Perl 6 - the future is here, just unevenly distributed

IRC log for #perl6, 2012-12-02

Perl 6 | Reference Documentation | Rakudo

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

All times shown according to UTC.

Time Nick Message
00:16 _jaldhar joined #perl6
00:18 _jaldhar joined #perl6
00:20 Pleiades` joined #perl6
00:29 pecastro_ joined #perl6
00:31 tokuhiro_ joined #perl6
00:34 colomon n: given 1 {␤when 1 {␤say "1";␤} default {␤ say "not 1";␤}␤}given 1 {␤when 1 {␤say "1";␤} default {␤ say "not 1";␤}␤}
00:34 p6eval niecza v22-44-g0247fe5: OUTPUT«[31m===[0mSORRY![31m===[0m��Strange text after block (missing comma, semicolon, comment marker?) at /tmp/aOjK_FPw4N line 4:�------> [32m}[33m�[31m default {[0m��Parse failed��»
00:34 colomon n: given 1 {␤when 1 {␤say "1";␤} default {␤ say "not 1";␤}␤}
00:34 p6eval niecza v22-44-g0247fe5: OUTPUT«[31m===[0mSORRY![31m===[0m��Strange text after block (missing comma, semicolon, comment marker?) at /tmp/6D2KJZMteb line 4:�------> [32m}[33m�[31m default {[0m��Parse failed��»
00:35 lue .oO(day 18 looks to be interesting)
00:36 colomon n: given 1 {␤when 1 {␤say "1";␤}␤ default {␤ say "not 1";␤}␤}
00:36 p6eval niecza v22-44-g0247fe5: OUTPUT«1␤»
00:36 colomon r: given 1 {␤when 1 {␤say "1";␤}␤ default {␤ say "not 1";␤}␤}
00:36 p6eval rakudo da6569: OUTPUT«1␤»
00:36 colomon there's your problem: missing newline
00:37 lue colomon: I know. I'm saying it shouldn't be an error, esp. considering the lack-of-a-newline works when using another when block instead of a default block.
00:38 lue rn: given 1 {␤when 1 {␤say "1";␤} when 2 {␤ say "not 1";␤}␤}
00:38 p6eval rakudo da6569: OUTPUT«1␤»
00:38 p6eval ..niecza v22-44-g0247fe5: OUTPUT«[31m===[0mSORRY![31m===[0m��Strange text after block (missing comma, semicolon, comment marker?) at /tmp/xgVDjUA1gM line 4:�------> [32m}[33m�[31m when 2 {[0m��Parse failed��»
00:39 lue (I personally like putting the first line of the next when/elsif/* block on the same line of the closing brace of the last one)
00:40 sorear lue: the fact that it works with two whens is a bg
00:40 sorear std: when 1 {} when 2 { }
00:40 p6eval std a8bc48f: OUTPUT«[31m===[0mSORRY![31m===[0mâ�¤Strange text after block (missing comma, semicolon, comment marker?) at /tmp/e_0QEKLmbU line 1:â�¤------> [32mwhen 1 {}[33mâ��[31m when 2 { }[0mâ�¤    expecting horizontal whitespaceâ�¤Parse failedâ�¤FAILED 00:00 42mâ�¤Â»
00:40 lue std: if 1 { } elsif 2 { }
00:40 p6eval std a8bc48f: OUTPUT«ok 00:00 42m␤»
00:41 lue ? why the inconsistency?
00:41 sorear if...elsif is a single statement
00:41 sorear when...when is two statements
00:41 sorear separating two statements always requires either a ; or a }\n
00:45 lue Aw. I think I prefer  } when 2 {  for similar reasons to why I don't like  sub declaration() \n { (I don't like the look of a brace on its own line, it seems)
00:45 sorear }; when 2 { then
00:48 lue :/ (/me goes to read up on the details of given/when to find a rationale for the "they're separate statements" decision)
00:52 lue r: given 1 {␤when 1 {␤say "1";␤} when * {␤ say "not 1";␤}␤}
00:52 p6eval rakudo da6569: OUTPUT«1␤»
01:51 gfldex joined #perl6
01:52 skids joined #perl6
01:59 tokuhiro_ joined #perl6
02:17 sorear phenny: tell japhb I have a theory... how does $acc ~= ($x ~ $y); do in rakudo, where $x and $y are loop-invariant but not "constant"s?
02:17 phenny sorear: I'll pass that on when japhb is around.
02:19 dayangkun joined #perl6
02:24 grondilu joined #perl6
02:25 grondilu r: role Foo {}; my $x = Int but Foo; say Int($x) ~~ Foo;
02:25 p6eval rakudo da6569: OUTPUT«True␤»
02:25 grondilu n: role Foo {}; my $x = Int but Foo; say Int($x) ~~ Foo;
02:25 p6eval niecza v22-44-g0247fe5: OUTPUT«Unhandled exception: Unable to resolve method postcircumfix:<( )> in type Int␤  at /tmp/BZhgrg1TWY line 1 (mainline @ 6) ␤  at /home/p6eval/niecza/lib/CORE.setting line 4211 (ANON @ 3) ␤  at /home/p6eval/niecza/lib/CORE.setting line 4212 (module-CORE @ 578…
02:26 grondilu r: role Foo {}; my $x = Int but Foo; say Int.bless($x) ~~ Foo;
02:26 p6eval rakudo da6569: OUTPUT«True␤»
02:26 grondilu How do I "un-role" an object?
02:32 grondilu For instance, how do I make a vanilla Int from Foo-enroled Int?
02:32 sorear 0 + $enrolled-int
02:33 sorear in the general case, though, there's no way
02:33 sorear role application is a one-way street
02:34 sorear jnthn will probably have more to say about this..
02:41 grondilu r: role Foo {}; my $x = 13 but Foo; say (0+$x)  ~~ Foo;
02:41 p6eval rakudo da6569: OUTPUT«False␤»
02:48 dalek niecza: 2ad9ba8 | sorear++ | lib/ (3 files):
02:48 dalek niecza: A bit of optimization to ~double the speed of ~; make .join non-quadratic
02:48 dalek niecza: review: https://github.com/sorear/niecza/commit/2ad9ba8a1f
02:50 grondilu In perl5 I could use the 'import' function to send a parameter when using a package.  Like 'package Foo { our $param; sub import { $param = shift } }' and then 'use Foo qw(bar);'   What's the equivalent of that in Perl 6?
03:07 popl joined #perl6
03:16 am0c joined #perl6
03:34 lue rn: say ?$~MAIN.defined
03:34 p6eval rakudo da6569: OUTPUT«===SORRY!===␤Non-declarative sigil is missing its name␤at /tmp/ugGZgbcO0G:1␤»
03:34 p6eval ..niecza v24-1-g2ad9ba8: OUTPUT«[31m===[0mSORRY![31m===[0mâ�¤â�¤Unhandled reference twigil ~ at /tmp/PrCDhhkBlV line 1:â�¤------> [32msay ?$~MAIN[33mâ��[31m.defined[0mâ�¤â�¤Unhandled exception: Check failedâ�¤â�¤  at /home/p6eval/niecza/boot/lib/CORE.setting line 1443 (die @ 5) â�¤  at /home/p6eval…
03:36 orafu joined #perl6
03:39 lue rn: https://gist.github.com/4186848
03:39 p6eval niecza v24-1-g2ad9ba8: OUTPUT«[31m===[0mSORRY![3​1m===[0mâ�¤â�¤GLOBAL::MAIN does not name any package at /tmp/5FsBGNYNFn line 5:â�¤------> [32maugment slang MAIN[33mâ��[31m {[0mâ�¤â�¤Augment requires a target at /tmp/5FsBGNYNFn line 5:â�¤------> [32maugment slang MAIN [33mâ��[31m{[0mâ�¤â�¤Illega…
03:39 p6eval ..rakudo da6569: OUTPUT«===SORRY!===␤You tried to augment slang MAIN, but it does not exist␤at /tmp/deaNnr5nXo:5␤»
03:44 rurban joined #perl6
03:47 rom1504 joined #perl6
03:51 sjohnson joined #perl6
04:06 lue do the terms in src/core/terms.pm get turned into dynamic variables (e.g. the %ENV in that file is %*ENV everywhere else)?
