6
04:34 REPLeffect joined #perl6
04:35 TimToady http://rosettacode.org/wiki/Date_manipulation#Perl_6
04:43 zby_home joined #perl6
04:57 japhb_ joined #perl6
05:05 masak good morning, #perl6
05:07 genehack joined #perl6
05:08 TimToady o/
05:08 masak TimToady: "(mostly due to a reticence to inflict Western cultural imperialism on other cultures...or maybe just due to laziness)" -- the latter, I'd say. also, that date format looks very US-centric to me, not just western :)
05:09 TimToady that's why I called the grammar US-DateTime
05:10 masak fairynuff.
05:11 TimToady I was unsuccessful in setting the DateTime timezone however; had to use .delta to get to Zulu
05:11 TimToady and looking at the code, it appears to ignore any named argument it doesn't know
05:14 TimToady maybe I just did it wrong
05:14 masak well, .delta is obviously wrong from a philosophical perspective.
05:14 masak and yes, the constructor probably should carp on unknown nameds.
05:15 TimToady yeah, I was just doing it rwong
05:18 robinsmidsrod joined #perl6
05:18 masak seems $timezone expects to be given in minutes or seconds, if I read the source right.
05:19 moritz \o
05:26 TimToady doesn't work with minutes, but does with seconds
05:27 TimToady anyway, fixed to use :$timezone now
05:34 masak \o/
05:34 * masak away until this evening, teaching
05:53 FROGGS_ joined #perl6
05:57 FROGGS_ joined #perl6
05:58 FROGGS o/
05:59 domidumont joined #perl6
06:02 dalek perl6-bench: b6cd4bd | (Geoffrey Broadwell)++ | analyze:
06:02 dalek perl6-bench: Continued improvements to html_plot output: relative info in plot tooltips, angled X-axis tick labels, undef rate protection, skip empty chart with --compare
06:02 dalek perl6-bench: review: https://github.com/japhb/perl6-bench/commit/b6cd4bd5d2
06:02 dalek perl6-bench: 9a0b2c7 | (Geoffrey Broadwell)++ | analyze:
06:02 dalek perl6-bench: Use a linear X axis when plotting linear-scaled tests, like rc-man-or-boy-test
06:02 dalek perl6-bench: review: https://github.com/japhb/perl6-bench/commit/9a0b2c7323
06:02 dalek perl6-bench: 17e7d2d | (Geoffrey Broadwell)++ | timeall:
06:02 dalek perl6-bench: Increase minimum run time for scalable tests to 3.0 seconds
06:02 dalek perl6-bench: review: https://github.com/japhb/perl6-bench/commit/17e7d2d8e3
06:02 dalek perl6-bench: 1454441 | (Geoffrey Broadwell)++ | timeall:
06:02 dalek perl6-bench: Properly handle linear-scaled tests in timeall
06:02 dalek perl6-bench: review: https://github.com/japhb/perl6-bench/commit/1454441482
06:02 dalek perl6-bench: 46452be | (Geoffrey Broadwell)++ | timeall:
06:02 dalek perl6-bench: Properly skip tests that aren't enabled
06:02 dalek perl6-bench: review: https://github.com/japhb/perl6-bench/commit/46452beb16
06:02 dalek perl6-bench: ce24c29 | (Geoffrey Broadwell)++ | / (2 files):
06:02 dalek perl6-bench: Factor mini-benchmark definitions out of timeall into minibenchmarks.pl; change default size of rc-forest-fire map from 30x20 to 16x16
06:02 dalek perl6-bench: review: https://github.com/japhb/perl6-bench/commit/ce24c29da7
06:04 dalek perl6-bench: 52656de | (Geoffrey Broadwell)++ | jqplot/plugins/jqplot.canvasAxisTickRenderer.min.js:
06:04 dalek perl6-bench: Add another required jqPlot plugin (canvasAxisTickRenderer), in minimized form as usual
06:04 dalek perl6-bench: review: https://github.com/japhb/perl6-bench/commit/52656de7ef
06:08 PacoAir joined #perl6
06:09 dmol joined #perl6
06:13 mikemol joined #perl6
06:14 domidumont joined #perl6
06:20 dalek v5: f42a6ab | (Tobias Leich)++ | lib/Perl5/Actions.nqp:
06:20 dalek v5: eager for loops, moritz++
06:20 dalek v5: review: https://github.com/rakudo-p5/v5/commit/f42a6abcb8
06:20 dalek v5: 276f51e | (Tobias Leich)++ | STATUS.md:
06:20 dalek v5: status update
06:20 dalek v5: review: https://github.com/rakudo-p5/v5/commit/276f51e1d5
06:22 FROGGS I'll start to prepare the nqp/rakudo compiler release in about 12 hours btw
06:25 dalek perl6-bench: 74a2fce | (Geoffrey Broadwell)++ | timeall:
06:25 dalek perl6-bench: Refactor timeall compiler definitions to not use VM{foo} all over the place
06:25 dalek perl6-bench: review: https://github.com/japhb/perl6-bench/commit/74a2fce738
06:26 dalek specs: 381e96a | larry++ | S32-setting-library/Containers.pod:
06:26 dalek specs: first should just return Nil on failure
06:26 dalek specs:
06:26 dalek specs: Don't throw an exception saying there's no value; that's what Nil means!
06:26 dalek specs: review: https://github.com/perl6/specs/commit/381e96a2d7
06:27 TimToady nr: say first * > 5, 1..3
06:27 camelia rakudo a50c2e: OUTPUT«No values matched␤  in method gist at src/gen/CORE.setting:10065␤  in method gist at src/gen/CORE.setting:893␤  in sub say at src/gen/CORE.setting:10963␤  in block  at /tmp/KeF1SaUNxa:1␤␤»
06:27 camelia ..niecza v24-51-g009f999: OUTPUT«Nil␤»
06:27 TimToady rakudobug :)
06:28 FROGGS bah :P
06:33 diakopter r: say first Nil
06:33 camelia rakudo a50c2e: OUTPUT«No values matched␤  in method gist at src/gen/CORE.setting:10065␤  in method gist at src/gen/CORE.setting:893␤  in sub say at src/gen/CORE.setting:10963␤  in block  at /tmp/LCx_5zdaGd:1␤␤»
06:35 kaleem joined #perl6
06:35 diakopter r: say first is last and last is first
06:35 camelia rakudo a50c2e: OUTPUT«[31m===[0mSORRY![31m===[0mâ�¤Undeclared routines:â�¤    and used at line 1. Did you mean '&rand', '&end', '&any'?â�¤    is used at line 1â�¤â�¤Â»
06:36 FROGGS r: say (ord("0") .. ord("A")).elems
06:36 camelia rakudo a50c2e: OUTPUT«18␤»
06:36 FROGGS r: say ("0" .. "A").elems
06:36 camelia rakudo a50c2e: OUTPUT«(timeout)»
06:36 FROGGS r: say ("A" .. "0").elems
06:36 camelia rakudo a50c2e: OUTPUT«0␤»
06:37 diakopter r: say (0 .. "A").elems
06:37 camelia rakudo a50c2e: OUTPUT«Cannot call 'Real'; none of these signatures match:␤:(Mu:U \v: Mu *%_)␤  in method Real at src/gen/CORE.setting:870␤  in method Real at src/gen/CORE.setting:2535␤  in method Real at src/gen/CORE.setting:870␤  in method new at src/gen/CORE.setting:5441␤  in method n…
06:37 FROGGS that could work too...
06:38 FROGGS I like the way you can do math using chars in C...
06:38 FROGGS but I guess "A" + 1 will never happen in perl
06:39 diakopter eval "A"+1
06:39 diakopter eval: "A"+1
06:39 FROGGS 2
06:39 diakopter p5eval: "A"+1
06:39 p5eval diakopter: 1
06:39 IKARi joined #perl6
06:39 FROGGS hmmm
06:39 FROGGS p5eval: -"A" - 1
06:39 p5eval FROGGS: -1
07:13 brrt joined #perl6
07:14 xinming joined #perl6
07:20 SamuraiJack joined #perl6
07:41 nwc10 jnthn: is there a way to get the perl6 compiler to spit out a Java class file that I can run standalone? (at least, stand alone enough that I don't have to load the perl 6 compiler?)
07:42 domidumont joined #perl6
07:43 daxim joined #perl6
07:47 domidumont joined #perl6
07:55 diakopter nwc10: standalone without the runtime? no. standalone without the compiler? no, because the runtime depends on the compiler because of string eval
07:57 diakopter also, there's not actually much of the compiler that's not inextricable from the runtime even if you were to unbind string eval... just the Perl 6 grammar and actions really
07:58 diakopter nwc10: are you thinking about load time or disk size or what?
