Perl 6 - the future is here, just unevenly distributed

IRC log for #perl6, 2011-11-05

Perl 6 | Reference Documentation | Rakudo

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

All times shown according to UTC.

Time Nick Message
00:12 dalek 6model: 2518ca6 | diakopter++ | lua/runtime/List.lua:
00:12 dalek 6model: array tests pass
00:12 dalek 6model: review: https://github.com/diakopter/6model/commit/2518ca6d3c
00:16 sorear colomon: You've convinced me not to do anything for now. :)
00:17 colomon drat!
00:17 colomon ;)
00:17 colomon though like I said, it certainly seems "good enough" already.
00:24 _jaldhar joined #perl6
00:54 wolfman2000 joined #perl6
01:12 dalek niecza: 54cf51a | sorear++ | lib/ (2 files):
01:12 dalek niecza: Eliminate now-unused BValue type
01:12 dalek niecza: review: https://github.com/sorear/niecza/commit/54cf51a4d9
01:12 dalek niecza: 750c097 | sorear++ | lib/ (3 files):
01:12 dalek niecza: Encapsulate setting of .v on existing StashEnts
01:12 dalek niecza: review: https://github.com/sorear/niecza/commit/750c097c0c
01:30 [Coke] \o/ # rt.perl.org/rt3 has a browser-friendly cert now.
01:32 sorear sweeet
01:39 dalek rakudo/nom: 9c28800 | (Geoffrey Broadwell)++ | src/core/Str.pm:
01:39 dalek rakudo/nom: Str.Numeric: Support Rats and Nums that start with a period (e.g. '.5')
01:39 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/9c28800294
01:40 sorear is that japhb?
01:41 japhb_ joined #perl6
01:41 japhb_ sorear, yes.  :-)
01:42 japhb_ decommuting &
01:53 sorear niecza: 5 * constant $x = 2;
01:53 p6eval niecza v11-14-g750c097: OUTPUT«Potential difficulties:â�¤  $x is declared but not used at /tmp/7uUQaWvm87 line 1:â�¤------> [32m5 * [33mâ��[31mconstant $x = 2;[0mâ�¤â�¤â�¤Unhandled Exception: assigning to readonly valueâ�¤  at /tmp/7uUQaWvm87 line 0 (mainline @ 0) â�¤  at /home/p6eval/niecza/lib/CO…
01:54 dalek niecza: 1491987 | sorear++ | / (3 files):
01:54 dalek niecza: Decouple constant initialization from the ability to rebind constants
01:54 dalek niecza: review: https://github.com/sorear/niecza/commit/149198748f
02:11 IngisKahn joined #perl6
02:13 plobsing joined #perl6
02:30 dalek niecza: eb2bc3c | sorear++ | lib/ (3 files):
02:30 dalek niecza: Implement a concept of constant common variables (half of #46)
02:30 dalek niecza: review: https://github.com/sorear/niecza/commit/eb2bc3ca02
02:40 lateau_ joined #perl6
02:43 jlaire .u 
02:43 phenny U+F8C7 (No name found)
02:52 diakopter .u money
02:52 phenny diakopter: Sorry, no results for 'money'.
02:56 dalek niecza: ccb1914 | sorear++ | lib/ (3 files):
02:56 dalek niecza: Remove capability of rebinding lexical constants (fixes #46)
02:56 dalek niecza: review: https://github.com/sorear/niecza/commit/ccb1914eb2
02:57 _jaldhar joined #perl6
02:57 sorear niecza: $@ + 5
02:57 p6eval niecza v11-15-g1491987: OUTPUT«[31m===[0mSORRY![31m===[0m��Unsupported use of $@ variable as eval error; in Perl 6 please use $! at /tmp/UJXlfbXx7T line 1:�------> [32m$@[33m�[31m + 5[0m��Parse failed��»
02:57 sorear niecza: $@
02:57 p6eval niecza v11-15-g1491987: OUTPUT«[31m===[0mSORRY![31m===[0mâ�¤â�¤Non-declarative sigil is missing its name at /tmp/ARW1f8iSxQ line 1:â�¤------> [32m<BOL>[33mâ��[31m$@[0mâ�¤â�¤Use of uninitialized value in string contextâ�¤  at /home/p6eval/niecza/boot/lib/CORE.setting line 773 (warn @ 2) â�¤  at …
03:05 mkramer joined #perl6
03:13 mkramer left #perl6
03:14 dalek niecza: 191508a | sorear++ | src/STD.pm6:
03:14 dalek niecza: Checking for $@ and $^X should use <!before \w> instead of <?before \W> so it works at string end (fixes #63)
03:14 dalek niecza: review: https://github.com/sorear/niecza/commit/191508a418
03:16 kaleem joined #perl6
03:27 kaleem joined #perl6
03:29 sorear perl6: say "23foo".subst(/\w/, {$/.uc})
03:29 p6eval rakudo 9c2880: OUTPUT«Method 'uc' not found for invocant of class 'Any'␤  in block <anon> at /tmp/MESSOCm_Ey:1␤  in method reify at src/gen/CORE.setting:3989␤  in method reify at src/gen/CORE.setting:3896␤  in method reify at src/gen/CORE.setting:3896␤  in method gimme at src/gen/CORE.s…
03:29 p6eval ..niecza v11-18-g191508a: OUTPUT«23foo␤»
03:29 p6eval ..pugs b927740: OUTPUT«Error eval perl5: "if (!$INC{'Pugs/Runtime/Match/HsBridge.pm'}) {␤    unshift @INC, '/home/p6eval/.cabal/share/Pugs-6.2.13.20111008/blib6/pugs/perl5/lib';␤    eval q[require 'Pugs/Runtime/Match/HsBridge.pm'] or die $@;␤}␤'Pugs::Runtime::Match::HsBridge'␤"␤*** '<HAND…
03:29 sorear perl6: say "+!foo".subst(/\w/, {$/.uc})
03:30 p6eval rakudo 9c2880: OUTPUT«Method 'uc' not found for invocant of class 'Any'␤  in block <anon> at /tmp/PtCAKvnnDj:1␤  in method reify at src/gen/CORE.setting:3989␤  in method reify at src/gen/CORE.setting:3896␤  in method reify at src/gen/CORE.setting:3896␤  in method gimme at src/gen/CORE.s…
03:30 p6eval ..niecza v11-18-g191508a: OUTPUT«+!Foo␤»
03:30 p6eval ..pugs b927740: OUTPUT«Error eval perl5: "if (!$INC{'Pugs/Runtime/Match/HsBridge.pm'}) {␤    unshift @INC, '/home/p6eval/.cabal/share/Pugs-6.2.13.20111008/blib6/pugs/perl5/lib';␤    eval q[require 'Pugs/Runtime/Match/HsBridge.pm'] or die $@;␤}␤'Pugs::Runtime::Match::HsBridge'␤"␤*** '<HAND…
03:30 sorear straw poll: is rakudo doing it wrong?
03:32 dalek niecza: 2c52e86 | sorear++ | / (2 files):
03:32 dalek niecza: Autothunkify left side of xx (fixes #70)
03:32 dalek niecza: review: https://github.com/sorear/niecza/commit/2c52e86ee2
03:33 drbean joined #perl6
03:39 sorear diakopter: could you take a look at https://github.com/sorear/niecza/issues/61 and https://github.com/sorear/niecza/issues/59 and tell me if both are still valid?