04:10 replore joined #perl6
04:15 sorear lue: yes, per spec even
04:36 Psyche^ joined #perl6
04:49 dalek mu: 2930b01 | colomon++ | misc/perl6advent-2012/schedule:
04:49 dalek mu: Update misc/perl6advent-2012/schedule
04:49 dalek mu:
04:49 dalek mu: Claim a couple of days.
04:49 dalek mu: review: https://github.com/perl6/mu/commit/2930b0193f
04:51 fgomez joined #perl6
04:52 lue where should I put tests for the $~ variables in S02?
04:54 tokuhiro_ joined #perl6
04:55 sorear o/ colomon
04:55 colomon \o
04:56 dalek niecza: 71a8938 | sorear++ | lib/CORE.setting:
04:56 dalek niecza: Specialize Range iteration for Numeric to avoid polymorphic cmp
04:56 dalek niecza: review: https://github.com/sorear/niecza/commit/71a89388c9
04:56 colomon about to sleep
04:56 colomon sorear++
04:56 sorear lue: I'd put them under S26, personally.
04:56 colomon sorear: I ran spectest today on niecza and got ~10 failing files.  Haven't had a chance to investigate them yet.
04:57 lue you mean S28, right?
05:24 dalek roast: 05ff1e5 | lue++ | S28-named-variables/slangs.t:
05:24 dalek roast: Added tests for $~ slang variables.
05:24 dalek roast:
05:24 dalek roast: Specifically the particular ones provided by Perl 6 by default.
05:24 dalek roast: review: https://github.com/perl6/roast/commit/05ff1e5462
05:24 dalek roast: a5c9497 | lue++ | S28-named-variables/slangs.t:
05:24 dalek roast: Fudge tests for $~ for rakudo.
05:24 dalek roast: review: https://github.com/perl6/roast/commit/a5c94970bb
05:38 sorear lue: I mean whichever one talks about pod
05:38 lue the $~ variables have nothing to do with Pod, though.
05:41 sorear oh I thought you were talking about $=
05:41 sorear my bad
06:30 wamba joined #perl6
06:31 SamuraiJack joined #perl6
06:58 moritz r: say 42 gcd 58
06:58 p6eval rakudo da6569: OUTPUT«2␤»
06:58 bapa joined #perl6
07:04 _jaldhar joined #perl6
07:17 azawawi joined #perl6
07:17 azawawi hi
07:17 phenny azawawi: 28 Nov 20:25Z <japhb> tell azawawi I'll add your any(@array) benchmark to perl6-bench. Any other micro-benchmarks you'd like to see added, or particular performance gotchas you've found?
07:17 sorear hello azawawi
07:18 azawawi hmm good
07:18 azawawi sorear++ # for faster junctions... :)
07:18 * azawawi takes a look @ perl6-bench
07:21 azawawi sorear: how can i embed a niecza REPL? any examples?
07:21 azawawi sorear: any API for that?
07:22 sorear I think you are looking for some kind of open3
07:35 dalek niecza: 1d0c020 | sorear++ | lib/ (2 files):
07:35 dalek niecza: Forward compatibility for extending the set of kernel-defined types
07:35 dalek niecza: review: https://github.com/sorear/niecza/commit/1d0c020009
07:35 dalek niecza: c85099a | sorear++ | lib/ (3 files):
07:35 dalek niecza: Some C# love for the numeric range hot path
07:35 dalek niecza: review: https://github.com/sorear/niecza/commit/c85099ac19
07:35 moritz phenny: ask masak if his advent calendar post is ready for publishing
07:35 phenny moritz: I'll pass that on when masak is around.
07:42 kaare__ joined #perl6
07:43 azawawi sorear: ok thx...
07:46 dalek niecza: 599cbcb | sorear++ | lib/Kernel.cs:
07:46 dalek niecza: Unroll the range iterator for further wins (I think this idea was from an old blog of colomon++ 's)
07:46 dalek niecza: review: https://github.com/sorear/niecza/commit/599cbcb326
07:47 sorear phenny: tell japhb 'for 1..20 { }' before: 757.56 ± 26.546 µs now: 53.345 ± 1.8765 µs
07:47 phenny sorear: I'll pass that on when japhb is around.
07:51 sorear phenny: tell japhb I'm shelving this for now but there's still much work to be done. [ loop (my $i = 1; $i <= 20; $i++) { } ] = 10.730 ± 0.0305 µs; [ my $i = -20; Nil while ++$i ] = 4.1028 ± 0.1603 µs
07:51 phenny sorear: I'll pass that on when japhb is around.
07:55 skids https://gist.github.com/4187527  # more bitops fixes, will post an RT tomorrow unless someone can find one it should belong to.
07:58 azawawi phenny: tell japhb i'll send pull requests with any performance bottlenecks I encounter. Thanks for making and pointing out your cool perl6-bench project. I will check it out later today.
07:58 phenny azawawi: I'll pass that on when japhb is around.
08:07 cedrvint joined #perl6
08:07 cedrvint moritz: I think masak's post can be published: http://irclog.perlgeek.de/​perl6/2012-12-01#i_6203594
08:22 am0c joined #perl6
08:23 lue phenny: ask jnthn how would someone go about turning %*LANG<MAIN> in Perl6/Compiler.pm into $~MAIN, along with the other $~ variables?
08:23 phenny lue: I'll pass that on when jnthn is around.
08:23 lue goodnight o/
08:24 cedrvint phenny: tell masak it seems auto-publishing has failed. The post is 8 hours late now, I publish it (and hope i'm not wrong)
08:24 phenny cedrvint: I'll pass that on when masak is around.
08:25 araujo joined #perl6
08:26 Kharec joined #perl6
08:27 wamba joined #perl6
08:35 GlitchMr joined #perl6
08:35 betterworld joined #perl6
09:04 FROGGS joined #perl6
09:04 FROGGS morning
09:06 sorear o/
09:15 GlitchMr Just wondering, can I get editor access on Perl 6 advent to put article for 12th day which I've already done?
09:22 kst` joined #perl6
09:34 tokuhiro_ joined #perl6
09:35 jnthn morning, #perl6
09:35 phenny jnthn: 08:23Z <lue> ask jnthn how would someone go about turning %*LANG<MAIN> in Perl6/Compiler.pm into $~MAIN, along with the other $~ variables?
09:38 sorear morning jnthn
09:39 jnthn phenny: tell lue Well, they are installed in some %?LANG but they will suffer the "not under Mu" issue. And I need to do some MOP stuff before turning to that.
09:39 phenny jnthn: I'll pass that on when lue is around.
09:40 jnthn phenny: tell lue but I think at a first guess $~MAIN would desugar to %?LANG<MAIN>
09:40 phenny jnthn: I'll pass that on when lue is around.
09:40 jnthn o/ sorear
09:42 * jnthn will be back to having regular-ish Perl 6 tuits again from tomorrow o/
09:46 GlitchMr 240 updates available... this gonna be fun
09:47 GlitchMr It appears I've to update every KDE app I've
09:52 cognominal joined #perl6
09:54 dalek perl6.org: 997054d | GlitchMr++ | source/documentation/index.html:
09:54 dalek perl6.org: Add Perl 6 Advent Calendar 2012 link.
09:54 dalek perl6.org: review: https://github.com/perl6/p​erl6.org/commit/997054da67
09:58 dalek perl6.org: e6eccff | GlitchMr++ | source/documentation/index.html:
09:58 dalek perl6.org: Revert "[documentation] warn about perl6-examples outdatedness, sbp++"
09:58 dalek perl6.org:
09:58 dalek perl6.org: It was updated lately by gerdr++.
09:58 dalek perl6.org:
09:58 dalek perl6.org: This reverts commit fc4bcf057b5fe916479f97984c2a3668dcb336b1.
09:58 dalek perl6.org:
09:58 dalek perl6.org: Conflicts:
09:58 dalek perl6.org: source/documentation/index.html
09:58 dalek perl6.org: review: https://github.com/perl6/p​erl6.org/commit/e6eccff68a
10:04 dalek perl6.org: 705305f | GlitchMr++ | source/about/index.html:
10:04 dalek perl6.org: Update Carl Mäsak website link.
10:04 dalek perl6.org: review: https://github.com/perl6/p​erl6.org/commit/705305f695
10:07 GlitchMr -Ofun
10:12 spider-mario joined #perl6
10:16 GlitchMr ok, now let's make something -Ofun :-)
10:17 GlitchMr glitchmr@feather ~> panda install Bailador
10:17 GlitchMr Mu()
10:17 GlitchMr resolve stage failed for Bailador: Bailador depends on HTTP::Easy, which was not found in the ecosystem
10:17 GlitchMr What?