07:59 diakopter afaik, load time is negligible compared to the jvm and its runtime
08:04 sorear niecza can create standalone runnable assemblies, I don't see the problem
08:04 sorear string eval (there is no other kind in perl 6) throws an exception
08:05 diakopter that was all I meant - you'd have to decouple it
08:05 sorear this is actually necessary because niecza is bootstrapped and we don't want an infinite regress of compilers
08:10 cibs joined #perl6
08:14 imIKARi joined #perl6
08:16 salv0 joined #perl6
08:42 berekuk joined #perl6
08:44 fhelmberger joined #perl6
08:44 berekuk joined #perl6
08:45 arlinius joined #perl6
08:50 dalek v5: ee62a77 | (Tobias Leich)++ | / (5 files):
08:50 dalek v5: moved more specialvars to Perl5::Terms
08:50 dalek v5: review: https://github.com/rakudo-p5/v5/commit/ee62a77af2
08:56 risou_awy joined #perl6
08:56 charsbar_______1 joined #perl6
08:58 Yappo_________ joined #perl6
09:03 woolfy left #perl6
09:08 dakkar joined #perl6
09:08 Nom- joined #perl6
09:12 hypolin joined #perl6
09:18 takesako_____ joined #perl6
09:18 Yappo_________ joined #perl6
09:19 dalek rakudo/nom: 627e17b | (Brent Laabs)++ | src/core/IO.pm:
09:19 dalek rakudo/nom: IO::Handle.slurp now does :enc/:bin, auto-opens prior .open; add .encoding
09:19 dalek rakudo/nom: * IO::Handle.encoding($enc?) added - get/set text encoding
09:19 dalek rakudo/nom: * IO::Handle.slurp no longer needs .open, always calls .close
09:19 dalek rakudo/nom: * sub slurp refactored into IO::Handle version
09:19 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/627e17b664
09:20 FROGGS labster++
09:20 labster Now I just need to spec all of that stuff :o)
09:23 FROGGS :o)
09:24 FROGGS and you need to add .input-line-separator, since I've got failing v5-tests *cough*
09:27 FROGGS mumble, okay, I think the next I do is to support 'use vars qw(...)' and then subroutine prototypes -.-
09:28 FROGGS after that I can apply TimToady++'s patch for indirect object notation *evenmorecough* *g*
09:28 daxim http://weathercrack.com/#!graphs;a=Poland/Warsaw
09:29 tadzik almost sunny
09:30 daxim I'll pack the long trousers :-<
09:31 FROGGS you can unpack them for austin though
09:33 tadzik btw, are you folks aware if trans-atlantic flights have AC power onboard, so I can charge my laptop during flights?
09:33 FROGGS no idea
09:33 FROGGS never left our nice continent :/
09:33 daxim only in first class, and only on very modern planes
09:33 daxim so forget it
09:33 FROGGS I though about bying another batterie pack though
09:34 tadzik I'll buy some extra batteries
09:34 tadzik orr... there's always a slot for a shaver in the toilet, no? :o
09:34 FROGGS *g*
09:36 FROGGS tadzik: there are mobile phone chargers with a crank... maybe there is something for you laptop model?
09:36 FROGGS you have to be very fast though wen compiling rakudo :/
09:36 berekuk joined #perl6
09:36 tadzik hehe
09:40 bonsaikitten tadzik: Emirates, for example
09:41 dalek roast: 2a5f7c4 | (Brent Laabs)++ | S32-io/ (2 files):
09:41 dalek roast: unfudge rakudo in slurp.t and io-path-unix.t, slurp encoding tests
09:41 dalek roast: review: https://github.com/perl6/roast/commit/2a5f7c4776
09:44 imIKARi joined #perl6
09:47 FROGGS I guess I'll a batterie pack for the drive bay and a regular on, so I can hot swap it
09:48 diakopter tadzik: check seatguru.com for your flight to see if there's power
09:52 sqirrel joined #perl6
09:52 tadzik Business Class offers a laptop power connection at every seat. There are no connections in economy class.
09:52 tadzik 9-cell battery it is then :0
10:02 mathw The image of someone sitting with a laptop desperately turning a handle to keep Rakudo compiling has put a big smile on my face this morning
10:05 hypolin joined #perl6
10:52 rindolf joined #perl6
11:00 FROGGS mathw: :P
11:02 tadzik or the handheld wind power plant, sticking your hand outside the airplane windows and yelling "FASTER! FASTER!"
11:06 brrt joined #perl6
11:44 sergot_ hi! o/
11:44 * sergot_ sergot
11:49 drbean joined #perl6
11:50 tadzik o/
12:05 PacoAir joined #perl6
12:05 Vlavv joined #perl6
12:10 aborazmeh joined #perl6
12:15 xinming joined #perl6
12:24 daxim I'm going to demo/walk through http://use.perl.org/use.perl.org/_masak/journal/40516.html at next λheads
12:37 timotimo meh, use.perl.org doesn't seem to load comments fully when you click on them and apparently the functionality is broken completely ... ?!
12:43 brrt left #perl6
12:49 daxim it is
12:49 daxim also, browsing through all my old u.p.o comments, damn, that was a good side
12:49 daxim b.p.o is crap in comparison
12:50 daxim site
12:51 [Coke] timotimo: sorry, which dash-e.t failures?
12:51 timotimo the ones in test-roast.t in the star testing repository
12:53 [Coke] ok. it's on feather - if you have an account there, you should be able to reproduce.
12:53 [Coke] oh.
12:53 timotimo i don't
12:53 [Coke] I suspect that's due to an env variable. See:
12:54 ajr joined #perl6
12:54 [Coke] https://github.com/coke/perl6-roast-data/blob/master/bin/rakudo.sh
12:54 [Coke] vs.
12:54 [Coke] https://github.com/coke/rakudo-star-daily/blob/master/bin/star.sh
12:55 [Coke] I wasn't even looking at roast failures in star, since I already have perl60-roast-data to catch those.
12:55 [Coke] *perl6-r
12:55 daxim bloggers, please revisit http://stackoverflow.com/questions/3135673/what-performance-increases-can-we-expect-as-the-perl-6-implementations-mature and post your 2013 findings
13:01 timotimo "There are also thoughts of porting Rakudo to other VMs, but that'll surely wait till after end of July." -- moritz 2010
13:03 FROGGS well, there is a july comming
13:05 tadzik well, it's after july 2010 now ;)
13:05 timotimo wasn't saying it's wrong
13:06 FROGGS (moritz is never wrong)
13:06 bluescreen10 joined #perl6
13:06 mtk joined #perl6
13:08 [Coke] r: say [+] 1 .. 100000;
13:08 moritz (except when I am)
13:08 camelia rakudo 627e17: OUTPUT«5000050000␤»
13:08 rhr joined #perl6
13:09 [Coke] ah, I remember optimizing something in re: ranges, but it wasn't that.
13:10 timotimo well, i suppose reduce on ranges could do without explicitly generating an actual range
13:10 timotimo er, an actual list
13:10 timotimo kind of like inlining the iteration into the reduce sub
13:10 timotimo no clue if that will make a noticable difference
13:11 sqirrel joined #perl6
13:13 moritz [Coke]: I think you optimizied .elems
13:14 moritz well, for this case you can also use a direct arithmetic expression
13:15 moritz r: multi prefix:<[+]>(Range $r where { $r.min ~~ Int && $r.max ~~ Int }) { ($r.max * ($r.max + 1) / 2).Int }
13:15 camelia rakudo 627e17:  ( no output )
13:15 moritz r: multi prefix:<[+]>(Range $r where { $r.min ~~ Int && $r.max ~~ Int }) { ($r.max * ($r.max + 1) / 2).Int }; say [+] 1 .. 100000
13:15 camelia rakudo 627e17: OUTPUT«===SORRY!===␤CHECK FAILED:␤Calling 'prefix:<[+]>' will never work with argument types (int) (lines 1, 1)␤    Expected any of:␤    :(Range $r)␤»
13:15 moritz r: multi prefix:<[+]>(Range $r where { $r.min ~~ Int && $r.max ~~ Int }) { ($r.max * ($r.max + 1) / 2).Int }; say [+] (1 .. 100000)
13:15 camelia rakudo 627e17: OUTPUT«5000050000␤»
13:15 moritz meh, precedence issue
13:15 moritz r: multi prefix:<[+]>(Range $r where { $r.min ~~ Int && $r.max ~~ Int }) { ($r.max * ($r.max + 1) / 2).Int }; say [+] (1.1 .. 5.1)
13:15 camelia rakudo 627e17: OUTPUT«Cannot call 'prefix:<[+]>'; none of these signatures match:␤:(Range $r)␤  in sub prefix:<[+]> at /tmp/mnf_NdsOey:1␤  in block  at /tmp/mnf_NdsOey:1␤␤»
13:16 moritz and that's the real problem :/
13:17 timotimo oh my :|
13:17 timotimo yeah, longest-token-matching is your friend
13:18 moritz that's not related to LTM
13:18 timotimo hm, so the [+] token isn't a longer token than the reduce + token?