03:39 drbean joined #perl6
03:42 diakopter interestingly, /61 now fails with the same error as /59
03:43 iktome joined #perl6
03:44 dalek niecza: b962c2f | sorear++ | t/run_spectests:
03:44 dalek niecza: Auto-checkout/update t/spec during make spectest (fixes #54)
03:44 dalek niecza: review: https://github.com/sorear/niecza/commit/b962c2f94d
03:44 iktome joined #perl6
03:45 diakopter sorear: I commented on /61
03:45 iktome left #perl6
03:45 sorear aargh
03:46 diakopter arg
03:46 diakopter h
03:46 diakopter I hope it comes back
03:48 dalek <sorear> Your comment on #61 implies that niecza on Win32 now cannot be loaded _at all_.
03:51 diakopter__ joined #perl6
03:51 diakopter__ sorear: I think that may be the case
03:53 diakopter__ oh noes; host04 went down too
04:12 drbean joined #perl6
04:12 diakopter___ joined #perl6
04:14 sorear joined #perl6
04:14 sorear diakopter___: here?
04:14 diakopter___ yes
04:15 sorear I'm going to have a zip for you to test in a few minutes
04:15 diakopter___ my niecza/cygwin installation is on another computer (where diakopter__ is logged in)
04:15 sorear replacing the particular .Invoke calls I was using with different ones closer to what MSDN says
04:17 sorear seems that MethodBase::Invoke(object, object[]) cannot be used to *allocate* a new object on .net
04:17 sorear in Mono, constructor.Invoke(null, args) will allocate a new object
04:18 sorear now trying it with constructor.Invoke(args) like the MSDN code samples had
04:18 sorear btw, do you use screen with irssi, diakopter___?
04:18 diakopter___ yeah, I did on host02
04:18 diakopter___ hopefully host02 comes back in a few hours or less
04:18 sorear I blame screen for the horrible display glitches I've been seeing, thinking I'll use this opportunity to replace it
04:19 sorear tmux or dtach or something
04:20 diakopter___ my screen/irssi seems to work okay through PuTTY
04:20 diakopter___ if there's ever a glitch, just resizing the window fixes it
04:20 diakopter___ ok I got niecza/cygwin set up here too
04:21 diakopter___ feel free to email the .zip
04:23 diakopter___ RUN_CLR=
04:23 diakopter___ CSC=csc
04:29 kaare_ joined #perl6
04:30 dalek joined #perl6
04:32 diakopter___ o_O
04:32 sorear I guess host04 is back up
04:32 sorear yep
04:32 diakopter___ it didn't reboot?
04:35 p6eval joined #perl6
04:37 diakopter___ std: 1
04:37 sorear joined #perl6
04:37 p6eval std be1f10e: OUTPUT«ok 00:01 118m␤»
04:37 sorear odd that host04 didn't reboot
04:37 sorear rakudo: pir::spawn__is("uptime")
04:37 p6eval rakudo 9c2880: OUTPUT«Use of uninitialized value in string context␤Use of uninitialized value in string context␤===SORRY!===␤error:imcc:syntax error, unexpected IREG, expecting '(' ('$I100')␤ in file '(file unknown)' line 12042681␤␤»
04:37 sorear rakudo: pir::spawn__IS("uptime")
04:38 p6eval rakudo 9c2880: OUTPUT«Use of uninitialized value in string context␤Use of uninitialized value in string context␤===SORRY!===␤error:imcc:syntax error, unexpected IREG, expecting '(' ('$I100')␤ in file '(file unknown)' line 12220154␤␤»
04:38 sorear rakudo: pir::spawn__vS("uptime")
04:38 p6eval rakudo 9c2880: OUTPUT«Use of uninitialized value in string context␤Use of uninitialized value in string context␤===SORRY!===␤error:imcc:syntax error, unexpected SREG, expecting '(' ('$S100')␤ in file '(file unknown)' line 138␤␤»
04:38 sorear rakudo: pir::spawnw__IS("uptime")
04:38 p6eval rakudo 9c2880: OUTPUT« 04:00:43 up 399 days,  9:52,  0 users,  load average: 0.10, 0.04, 0.01␤»
04:40 diakopter___ I and another tried all kinds of failed debian upgrades on host02; I hope it reboots successfully.
04:41 diakopter___ sorear: you'll email me a .zip?
04:42 drbean joined #perl6
04:44 sorear diakopter___: yeah if it works this time
04:44 diakopter___ once I get the zip, how would I use it
04:44 diakopter___ just unzip it into boot/ in the build dir?
04:49 sorear yes, deleting what's there first
04:49 sorear see if it runs at all (boot/run/Niecza.exe --help)
04:51 sorear sent
04:54 diakopter___ gmail probably blocked the email b/c it had a .zip; rename the file to .piz and it should work
04:55 japhb How do you get STD to spit out a parse tree, rather than just ok or FAILED?
04:55 diakopter___ ./viv has some options
04:56 diakopter___ yaml output is one
04:56 sorear diakopter___: wtf
04:57 japhb diakopter__: thank you
04:57 japhb I was stupidly looking in tools/, since that's where the README seems to point
04:57 sorear diakopter___: both that the email was blocked, and that google isn't using file(1) or equivalent
04:57 diakopter___ yeah :)
04:57 sorear tools/ is for supplementary tools
04:57 sorear the real important stuff is in .
04:57 sorear like the README itself
04:58 sorear diakopter___: done
04:58 diakopter___ sorear: got the piz
04:59 japhb methinks the README could use some love; I'm not (yet) qualified to do that.
05:00 sorear japhb: your English isn't that bad.
05:00 sorear I actually thought you were native.
05:00 * japhb laughs
05:01 diakopter___ boot/run/Niecza.exe --help  worked
05:01 japhb I am native.  I meant that I was not qualified in the sense of "I don't know the correct use and current state of all of the pieces, so I can't make the README fully correct."
05:02 * japhb now wonders if sorear, despite his location *isn't* a native speaker, based on some variant of the "He who looks behind the door has stood there before" principle.
05:06 sorear sort of.  English is my native language but I'm 2.5-lingual
05:07 diakopter___ what other languages?
05:07 sorear es, jp
05:07 sorear ja
05:07 diakopter___ how might I use your piz instead of the one downloaded by the Makefile
05:08 sorear I wonder if putting a file:// URL in FETCH_URL would work
05:09 sorear if not, follow the steps given in the .fetch-stamp: makefile rule
05:09 sorear making sure to touch at the end
05:09 sorear_ joined #perl6
05:09 sorear_ joined #perl6
05:10 diakopter joined #perl6
05:10 sorear o/ diakopter
05:11 diakopter whew
05:11 diakopter___ left #perl6
05:11 diakopter nqplua passes a bunch of tests
05:11 sorear http://mail.google.com/support/bin/answer.py?answer=6590
05:12 sorear "Gmail does not allow you to send or receive files that are corrupted."
05:12 diakopter heh
05:13 orafu joined #perl6
05:14 sorear diakopter: any luck w/ niecza?
05:15 diakopter well after the Makefile deleted boot/ a couple of times
05:19 diakopter sorear: odd
05:22 diakopter Strange text after block (missing comma, semicolon, comment marker?) at C:\Users\mwilson\src\niecza\src\CClass.pm6 line 15:
05:24 sorear you need to touch .fetch-stamp if you don't want the makefile to delete boot/
05:26 diakopter sorear: I got the above error
05:26 diakopter when running cd src && NIECZA_KEEP_IL=1  ../boot/run/Niecza.exe -C CClass
05:27 sorear diakopter: I can repro that here - the parser doesn't like \r\n much
05:27 sorear diakopter: you could try changing *.pm6 *.setting to Unix line endings, or you can wait for me to fix the compiler
05:29 sorear rakudo: say "\r\n" ~~ /\n//
05:29 p6eval rakudo 9c2880: OUTPUT«===SORRY!===␤Confused at line 1, near "say \"\\r\\n\""␤»
05:29 sorear rakudo: say "\r\n" ~~ /\n/
05:29 p6eval rakudo 9c2880: OUTPUT«=> <
05:29 sorear What?