10:23 rindolf joined #perl6
10:24 sorear sleep&
10:38 grondilu joined #perl6
10:39 grondilu Is panda supposed to be compliant with rakudo 2012.11?
10:39 spider-mario_ joined #perl6
10:40 grondilu Caus' it's not.  Probably since the disappearance of %*ENV{HOME}/.perl6/lib and all
10:41 GlitchMr It hasn't disapperead
10:41 GlitchMr It will disappear in 2012.12
10:41 GlitchMr disappeared*
10:41 GlitchMr Besides, it's compatible with it
10:41 tokuhiro_ joined #perl6
10:41 GlitchMr It saves to path shown by %*CUSTOM_LIB
10:42 grondilu I don't understand.  I ran ./rebootstrap.pl once again, and now it seems to work fine.  I don't know what happened.
10:43 grondilu Please ignore my previous rant anyway.
10:43 birdwindupbird joined #perl6
10:44 grondilu Ah, hang on.  There is indeed an error.  It just appeared a bit later than I thought:  ===SORRY!===
10:44 grondilu Could not find File::Find in any of: /
10:44 grondilu (+long list of directories)
10:44 * jnthn knows there was a branch to update it to the custom lib stuff, but ain't sure if it was merged
10:45 jnthn guess I should go do a little sightseeing while there's sun... :)
10:45 jnthn bbl
10:46 GlitchMr I've weird problem with Bailador
10:46 GlitchMr Any header other than Content-Type doesn't work
10:46 GlitchMr Weird...
10:46 GlitchMr I wonder why it happens
10:48 FROGGS_ joined #perl6
10:50 GlitchMr This is so confusing
10:50 GlitchMr What's so special about Content-Type that it works...
10:51 GlitchMr (hmmm... perhaps Perl 6 debugger will help me)
10:51 FROGGS_ joined #perl6
11:01 GlitchMr oh
11:01 GlitchMr I found a stupid problem
11:01 GlitchMr HTTP::Easy::PSGI supports only one HTTP header
11:01 GlitchMr WHAT?
11:02 GlitchMr |   for @($result[1]) -> $header
11:02 GlitchMr > $result
11:02 GlitchMr [303, "Content-Type" => "text/plain", "Nothing-Lol" => "nowhere"]
11:02 GlitchMr That... doesn't feel rite
11:05 Su-Shee_ joined #perl6
11:16 GlitchMr http://www.perlcabal.org:1337/ -ofun
11:16 GlitchMr s/-o/-O/
11:31 MayDaniel joined #perl6
11:35 GlitchMr *** glibc detected *** perl6: double free or corruption (out): 0x0b1fe0a0 ***
11:35 GlitchMr What... the... hell?
11:35 GlitchMr That shouldn't happen, should it?
11:35 GlitchMr https://gist.github.com/4188309
11:35 GlitchMr There is the log, but not sure how useful it is
12:06 dalek perl6-examples: 4ed4ade | (Gerhard R)++ | euler/prob047-gerdr. (2 files):
12:06 dalek perl6-examples: [euler/prob047] add version using NativeCall
12:06 dalek perl6-examples: review: https://github.com/perl6/perl​6-examples/commit/4ed4ade5db
12:06 dalek perl6-examples: 8b1cc51 | (Gerhard R)++ | euler/prob053-gerdr.pl:
12:06 dalek perl6-examples: [euler/prob053] bikeshedding to align with coming advent post
12:06 dalek perl6-examples: review: https://github.com/perl6/perl​6-examples/commit/8b1cc518b7
12:06 dalek perl6-examples: 50cd5aa | (Gerhard R)++ | euler/prob002-gerdr.pl:
12:06 dalek perl6-examples: [euler/prob002] respect proper bound
12:06 dalek perl6-examples: review: https://github.com/perl6/perl​6-examples/commit/50cd5aa6b4
12:06 dalek perl6-examples: ea70e6e | (Gerhard R)++ | euler/prob029-gerdr.pl:
12:06 dalek perl6-examples: [euler/prob029] replace loop with bulk hash assignment
12:06 dalek perl6-examples: review: https://github.com/perl6/perl​6-examples/commit/ea70e6eb62
12:17 felher masak++ #blog post (haven't read it yet. But will after lunch)
12:36 moritz nqp: nqp::findmethod(1, 'foo')
12:36 p6eval nqp: OUTPUT«Method 'foo' not found for invocant of class 'Integer'␤current instr.: '' pc 43 ((file unknown):38) (/tmp/AColaPrHet:1)␤»
12:40 grondilu rn: role Foo { has $.x; method succ { self.new: :x($.x eq "foo" ?? "bar" !! "foo") } }; my $f = "foo" but Foo; say $f++; say $f++;
12:40 p6eval niecza v24-5-g599cbcb: OUTPUT«Use of uninitialized value in string context␤  at /home/p6eval/niecza/lib/CORE.setting line 1287 (warn @ 5) ␤  at /home/p6eval/niecza/lib/CORE.setting line 266 (Mu.Str @ 15) ␤  at <unknown> line 0 (ExitRunloop @ 0) ␤  at /tmp/svWveTRh9O line 1 (Foo.succ @ 4…
12:40 p6eval ..rakudo da6569: OUTPUT«use of uninitialized value of type Any in string context  in method succ at /tmp/GbDMwdJboV:1␤␤foo␤␤»
12:41 grondilu rn: role Foo { has $.x; method succ { self.new: :x($.x eq "foo" ?? "bar" !! "foo") } }; my $f = "foo" but Foo; $f.x = "foo"; say $f.x; $f++; say $f.x;
12:41 p6eval rakudo da6569: OUTPUT«Cannot assign to a readonly variable or a value␤  in block  at /tmp/0TQMwkCkum:1␤␤»
12:41 p6eval ..niecza v24-5-g599cbcb: OUTPUT«foo␤Unhandled exception: Excess arguments to Str.new, unused named x␤  at <unknown> line 0 (ExitRunloop @ 0) ␤  at /tmp/F58EYLv3ny line 1 (Foo.succ @ 5) ␤  at <unknown> line 0 (ExitRunloop @ 0) ␤  at /tmp/F58EYLv3ny line 1 (mainline @ 10) ␤  at /home/p6eval…
12:45 moritz aaaarrrgh
12:45 moritz parrot has a VTABLE_find_method, and an op called find_method
12:45 moritz and guess what, the vtable returns PMCNULL when it can't find the method being looked for
12:45 moritz and the op throws an exception
12:45 moritz fun, innit?
12:47 bonsaikitten lol
12:51 grondilu rn: role Foo {}; multi infix:<+>(Foo $, Foo $) { !!! "nyi" }; say (1 but Foo) + (2 but Foo);
12:51 p6eval niecza v24-5-g599cbcb: OUTPUT«Unhandled exception: >>>Stub code executed␤  at /tmp/a4uxlfMiGU line 1 (infix:<+> @ 4) ␤  at /tmp/a4uxlfMiGU line 1 (mainline @ 7) ␤  at /home/p6eval/niecza/lib/CORE.setting line 4208 (ANON @ 3) ␤  at /home/p6eval/niecza/lib/CORE.setting line 4209 (module-C…
12:51 p6eval ..rakudo da6569: OUTPUT«Ambiguous call to 'infix:<+>'; these signatures all match:␤:(Int:D \a, Int:D \b)␤:(Foo , Foo )␤␤  in block  at /tmp/Lp87STYsnE:1␤␤»
12:58 GlitchMr http://www.perlcabal.org:1337/
12:58 GlitchMr I've made a wiki in Perl 6
12:58 GlitchMr I know, lame
12:59 moritz \o/ Decemeber
12:59 GlitchMr I actually like that name :-)
13:00 GlitchMr I will use December as name, ok?
13:00 moritz I have no objections :-)
13:01 GlitchMr It's just quick experiment with Perl 6
13:01 grondilu GlitchMr: can we see the code?
13:01 GlitchMr Yeah, sure. It's in /home/glitchmr/WorstWikiEver on feather
13:01 GlitchMr I can upload it if somebody doesn't have access to feather
13:02 moritz not everybody has :-)
13:02 Khisanth joined #perl6
13:02 GlitchMr https://gist.github.com/4188536
13:02 GlitchMr I know, it looks bad.