13:18 moritz the problem is that the meta operator generates a prefix:<[+]> with same precedence as a listop
13:20 moritz timotimo: it is, which is why the approach to optimizing meta ops can work at all
13:20 timotimo so, the original meta op will still be a candidate?
13:20 timotimo that's nice
13:20 moritz well, that's the second problem
13:20 moritz the meta op only creates a subroutine if there's none yet
13:21 moritz so once you define your own multi, the normal mechanism doesn't kick in anymore
13:21 timotimo oh, whoops
13:21 moritz r: multi prefix:<[+]>(Range $r where { $r.min ~~ Int && $r.max ~~ Int }) { ($r.max * ($r.max + 1) / 2).Int }; say [+] (1.1 .. 5.1)
13:21 camelia rakudo 627e17: OUTPUT«Cannot call 'prefix:<[+]>'; none of these signatures match:␤:(Range $r)␤  in sub prefix:<[+]> at /tmp/O6a0IyNUvs:1␤  in block  at /tmp/O6a0IyNUvs:1␤␤»
13:21 moritz that's why this one fails
13:21 timotimo right
13:21 aborazmeh joined #perl6
13:21 moritz r: [+] 1, 2; do { multi prefix:<[+]>(Range $r where { $r.min ~~ Int && $r.max ~~ Int }) { ($r.max * ($r.max + 1) / 2).Int }; say [+] (1.1 .. 5.1) }
13:21 camelia rakudo 627e17: OUTPUT«Cannot call 'prefix:<[+]>'; none of these signatures match:␤:(Range $r)␤  in sub prefix:<[+]> at /tmp/OJ3gIVHPFd:1␤  in block  at /tmp/OJ3gIVHPFd:1␤␤»
13:21 timotimo hm, but wouldn't that only be acceptable if the user didn't override + in that lexical scope?
13:22 moritz proper checks must be made while optimizing such things, yes
13:22 timotimo r: sub infix:<+>($a, $b) { $a - $b }; say [+] (1 ... 50)
13:22 camelia rakudo 627e17: OUTPUT«-1273␤»
13:22 moritz like &infix:<+> === &CORE::infix:<+> or so
13:23 timotimo so it would indeed be something the optimizer would have to do? insert a correct optimised-metaop in the tightest scope that makes it still work or something?
13:24 moritz I'm not sure
13:28 kivutar joined #perl6
13:29 Ulti_ daxim most of the time is just Rakudo startup
13:29 daxim needs more lazy loading of features, then
13:30 Ulti or just dont worry about the half second it takes something to run
13:30 Ulti and use perl5 for your dirty one liners
13:31 Ulti the point at which perl5 use case falls over and needs perl6 is large systems that aren't limited by the startup time of the interpreter
13:32 Ulti its as if sysadmins and ex-sed/awk users believe thats the only thing perl is for
13:32 timotimo i'm still hoping the startup time will improve dramatically some day
13:32 ajr_ joined #perl6
13:32 Ulti I use perl to run calculations that take months to finish, the runtime isnt even language limited using C wouldnt be much faster
13:33 Ulti I'm not like OMG 0.005s startup this is the best language for the task
13:33 Ulti if you want fast startup time buy an SSD for your computer
13:33 timotimo i don't want to learn perl 5 just to make fast one-liners :P
13:34 timotimo i have an SSD, in fact
13:34 timotimo perl6 -e 'say 10' -> 1.68user 0.32system 0:02.07elapsed 96%CPU (0avgtext+0avgdata 179716maxresident)k
13:35 Ulti perl6 -e 'exit' real 0m 0.433s
13:35 timotimo your ssd must be much better than mine
13:35 Ulti perl -e 'exit' real 0m 0.005s
13:36 Ulti perl6 -e 'my $i += $_ for 1 .. 100; say $i;' real 0m 0.461s
13:36 timotimo what kind of beast machine is this? :)
13:36 Ulti perl -e '$i += $_ for 1 .. 100; print $i;' real 0m 0.007s
13:36 timotimo oh, i may have set my processor to powersave ...
13:36 Ulti guess I should have used strict :)
13:37 timotimo okay, it's 0.71s now for -e exit
13:37 Ulti < timotimo> what kind of beast machine is this? :)   <--- a MacBook Air 1.7GHz i5
13:37 Ulti >:3
13:40 Ulti I'm just waiting for threaded loops in perl6 so I can do a super fair benchmark of rakudo against perl5 on my 24 core dual xeon machine :)
13:40 Ulti lets see who wins then
13:41 timotimo bwahaha :)
13:42 Ulti seriously though perl5 vs perl6 "debate" has far too many perl5 sysadmin voices
13:43 geekosaur (hey now...)
13:43 Ulti as a Bioinformatician there is essentially nothing wrong with the state of perl6 right now apart from the limited port of BioPerl libraries, which I plan to help out with once I finish my thesis
13:44 bamboo joined #perl6
13:44 timotimo what about those rosalind exammmples that are too slow to even finish? ;(
13:44 timotimo (in one patient-person-time)
13:44 Ulti orly
13:44 Ulti the only thing I'm :'( about is the runtime of grammars on big data files
13:45 Ulti but that really comes from being greedy and objectafying everything rather than just writing the simple parser you would have otherwise written
13:46 ajr_ I golfed the startup down to perl[6 ] -e ""
13:46 ajr_ Is anyone interested in the results?
13:47 geekosaur although, as a perl5 sysadmin, one of my big regrets is not having the time to sit down and figure out perl6 properly. wonder how much of that is in the background of my colleagues?
13:47 moritz ajr_: yes
13:48 ajr_ time perl -e ""
13:48 ajr_ 0.00user 0.00system 0:00.00elapsed 0%CPU (0avgtext+0avgdata 5744maxresident)k
13:48 ajr_ 0inputs+0outputs (0major+422minor)pagefaults 0swaps
13:48 timotimo holy hell, that's fast
13:48 ajr_ time perl6 -e ""
13:48 ajr_ 0.57user 0.15system 0:00.73elapsed 98%CPU (0avgtext+0avgdata 392480maxresident)k
13:48 ajr_ 0inputs+0outputs (0major+28073minor)pagefaults 0swaps
13:49 Ulti geekosaur: I'm a "bash sysadmin" on the side... why perl?
13:49 geekosaur watch out for cache effects when timing stuff like that
13:49 timotimo 60x as much memory usage, too
13:50 geekosaur Ulti, I'm faitly familiar with bash, but it still doesn't take much for me to decide that it'll be easier or more readable in perl
13:50 geekosaur that said I will do in shell what can be done in shell
13:50 Ulti I guess most of our cron stuff is all perl, but I wouldnt call what runs "sysadmin"
13:50 ajr_ The 5 job was the first one. I ran the 6 several times to let the cache effects settle. (The first one was
13:50 ajr_ time perl6 -e ""
13:50 ajr_ 2.11user 0.49system 0:03.49elapsed 74%CPU (0avgtext+0avgdata 392512maxresident)k
13:50 ajr_ 47640inputs+0outputs (31major+28045minor)pagefaults 0swaps
13:51 geekosaur but about the time I need some kind of map/dictionary structure that's sane, or the quoting gets funky, I switch to something sane
13:51 geekosaur which is mostly perl in my case although sometimes python for coworkers' sakes
13:52 geekosaur (said coworkers don't know shell very well either...)
13:52 Ulti I installed rakudo with the prefix /dev/shm that was the right thing to do, right? ;P
13:52 Ulti geekosaur: yeah I agree with that
13:53 ajr_ I did some straces to see if it was purely file finding and reading time, but if I've interpreted the results correctly,
13:53 timotimo shell syntax is very off-putting to me, tbh
13:53 ajr_ there's some heavy thinking going on.
13:53 Ulti geekosaur or as soon as you need anything other than integer arithmetic
13:53 Ulti :)
13:54 geekosaur enh
13:54 geekosaur I have been known to pipe to dc in a shell script, when nothing else was available
13:54 geekosaur (hello older solaris...)