05:30 sorear rakudo: say ("\r\n" ~~ /\n/).gist.perl
05:30 diakopter :O)
05:30 p6eval rakudo 9c2880: OUTPUT«"=> <\r\n>\n"␤»
05:30 sorear rakudo: say ("AB" ~~ /B/).gist.perl
05:30 p6eval rakudo 9c2880: OUTPUT«"=> <B>\n"␤»
05:30 sorear hrm. ok.
05:31 sorear I guess \n in Rakudo can match \cM\cJ
05:31 sorear rakudo: say ("\r\n" ~~ /<[\n]>/).gist.perl
05:31 p6eval rakudo 9c2880: OUTPUT«"#<failed match>"␤»
05:31 sorear I wonder what <[\n]> should do, or if it should be outlawed
05:32 sorear if I'm going to make \n match potentially more than one character
05:32 diakopter what does p5 do
05:42 sorear stefan@stefans:~$ perl -E '"\015\012" =~ /\n/; say length($&), " ", length($`)'
05:42 sorear 1 1
05:42 sorear \n matches \012 only
05:42 diakopter maybe it behaves differently on windows
05:42 diakopter nope; same
05:43 sorear p5, like C, does line ending mangling on I/O
05:43 sorear I wonder if niecza should do the same
05:45 * sorear checks UTS18
05:50 sorear hrm, http://www.unicode.org/reports/tr18/#Partial_Matches and the succeeding section are problematic for Perl 6 regexes
05:52 diakopter so pie-in-the-sky
05:53 sorear I don't see anything in tr18 about \n
05:56 sorear hmm, seems STD has a deeper assumption on \n separating lines
05:59 diakopter Perl 6 source code exists on Windows too
05:59 diakopter (or does it?)
06:00 sorear phenny: tell TimToady I would like to extract knowledge from you on how Perl 6 should treat \n and text-mode file IO on Windows
06:00 phenny sorear: I'll pass that on when TimToady is around.
06:00 sorear diakopter: apparently it does, or what is CClass.pm6?
06:00 sorear :)
06:04 diakopter alternatively, I could switch my git to not translate line endings
06:08 sorear diakopter: that would be a good idea, since I'm unlikely to get \n sorted with TimToady tonight
06:10 Woodi me wa
06:10 * Woodi was readin http://www2.parc.com/csl/groups/sda/projects/oi/workshop-94/foil/main.html yesterday
06:10 sorear submitted #78
06:10 * Woodi need coffe or Mg probably...
06:11 Woodi any aspects of Aspect-oriented programming in Perl6 ?
06:15 daniel-s joined #perl6
06:16 sorear Magnesium?
06:16 * sorear imagines Woodi munching on a NeXT cube
06:22 sorear interesting link, Woodi
06:22 sorear Woodi: &wrap is supposedly like AOP
06:31 Woodi realy not dig yet into AOP, just read somewhere mr. G.Kiczales (CLOS, OI) moved to AOP
06:32 Woodi an maybe I started to like grammars in Perl6 as somethink MOP/OI thing :)
06:46 sorear diakopter: without line ending mangling, how far does it get?
06:51 diakopter I'll try it
06:51 [Sec] joined #perl6
06:58 diakopter warning CS1668: Invalid search path '/usr/local/lib:/usr/lib:/lib' specified in 'LIB environment variable' -- 'The filename, directory name, or volume label syntax is incorrect. '
06:59 diakopter error CS0006: Metadata file 'Run.Kernel' could not be found
06:59 diakopter make: *** [boot/obj/Run.CompilerBlob.dll] Error 1
07:02 moritz nom: say $*VERSION
07:02 p6eval nom 9c2880: OUTPUT«Failure.new()␤»
07:03 moritz nom: say $*PERL
07:03 p6eval nom 9c2880: OUTPUT«Failure.new()␤»
07:03 moritz nom: say $*VM
07:03 p6eval nom 9c2880: OUTPUT«("name" => "parrot", "config" => {"git_describe" => "RELEASE_3_9_0-222-g305b704", "sha1" => "305b704a6b79f86ce0919bc258be6e9844739218", "a" => ".a", "ar" => "ar", "ar_extra" => "", "ar_out" => "", "archname" => "x86_64-linux-gnu-thread-multi", "arflags" => "cr", "as" …
07:03 moritz b: say $*VM<config><git_describe>
07:03 p6eval b 1b7dd1: OUTPUT«RELEASE_3_6_0␤»
07:03 moritz nom: say $*VM<config><git_describe>
07:03 p6eval nom 9c2880: OUTPUT«RELEASE_3_9_0-222-g305b704␤»
07:05 sorear diakopter: that doesn't sound like a niecza problem to me :)
07:10 mberends joined #perl6
07:13 mberends good *
07:13 phenny mberends: 28 Oct 05:03Z <sorear> tell mberends /serialize is merged
07:13 mberends ooh! sorear!
07:14 * mberends merges locally
07:15 sorear mberends! how are you!
07:17 sorear mberends?
07:17 mberends hi sorear, I'm mentally rested after about 2 weeks offline. There's no way I'm going to catch up reading IRC backlogs, but it looks like lots has been happening again :)
07:18 mberends I've just been picking away a little more at 6model/C
07:18 sorear heh
07:18 * sorear is picking at the niecza bug queue
07:19 mberends sorear: it looks like your serialize branch ended well :)
07:19 sorear I'd like to get it building again on Win32 but there's a lot of back and forthing involved
07:19 sorear seems diakopter++ has an environment variable issue now
07:20 sorear yeah, merged serialize, very few spectest regressions
07:20 sorear niecza: BEGIN { say "Hello, world"; }; 1 1
07:21 p6eval niecza v11-20-gb962c2f: OUTPUT«Hello, world�[31m===[0mSORRY![31m===[0m��Two terms in a row at /tmp/Evq_iY3Rm4 line 1:�------> [32mBEGIN { say "Hello, world"; }; 1 [33m�[31m1[0m��Parse failed��»
07:21 sorear look :)
07:21 mberends yes!
07:21 mberends I'll have the use of some Win32 hosts from about Wednesday, for testing if it helps
07:22 sorear couldn't hurt, hopefully by Wed TimToady will be back online
07:23 replore_ joined #perl6
07:24 sorear mberends: what is the implementation level of 6model/c?
07:25 sorear mberends: what if I were, hypothetically, to steer 6model/c in a direction where it could be used as a niecza backend?
07:25 sorear I'll need a 6model-aware garbage collector and 6model object implementation for my niecza/c ideas
07:27 mberends sorear: there is too little of it completed so far use it right now, and I'm very slow at progressing it. The thought of it serving as a backend had not occurred to me, but I'll bear it in mind while I work on it.
07:28 * sorear is also pondering now how roles and ClassHOW will look
07:28 sorear (in niecza/clr)
07:29 sorear mberends: well, welcome back to #perl6, but I need to sleep now
07:29 sorear bye
07:29 mberends good night
07:33 Woodi mberends: what exactly you would like to do with 6model/C ?
07:34 mberends Woodi: I'd like to implement the 6model core in C with no external dependencies.
07:35 Woodi <exclude parrot.h> >
07:35 Woodi ?
07:35 Woodi so PMC implementation probably
07:36 mberends Woodi: yes. one of the main reasons for doing that is to more easily understand how 6model works
07:36 Woodi just was thinking about this too...
07:37 Woodi 'representations' sugests native mapings and 'model' kind of MOP ?
07:37 mberends that's right
07:38 Woodi some other functionality is there maybe ?