13:03 grondilu Oh you actually use git for the versionning system.
13:04 GlitchMr Is there anything wrong with it? It just works.
13:04 grondilu No, it actually makes sense to do so.  I wouldn't have thought about it.
13:05 pmurias joined #perl6
13:05 GlitchMr Now, how can I read the file from revision without actually checkouting that revision.
13:05 pmurias jnthn: ping
13:06 GlitchMr git show perhaps
13:06 dalek rakudo/nom: 5481c6e | moritz++ | src/Perl6/Metamodel/MethodDelegation.pm:
13:06 dalek rakudo/nom: fix find_method for types with delegation
13:06 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/5481c6efa9
13:06 moritz GlitchMr: git show $revision:$filename
13:06 GlitchMr Thanks :-)
13:10 Moukeddar joined #perl6
13:13 tokuhiro_ joined #perl6
13:17 bowtie joined #perl6
13:19 not_gerd joined #perl6
13:19 not_gerd hello, #perl6
13:20 FROGGS hi not_gerd
13:20 not_gerd I believe I'm done with my advent post: https://gist.github.com/73e4f265c2c0b500b568
13:20 FROGGS the problem you posted yesterday is in an older version of Inline::C, right?
13:21 FROGGS the current versions shouldn't work at, without changes to NativeCall btw
13:21 FROGGS so there is no working version yet
13:22 not_gerd FROGGS: I believe Inline::C was a fresh checkout, but I don't remember if I updated Rakudo/NativeCall
13:22 moritz not_gerd: "However, no one forces you to go crazy with operators - there's nothing wrong with vanilla imperative code:"
13:22 moritz not_gerd: either that colon should be a dot, or there should be imperative code after it
13:22 moritz oh, you're talking about the next problem
13:23 dalek Heuristic branch merge: pushed 31 commits to rakudo/qast-sink-1 by moritz
13:24 jnthn moritz: There's a find_method_null_ok op in perl6.ops, if that helps
13:25 jnthn pmurias: pong
13:26 GlitchMr http://www.perlcabal.org:1337/M​ain%20page/log/ff3f0dd/378a2d0
13:28 moritz jnthn: then I guess 5481c6e should have used that, right?
13:28 jnthn moritz: It'd be more efficient.
13:29 moritz jnthn: is there an nqp:: op for that?
13:31 jnthn moritz: not yet
13:32 moritz https://gist.github.com/4109216 # updated qast-sink-1 spectest summary
13:33 moritz jnthn: can you outline how I could fix nqp::can(role { }, 'sink') ?
13:34 jnthn Hm, did I get to the bottom of what was going on there?
13:34 jnthn oh...I think I remember.
13:34 jnthn Once you get the pun
13:34 jnthn Check if it can do the method
13:35 jnthn If yes return the closure we currently do
13:35 jnthn If not return nqp::null() or some type object
13:36 jnthn Should just need a those relatively small changes in rolepuning.pm
13:36 moritz I see
13:38 moritz r: say (role { }).HOW.^name
13:38 p6eval rakudo 5481c6: OUTPUT«Perl6::Metamodel::ParametricRoleHOW␤»
13:40 * moritz tests a patch
13:40 grondilu is ':$.foo' a valid syntax for ':foo($.foo)' ?
13:41 moritz yes
13:41 grondilu ok
13:41 moritz r: class A { method x() { 42 }; method y { :$.x } }; say A.y.perl
13:42 p6eval rakudo 5481c6: OUTPUT«"x" => 42␤»
13:42 moritz r: class A { method y { :$.^name } }; say A.y.perl
13:42 grondilu thanks
13:42 p6eval rakudo 5481c6: OUTPUT«===SORRY!===␤Unable to parse expression in block; couldn't find final '}' at line 2, near ":$.^name }"␤»
13:43 moritz std: class A { method y { :$.^name } }; say A.y.perl
13:43 p6eval std a8bc48f: OUTPUT«[31m===[0mSORRY![31m===[0m�Bogus statement at /tmp/gZ9uKVpr7I line 1:�------> [32mclass A { method y { [33m�[31m:$.^name } }; say A.y.perl[0m�Parse failed�FAILED 00:00 43m�»
13:43 moritz std: class A { method x { $.^foo } }
13:43 p6eval std a8bc48f: OUTPUT«[31m===[0mSORRY![31m===[0m�Unsupported use of $. variable; in Perl 6 please use the filehandle's .line method at /tmp/l9yTEWBzYY line 1:�------> [32mclass A { method x { $.[33m�[31m^foo } }[0m�Parse failed�FAILED 00:00 42m�»
13:47 moritz jnthn: the fix for nqp::can seems to work; spectesting now
13:49 moritz or not :(
13:49 moritz it works for ParametricRoleHOW, but not for ParametricRoleGroupHOW
13:51 moritz because the latter doesn't do MethodContainer
13:51 not_gerd the advent posts miss the Day XX prefix - should that be fixed (with or without breaking the URL?)
13:51 moritz without breaking the URLs please :-)
13:53 not_gerd will do
14:08 masak g'day, #perl6
14:08 phenny masak: 07:35Z <moritz> ask masak if his advent calendar post is ready for publishing
14:08 phenny masak: 08:24Z <cedrvint> tell masak it seems auto-publishing has failed. The post is 8 hours late now, I publish it (and hope i'm not wrong)
14:08 masak oh, you did the right thing.
14:08 masak huh, I probably should've taken it from "draft" to "to be published" status or something.
14:09 masak the GUI made it seem to me everything was in order. :/
14:10 pmurias jnthn: would it be ok if I commited small expansions to the nqp test suit (which nqp-parrot passes without change) directly to the nqp repo? as updating an nqp fork is a bit annoying
14:12 masak lol, "December". :)
14:12 masak GlitchMr:
14:12 masak GlitchMr++ # https://gist.github.com/4188536
14:14 GlitchMr It's very simple wiki :-)
14:14 GlitchMr But hey, it already "sorta" works
14:14 GlitchMr http://www.perlcabal.org:1337/
14:15 masak GlitchMr: have you considered what happens on line 60, when a revert results in a conflict?
14:15 GlitchMr That was old code
14:15 GlitchMr It was already fixed... sorta
14:16 pmurias jnthn: is it correct if in QAST a variable is first declared as 'contextual' and then accessed as 'lexical'?
14:16 GlitchMr https://gist.github.com/4188536
14:16 cedrvint joined #perl6
14:16 masak GlitchMr: that looks identical to me. what changed?
14:16 GlitchMr huh?
14:17 GlitchMr I'm confused too
14:17 GlitchMr oh, right
14:17 GlitchMr That paste was newer
14:17 GlitchMr Nothing
14:17 GlitchMr if $content ~~ /'<<<<<<< HEAD'/ {
14:17 masak o.O
14:17 GlitchMr I know, it's lame hack
14:17 masak surely there's a better way...
14:18 GlitchMr For now I wanted it to work.
14:18 GlitchMr Before doing more complex stuff
14:18 masak *nod*
14:18 masak well, been there, done that.
14:18 masak good luck, sur.
14:18 masak sir*
14:19 cedrvint hello #perl6, the post for day 3 needs proof-reading. Any volunteers?
14:19 masak I'll gladly have a look.
14:19 masak url?
14:20 cedrvint masak: in wordpress, so you can make changes directly if you wish
14:21 masak ah, ok.
14:21 * masak looks
14:23 not_gerd same for day 4. btw
14:27 masak cedrvint: nice post.
14:27 masak cedrvint++
14:28 cedrvint thanks :)
14:34 dalek perl6-roast-data: ca91a62 | coke++ | / (3 files):
14:34 dalek perl6-roast-data: today (automated commit)
14:34 dalek perl6-roast-data: review: https://github.com/coke/perl6​-roast-data/commit/ca91a62305
14:34 dalek perl6-roast-data: b8554ca | coke++ | / (4 files):
14:34 dalek perl6-roast-data: today (automated commit)
14:34 dalek perl6-roast-data: review: https://github.com/coke/perl6​-roast-data/commit/b8554ca78a
14:35 jnthn pmurias: If the tests pass with nqp-parrot, go right ahead
14:36 jnthn pmurias: Not sure if you can declare a contextual. But anyway, they are ordinary lexicals storage wise, it's just that we look them up dynamically. So no problems there.