13:55 Ulti error token is ".0 "  <--- dont lie bash, you have never tokenised input
13:55 geekosaur it does tokenize, actually. they're just not what we're used to thinking of as tokens
13:55 Ulti :)
13:56 geekosaur also the builtins are not parsed *as* builtins, each one does its own parse
13:56 geekosaur (aside from e.g. control structures)
13:57 ajr_ perl 5 actually got slightly worse on subsequent tries:
13:57 ajr_ perl -e ""
13:57 ajr_ 0.00user 0.01system 0:00.01elapsed 90%CPU (0avgtext+0avgdata 5728maxresident)k
13:57 ajr_ 0inputs+0outputs (0major+421minor)pagefaults 0swaps
13:59 Ulti geekosaur: if you want a "fun" sysadmin perl6 challenge use the perl6 IRC module (which is really quite nice) to stream filtered logs from a load of servers to an IRC channel, then have it so you can just chat to the server to run commands
13:59 timotimo that's not just slightly, that's infinity times slower!
13:59 Ulti also I want that tool it would be p. useful
14:00 timotimo accepting commands from IRC, what could possibly go wrong
14:00 Ulti obviously on your own server with encryption jees
14:00 Ulti also the commands dont have to be arbitrary
14:02 FROGGS joined #perl6
14:03 Ulti heh the more I think about this the more I might do it myself just as a research tool, just type SQL into IRC and have whichever server finds it has the database run the query and give me the result
14:03 timotimo :)
14:03 dalek v5: 859212e | (Tobias Leich)++ | / (3 files):
14:03 dalek v5: fixed: require VERSION
14:03 dalek v5: review: https://github.com/rakudo-p5/v5/commit/859212e8e7
14:04 Ulti or have servers bid on chunks of a data file to process and have a prallel job just work by itself without me thinking
14:05 Ulti install some android push IRC bridge to my phone... and I dont even need to do work anymore
14:08 PerlJam Add the voice-to-text interface and you can just call your system and tell it what to do just like any other minion
14:09 GlitchMr rn: / #`[A comment] /
14:09 domidumont joined #perl6
14:09 camelia rakudo 627e17: OUTPUT«[31m===[0mSORRY![31m===[0mâ�¤Regex not terminatedâ�¤at /tmp/W2dH483kaE:1â�¤------> [32m/ #`[A comment] /[33mâ��[31m<EOL>[0mâ�¤    expecting any of:â�¤        statement listâ�¤        prefix or termâ�¤        prefix or meta-prefixâ�¤        infix stopperâ�¤Â»
14:09 camelia ..niecza v24-51-g009f999:  ( no output )
14:09 GlitchMr std: / #`[A comment] /
14:09 camelia std 0336087: OUTPUT«[31m===[0mSORRY![31m===[0mâ�¤Null pattern not allowed at /tmp/6DjTukt9xb line 1:â�¤------> [32m/ #`[A comment] [33mâ��[31m/[0mâ�¤    expecting any of:â�¤     statement endâ�¤  statement listâ�¤Parse failedâ�¤FAILED 00:00 41mâ�¤Â»
14:09 GlitchMr Niecza allows null pattern pattern
14:11 colomon weird
14:14 PerlJam That's a rakudo buglet that #`[] isn't recognized in regex, isn't it?   (at least that's what I think is causing the "Regex not terminated" error)
14:15 moritz PerlJam: correct
14:18 yx joined #perl6
14:22 domidumont joined #perl6
14:23 FROGGS rn: / #`[A comment]␤ /
14:23 camelia rakudo 627e17, niecza v24-51-g009f999:  ( no output )
14:24 FROGGS rn: say "a" ~~ / #`[A comment]␤ /
14:24 camelia rakudo 627e17, niecza v24-51-g009f999: OUTPUT«「」␤␤»
14:24 FROGGS rakudo should complain too about null pattern
14:25 moritz it's null pattern detection is known to be very incomplete
14:25 FROGGS yesh
14:25 moritz *its
14:27 bluescreen10 joined #perl6
14:34 prevost joined #perl6
14:36 xilo joined #perl6
14:37 SamuraiJack joined #perl6
14:43 spider-mario joined #perl6
14:48 imIKARi joined #perl6
14:50 zby_home_ joined #perl6
14:51 rindolf joined #perl6
14:54 lizmat joined #perl6
14:54 FROGGS t/spec/S32-hash/delete-adverb.rakudo                    (Wstat: 0 Tests: 108 Failed: 1)
14:54 FROGGS Failed test:  27
14:55 FROGGS not ok 27 - slice unexisting single key/value   # got: Any   # expected: $("e", Any)
14:57 FROGGS nr: my %h; say %h<e>:!kv:delete # why should this give ('e', Any) ?
14:57 camelia rakudo 627e17, niecza v24-51-g009f999: OUTPUT«(Any)␤»
14:58 FROGGS that is the only error I see, so it is going to be a nice release day o)
14:58 FROGGS :o)
14:59 lizmat FROGGS: that wasn't marked as todo ?
15:00 lizmat it should, because the key did not exist, and you're asking for a key/value even if it doesn't exist (because of the ! in :!kv)
15:00 FROGGS lizmat: looks like the test count is wrong
15:01 FROGGS 6 todo for 4 is_deeply checks
15:01 lizmat let me check
15:01 * lizmat just arrived in the east of Germany  :-)
15:01 FROGGS where exactly?
15:01 lizmat Querfurt
15:02 lizmat in the Eisleben / Halle area
15:02 FROGGS ahh
15:04 lizmat hmmm.. that file doesn't have a plan, so I can't get the number of tests wrong
15:04 FROGGS lizmat: looks like *I* cant count
15:04 FROGGS there is '6 todo', followed by 2xok, and 4xis_deeply
15:05 woolfy joined #perl6
15:06 FROGGS lizmat: the 'my' declaration gets the todo, so the last test doesnt
15:06 lizmat argh, it's # lines, not # of tests...
15:07 kaare_ joined #perl6
15:07 dalek roast: 298edbd | (Elizabeth Mattijsen)++ | S32-hash/delete-adverb.t:
15:07 dalek roast: Fix fudging, it's # of lines, *NOT* number of tests, stupid!  (that would be me)
15:07 dalek roast: review: https://github.com/perl6/roast/commit/298edbdab5
15:12 lizmat FROGGS: are we ok now?
15:13 bamboo joined #perl6
15:15 FROGGS I'll check
15:16 lizmat tadzik: "or the handheld wind power plant, sticking your hand outside the airplane windows and yelling "FASTER! FASTER!""
15:17 lizmat that is actually exactly how the fallback fallback power supply of many planes work:
15:17 tadzik haha
15:17 lizmat when needed, a small propellor comes out of the hull in the back and generates electricity
15:18 tadzik I just had an idea that it could generate electricity from braking, like a car does. That may not work though.
15:18 lizmat that's what a Prius does already
15:18 tadzik yes
15:18 tadzik may be tricky with a plane
15:18 lizmat ah, you mean the plane
15:19 lizmat generally, they're glad if they don't smoke the tires
15:19 lizmat also: the energy flow would be very peaky: a fast brake in a Prius will also not load the battery, as it wouldn't be able to cope with that much energy in a short period
15:20 daxim http://enwp.org/regenerative_braking
15:22 lizmat what I said :-)
15:27 bamboo joined #perl6
15:31 masak evenin', #perl6
15:32 tadzik it's masak! \o/
15:32 * masak double-checks, just to be sure
15:32 masak yup.
15:34 lizmat masak+
15:35 lizmat at least one of them, right?