07:38 mberends method dispatch
07:41 Woodi any idea how to plug connect eg. Boehem GC ?
07:41 Mowah joined #perl6
07:42 mberends Woodi: you would have to make a "plug connect" system yourself. That is non trivial, I think.
07:44 Woodi mberends: I hear some good about jemalloc and realy think how to plug in many other external components :) but for example Bohem GC and jemalloc exclude on each other :)
07:46 mkramer joined #perl6
08:09 mberends Woodi: jemalloc is new to me, I'll find its docs and see how it compares. For 6model/C I would prefer a separate implementation of a recognized algorithm, in order to avoid adding an external dependency. I'm not in favour of Boehm because of its fundamental design, which is for "non-cooperative" environments, because it makes much more sense for new software to be written to cooperate.
08:18 drbean joined #perl6
08:30 Woodi mberends: Boehm GC is about portability so I like it. And it have atomic_ops library included which is nice for portability too. But Perl6 probably can wait for C++0x/C 0x memory model implementation :)
08:35 molaf joined #perl6
08:38 mberends Woodi: yes, its portability is good, and several VM designs used it before choosing something else for performance reasons. Efficient memory management seems to remain an enduring problem problem or challenge.
08:40 mberends the trick is for each application to find the memory manager with the least drawbacks for a given situation :)
09:12 moritz \o mberends
09:15 mberends o/ moritz :-D
09:15 tadzik hello o/
09:15 ab5tract joined #perl6
09:18 zby_home_ joined #perl6
09:29 Bucciarati joined #perl6
09:47 ethndbst joined #perl6
09:47 ethndbst joined #perl6
09:54 not_gerd joined #perl6
09:55 not_gerd hello, #perl6
09:55 not_gerd
09:55 not_gerd NQP .gitignore fix for MSYS
09:55 not_gerd ^ https://gist.github.com/1341331
09:56 not_gerd one way to make --gen-parrot --gen-nqp work on MSYS: https://gist.github.com/1340539
09:57 Woodi heh, looks nice and quick :)
09:57 moritz where does that need to go?
09:58 not_gerd Makefile.in
09:58 not_gerd is there a standard way to handle platform-specific code?
09:58 not_gerd didn't really investigate, just threw something together that seems to work...
09:59 not_gerd (Makefile.in for both Rakudo and NQP, that is...)
09:59 dalek nqp/bigint: 45d5eed | moritz++ | .gitignore:
09:59 dalek nqp/bigint: fix .gitignore for msys, not_gerd++
09:59 dalek nqp/bigint: review: https://github.com/perl6/nqp/commit/45d5eedfc8
10:00 moritz not_gerd: I'd like to have jnthn++ test that first on msvc before applying that
10:00 not_gerd moritz: should that have gone to /biging?
10:00 not_gerd ^bigint
10:01 not_gerd ls
10:04 moritz not_gerd: not really :/
10:04 not_gerd on non-MSYS Windows, Configure.pl just copies libparrot.dll into the Rakudo and NQP build directories
10:04 not_gerd perhaps I should do that for consistency...
10:18 grondilu joined #perl6
10:19 grondilu perl6: sub ∫() {...} # trying to use the unicode Integral character (0x222b) for a sub
10:19 p6eval rakudo 9c2880: OUTPUT«===SORRY!===␤Malformed block at line 1, near "\u222b() {...} "␤»
10:19 p6eval ..niecza v11-20-gb962c2f: OUTPUT«[31m===[0mSORRY![31m===[0m��Malformed block at /tmp/MaN9zW9by_ line 1:�------> [32msub [33m�[31m∫() {...} # trying to use the unicode In[0m��Parse failed��»
10:19 p6eval ..pugs b927740: OUTPUT«decodeUTF8': bad data: '\8747'␤*** ␤    Unexpected "\8747()"␤    expecting "=", subroutine parameters, trait or block␤    at /tmp/akkO5aULpV line 1, column 5␤»
10:20 grondilu shouldn't this be working?!
10:22 ethndbst joined #perl6
10:25 Kivutarrr joined #perl6
10:26 grondilu nom: sub ∫() {...} # trying to use the unicode Integral character (0x222b) for a sub
10:26 p6eval nom 9c2880: OUTPUT«===SORRY!===␤Malformed block at line 1, near "\u222b() {...} "␤»
10:38 drbean joined #perl6
10:46 moritz geekosaur: nope, it's not an alphanumeric character
10:47 tokuhirom3 joined #perl6
10:49 Woodi will work for prefix op ?
11:35 not_gerd joined #perl6
11:45 tokuhiro_ joined #perl6
11:59 colomon mberends: ping?
12:00 mberends colomon: pong
12:01 colomon mberends: I'm trying to see if I can use TagLib# from niecza
12:02 colomon but right now I'm getting
12:02 colomon Unhandled Exception: Unable to resolve method Create in class CLR::TagLib::File,taglib-sharp,  Version=2::0::4::0, Culture=neutral, PublicKeyToken=db62eba44689b5b0
12:02 colomon I was hoping you might have some insight as to what I should be doing to track down the problem.  I've been trying to use monop without any success at all.
12:03 mberends colomon: that looks like a complicated way to say "library not found" :(
12:03 colomon Hmmm.... gacutil doesn't seem to list it?
12:03 colomon but it sure looks like it was installed:
12:04 colomon Package exported to: /usr/local/lib/mono/taglib-sharp/taglib-sharp.dll -> ../gac/taglib-sharp/2.0.4.0__db62eba44689b5b0/taglib-sharp.dll
12:04 colomon Installed taglib-sharp.dll into the gac (/usr/local/lib/mono/gac)
12:04 colomon shoot, I've got to go deal with a singing 3-year-old.  but any suggestions you can think of would be very appreciated.
12:05 mberends there may be a confusing way in which mono accesses different GACs at different times, I think diakopter had that for a while on a different platform.
12:11 Psyche^ joined #perl6
12:15 colomon hmmm... taglib-sharp is installed in /usr/local/lib/mono/gac -- and it's the only thing installed here
12:15 moritz so maybe it's not in the search path?
12:15 * moritz doesn't know anything about mono, just guesses wildly
12:17 colomon looks rightish
12:17 colomon gacutil -l -gacdir /usr/local/
12:17 colomon The following assemblies are installed into the GAC:
12:17 colomon policy.2.0.taglib-sharp, Version=0.0.0.0, Culture=neutral, PublicKeyToken=db62eba44689b5b0
12:17 colomon taglib-sharp, Version=2.0.4.0, Culture=neutral, PublicKeyToken=db62eba44689b5b0
12:17 colomon Number of items = 2
12:20 mberends colomon: I'll be afk for a few hours, but will attempt a similar setup later on Debian or Ubuntu
12:20 colomon mberends++
12:20 colomon think I might have it now
12:21 colomon I'm installing it by hand
12:22 colomon and now I get a different error!
12:22 mberends different error == progress
12:23 colomon and that was because my imaginary filename wasn't in a valid tag format.  error gone now that I have real MP3 in place to read.  \o/
12:23 moritz \o/
12:25 mberends excellent!
12:26 colomon sweet!  it works!
12:26 colomon my $file = TagLib-File.Create("Fathers_Jig.mp3");
12:26 colomon say $file.Tag.Title;
12:26 colomon Father's Jig
12:29 GlitchMr joined #perl6
12:31 mberends nice one colomon
12:32 colomon now if I can just figure out where the TagLib# documentation lives...
12:49 icovnik joined #perl6
12:56 lestrrat joined #perl6
13:01 colomon hmmm... how to convert a System.String[] to something useful?