14:39 pmurias jnthn: may I have a nqp commit bit?
14:40 smallfish joined #perl6
14:40 smallfish left #perl6
14:41 jnthn pmurias: done
14:41 pmurias jnthn: in perl6 lexicals must be declared with is dynamic to be looked up with OUTER::
14:41 jnthn pmurias: Is it "is context" these days?
14:42 jnthn pmurias: is it true for OUTER:: as well as CALLER:: and DYNAMIC::?
14:42 jnthn Rakudo doesn't enforce this just yet.
14:47 am0c joined #perl6
14:49 pmurias jnthn: is dynamic
14:50 pmurias jnthn: I think it's not required for OUTER::
14:51 jnthn ok
14:51 pmurias jnthn: the reason I'm concerned with that is that contextuals carry a performance hit on javascript
14:51 pmurias as I don't have direct control over the native lexical pads
14:54 xilo_ joined #perl6
14:57 not_gerd bye, #perl6
14:57 not_gerd left #perl6
14:57 geekosaur joined #perl6
15:02 felher Is there a way to test perl6.org on my machine?
15:02 moritz felher: cpanm App::mowyw; cd perl6.org; mowyw
15:02 felher moritz: thanks :)
15:03 moritz and then x-www-browser online/index.html
15:03 moritz or set up a local apache installation with perl6.org/online as a separate virtual host
15:04 moritz thre's also an app.psgi which I suppose also serves the files
15:09 PacoAir joined #perl6
15:12 GlitchMr moritz: perl6.org uses URLs with / at beginning
15:12 GlitchMr I usually use php -S localhost:8080 to test perl6.org
15:14 felher moritz: works, thnx
15:15 GlitchMr Surprisingly, PHP builtin server works well when testing static websites.
15:16 GlitchMr I think that httpd in busybox is fine to, but not sure
15:16 GlitchMr Like "busybox httpd -p 8080", but it doesn't have nice debugging info :-(
15:18 GlitchMr I think I've once used httpd in Busybox to serve webpages from Android
15:26 pmurias jnthn: does the order in which we intermix named and flat named arguments matter?
15:26 pmurias jnthn: as in foo(:a(1),|%b,:c(3),|%d)?
15:30 timotimo enjoying the prel6advent so far :)
15:30 masak \o/
15:33 GlitchMr Just wondering, there is no difference between List.roll and List.pick when called without arguments.
15:36 wamba joined #perl6
15:36 rurban joined #perl6
15:38 fgomez joined #perl6
15:45 felher I find the Advent Calendar to be a really nice thing for people just starting with perl6. Maybe it's special enough to promote it on the main perl6.org site while it is running? (Something like http://fasel.nerd2nerd.org/perl6.png maybe?)
15:46 felher GlitchMr: as far as i know, yes
15:48 GlitchMr felher: I was considering putting it in "Documentation" section during Perl 6 Advent Calendar
15:48 GlitchMr But it's already full of things
15:48 GlitchMr But perhaps
15:49 GlitchMr felher: I like it, but for me it looks like a banner, so not sure, really.
15:49 felher GlitchMr: yeah, i though about putting in doc, too. But as you said: it's already full of thinks and may to easily overlooked there.
15:50 felher GlitchMr: yeah. I really suck at designing :) It's mainly an attempt to get someone to do it right ;)
15:50 jnthn pmurias: Certainly shouldn't in NQP
15:51 jnthn pmurias: oh, wait...if there's duplicate ones...guess that gets interesting.
15:51 jnthn In the absence of dupes...no difference.
15:51 GlitchMr That's the problem with current Perl 6 design
15:51 masak GlitchMr: there's no difference between .roll(1) or .pick(1), no. (or between the argumentless forms which default to this)
15:51 GlitchMr It's full of links
15:51 GlitchMr But really, I don't think I could think of anything else, as I suck at designing
15:52 masak GlitchMr: pick is "selection without replacement", roll is "selection with replacement". but with only one trial, the difference isn't noticeable.
15:52 GlitchMr felher: also, you have really wide screen :-).
15:53 felher GlitchMr: right :)
15:54 GlitchMr Perhaps I could redesign the Perl 6 page, but that could be difficult.
15:56 felher I have no problem with the design of Perl6.org. I just wanted to place the Advent Calendar prominently :)
15:57 GlitchMr Except if I would redesign the site, it would look as bad as http://dl.dropbox.com/u/63913412/perl6lol.png
15:57 GlitchMr So I don't know
15:57 GlitchMr I'm not designer.
15:58 felher :D
15:58 SamuraiJack joined #perl6
15:58 GlitchMr But perhaps I could try to come with some design
15:58 GlitchMr It looks ugly for Poincare from GitHub
15:59 GlitchMr But I don't like the design Poincare proposed - http://ge.tt/4ich05L/v/0
16:00 GlitchMr Camelia will stay in my design, that's for sure. But as for other things...
16:00 GlitchMr (I know that my design would have 0% chances of accepting it without Camelia)
16:03 SHODAN just use sharper corners for the boxes, use less colors, make information and action boxes look the same and so on
16:03 SHODAN that's the new thing
16:04 felher Well, I won't get involved in any design issues. First of all my taste is quite different from most people and perl6.org should look like what most people like. Second of all I have dyschromatopsia, which makes me very bad at choosing colors :)
16:28 rurban joined #perl6
16:28 cognominal joined #perl6
16:35 moritz jnthn: https://gist.github.com/4189639 that's my attempt to fix nqp::can + role
16:35 moritz (backtrace further down the page)
16:36 moritz it seems to work fine with normal roles
16:38 moritz but with CachedCompiledRegex I run into bootstrappy issues
16:38 moritz currently it fails in dispatch:<.^>
16:39 moritz and I can change infix:<does> not to use that, then it runs into some other problem (.attributes not being there, or something)
16:40 moritz jnthn: is there a good way to apply the bootstrap chainsaw somewhere?
16:42 Ulti has anyone tried Rakudo* under Windows 8 yet? I'm unfortunate enough to be in a position to
16:43 Ulti heh well first up the .msi isnt signed so you have to click some tiny "more info" text in Windows8 to force the install :(
16:44 Ulti cool it runs ok though
16:45 GlitchMr Ok, so I've put perl6lol.png picture of "how bad this website would look if I would design it".
16:46 GlitchMr Now time for serious "how bad this website would look if I would design it" screenshot.
16:46 GlitchMr http://dl.dropbox.com/u/63913412/perl6lol2.png
16:46 GlitchMr I know, it looks bad
16:47 GlitchMr perl6lol.png was close enough ;-)
16:48 Ulti GlitchMr: are you making a new site?
16:48 GlitchMr I'm currently experimenting with trying to redesign it
16:48 GlitchMr But probably old website will stay
16:50 Ulti is it heresy to say that I like Sebastian Reidel's Perl6 logo? http://blog.kraih.com/a-logo-for-perl
16:51 dalek rakudo-js: 96abb65 | (Paweł Murias)++ | / (3 files):
16:51 dalek rakudo-js: [nqp] pass test 41
16:51 dalek rakudo-js: review: https://github.com/pmurias/​rakudo-js/commit/96abb65001
16:51 GlitchMr I like it too
16:51 GlitchMr But well, Larry Wall really wants Camelia
16:52 GlitchMr At least for now, Camelia has to stay.
16:52 Ulti even if you put Camelia ontop of the 6 I like the idea of the flight path being the 6
16:54 masak me too.
16:54 masak and sri's design looks more polished, indeed.
16:54 GlitchMr I really would like Sebastian Reidel's Perl 6 logo
16:54 masak though anyone who aims to improve on Camelia has to take into account that Camelia is explicitly *meant* not to look polished.
16:55 masak it's meant to look like it's made by a young person.
16:55 GlitchMr And I think it's used as cover of "Using Perl 6" book.
16:55 Ulti well I didnt want to trample on camelia ;) but yeah amelia looks a bit more hipster-cool
16:55 GlitchMr But, I think that Camelia is there to stay.
16:55 Ulti rather than fisher price bold
16:55 Ulti camelia will certainly get children into I guess
16:56 Ulti *in too
16:57 GlitchMr I really would like to use Riedel's logo.
16:57 masak GlitchMr: by all means, use it all you like.
16:57 masak you may want to ask sri, but apart from that... go right ahead.