15:35 masak nr: say first * > 5, 1..3
15:35 camelia niecza v24-51-g009f999: OUTPUT«Nil␤»
15:35 camelia ..rakudo 627e17: OUTPUT«No values matched␤  in method gist at src/gen/CORE.setting:10065␤  in method gist at src/gen/CORE.setting:893␤  in sub say at src/gen/CORE.setting:10963␤  in block  at /tmp/gR86f5feCs:1␤␤»
15:36 * masak submits TimToady++'s rakudobug from earlier
15:36 masak <TimToady> rakudobug :)
15:37 FROGGS t/spec/S32-hash/delete-adverb.rakudo  TODO passed:   17, 22-23, 25, 28-29, 31, 34-37, 39, 54, 68
15:37 FROGGS Result: PASS
15:37 FROGGS lizmat: ^^
15:38 FROGGS nr: say first Nil
15:38 camelia rakudo 627e17: OUTPUT«No values matched␤  in method gist at src/gen/CORE.setting:10065␤  in method gist at src/gen/CORE.setting:893␤  in sub say at src/gen/CORE.setting:10963␤  in block  at /tmp/4oL6SnCZ4n:1␤␤»
15:38 camelia ..niecza v24-51-g009f999: OUTPUT«Unhandled exception: No value for parameter '$test' in 'first'␤  at /home/p6eval/niecza/lib/CORE.setting line 0 (first @ 1) ␤  at /tmp/mhKl5YSEUm line 1 (mainline @ 3) ␤  at /home/p6eval/niecza/lib/CORE.setting line 4331 (ANON @ 3) ␤  at /home/p6eval/niecz…
15:39 TimToady that's smartmatching with Nil, which is always false
15:40 woolfy left #perl6
15:40 TimToady nr: say Nil ~~ Nil
15:40 camelia rakudo 627e17, niecza v24-51-g009f999: OUTPUT«True␤»
15:40 masak nr: say grep * > 5, 1..3
15:40 camelia rakudo 627e17, niecza v24-51-g009f999: OUTPUT«␤»
15:40 TimToady two bugs for the price of one :)
15:40 lizmat afk until maybe later tonight / tomorrow
15:41 masak TimToady: I see one: 'Nil ~~ Nil' should be False.
15:41 masak what's the other one?
15:41 TimToady nr is two things
15:42 masak TimToady: waitwait, why was 'nr: say first * > 5, 1..3' => 'No values matched' a bug in Rakudo?
15:43 masak I almost pushed the "Send" button, but now I'm not so sure.
15:43 flussence probably because there's a backtrace in there for no apparent reason
15:43 moritz masak: I think TimToady meant that Nil~~Nil being true is a bug both in Rakudo and Niecza, so two bugs
15:44 flussence oh wait
15:44 masak flussence: it's a Failure, which gets triggered because we try to print the (non-existent) first value.
15:44 flussence never mind, I misread your line as the actual code
15:44 masak moritz: ok. then I see why I was confused; I only report Rakudobugs ;)
15:44 masak (except when I don't)
15:45 flussence nr: say first * > 5, 1..3 => 'No values matched' # just ooc, I wonder what this'll do
15:45 camelia niecza v24-51-g009f999: OUTPUT«Unhandled exception: Cannot use value like Pair as a number␤  at <unknown> line 0 (ExitRunloop @ 0) ␤  at /home/p6eval/niecza/lib/CORE.setting line 293 (Any.Numeric @ 6) ␤  at <unknown> line 0 (ExitRunloop @ 0) ␤  at /tmp/_GYD2ok9bg line 0 (ANON @ 1) ␤  at…
15:45 camelia ..rakudo 627e17: OUTPUT«Cannot call 'Real'; none of these signatures match:␤:(Mu:U \v: Mu *%_)␤  in method Real at src/gen/CORE.setting:870␤  in sub infix:<>> at src/gen/CORE.setting:3047␤  in sub infix:<>> at src/gen/CORE.setting:3045␤  in block  at /tmp/MDwtBBP4Kx:1␤␤»
15:45 masak TimToady: timing out; unsubmitting bug.
15:47 masak if anything, that's a Nieczabug.
15:49 masak no, I change my mind again: the behavior is *unspec'd*, and Rakudo and Niecza are just exposing two reasonable implementations of the hole in the spec.
15:49 masak so, which one do we like better? :)
15:49 masak let's fill the hole in the spec with our preferred implementation.
15:50 TimToady which behavior are you referring to?
15:50 moritz niecza's is more user friendly
15:50 moritz nr: say (42 => 5) > 1
15:50 camelia niecza v24-51-g009f999: OUTPUT«Unhandled exception: Cannot use value like Pair as a number␤  at <unknown> line 0 (ExitRunloop @ 0) ␤  at /home/p6eval/niecza/lib/CORE.setting line 293 (Any.Numeric @ 6) ␤  at <unknown> line 0 (ExitRunloop @ 0) ␤  at /tmp/HXsebyRmh7 line 1 (mainline @ 3) ␤…
15:50 camelia ..rakudo 627e17: OUTPUT«Cannot call 'Real'; none of these signatures match:␤:(Mu:U \v: Mu *%_)␤  in method Real at src/gen/CORE.setting:870␤  in sub infix:<>> at src/gen/CORE.setting:3047␤  in sub infix:<>> at src/gen/CORE.setting:3045␤  in block  at /tmp/X9yaimO7bk:1␤␤»
15:53 masak TimToady: I'm referring to the rakudobug you reported and I just unsubmat.
15:53 masak nr: say first * > 5, 1..3
15:53 camelia niecza v24-51-g009f999: OUTPUT«Nil␤»
15:53 camelia ..rakudo 627e17: OUTPUT«No values matched␤  in method gist at src/gen/CORE.setting:10065␤  in method gist at src/gen/CORE.setting:893␤  in sub say at src/gen/CORE.setting:10963␤  in block  at /tmp/YvhvRLAF2u:1␤␤»
15:53 masak that one.
15:53 masak the spec doesn't say what to do in this situation.
15:53 masak nr: say first 5, ()
15:53 TimToady yes it does :)
15:53 camelia niecza v24-51-g009f999: OUTPUT«Nil␤»
15:53 camelia ..rakudo 627e17: OUTPUT«No values matched␤  in method gist at src/gen/CORE.setting:10065␤  in method gist at src/gen/CORE.setting:893␤  in sub say at src/gen/CORE.setting:10963␤  in block  at /tmp/mAqDm3pDCS:1␤␤»
15:53 masak TimToady: where?
15:53 masak I went to S32/Containers and looked for 'first' there.
15:54 masak maybe I looked in the wrong place?
15:54 TimToady did you git pull?
15:55 * masak does so now
15:55 * masak submits rakudobug
15:56 TimToady that's why there was a smiley on it, because I'd just done that
15:56 masak ah.
15:56 masak sorry, I'm too exhausted to get subtle jokes ;)
15:56 TimToady in general, I'm of the opinion that subscript failure should probably just return Nil, not failure
15:57 TimToady (or a WHENCE when it might be an lvalue)
15:57 TimToady but lists can't do that, so Nil
15:58 TimToady (arrays probably return their default value instead of Nil)
15:58 moritz oh wow
15:58 moritz that means that stuff like @a[2..*] won't DWIM anymore
15:59 moritz because it returns a trailing, infinite list of Nil's
15:59 TimToady ranges can still be special
16:00 TimToady in the same way we're already distinguishing it such that it's currently returning a failure
16:00 saxx11 joined #perl6
16:00 PacoAir joined #perl6
16:00 TimToady Nil is just the most benign sort of failure, and I think subscripting is a good place to keep it simple
16:02 TimToady or going the other way, one can think of a failure as Nil with extra baggage
16:04 leeder-krenon joined #perl6
16:05 AMENDEES_ joined #perl6
16:10 FROGGS $ perl6 -e 'my ($a, $c); ($a, $, $c) = 1..3'
16:10 FROGGS Non-declarative sigil is missing its name
16:10 FROGGS ------> my ($a, $c); ($a, $⏏, $c) = 1..3
16:11 FROGGS hmmmm
16:11 masak I'm not sure it makes much sense to use '$' outside of a signature.
16:12 FROGGS but it might be still better to treat it as '$' and not '$,' there, right?
16:12 TimToady nr: my ($a, $c); ($a, *, $c) = 1..3;
16:12 camelia rakudo 627e17, niecza v24-51-g009f999:  ( no output )
16:12 TimToady FROGGS: that's why I did the STD patch
16:13 TimToady std: my ($a, $c); ($a, $, $c) = 1..3
16:13 camelia std 0336087: OUTPUT«[31m===[0mSORRY![31m===[0m�Non-declarative sigil is missing its name at /tmp/wTvOaI3RgM line 1:�------> [32mmy ($a, $c); ($a, [33m�[31m$, $c) = 1..3[0m�Use of uninitialized value $first in string eq at STD.pm line 66016.�Use of uninitialized value $first in s…
16:14 FROGGS TimToady: ahh, there already is a patch *g*
16:17 TimToady std: my ($a, $c); ($a, $(), $c) = 1..3;  # curious
16:17 camelia std 0336087: OUTPUT«ok 00:00 47m␤»
16:17 TimToady npr: my ($a, $c); ($a, $(), $c) = 1..3;  # curious
16:17 camelia niecza v24-51-g009f999: OUTPUT«Unhandled exception: Unable to resolve method ast in type Any␤  at /tmp/gp1hCMi0wY line 1 (mainline @ 4) ␤  at /home/p6eval/niecza/lib/CORE.setting line 4331 (ANON @ 3) ␤  at /home/p6eval/niecza/lib/CORE.setting line 4332 (module-CORE @ 582) ␤  at /home/p6…
16:17 camelia ..pugs:  ( no output )
16:17 camelia ..rakudo 627e17: OUTPUT«No such method 'ast' for invocant of type 'Any'␤  in block  at /tmp/jORWahGc1p:1␤␤»
16:18 TimToady ah, $() is special already
16:29 SamuraiJack joined #perl6
16:36 dalek rakudo/nom: 25bfa5f | (Tobias Leich)++ | src/Perl6/Grammar.nqp:
16:36 dalek rakudo/nom: by TimToady++: only carp about P5 $, variable where it is obvious
16:36 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/25bfa5fa80
16:42 dalek rakudo/nom: d3b36e4 | jonathan++ | README:
16:42 dalek rakudo/nom: Fix year in README.