13:07 amkrankruleuen joined #perl6
13:07 colomon nom: say @ARGS
13:07 p6eval nom 9c2880: OUTPUT«===SORRY!===␤Variable @ARGS is not predeclared at line 1, near ""␤»
13:08 colomon nom: say @*ARGS
13:08 p6eval nom 9c2880: OUTPUT«␤»
13:14 Woodi just read http://en.wikipedia.org/wiki/Perl_6 and some info predates git :) no niecza info and no www.perl6.org link...
13:17 huf it links to perl6.org right there in the little infobox
13:21 ethndbst joined #perl6
13:24 moritz Woodi: feel free to expand and correct as you see fit
13:35 masak hi from TCPW again, #perl6
13:35 szabgab hi masak
13:35 masak general poll: will give an '11 years of Perl 6' lightning talk later today -- what should I put the focus on?
13:36 szabgab "of"
13:36 masak I've been asking fivers this question and gotten interesting suggestions. maybe you will have some too.
13:36 masak szabgab: :P
13:36 szabgab oh I just managed to find the backlog
13:37 szabgab but the OP for Padre isn't around
13:38 szabgab IMHO the biggest impact might be the things that got back to Perl 5
13:39 szabgab btw moritz masak do look at the logs of your blogs sometimes? do you check how people find your sites?
13:40 moritz szabgab: planetsix, reddit, other perl blogs, other perl blog aggregators, hacker news, google (roughly in that order)
13:44 szabgab moritz: I mean searches
13:44 szabgab what search terms are people using to find your pages
13:45 szabgab I just noticed people searching for "uniq perl array" or some similar and arriving to my blog entry where I describe the uniq of Perl 6
13:45 szabgab that's not what they are looking for and I guess that just confuses them
13:45 moritz szabgab: I have similar experience with perl-6.de
13:45 huf put a little blurb at the top of the entry that directs them to list::moreutils or summit
13:46 moritz I have a little (German) Perl 6 tutorial with separate pages for if, while, arrays, for etc.
13:46 moritz and people search for "perl if", and find it
13:46 szabgab that's the reason I jumped ahead in my perl 5 tutorial and wrote the unique entry even before discussing the plain array
13:46 szabgab so I can put a link from the perl 6 entry to that new page
13:47 moritz szabgab: the non-perl 6 parts of my site have much higher search volume than the p6 parts
13:47 moritz for the p6 parts, my highest ranking search referral is "perl6 grammar"
13:50 szabgab I am worried only about people who look for perl 5 solution and find by mistake the perl 6 one
13:51 szabgab we could point them to the respective perl 5 tutorials
13:59 masak oh my, we still have &roundrobin in S32! I had completely forgotten about it!
13:59 jnthn huh, you talked about it in le pub last night :P
14:00 masak yeah, I had forgotten about it up until then, I mean :P
14:00 jnthn oh...was gonna say, the beer was that much :P
14:01 masak it was good beer
14:09 not_gerd jnthn: do you know how the command in $config{'win32_libparrot_copy'} set by Rakudo's/NQP's Configure.pl actually gets executed?
14:10 not_gerd (trying to make --gen-parrot --gen-nqp work on MSYS)
14:11 MayDaniel joined #perl6
14:15 Woodi masak: about '11 years': me thinks 'Perl6' concept makes great promises - mantras that all we know - PR was/is realy strong Perl6 side :) Promises: EASY integrating p5 code - just 'use5'! or by grammars or by Perl5-on-Parrot; much better concurency/threading; including other languages via Parro, eg Python; cleaning Perl...
14:15 masak szabgab: I sometimes look at the blog statistics, but I often forget. I should have an alarm to do it at the end of them month, because it clears with each new month.
14:17 Woodi but I think Pugs incident is realy important 'turning point' into functional functionalities... i would call it even 'restarting' or Phase 2 or 3...
14:17 masak Woodi: I'd much rather focus on things that happened than on things that didn't -- but yes, the general expectations/promises are important too. :)
14:17 masak Pugs was an inflection point, yes.
14:18 Woodi masak: the biggest point for me personally was/is concurency...
14:19 masak from ten thousand feet, the things you can see in the 11-year history are: (1) announcement of Perl 6, (2) Pugs, (3) general silver-age activity, maybe Rakudo Star release if you squint
14:19 masak I'm not saying Rakudo or Niecza haven't been influential -- they have been, and are.
14:19 masak but in terms of things that have clearly reached out of the echo chamber, those are the three that come to mind.
14:21 Woodi from ten thousand you see just distinguishing points in non-discret space
14:21 Woodi everyday work matters too
14:22 Woodi from other angle: you can say things that ARE implemented
14:23 Woodi (4) 6model is influent probably
14:24 * Woodi is afraid to modify wikipedia article for Perl6 :)
14:26 tadzik don't be
14:26 tadzik whatcha wanna change?
14:27 Woodi that Perl6 is developed in SVN repo :)
14:27 tadzik ;)
14:27 gerdr joined #perl6
14:32 Woodi masak: 'anything that pass official test suite is Perl6' - there is test suite and STD now, probably stable on all normal languages functionalities :)
14:33 ksi joined #perl6
14:46 masak yeah, STD is generally considered "part of spec".
14:47 masak TimToady is slightly ahead of the curve though and sometimes mumbling about STD having "filled its purpose" or something like that.
14:48 colomon Woodi: the test suite isn't quite stable on normal language functionalities, alas.
14:49 colomon for instance, the "munch" problem pointed out last week.
14:49 Woodi colomon: munch ?
14:49 mishin joined #perl6
14:50 mishin hi perl 6 what compiler i need to compile racudo under win7?
14:51 kmwallio joined #perl6
14:52 Woodi that one http://www.artcyclopedia.com/artists/munch_edvard.html ?
14:55 colomon munch is a list-related method which is used for tests in seven or eight files of the test suite
14:55 colomon it's never been spec'd
14:56 colomon and no one knows if it is going to go away or eventually become part of the spec
14:56 colomon (as that entire section of the spec is scheduled to be rewritten)
14:57 masak hi mishin -- it should just work, I think. jnthn will know more, but he's giving a talk right now.
14:58 moritz mishin: I think msvc should be fine
14:58 Woodi I do not thing such thing is standard functionality. anyway I just wrote Perl6 module with semi-Perl5-features and it works
14:59 moritz mishin: I think msvc should be fine
14:59 Woodi probably Perl5 installation is needed too. and there are binaries for Win
15:00 masak heh -- I ended up with exactly 100 slides for my "small languages, less accidentally" talk :)
15:00 Woodi animation ?
15:08 masak well, gradual slides, yes.
15:08 masak some of them.
15:09 masak I see 2011 as the year where Rakudo (thanks to jnthn's optimizer branch) starts to clearly do things that Perl 5 doesn't and probably can't.
15:13 Woodi masak: just remember: 'Perl6 is here!' :)
15:14 Woodi what shell or awk functionality missing in Perl6 ?
15:14 masak the spec? that's a kinda vague question.
15:14 Woodi no, here I mean Rakudo and Niecza
15:15 Woodi even Pugs
15:15 masak I don't think Rakudo does <<...>> yet, but I might be wrong.
15:15 masak Pugs does it.
15:16 Woodi excluding speed Perl6 compilers can do real work
15:16 masak and jnthn++ is working on the speed bit. and Niecza already does quite well, I think.
15:17 Woodi what is status NCI in b ?
15:17 Woodi MYSQL was working ?