16:57 GlitchMr But I think it isn't negotable.
16:58 masak *you* using it doesn't have to be negotiable.
16:58 masak replacing Camelia as official mascot... is possible, but you'd have to fulfill a rather long list of criteria.
16:58 masak see the end of https://raw.github.com/perl6​/mu/master/misc/camelia.txt
16:59 GlitchMr I think that Riedel's logo fits all of those.
16:59 GlitchMr And it's already used as unofficial logo, even in the cover of "Using Perl 6" book
17:00 __sri you can use the png version of amelia that has been released under CC-SA
17:01 __sri i never actually released a svg version since the interest was so low
17:01 moritz __sri: I think you provided me the SVG for the book cover
17:01 moritz (though that's not the same as a public release)
17:01 * __sri doesn't remember
17:01 masak __sri: I dunno about low interest. people have been pretty receptive wrt your redesign.
17:02 __sri but yea, i guess the book cover version should have a good license too
17:02 masak all I ever hear are good things about it.
17:02 masak you might be confusing "low interest" with "small community" :)
17:02 __sri perhaps :)
17:02 edsiper joined #perl6
17:03 edsiper what means if($if =~ "something") ?
17:03 timotimo call the function "if" with a match result, no?
17:03 edsiper whats the difference between == and =~ ?
17:04 timotimo hold on, isn't =~ like "use a regular expression to transform the name on the left"?
17:04 hash_table joined #perl6
17:04 masak edsiper: =~ is not valid in Perl 6.
17:04 timotimo r: my $a = "hello world"; $a =~ s/world/blob/; say $a;
17:04 p6eval rakudo 5481c6: OUTPUT«===SORRY!===␤Unsupported use of =~ to do pattern matching;  in Perl 6 please use ~~␤at /tmp/rb7Ryqwb0l:1␤»
17:04 masak std: 2 =~ 2
17:04 p6eval std a8bc48f: OUTPUT«[31m===[0mSORRY![31m===[0m�Unsupported use of =~ to do pattern matching; in Perl 6 please use ~~ at /tmp/xT70gXtI60 line 1:�------> [32m2 =~[33m�[31m 2[0m�Parse failed�FAILED 00:00 41m�»
17:04 timotimo r: my $a = "hello world"; $a ~~ s/world/blob/; say $a;
17:04 p6eval rakudo 5481c6: OUTPUT«hello blob␤»
17:04 __sri moritz: did i really give you an svg? i don't remember fixing the image for vector export :o
17:04 masak edsiper: the channel is #perl6. note the '6' at the end.
17:06 edsiper thanks timotimo
17:06 timotimo r: sub if($a, &blk) { &blk() if $a }; if("hello" ~~ /ll/, { say "whoa!" });
17:06 p6eval rakudo 5481c6: OUTPUT«whoa!␤»
17:06 GlitchMr http://dl.dropbox.com/u/63​913412/itsactuallygood.png
17:07 __sri eeeeep
17:07 GlitchMr I've replaced (as a test) Larry's Camelia with Riedel's logo. It's good :-).
17:07 spider-mario joined #perl6
17:07 * __sri hates the purple one
17:07 timotimo i think the page looks not minimal enough to be minimal-goodlooking, but not fleshed out enough to look good otherwise
17:08 GlitchMr I'm still working on it
17:08 timotimo that would explain it :)
17:08 __sri page needs good typography
17:08 GlitchMr Yeah
17:08 timotimo the current perl6 page looks a bit strange with the out-of-place looking download button i find
17:08 GlitchMr Perl 6 is sans-serif, so rest of page should be sans-serif too
17:08 au Source Sans Pro ?
17:08 timotimo is the page you're working on going to replace that page?
17:08 au # https://typekit.com/fonts/source-sans-pro
17:08 __sri Helvetica Neue imo
17:09 __sri Source Sans Pro is not bad either
17:10 masak "Brawny gods just flocked up to quiz and vex him" :)
17:12 GlitchMr http://dl.dropbox.com/u/63913412/sansserif.png
17:12 au "The five boxing wizards jump quickly. --More--" #nethack
17:12 GlitchMr It looks nicer with sans-serif font.
17:12 au that's Ubuntu?
17:12 GlitchMr Well, yes
17:12 GlitchMr I haven't yet specified what font exactly
17:13 GlitchMr oh wait, not
17:13 GlitchMr no*
17:13 GlitchMr It's DejaVu Sans.
17:13 au aha.
17:14 moritz __sri: I have a usingperl6.svg, which is the whole cover including title and amelia
17:14 * flussence suggests Oxygen Sans, just to be different
17:16 au oxygen++ # just two weight-levels, but looks nice
17:17 GlitchMr Well, actually I'm not designer. I just wanted to see how nicely perl6.org would look if I would "design" it.
17:18 masak keep up the good work.
17:19 jnthn moritz: I didn't mean to write a declares_method
17:19 GlitchMr But i start to like my design
17:19 __sri think i'll kill the deep purple version from the blog post... seeing the unfinished image used over and over makes me feel bad
17:19 jnthn moritz: I meant to nqp::can($!pun, $name)
17:20 jnthn moritz: The reason you get the error is because the role's method list doesn't have dispatch:<.^>
17:20 jnthn moritz: It comes from Mu via Any, the default base class for the role.
17:20 jnthn If you do the check in $!pun then you (a) have less code, and (b) it should work
17:20 jnthn moritz: Sorry for unclear explanation of what to do.
17:22 GlitchMr Still, I think that my project for perl6.org is better than https://www.djangoproject.com/...
17:22 GlitchMr But I think that djangoproject.com is one of ugliest webpages on the internet
17:22 GlitchMr (not counting Larry Wall's webpage)
17:24 birdwindupbird joined #perl6
17:28 masak GlitchMr: I see honesty is not one of the issues you're struggling with. :)
17:28 masak s/one of the issues/an issue/
17:31 rurban_ joined #perl6
17:34 GlitchMr http://dl.dropbox.com/u/63913412/community.png
17:37 masak GlitchMr: the page feels much less "alive" than the one we have now. if you see what I mean.
17:38 GlitchMr hmmm, yeah
17:38 masak it feels almost one-dimensional compared to the boxes on the current perl6.org page.
17:38 GlitchMr that's good argument :-)
17:38 masak I think there is some wisdom in presenting options to people in big noticeable boxes.
17:39 masak you're presenting them as headings. might as well send them a Word document.
17:39 GlitchMr Also http://dl.dropbox.com/u/63913412/mobile1.png and http://dl.dropbox.com/u/63913412/mobile2.png
17:40 GlitchMr But well, it doesn't matter much
17:40 GlitchMr (well, most of users will use desktop computer)
17:40 GlitchMr s)
17:40 masak GlitchMr: compare, for example, github.com when you're signed out. big blue "Plans, Pricing and Signup" button front and center. below that, four nice *areas* side-by-side about good things about GitHub.
17:41 masak or take http://37signals.com/ as an example. also very two-dimensional, with things side-by-side.
17:41 masak that just seems to be good use of space on a page, and makes the page more interesting.
17:41 GlitchMr Designing is difficult, I know
17:41 masak yes, but that was not my point. :)
17:42 masak I don't know how these two examples look on a mobile browser. would be interesting to find out.
17:43 GlitchMr Well, my mobile phone has very small screen size
17:45 __sri http://blog.kraih.com/a-logo-for-perl # the deep purple butterfly has been replaced
17:45 __sri GlitchMr: please use the orange one :)
17:45 GlitchMr Thanks
17:45 GlitchMr Going to change it now
17:46 au for small mobile screens, starting from http://foundation.zurb.com/ or http://twitter.github.com/bootstrap/ could be easier than hand-coding responsive styles
17:46 birdwindupbird pwd
17:48 GlitchMr http://dl.dropbox.com/u/63913412/yellow.png
17:49 GlitchMr I'll start a new branch when every page will be ported to this layout
17:51 masak birdwindupbird: /home/birdwindupbird
17:55 * moritz agrees
17:56 moritz erm, replying to stale chat :(
17:56 masak :)
17:57 masak irssi should be more conspicuous about that scenario.
17:58 FROGGS does somebody know a tutorial about grammars or so? I'd like to parse the module strings: https://gist.github.com/4190173
17:59 moritz FROGGS: the Using Perl 6 book has a chapter on regexes and one on grammars
18:01 FROGGS moritz: thanks, will read now ;o)
18:02 japhb joined #perl6
18:02 japhb Uggh.