16:42 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/d3b36e40f3
16:42 yx joined #perl6
16:43 timotimo fix the whole year? wow.
16:43 FROGGS 2013 is safe now man
16:44 jnthn you're welcome
16:44 timotimo jnthn and his fabulous cape save the day^Wyear again
16:45 timotimo (jnthns cape is actually shaped like two +es)
16:46 dmol joined #perl6
16:48 denisboyun joined #perl6
16:51 FROGGS so like a # ?
16:52 TimToady that's four
16:52 masak so, like a ‡ ?
16:53 FROGGS I'd expected to see a 'qnthn' on his cape
16:53 TimToady
16:53 FROGGS or maybe the 'j' is on purpose and the jvm port was destiny?
16:54 TimToady
16:56 ajr joined #perl6
16:57 japhb_ joined #perl6
16:58 masak TimToady: ☨
16:58 timotimo :D
16:58 masak "cross of jnthainne"
17:00 dalek Heuristic branch merge: pushed 32 commits to rakudo/jvm-support by jnthn
17:11 dalek rakudo/jvm-support: 4e74ee5 | jonathan++ | tools/build/ (2 files):
17:11 dalek rakudo/jvm-support: Tweak some memory flags.
17:11 dalek rakudo/jvm-support: review: https://github.com/rakudo/rakudo/commit/4e74ee5905
17:12 timotimo "tweak JVM to use less RAM"
17:13 jnthn Well, "tweak it to use enough RAM to get through the setting" too :)
17:13 jnthn (Its default limits vary by system.)
17:14 timotimo hehe
17:14 timotimo is that the "code object too large" problem?
17:15 jnthn timotimo: that was in NQP and I did the patch last night
17:15 jnthn I will merge stuff after the Rakudo release today though
17:15 timotimo oh, ok
17:22 Util jnthn: who is cutting the Rakudo release today?
17:22 jnthn 2013-05-23   Rakudo #64 "Austin"             FROGGS
17:23 rindolf joined #perl6
17:24 jnthn Looks like masak gets to relesae the first Rakudo with some amount of JVM support :)
17:25 masak yesssss! \o/
17:25 jnthn (Amount to be determined ;))
17:26 Util masak: Be advised that Parrot 5.4.0, which I released 2 days ago, had a problem that causes it to fail the .t file for Parrot::Distribution. It is a harmless failure.
17:27 denisboyun joined #perl6
17:28 Util The problem is my fault, and I can release 5.4.1 if the `make test` failure is screwing up your release process.
17:28 Util The problem is already fixed in HEAD.
17:31 jnthn Nearly home...time to detrain...
17:31 jnthn bbl
17:32 Util FROGGS: ^^^
17:34 timotimo Util: would you be receptible to worries about the readline implementation?
17:35 timotimo if i told you readline-ing from a socket will sometimes give me a not-complete line and leave me with no way to figure out what is a full line and what is bugged, would you consider that something that should be fixed?
17:38 REPLeffect joined #perl6
17:45 kaleem joined #perl6
17:45 Util timotimo: yes. Is there an open ticket on this issue?
17:46 timotimo i didn't actually make one yet :(
17:46 timotimo i think i'll have to make a reproducible, small, self-contained test case
17:51 rindolf joined #perl6
18:08 nyuszika7h joined #perl6
18:14 dalek roast: ab1d3b7 | dagurval++ | S32-io/spurt.t:
18:14 dalek roast: unfudge rakudo, fudge niecza
18:14 dalek roast: review: https://github.com/perl6/roast/commit/ab1d3b76f4
18:15 * colomon doesn't like the sounds of that
18:18 timotimo dagurval: it doesn't seem to have a plan now, is that intended?
18:19 timotimo oh, maybe it has a "done" at the end
18:19 timotimo indeed it does
18:19 Util timotimo: Looks like not_gerd was working on the readline issue: http://irclog.perlgeek.de/parrotsketch/2013-03-12
18:19 dagurval timotimo: it was kind of difficult with the fudging :).
18:20 SamuraiJack joined #perl6
18:20 timotimo it doesn't sound like the issue i was getting, though
18:21 FROGGS Ulti: whats up?
18:24 FROGGS Ulti: I won't upgrade parrot for today's release
18:25 masak FROGGS: Util, not Ulti.
18:25 FROGGS ohh
18:26 * masak is such a big help :)
18:26 FROGGS Util^^ sorry for that
18:26 FROGGS masak++ # indeed :o)
18:27 Util FROGGS: No problem. You are not the first to switch us up.
18:27 FROGGS yeah, I should stop reading just the first and the last letter :o)
18:28 FROGGS err, well, you know what I mean *g*
18:29 dagurval r: "/tmp/asdf".IO.spurt("hello").WHAT.say
18:29 camelia rakudo d3b36e: OUTPUT«(Bool)␤»
18:29 dagurval I wonder what spurt should return, perhaps self?
18:29 masak FROGGS: I bet you've seen http://dan.hersam.com/2005/01/27/reading-jumbled-letters/
18:30 dalek rakudo/nom: 90daa18 | (Tobias Leich)++ | src/Perl6/Grammar.nqp:
18:30 dalek rakudo/nom: by TimToady++: allow $% and $@ itemizers to align with STD
18:30 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/90daa18887
18:30 FROGGS masak: not this one, but a similar article
18:32 masak yeah.
18:35 dalek specs: c2ac9b8 | dagurval++ | S32-setting-library/IO.pod:
18:35 dalek specs: Updated IO::Handle.slurp and added IO::Handle.spurt
18:35 dalek specs: review: https://github.com/perl6/specs/commit/c2ac9b87be
18:38 denisboyun_ joined #perl6
18:39 alester joined #perl6
18:52 berekuk joined #perl6
18:58 Bzek joined #perl6
19:01 dagurval I think this is a bug:
19:01 dagurval r: "/tmp/asdf".IO.open(:w).print("hello world"); say slurp("/tmp/asdf")
19:01 camelia rakudo d3b36e: OUTPUT«␤»
19:01 dagurval r: my $a = "/tmp/asdf".IO; $a.open(:w); $a.print("hello world"); $a.close; say slurp("/tmp/asdf")
19:01 camelia rakudo d3b36e: OUTPUT«hello world␤»
19:01 dagurval Has top be closed explicitly. Does it have to do with GC?
19:02 massimino joined #perl6
19:02 PerlJam buffering I'd wager.
19:02 moritz yes, and with the lack of destroy methods
19:03 dagurval ok, so it's something that will be fixed when destroy methods arrive?
19:03 PerlJam or, autoflush on the filehandle.
19:04 moritz dagurval: yes
19:04 * masak is not so sure
19:05 masak the thing with destroy methods is that they are not timely, no?
19:05 masak can you even guarantee that they run before the program exits normally?
19:05 masak I guess you can, though Java seems to have had issues in that area, for some reason.
19:06 * masak mislikes DESTROY methods and would much prefer to see a total transition to the Dispose pattern
19:06 dagurval r: "/tmp/asdf".IO.open(:w).print("hello world");
19:06 camelia rakudo d3b36e:  ( no output )
19:06 dagurval this seems to never actually be written
19:06 dagurval r: say slurp("/tmp/asdf")
19:06 camelia rakudo d3b36e: OUTPUT«␤»
19:08 PerlJam r: my $fh = "/tmp/asdf".IO.open(:w); $fh.print("hello world"); $fh.flush();  say slurp("/tmp/asdf");
19:08 dalek nqp: 7305cf5 | (Tobias Leich)++ | docs/release_guide.pod:
19:08 dalek nqp: fixed typo
19:08 dalek nqp: review: https://github.com/perl6/nqp/commit/7305cf5e15
19:08 camelia rakudo d3b36e: OUTPUT«hello world␤»
19:09 grondilu rn: say 0/0
19:09 camelia rakudo d3b36e: OUTPUT«Divide by zero␤  in sub infix:<div> at src/gen/CORE.setting:3347␤  in sub infix:<div> at src/gen/CORE.setting:2998␤  in sub DIVIDE_NUMBERS at src/gen/CORE.setting:7992␤  in sub infix:</> at src/gen/CORE.setting:8124␤  in sub infix:</> at src/gen/CORE.setting:2993␤ …
19:09 camelia ..niecza v24-51-g009f999: OUTPUT«0␤»
19:09 PerlJam dagurval: you can't trust that camelia won't clean up after itself between invocations.