15:23 sorear good * #perl6
15:28 _jaldhar joined #perl6
15:28 moritz perl6: sub f($x) { return if $x == 1; state %h = a => 1; say %h.perl }; f 1; f 2;
15:28 p6eval niecza v11-20-gb962c2f: OUTPUT«{"a" => 1}.hash␤»
15:28 p6eval ..pugs b927740: OUTPUT«{("a" => 1),}␤»
15:28 p6eval ..rakudo 9c2880: OUTPUT«().hash␤»
15:28 mishin moritz: tahnks, i try http://www.microsoft.com/visualstudio/en-us/products/2010-editions/visual-cpp-express
15:28 * moritz thinks it's a rakudobug
15:31 moritz buubot_backup: eval: sub f { return if $_[0]; state $x = 3; $x }; f 1; f
15:31 buubot_backup moritz: ERROR: Can't call method "state" on an undefined value at (eval 20) line 1.
15:31 moritz buubot_backup: eval: use 5.010; sub f { return if $_[0]; state $x = 3; $x }; f 1; f
15:31 buubot_backup moritz: 3
15:33 Woodi nom: my %cmds = { 'a' => (0,1) }; my $s = %cmds{ 'a' }; say $s
15:33 p6eval nom 9c2880: OUTPUT«0 1␤»
15:34 Woodi $scalar contains @list... Parcel but strange...
15:35 Woodi different then Perl5
15:35 moritz .oO( Woodi but Strange )
15:37 Woodi :) no exact idea what you think but it differs from Perl5
15:37 Woodi found it in module i work on 2 months....
15:37 moritz yes, we figured it makes sense not to lose information just because you put somethin in scalar context
15:38 Woodi work nice
15:38 dalek faq: 645ea3e | moritz++ | answers.md:
15:38 dalek faq: fix markup
15:38 dalek faq: review: https://github.com/perl6/faq/commit/645ea3ed6f
15:38 Woodi so this parcel means ?
15:38 moritz can you pleaes rephrase your question?
15:39 moritz I don't understand it
15:39 Woodi first think: it can have separate sigil.
15:39 Woodi ok. so parcel means: not to lose information just because you put somethin in scalar context
15:40 moritz I don't see how that's related to Parcels
15:40 Woodi nom: my %cmds = { 'a' => (0,1) }; my $s = %cmds{ 'a' }; say $s.WHAT
15:40 p6eval nom 9c2880: OUTPUT«Parcel()␤»
15:40 moritz (0, 1) is a Parcel
15:41 moritz (though maybe it should become a List in item context, the spec is kinda vague on that point)
15:41 Woodi so parcels are kind a pointer ?
15:41 moritz no
15:41 Woodi in Java sense
15:41 moritz nearly everything is a pointer in the java sense in Perl 6
15:42 uv_ joined #perl6
15:42 moritz we just lifted the p5 restriction that you can't directly put an array or a hash into a scalar variable
15:42 Woodi moritz: me thinks this is worth separate blogpost :)
15:42 moritz Woodi: http://faq.perl6.org/#ref
15:45 Woodi this is what i mean 'Perl6 is here!' - I write Perl6 module just in semi-Perl5 and have no idea about real Perl6...
15:45 dalek faq: fa74992 | moritz++ | / (2 files):
15:45 dalek faq: report last update time in UTC
15:45 dalek faq: review: https://github.com/perl6/faq/commit/fa74992a88
16:08 moritz perl6: sub f { return eval '$x'; state $x = 3 }; say f
16:08 p6eval niecza v11-20-gb962c2f: OUTPUT«Potential difficulties:â�¤  $x is declared but not used at /tmp/Zxv8KvQB05 line 1:â�¤------> [32msub f { return eval '$x'; state [33mâ��[31m$x = 3 }; say f[0mâ�¤â�¤Any()â�¤Â»
16:08 p6eval ..rakudo 9c2880: OUTPUT«Any()␤»
16:08 p6eval ..pugs b927740: OUTPUT«␤»
16:10 dalek roast: 633e3c3 | moritz++ | S04-declarations/state.t:
16:10 dalek roast: test state var initialization when the first run of the routine does not reach the initialization
16:10 dalek roast: review: https://github.com/perl6/roast/commit/633e3c374f
16:12 * moritz submits rakudobug
16:22 jnthn moritz: what's the bug?
16:24 icwiener joined #perl6
16:25 diakopter__ left #perl6
16:29 moritz perl6: sub f($x) { return if $x; state $a = 5; say $a }; f 1; f 0
16:29 p6eval pugs b927740, niecza v11-20-gb962c2f: OUTPUT«5␤»
16:29 p6eval ..rakudo 9c2880: OUTPUT«Any()␤»
16:29 moritz jnthn: that's the bug
16:29 moritz (not related to the eval thing, I just wanted to check how the different implementations do it)
16:30 moritz submitted as RT #102994
16:30 jnthn Hm. It's behaving as I expect
16:30 jnthn START is not ENTER.
16:30 jnthn Or a one time ENTER
16:30 diakopter sorear: oh
16:30 mishin moritz: well, i run  perl Configure.pl --gen-parrot https://gist.github.com/1341665 but it doesn't help me compile rakudo, i don't know what i should do under windows
16:32 mishin i think i need remove previos rakudo installation
16:33 moritz jnthn: S04 says about START: "on first ever execution, once per closure clone"
16:33 moritz jnthn: but rakudo never seems to initialize $a
16:33 moritz mishin: yes, you should
16:33 jnthn moritz: It says it does it inline
16:34 jnthn (the spec)
16:34 jnthn (S04)
16:34 jnthn masak and I are just looking at it
16:34 moritz how does doing it inline comes out as not doing it at all?
16:36 thou joined #perl6
16:38 jnthn moritz: When we first enter a block, we flag it as started
16:38 jnthn That flag gets checked when deciding whether to execute STARTs
16:43 diakopter nom: my $a = (sub { my $b; return sub { START { $b++ }; return $b; }; })(); say $a(); say $a();
16:43 p6eval nom 9c2880: OUTPUT«===SORRY!===␤CHECK FAILED:␤Undefined routine '&START' called (line 1)␤»
16:43 moritz ... and?
16:43 * moritz fails to see the point
16:43 * diakopter too
16:44 moritz jnthn: when it's implemented like that, all START phasers have to first block enter, even if they aren't reached textually
16:45 moritz just like all other phasers
16:45 moritz nom: exit; BEGIN say 1
16:45 p6eval nom 9c2880: OUTPUT«1␤»
16:45 moritz same here; the control flow never reaches the BEGIN, but the phaser still fires
16:46 moritz nom: sub f { return; ENTER { say "in f" } }; f
16:46 p6eval nom 9c2880: OUTPUT«===SORRY!===␤ENTER phaser not yet implemented at line 1, near " }; f"␤»
16:46 dalek 6model: cd9435d | diakopter++ | lua/runtime/List.lua:
16:46 dalek 6model: minor bugfix
16:46 dalek 6model: review: https://github.com/diakopter/6model/commit/cd9435d6e2
16:49 moritz on an unrelated matter, do we plan to do a perl 6 advent calendar this year too?