18:02 phenny japhb: 02:17Z <sorear> tell japhb I have a theory... how does $acc ~= ($x ~ $y); do in rakudo, where $x and $y are loop-invariant but not "constant"s?
18:02 phenny japhb: 07:47Z <sorear> tell japhb 'for 1..20 { }' before: 757.56 ± 26.546 µs now: 53.345 ± 1.8765 µs
18:02 * masak submitted a bug report to Github about a corner case when editing gists :)
18:02 phenny japhb: 07:51Z <sorear> tell japhb I'm shelving this for now but there's still much work to be done. [ loop (my $i = 1; $i <= 20; $i++) { } ] = 10.730 ± 0.0305 µs; [ my $i = -20; Nil while ++$i ] = 4.1028 ± 0.1603 µs
18:02 phenny japhb: 07:58Z <azawawi> tell japhb i'll send pull requests with any performance bottlenecks I encounter. Thanks for making and pointing out your cool perl6-bench project. I will check it out later today.
18:02 japhb Woah, clearly I have some backlogging to do.
18:03 * japhb is digging out of home network unhappiness because of power outage that outlasted UPS.
18:04 FROGGS japhb: welcome back then
18:05 japhb Hi FROGGS!  How are things in your part of the world?
18:05 FROGGS japhb: well, good actually
18:05 japhb That's good to hear.
18:05 FROGGS japhb: have two weeks off now, trying to parse the examples in S11/versioning
18:06 japhb Ah nice!
18:06 FROGGS after that, I want to provide a "test suite" which somebody can test her/his implementation
18:06 japhb Any big ideas since I disappeared?
18:06 FROGGS (and I want to start my own implementation)
18:06 japhb Good ideas, both.
18:07 FROGGS some rumor, but nothing concrete
18:07 FROGGS it is always the same back and forth
18:07 japhb Do tell ...
18:08 japhb .oO( Judging by my repo pulls, it's been an active 2-3 days.  Especially Perlito5, wow. )
18:09 dalek rakudo/nom: e2f2dc4 | moritz++ | src/Perl6/Metamodel/RolePunning.pm:
18:09 dalek rakudo/nom: fix nqp::can(role { }, "anything"). jnthn++
18:09 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/e2f2dc436d
18:11 FROGGS I've forgotten who was it last night, but it was almost the same viewpoint like yours, but he didnt cared <auth>, just <ver>, IIRC
18:12 FROGGS and there is another thing we have to care: there if a from:<> tag that allows one to import Perl 5 stuff for example
18:13 FROGGS don't know how this could work, I guess we just care about auth+ver and storage right now
18:13 japhb Yeah, but then I'm not sure any of the other stuff matters.  With :from<> I think we essentially have to pass a name string and a capture to the proxy for the foreign module system
18:15 FROGGS well, Perl 5 modules have a version, but as long as you use Perl 5's cpan clients to install these modules, you dont have an auth
18:16 FROGGS IMO you would have to use panda for this stuff too, so that these modules get installed in a nown path
18:16 japhb Well, Perl 5 module version handling is done with magic; you would generally have to just pass that through to the proxy loader.
18:16 FROGGS and that XS modules are compiled the "right way"
18:17 japhb (not technically magic in the perl5 guts sense, I mean magic as in "funky logic involving prestidigitation and gnomes")
18:18 japhb Perl 5 modules should be treated as Perl 5 modules; not Perl 5 modules that got eaten by the Perl 6 beast.
18:27 tadzik good evening #perl6
18:28 masak tadzik! \o/
18:28 tadzik \o/
18:30 FROGGS hi tadzik
18:32 japhb tadzik, masak, hi!
18:39 rindolf joined #perl6
18:44 japhb phenny, ask azawawi Would you mind using a different pastebin than the one at paste.debian.net (or configuring it to keep the pastes for longer), because they are expiring quite quickly (sometimes faster than I can backlog if I've been AFK for a couple days)?
18:44 phenny japhb: I'll pass that on when azawawi is around.
18:47 ifim joined #perl6
18:48 GlitchMr I've done an experiment about parsing archives: http://dl.dropbox.com/u/63913412/parsearchives.png
18:48 GlitchMr Currently, they aren't feeling like part of the website.
18:49 masak gists are so superior as a pastebin, I don't see why more aren't using them.
18:50 GlitchMr Perhaps s/This file/This page/...
18:50 GlitchMr "file" is implementation detail.
18:50 japhb masak, I very much agree
18:50 GlitchMr masak: I agree too
18:52 tadzik GlitchMr: this is very nice
18:52 wooden joined #perl6
18:52 wooden joined #perl6
18:57 GlitchMr ok, soon I will upload the new design branch
19:12 moritz 32 test files failing in the qast-sink-1 branch
19:12 moritz of 717 files in total
19:13 sorear good * #perl6
19:13 moritz \o sorear
19:13 sorear au!
19:14 sorear o/ moritz
19:15 GlitchMr ok... here goes
19:15 dalek perl6.org/new-design: 173c755 | GlitchMr++ | / (613 files):
19:15 dalek perl6.org/new-design: New Perl 6 layout
19:15 dalek perl6.org/new-design: review: https://github.com/perl6/p​erl6.org/commit/173c75510b
19:22 dalek perl6.org/new-design: 7d2139d | GlitchMr++ | source/style.css:
19:22 dalek perl6.org/new-design: Add updated styles I forgot previous time.
19:22 dalek perl6.org/new-design: review: https://github.com/perl6/p​erl6.org/commit/7d2139d738
19:22 au o/ sorear
19:23 masak sorear! \o/
19:23 GlitchMr hi, sorear
19:25 sorear japhb: ping
19:25 sorear o/ masak, GlitchMr
19:26 sorear japhb: when you have the tuits, I'd appreciate an update on standing now that for 1..20 {} is 15 times faster; also smaller gains for ~
19:28 moritz r: my $x := 1; say do while $x { $x-- }
19:28 p6eval rakudo e2f2dc: OUTPUT«Cannot assign to a non-container␤  in sub postfix:<--> at src/gen/CORE.setting:3082␤  in code  at /tmp/G6qMBsShlJ:1␤  in block  at /tmp/G6qMBsShlJ:1␤␤»
19:28 moritz r: my $x = 1; say do while $x { $x-- }
19:28 p6eval rakudo e2f2dc: OUTPUT«0␤»
19:31 GlitchMr http://dl.dropbox.com/u/63913412/randomcode.png
19:32 GlitchMr Now main page has random example code
19:32 moritz I hope it doesn't rely on javascript
19:32 GlitchMr Nope
19:32 GlitchMr Actually it already exists
19:32 moritz \o/
19:32 GlitchMr But
19:32 GlitchMr It was unused
19:32 moritz i know
19:32 moritz I implemented it, and masak promised to incorporate it
19:32 GlitchMr http://perl6.org/TEMP-example.html
19:34 dalek perl6.org/new-design: 0df6ae4 | GlitchMr++ | source/ (3 files):
19:34 dalek perl6.org/new-design: Random code
19:34 dalek perl6.org/new-design: review: https://github.com/perl6/p​erl6.org/commit/0df6ae4516
19:37 RandalSchwartz joined #perl6
19:37 RandalSchwartz joined #perl6
19:46 [Coke] GlitchMr: is your new design checked out somewhere we can poke at?
19:47 GlitchMr Not really
19:47 GlitchMr If you want, download new-design branch
19:47 GlitchMr But I haven't set site (aside of localhost) to check it
19:47 GlitchMr But, ok, I will put it on my feather account if you want
19:48 GlitchMr or not
19:48 GlitchMr it depends on being on root of the website :-(...
19:48 [Coke] eh. I'm lazy. easy to click on a URL.
19:48 [Coke] ah. no worries, then. I'm actually working for $DAYJOB atm anyway. :P
19:50 dalek perl6.org/new-design: 555c1b7 | GlitchMr++ | source/archive/ (78 files):
19:50 dalek perl6.org/new-design: Fix Unicode
19:50 dalek perl6.org/new-design: review: https://github.com/perl6/p​erl6.org/commit/555c1b702b
19:51 [Coke] GlitchMr: looks to me like you broke unicode.
19:51 [Coke] (clicking on that link and looking on github.