19:09 * grondilu was expecting NaN
19:10 dagurval PerlJam: it doesn't clean the files I create with IO
19:10 dagurval r: "/tmp/asdf".IO.open(:w).spurt("hello world");
19:10 camelia rakudo d3b36e:  ( no output )
19:10 dagurval r: say slurp("/tmp/asdf")
19:10 camelia rakudo d3b36e: OUTPUT«hello world␤»
19:10 masak r: say 0.0 / 0.0
19:10 camelia rakudo d3b36e: OUTPUT«Divide by zero␤  in sub infix:<div> at src/gen/CORE.setting:3347␤  in sub infix:<div> at src/gen/CORE.setting:2998␤  in sub DIVIDE_NUMBERS at src/gen/CORE.setting:7992␤  in sub infix:</> at src/gen/CORE.setting:8103␤  in sub infix:</> at src/gen/CORE.setting:2993␤ …
19:11 dalek nqp: 6bbc8bd | (Tobias Leich)++ | VERSION:
19:11 dalek nqp: bump VERSION to 2013.05
19:11 dalek nqp: review: https://github.com/perl6/nqp/commit/6bbc8bd929
19:11 jnthn r: say 0e0 / 0e0
19:11 camelia rakudo d3b36e: OUTPUT«Divide by zero␤  in sub infix:</> at src/gen/CORE.setting:3801␤  in sub infix:</> at src/gen/CORE.setting:2993␤  in block  at /tmp/jF1SKLvONe:1␤␤»
19:13 PerlJam r: say $*PID;
19:13 camelia rakudo d3b36e: OUTPUT«24667␤»
19:13 PerlJam r: say $*PID;
19:13 camelia rakudo d3b36e: OUTPUT«24673␤»
19:19 massimino Good evening. I've a problem installing modules with panda Can someone help me please? The problem is https://gist.github.com/anonymous/5638586
19:19 masak good evening massimino.
19:20 massimino Hi masak
19:20 masak everyone: is this panda problem known?
19:20 masak can someone reproduce it on a clean Star build?
19:21 masak is 2013.02 known-bad? I think not.
19:21 masak or is it known-bad for Opensuse?
19:21 sorear o/
19:21 massimino I downloaded rakudo star yesterday. It's 2013.2 and panda one hour ago
19:21 jnthn Doesn't look like a familiar error to me.
19:22 jnthn Does R* not come with a Panda?
19:22 jnthn (May just be some kinda version skew...)
19:22 masak massimino: how did you run panda? did you download it separately from Rakudo Star?
19:23 massimino yes masak
19:23 massimino I downloaded rakudo star and i've change my path. Then I downloaded panda and changed my path again
19:24 masak :/
19:24 massimino after run bootstrap.pl
19:24 masak that... may not work.
19:24 masak Rakudo Star already has a panda. use that.
19:24 masak er, I mean "I suggest you use that" :)
19:24 massimino Ok I try
19:25 masak shouldn't a full 'make' of Rakudo Star already invoke panda, even?
19:25 masak or is that a diff'rent 'make' target?
19:25 jnthn Not sure...I recall that something was done a while back to make the Panda with star recognize the already-supplied module...
19:29 tadzik massimino: does 'panda update' change anything?
19:29 massimino The error seems to be the panda command in rakudo.
19:30 massimino because when I launch
19:30 massimino panda on diretory I've downloaded from git-hub I see the list of modules
19:31 massimino but there is another error: https://gist.github.com/anonymous/5638777
19:33 massimino I'm trying panda update
19:35 massimino Error again: no such method 'update' for invocant of type 'Any'
19:35 tadzik oh shit
19:35 tadzik how new is that panda?
19:35 tadzik oh, I see it
19:35 tadzik this has been fixed some time ago
19:35 tadzik like, 5 days ;)
19:35 tadzik sorry for that
19:36 fgomez joined #perl6
19:36 massimino I've downloaded rakudo yesterday. But panda 1:30 ago
19:36 tadzik hmm
19:36 tadzik that shouldn't be happening, since...
19:37 tadzik https://github.com/tadzik/panda/commit/02765c9329 since this
19:37 massimino I don't know but the first error is when I run panda that comes with rakudo.
19:38 tadzik massimino: I did some serious refactoring with panda some time ago, I may have broken stuff :( sorry for that
19:39 FROGGS maybe the old panda that is flying around should be removed from disk?
19:39 massimino Yes but when I run panda inside that I've downloaded from git-hub I can see the list of modules  but I've another error: https://gist.github.com/anonymous/5638777
19:39 massimino I don't know what directory .work is
19:42 tadzik it's the same one
19:43 TimToady http://rosettacode.org/wiki/Subset_sum_problem#Perl_6
19:45 TimToady btw, that's a pretty fast combinations routine in there, compared to the recursive version
19:47 tadzik massimino: did you try the very fresh panda? How do you install it?
19:47 tadzik I suck at maintaining this
19:49 massimino massimino: I've three executables of panda: one in rakudo-star-2013.02/install/bin, another in rakudo-star-2013.02/install/lib/parrot/4.10.0/languages/perl6/site/bin and another in ~/panda (I've downloaded it today)
19:50 massimino I run certainly panda in the first directory
19:50 tadzik did you boostrap it?
19:50 tadzik like, 'perl6 boostrap.pl'
19:50 FROGGS massimino: what does 'which panda' tell you?
19:50 massimino yes
19:50 massimino I've tried a thing.
19:51 massimino inside rakudo-star-2013.02/install/lib/parrot/4.10.0/languages/perl6/site/bin i've created the directory work
19:51 massimino then I launched ./panda list
19:52 massimino All was ok
19:52 tadzik aww
19:53 tadzik massimino: is this the star panda, that requires the .work directory to work?
19:53 tadzik That shouldn't have ever got in any release
19:53 massimino In rakudo star there are two panda commands
19:53 konundra joined #perl6
19:54 tadzik really?
19:54 tadzik at least htat's not my faul...
19:54 massimino the second I think was created when I did perl6 bootstrap.pl
19:54 tadzik ah
19:54 tadzik okay, again
19:55 tadzik you installed star, and the bundled panda didn't work ootb?
19:55 massimino because the last message was that I had to modify my path
19:55 massimino I don't know... after installed rakudo I downloaded panda and I did perl6 bootstrap.pl
19:56 tadzik ah. That was not necessary
19:56 tadzik now you have two pandas
19:56 tadzik panda is included in star
19:57 tadzik you installed another, and they interact in the bad way
19:57 tadzik the git version is a bit untested, there's a reason it's not released yet
19:57 tadzik please try removing rakudo-star-2013.02/install/lib/parrot/4.10.0/languages/perl6/site altogether
19:57 tadzik then you should still have the star panda
19:58 massimino Ok tadzik, I try
19:58 tadzik unless star installs to site. I hope not
19:59 massimino I will try to cancel my rakudo star installation and reinstall it to have a clean installation
20:00 tadzik I'm sorry panda gave you trouble
20:00 sorear .o( I will cancel my rakudo star subscription!!! )
20:01 massimino Don't worry.
20:02 tadzik it's a bit of an unruly child
20:03 tadzik rebellious age and all that
20:03 arnsholt tadzik: We'll have to meet up and discuss our unruly children in Austin
20:03 arnsholt NativeCall is a bit of a handful as well =)
20:03 sorear terrible twos?
20:04 arnsholt Yeah, something like that
20:04 sorear arnsholt: you're making it to austin?
20:04 sorear in spite of the INS?
20:04 arnsholt Hehe
20:04 arnsholt Norway is part of the visa waiver program, so it wasn't as hard as it was for tadzik
20:04 jnthn .oO( LET ME INS! )
20:06 tadzik arnsholt: good :0
20:07 tadzik maybe panda should warn if it's already installed?