16:49 moritz s/too/again/
16:50 moritz I'm certainly for it, and would write one or more posts, but I'd like support from all you awesome guys and gals too :-)
16:50 dalek 6model: ed7a683 | diakopter++ | .gitignore:
16:50 dalek 6model: update .gitignore
16:50 dalek 6model: review: https://github.com/diakopter/6model/commit/ed7a683d4d
16:56 huf joined #perl6
16:58 dalek 6model: d4978a3 | diakopter++ | lua/compiler/Makefile:
16:58 dalek 6model: Makefile doesn't blow away LocalsOptimizer.lua
16:58 dalek 6model: review: https://github.com/diakopter/6model/commit/d4978a31c7
17:01 TimToady joined #perl6
17:01 diakopter TimToady: wb
17:01 TimToady thanks
17:01 phenny TimToady: 06:00Z <sorear> tell TimToady I would like to extract knowledge from you on how Perl 6 should treat \n and text-mode file IO on Windows
17:01 diakopter re: that, STD doesn't seem to like \r\n
17:02 uv_ perl6: my $abc='ABC'; say $abc; $abc.=lc.=ucfirst; say $abc;
17:02 p6eval rakudo 9c2880: OUTPUT«ABC␤Cannot assign to a readonly variable or a value␤  in method dispatch:<.=> at src/gen/CORE.setting:730␤  in block <anon> at /tmp/tI7h0urOWk:1␤  in <anon> at /tmp/tI7h0urOWk:1␤»
17:02 p6eval ..pugs b927740, niecza v11-20-gb962c2f: OUTPUT«ABC␤Abc␤»
17:03 TimToady probably not treating \n as logical newline
17:03 TimToady I'm fine with continuing the fine tradition of turning binary \r\n into \n internally on Windows
17:03 TimToady if anything, P6 has a much larger text vs binary distinction
17:04 TimToady unlike P5, P6 defaults to text
17:04 TimToady including for unicode
17:04 TimToady P5 defaults to both :)
17:06 diakopter phenny: tell sorear see http://irclog.perlgeek.de/perl6/2011-11-05#i_4665299 ff.
17:06 phenny diakopter: I'll pass that on when sorear is around.
17:10 wk joined #perl6
17:20 dalek 6model: 0ed7989 | diakopter++ | / (2 files):
17:20 dalek 6model: blow away P6Objects.pm; lots more tests pass.
17:20 dalek 6model: review: https://github.com/diakopter/6model/commit/0ed7989a0f
17:21 mishin joined #perl6
17:25 dalek mu: a0a224f | moritz++ | misc/perl6advent-2011/ (2 files):
17:25 dalek mu: initial topic brainstomring for the Perl 6 advent calendar 2011
17:25 dalek mu: review: https://github.com/perl6/mu/commit/a0a224f211
17:26 colomon brainstorming already, oh my!
17:27 bluescreen10 joined #perl6
17:31 moritz colomon: I've got an email asking if were planning one for 2011
17:32 moritz and since I'll be on vacation for one week in November, it's better to start a bit earlier
17:38 skangas joined #perl6
17:55 TimToady .oO(Is Perl 6 done yet?  Perl 6 is the Future, so of course it's unevenly distributed!)
17:57 MayDaniel joined #perl6
18:02 moritz .oO( Perl 6 is the future, and TimToady its prophet )
18:05 thou joined #perl6
18:24 mishin moritz: i couldn't compile perl6 for win https://plus.google.com/u/0/116733219688154342581/posts/TsqbKTHKXPX
18:29 moritz mishin: I don't know what to do, since I don't develo on windows :/
18:29 moritz maybe jnthn knows
18:32 LarryB joined #perl6
18:39 diakopter I was just about to tell mishin of my misfortune trying to build parrot on windows
18:40 diakopter it failed on cygwin/perl/gcc, windows/strawberry/gcc, windows/activeperl/msvc, all three with very different errors
18:41 mishin joined #perl6
18:41 diakopter mishin: I tried building rakudo/parrot on windows the other day; if failed on cygwin/perl/gcc, windows/strawberry/gcc, windows/activeperl/msvc, all three with very different errors
18:42 diakopter mishin: for the time being, you might try the parrot 3.9.0 installer from sourceforge
18:42 vmspb joined #perl6
18:42 diakopter http://sourceforge.net/projects/parrotwin32/files/parrotwin32%20setup/Parrot-3.9.0/setup-parrot-3.9.0.exe/download
18:43 mishin diakopter: thanks for info and for link, may be it help's me
18:44 diakopter mishin: unfortunately I don't know the command line option to rakudo's Configure to tell it to use an installed parrot
18:45 not_gerd joined #perl6
18:46 moritz perl Configure --gen-nqp --with-parrot=path/to/parrot.exe
18:48 mishin diakopter:I will try to put a, moritz:thanks
18:56 vmspb joined #perl6
19:02 dalek 6model: e35833a | diakopter++ | lua/runtime/Runtime/Ops/P6mapping.lua:
19:02 dalek 6model: implement missing P6mapping Ops
19:02 dalek 6model: review: https://github.com/diakopter/6model/commit/e35833ad8b
19:14 dalek 6model: 1affb51 | diakopter++ | lua/runtime/Runtime/ (2 files):
19:14 dalek 6model: moar fixes
19:14 dalek 6model: review: https://github.com/diakopter/6model/commit/1affb510e3
19:15 sorear hello
19:15 phenny sorear: 17:06Z <diakopter> tell sorear see http://irclog.perlgeek.de/perl6/2011-11-05#i_4665299 ff.
19:22 sorear TimToady: ping
19:22 sorear TimToady: if \n should match a logical newline, where does that leave <[ \ \t\n ]> ?  obviously(?) a character class should never match more than one grapheme
19:25 sorear TimToady: I'm also interested in knowing your thoughts on http://www.unicode.org/reports/tr18/#Partial_Matches
19:29 moritz sorear: in perl 5, 'SS' =~ /^[ß]$/i
19:35 moritz nom: say (104729, 86653, * % * ... 1).perl
19:35 p6eval nom 9c2880: OUTPUT«(104729, 86653, 18076, 14349, 3727, 3168, 559, 373, 186, 1).list␤»
19:36 moritz nom: say (104729, 86653, * div * ... 0).perl
19:36 p6eval nom 9c2880: OUTPUT«(104729, 86653, 1, 86653, 0).list␤»
19:36 cooper joined #perl6
19:40 sorear Should "\n" always be taken to mean "\cJ" ?
19:40 sorear Should there be any wildcarding magic in / "\n" / ?
19:41 colomon joined #perl6
19:42 sorear o/ colomon
19:43 mishin moritz: it doesn't help https://gist.github.com/1341885
19:44 moritz :(
19:46 colomon \o
19:46 colomon sorear:
19:46 colomon If the C# accessor code looks like this:
19:46 colomon public virtual string [] Performers {
19:46 colomon get {return new string [] {};}
19:46 colomon set {}
19:46 colomon }
19:47 colomon How do I get at it from Niecza?
19:47 colomon (Don't know if you saw in the backlog, I'm trying to get TagLib# to work with Niecza.)
19:47 sorear my $string-arr = $obj.Performers;
19:48 sorear $obj.Performers = $string-arr;
19:48 sorear $string-arr will be a CLR::("System.String[]") object; there isn't an unambiguous best way to marshal them, so they're left alone by the interface
19:49 colomon so what can you do with CLR::System::String[]?
19:49 colomon (as it calls itself.)
19:50 colomon (in .WHAT, I mean)
19:51 sorear monop 'System.String[]'
19:51 sorear in particular .Length() and .Get($index)
19:52 * diakopter cranially dents a brick wall
19:53 * sorear wonders what would happen if you tried to $string-arr.Address($index) from Perl 6 code
19:53 * sorear tries it
19:53 colomon sorear: gotcha.  BTW, on second thought, why isn't there an unambiguous best way to marshall?
19:53 sorear Aborted (core dumped)
19:53 mishin moritz: returning to ubuntu;))
19:53 sorear colomon: because CLR arrays are inescapably mutable objects
19:54 colomon sorear: ah, and we'd have to break that connection, so to speak?
19:54 sorear yes
19:55 sorear it's even more of a problem going the other way - CLR arrays cannot be tied, so there's no fully correct way to pass an Array[Int] as a CLR::System::Int32[]
19:55 sorear coming *from* we could probably fake marshalling with a tie
19:55 moritz I have a lazy list, say @a
19:56 moritz and want to use its element one after the other in the generator of the series operator
19:56 moritz is there a neat way to do it?
19:56 colomon sequence operator?