19:52 GlitchMr It's because GitHub assumes ISO-8859-1 by default
19:52 GlitchMr (or something)
19:53 GlitchMr Or rather, the diff doesn't contain valid UTF-8 (because it contains removed ISO-8859-1), so GitHub assumes ISO-8859-1.
19:53 [Coke] hokay.
19:53 GlitchMr If you will view any of those files directly, everything will be correct
19:58 dalek perl6.org/new-design: 88851a9 | GlitchMr++ | / (2 files):
19:58 dalek perl6.org/new-design: Restore online directory
19:58 dalek perl6.org/new-design: review: https://github.com/perl6/p​erl6.org/commit/88851a9e32
20:11 moritz r: say (1, 2).map(*+1).WHAT
20:11 p6eval rakudo e2f2dc: OUTPUT«List()␤»
20:12 pmurias re redesign of perl6 wouldn't just replacing the ugly download button make it look much better?
20:12 masak "...is not delivering the quantity of hedons per second I have come to expect..." -- today's measurement unit: hedons per second. http://hpmor.com/notes/progress-12-12-0/
20:13 masak pmurias: that may help a lot, yes.
20:13 dalek perl6.org/new-design: da496c9 | GlitchMr++ | source/archive/ (18 files):
20:13 dalek perl6.org/new-design: More cleaning
20:13 dalek perl6.org/new-design: review: https://github.com/perl6/p​erl6.org/commit/da496c91b4
20:14 masak though I must say that reading Charles Stross's work tends to be fun in the way lesswrong feels he's missing.
20:15 masak ditto Neal Stephenson.
20:22 kaare_ joined #perl6
20:22 [Coke] I would be happy with this button: http://twitter.github.com/boo​tstrap/base-css.html#buttons
20:22 dalek perl6.org/new-design: 43b14a4 | GlitchMr++ | source/archive/rfc/183.html:
20:22 dalek perl6.org/new-design: Fix RFC 183 title
20:22 dalek perl6.org/new-design: review: https://github.com/perl6/p​erl6.org/commit/43b14a490c
20:23 dalek perl6.org/new-design: f400a18 | GlitchMr++ | source/archive/rfc/183.html:
20:23 dalek perl6.org/new-design: Fix RFC 183 title again
20:23 dalek perl6.org/new-design: review: https://github.com/perl6/p​erl6.org/commit/f400a18cf9
20:23 moritz phenny: de en "lispeln"?
20:23 phenny moritz: "lisp" (de to en, translate.google.com)
20:24 moritz somehow that meaning of "lisp" was never obvious to me :-)
20:24 diakopter GlitchMr: I think we need to keep the spokesbug as TimToady rendered it
20:26 moritz I don't think so
20:27 moritz TimToady has mentioned that it was intended to decolorize a bit
20:27 diakopter I'm pretty sure he meant on pages other than perl6.org
20:28 moritz let's just ask him, shall we?
20:28 diakopter :)
20:28 GlitchMr Well, I can restore Camelia if Larry Wall won't agree.
20:29 GlitchMr It is just a branch currently.
20:29 GlitchMr It's not actual perl6.org, just a design I'm making.
20:32 GlitchMr Personally, I would like to see Riedel's logo. Larry Wall's logo may be cute, but I think it could hurt "first-impression".
20:33 GlitchMr People even think it's a bee - https://en.wikipedia.org/wiki/Talk:Perl_6#Camelia.3F
20:37 Chillance joined #perl6
20:39 masak "people even think it's a bee" seems a tame argument next to the things detractors usually say about Camelia.
20:39 dalek perl6.org/new-design: 8089137 | GlitchMr++ | source/favicon.ico:
20:39 dalek perl6.org/new-design: New favicon
20:39 dalek perl6.org/new-design: review: https://github.com/perl6/p​erl6.org/commit/8089137a5c
20:39 moritz it's a "somebody is WRONG on the internet" kind of argument :-)
20:40 tadzik (:
20:40 GlitchMr http://xkcd.com/386/
20:40 diakopter masak: follow the bee link (if you haven't); it's not just about "bee"
20:41 masak diakopter: url?
20:41 diakopter the one GlitchMr mentioned on wikipedia
20:41 masak or do you simply mean the talk page? I'm reading it.
20:41 masak "If you worry that Perl isn't wearing a tie, you're not the person Perl is trying to impress." -- from that talk page.
20:41 rurban joined #perl6
20:47 diakopter masak: yes, the Camelia section of the talk page
20:48 masak it seems evenly divided between people for and people against.
20:48 masak much like most of the Camelia discourse in other places.
20:57 Chillance joined #perl6
20:59 moritz more fun from them sink department:
20:59 moritz %a = 1, 2;
20:59 moritz compiles to  %a.STORE(1, 2)
20:59 GlitchMr Please don't merge my new layout yet as main page isn't ready yet ;-). Also, bye.
21:00 moritz so that's a methodcall, and thus currently we call .sink on the result
21:00 moritz erm, not a problem for hashes
21:00 moritz but a problem for lists
21:12 snearch joined #perl6
21:13 jnthn moritz: Hm, ouch...
21:13 jnthn moritz: Special-casing the name STORE is one option... :0
21:13 jnthn *:)
21:13 japhb sorear, sorry, I've got a lot going on today.  As soon as I can spare a few minutes, I'll toss them in the direction of perl6-bench.  :-)
21:13 japhb afk again &
21:14 moritz jnthn: yes, I'm just introducing $past<nosink>
21:14 dalek rakudo/qast-sink-1: 1c84805 | moritz++ | src/Perl6/Metamodel/RolePunning.pm:
21:14 dalek rakudo/qast-sink-1: fix nqp::can(role { }, "anything"). jnthn++
21:14 dalek rakudo/qast-sink-1: review: https://github.com/rakudo/rakudo/commit/1c8480581f
21:14 dalek rakudo/qast-sink-1: 17bcd96 | moritz++ | src/core/Parameter.pm:
21:14 dalek rakudo/qast-sink-1: switch to real constants in Parameter.pm
21:14 dalek rakudo/qast-sink-1: review: https://github.com/rakudo/rakudo/commit/17bcd96493
21:14 dalek rakudo/qast-sink-1: 36e4477 | moritz++ | src/Perl6/Actions.pm:
21:14 dalek rakudo/qast-sink-1: do not sink list assignment
21:14 dalek rakudo/qast-sink-1: review: https://github.com/rakudo/rakudo/commit/36e4477e7c
21:17 moritz ... and now everything segfaults
21:17 moritz that is, ./perl6 -e '' segfaults
21:18 * moritz calls it a day
21:18 jnthn moritz++ # sink hacking
21:18 jnthn moritz: I should have tuits to help out with it a bit this week, if you like.
21:19 moritz jnthn: that would be greatly appeciated
21:32 rindolf Hi all.
21:36 lue hello o/
21:36 phenny lue: 09:39Z <jnthn> tell lue Well, they are installed in some %?LANG but they will suffer the "not under Mu" issue. And I need to do some MOP stuff before turning to that.
21:36 phenny lue: 09:40Z <jnthn> tell lue but I think at a first guess $~MAIN would desugar to %?LANG<MAIN>
21:37 lue Ah. I had a feeling it was more complicated than "Oh, just return the hash value from Actions.pm" or something like that :) .
21:40 Kharec joined #perl6
21:43 lue the logo on kraih.com (linked from the talk page) feels too corporate for me :P
21:43 sevin joined #perl6
21:46 * __sri starts working on a version of the amelia logo with suit and briefcase
21:48 lue .oO(It's a terrific logo, but it just doesn't feel right for P6.)
21:56 bruges_ joined #perl6
21:57 skids joined #perl6
21:59 kurahaupo joined #perl6
22:01 cognominal joined #perl6
22:34 __sri amelia with pinstripe wings flying out of a big briefcase is actually not a bad logo for a perl 6 enterprise edition :)
22:38 colomon joined #perl6
22:45 rurban joined #perl6
23:20 cognominal_ joined #perl6
23:34 cognominal_ joined #perl6
23:38 sevin joined #perl6
23:46 dalek rakudo-js: cbdb432 | (Paweł Murias)++ | / (3 files):
23:46 dalek rakudo-js: [nqp] pass test 34
23:46 dalek rakudo-js: review: https://github.com/pmurias/​rakudo-js/commit/cbdb432684
23:52 popl joined #perl6

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

Perl 6 | Reference Documentation | Rakudo