20:07 sorear ETOOMANYPANDAS
20:08 tadzik eggzactly
20:08 sorear PANDA CARRYING CAPACITY EXCEEDED
20:09 nebuchadnezzar joined #perl6
20:29 masak PANDA TO ŻUBR RATIO OUT OF BOUNDS
20:29 eternaleye joined #perl6
20:31 jnthn .oO( Just teething problems... )
20:33 sorear never heard of that before
20:37 bbkr joined #perl6
20:38 * masak almost deadpans an explanation in sorear's direction about teeth
20:38 fhelmberger joined #perl6
20:39 masak sorear: by the way, today's weird mathematical object: https://en.wikipedia.org/wiki/Empty_semigroup
20:40 masak or not so much "weird" as "amusingly useless".
20:41 eternaleye joined #perl6
20:42 sorear masak: was actually referring to /Bison bonasus/
20:43 masak aye.
20:43 masak sorear: I've seen them. they are very docile.
20:43 masak almost went extinct.
20:43 masak very hard to rebuild from scratch after they do.
20:44 massimino tadzik: I've reinstalled rakudo 2013.2 and now panda works fine thank you for everything
20:44 FROGGS O.o
20:44 jnthn Gee, I hope I didn't accidentally eat one in Belarus...
20:44 jnthn (I was on the other side of the border at the same national park, and ate some wild meat that I never worked out what actually was :))
20:45 masak jnthn: nah. probably just dodo. :P
20:45 jnthn :P
20:45 jnthn oh bother. Seems that when I did the git merge nom on the train into jvm-support, I busted it.
20:46 FROGGS jnthn: np, I'll add that as a quote to the changelog *g*
20:48 jnthn FROGGS: It's in a branch :P
20:48 jnthn I've got until the next release to fix it. :P
20:48 FROGGS ha, I'll do two releases!!
20:49 masak nooo
20:54 eternaleye joined #perl6
21:00 dalek rakudo/nom: 781a32b | (Tobias Leich)++ | docs/ (3 files):
21:00 dalek rakudo/nom: prepared current release
21:00 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/781a32bdc3
21:01 timotimo ++ underlaying nqp layer support parrot and JVM as backend <- should be "underlying"
21:02 timotimo would like to see the line of contributors hard wrapped
21:03 jnthn Dunno whether to mention that in Rakudo release; people may not read it properly and think this release has the some Rakudo on JVM support... :)
21:07 FROGGS timotimo: will fix both, thanks for reporting
21:17 timotimo oh, indeed
21:23 FROGGS jnthn: but it is somehow important to mention me thinks
21:28 labster JVM support is nigh!
21:29 * timotimo holds up signs
21:31 dalek rakudo/jvm-support: 2778b30 | jnthn++ | src/vm/jvm/runtime/org/perl6/rakudo/ (2 files):
21:31 dalek rakudo/jvm-support: Make (|c) signatures work in binder.
21:31 dalek rakudo/jvm-support:
21:31 dalek rakudo/jvm-support: Means .^foo now works, which somehow we'd come to hit during setting
21:31 dalek rakudo/jvm-support: loading. Thus, basic stuff works again.
21:31 dalek rakudo/jvm-support: review: https://github.com/rakudo/rakudo/commit/2778b30470
21:31 dalek nqp/rak-jvm-support: 8c43ba4 | jnthn++ | src/vm/jvm/runtime/org/perl6/nqp/runtime/ (3 files):
21:31 dalek nqp/rak-jvm-support: Some error reporting fixes.
21:31 dalek nqp/rak-jvm-support: review: https://github.com/perl6/nqp/commit/8c43ba4c32
21:32 jnthn FROGGS: Yeah, maybe just make it super clear ("in preparation for JVM support in a future Rakudo release" or so maybe...)
21:32 FROGGS will do
21:34 dalek rakudo/nom: 80c00e8 | (Tobias Leich)++ | docs/ChangeLog:
21:34 dalek rakudo/nom: make clear what JVM support is about
21:34 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/80c00e8dea
21:34 dalek rakudo/nom: b79a19c | (Tobias Leich)++ | docs/announce/2013.05.md:
21:34 dalek rakudo/nom: added contributors
21:34 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/b79a19c35a
21:43 berekuk joined #perl6
21:43 bruges joined #perl6
21:44 dalek rakudo/nom: 264859c | (Tobias Leich)++ | tools/build/NQP_REVISION:
21:44 dalek rakudo/nom: [release] bump NQP revision
21:44 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/264859cf02
21:44 dalek rakudo/nom: f7bfee6 | (Tobias Leich)++ | VERSION:
21:44 dalek rakudo/nom: [release] bump VERSION
21:44 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/f7bfee6cd6
21:46 lizmat joined #perl6
21:47 colomon_ joined #perl6
21:52 colomon psst -- http://71.10.146.50:3001/report -- very primitive and temporary preview of module smoke test reports
21:53 woolfy joined #perl6
21:53 FROGGS colomon: the columns are the last n days?
21:54 colomon FROGGS: yes
21:54 colomon with the latest on the right
21:54 colomon obviously very crude at the moment
21:55 FROGGS but you already see what is going on
21:55 FROGGS colomon++
21:56 jnthn colomon++
21:56 colomon I'm planning on sorting it so the most recently changed modules go on top
21:56 lizmat colomon++
21:56 colomon and making the module name links go to a useful page.
21:56 colomon my first ever Dancer app
21:56 FROGGS :o)
21:58 labster nice colomon++
21:58 sorear colomon++
22:05 colomon okay, turned off outside access to it again.  :)
22:05 FROGGS meh
22:05 grondilu what's "constant folding"?
22:06 * grondilu has rred this expression several times lately.
22:07 Tene grondilu: an optimizer pass to evaluate expressions only involving constants at compile time
22:07 Tene for example, replacing 5+3 with 8
22:08 grondilu ok.  thanks
22:09 dalek nqp/rak-jvm-support: c8bbb9a | jnthn++ | src/vm/jvm/runtime/org/perl6/nqp/runtime/IndyBootstrap.java:
22:09 dalek nqp/rak-jvm-support: Missing decont in indcall.
22:09 dalek nqp/rak-jvm-support: review: https://github.com/perl6/nqp/commit/c8bbb9a799
22:09 dalek nqp/rak-jvm-support: 086cb6f | jnthn++ | src/vm/jvm/runtime/org/perl6/nqp/runtime/Ops.java:
22:09 dalek nqp/rak-jvm-support: Implement some missing exception related ops.
22:09 dalek nqp/rak-jvm-support: review: https://github.com/perl6/nqp/commit/086cb6fed6
22:18 kurahaupo joined #perl6
22:21 lizmat goodnight #perl6
22:22 FROGGS gnight
22:24 snearch joined #perl6
22:27 jnthn Would love to bughunt some more, but I'm making too many silly mistakes to come out with more fixed that broke...
22:27 jnthn 'night o/
22:28 timotimo http://tellaparteng.tumblr.com/post/51142273614/presenting-commandr-an-easy-to-use-and-automatic-cli - cool, kind of like MAIN in perl6
22:28 masak 'night, #perl6
22:28 * grondilu is not sure about the semantics of List.combinations
22:28 grondilu "Returns all combinations of the list taken N at a time (specified by the $of parameter)."
22:30 sorear so [1,2,3,4,5]; 2  -->  [[1,2], [1,3], [1,4], ... [4,5]]
22:30 sorear https://en.wikipedia.org/wiki/Combinations
22:31 grondilu yeah that's how I understood it first.  But "taken N at a time" is a terrible way to phrase it imho.
22:32 * grondilu would have written: "Returns all combinations of N elements of the list"
22:32 geekosaur that's combinatorics lingo, I believe
22:32 massimino left #perl6
22:33 xilo joined #perl6
22:33 sorear yeah, fairly standard wording in the english-language literature
22:33 grondilu oh indeed the expression "at a time" is in the wikipedia article.  I had no clue.
22:34 grondilu it's all good then.  My bad.
22:37 dalek rakudo/nom: 72eff27 | (Tobias Leich)++ | docs/announce/2013.05.md:
22:37 dalek rakudo/nom: forgot to update short changelog -.-
22:37 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/72eff27b5b
22:42 FROGGS (pmichaud, jnthn, masak, tadzik, moritz, PerlJam: can someone please upload this to rakudo.org? froggs.de/rakudo-2013.05.tar.gz froggs.de/nqp-2013.05.tar.gz
22:42 berekuk joined #perl6
22:53 woolfy left #perl6
23:00 FROGGS gnight all
23:11 diakopter yapc austin registrations: http://i.imgur.com/0D2q3nr.png
23:17 jerome__ joined #perl6
23:17 REPLeffect joined #perl6
23:53 adu joined #perl6

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

Perl 6 | Reference Documentation | Rakudo