19:56 colomon @a ... whatever
19:56 moritz something like $a, $b, { $^b + $next-item-from-a * $^a}
19:56 colomon ah
19:56 moritz s/series/sequence/, yes
19:57 moritz I'm currently useing @a.shift, but it's destructive and feels rather cheaty
19:57 colomon seems like that might not be the best candidate for a sequence operator, off the top of my head
19:57 moritz hm
19:58 moritz a zip seems better for interleaving, but then the second-old element from the to-be-generated sequence isn't available
19:59 colomon sorear: so something like (^$a.Length).map({ $a.Get($_) }) converts the string array to a p6 list of Str... errr, probably a list of C# strings.
19:59 sorear this is a classic use case for unfoldr from the Haskell Prelude.
19:59 sorear colomon: C# strings are immutable so they can be transparently mapped to p6 Strs (and they are)
20:00 * sorear ponders writing a CLRUtilities.pm6 module with facilities for list marshalling, etc
20:00 diakopter +1
20:00 colomon sorear++ # quite right
20:00 colomon afk
20:00 sorear which for?
20:01 diakopter jnthn: ping
20:04 diakopter phenny: ask jnthn my eyes bleed from debugging nqplua; it needs a fresh set of eyes
20:04 phenny diakopter: I'll pass that on when jnthn is around.
20:05 colomon sorear: ++ was for correctly calling (and therefore previously having implemented) the conversion to Str.
20:05 colomon but +1 on CLRUtilities.pm6, too
20:06 sorear fwiw, there's a possibility that the string stuff may have to change with universal NFG in the coming months
20:06 colomon .... universal NFG?
20:07 sorear right now in niecza .chars means .codes
20:07 sorear someday I need to fix that and make it possible for strings to be indexed by grapheme index
20:08 sorear currently niecza's Str wraps a C# String, so converting between the forms is O(1)
20:08 colomon oh.
20:09 colomon gotcha.
20:10 colomon I've got to run work on the leaves in the yard, but I'm happy to boldly stumble where few have gone before and come up with routines that would be great to have in CLRUtilities.pm6.
20:10 colomon afk
20:17 * sorear prepares a bug report for the Address thing :)
20:19 saaki joined #perl6
20:28 ethndbst joined #perl6
20:29 Chillance joined #perl6
20:45 cooper joined #perl6
20:52 vmspb perl6: $_='aAa';  say qf/$_.uc $_.lc/;
20:52 p6eval pugs b927740, niecza v11-20-gb962c2f: OUTPUT«$_.uc $_.lc␤»
20:52 p6eval ..rakudo 9c2880: OUTPUT«===SORRY!===␤Confused at line 1, near "say qf/$_."␤»
20:54 vmspb perl6: $_='aAa';  say qq/$_.uc $_.lc/;
20:54 p6eval pugs b927740, rakudo 9c2880, niecza v11-20-gb962c2f: OUTPUT«aAa.uc aAa.lc␤»
20:54 vmspb perl6: $_='aAa';  say q/$_.uc $_.lc/;
20:54 p6eval pugs b927740, rakudo 9c2880, niecza v11-20-gb962c2f: OUTPUT«$_.uc $_.lc␤»
20:54 sorear perl6: $_='aAa';  say qq/$_.uc() $_.lc()/;
20:54 p6eval pugs b927740, rakudo 9c2880, niecza v11-20-gb962c2f: OUTPUT«AAA aaa␤»
20:55 sorear to reduce the false positive rate, parens are required on all interpolated method calls
21:06 simcop2387 joined #perl6
21:13 worr joined #perl6
21:24 vmspb sorear: tnx
21:24 vmspb perl6: my $a='aAa'; $_='bBb';  say <<1 $a.=uc().=lc()>>;
21:24 p6eval rakudo 9c2880: OUTPUT«1 $a.=uc().=lc()␤»
21:24 p6eval ..niecza v11-20-gb962c2f: OUTPUT«1 aaa␤»
21:24 p6eval ..pugs b927740: OUTPUT«1aAa.=uc().=lc()␤»
21:24 vmspb perl6: my $a='aAa'; $_='bBb';  say <<2 $a.=uc() .=lc()>>;
21:24 p6eval niecza v11-20-gb962c2f: OUTPUT«2 AAA .=lc()␤»
21:24 p6eval ..pugs b927740: OUTPUT«2aAa.=uc().=lc()␤»
21:24 p6eval ..rakudo 9c2880: OUTPUT«2 $a.=uc() .=lc()␤»
21:24 vmspb perl6: my $a='aAa'; $_='bBb';  say <<3 $a.=uc() $_.=lc()>>;
21:24 p6eval rakudo 9c2880: OUTPUT«3 $a.=uc() $_.=lc()␤»
21:24 p6eval ..pugs b927740: OUTPUT«3aAa.=uc()bBb.=lc()␤»
21:24 p6eval ..niecza v11-20-gb962c2f: OUTPUT«3 AAA bbb␤»
21:25 vmspb I think variant 2 and 3 should be the same
21:25 sorear vmspb: what exactly are you trying to do?
21:25 Reaganomicon joined #perl6
21:25 flussence he's doing a good job of digging up bugs :)
21:26 sorear vmspb: no, 2 and 3 should not be the same
21:29 vmspb http://perlcabal.org/syn/S02.html#Implicit_Topical_Method_Calls
21:29 sorear vmspb: that applies to code, not quotes
21:30 mj41 joined #perl6
21:31 sorear contrary to popular belief, you can't just put arbitrary Perl 6 code in the middle of a double-quoted string and have it work.
21:33 moritz well, you can, if you put {...} around it
21:33 moritz :-)
21:34 dalek 6model: c42173e | diakopter++ | lua/compiler/try.sh:
21:34 dalek 6model: update try.sh (untested) for nqplua; sorear++ for finding
21:34 dalek 6model: review: https://github.com/diakopter/6model/commit/c42173ea08
21:50 fsergot joined #perl6
22:04 MayDaniel joined #perl6
22:05 bluescreen10 joined #perl6
22:11 soh_cah_toa joined #perl6
22:15 localhost joined #perl6
22:21 PacoLinux joined #perl6
22:32 soh_cah_toa rakudo: my :($foo, $bar)
22:32 p6eval rakudo 9c2880: OUTPUT«===SORRY!===␤Malformed my at line 1, near ":($foo, $b"␤»
22:33 soh_cah_toa well, even though that failed, what does that ^ kind of declaration do? i've never seen the : used before
22:33 sorear std: my :($foo, $bar)
22:33 p6eval std be1f10e: OUTPUT«[31m===[0mSORRY![31m===[0mâ�¤Malformed my at /tmp/tzmhRV4Nei line 1:â�¤------> [32mmy [33mâ��[31m:($foo, $bar)[0mâ�¤    expecting scoped declaratorâ�¤Parse failedâ�¤FAILED 00:01 118mâ�¤Â»
22:43 dalek 6model: 88e7252 | sorear++ | lua/compiler/try.sh:
22:43 dalek 6model: try.sh needs to be marked executable to work on Unix
22:43 dalek 6model: review: https://github.com/diakopter/6model/commit/88e7252f77
22:52 Ingis joined #perl6
23:03 vmspb left #perl6
23:24 kst joined #perl6
23:26 dorlamm joined #perl6
23:26 dalek 6model: 8763433 | sorear++ | lua/runtime/Metamodel/KnowHOW/KnowHOWBootstrapper.lua:
23:26 dalek 6model: Fix nil concatenate secondary error on method-not-found
23:26 dalek 6model: review: https://github.com/diakopter/6model/commit/8763433160
23:48 am0c joined #perl6

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

Perl 6 | Reference Documentation | Rakudo