Perl 6 - the future is here, just unevenly distributed

IRC log for #perl6, 2010-10-11

Perl 6 | Reference Documentation | Rakudo

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

All times shown according to UTC.

Time Nick Message
00:02 am0c^ joined #perl6
00:02 sorear A language like C++, because of its weak metamodel, only allows a single class per method, and thus a method can appear at most once in a dispatch list and (__function__, typeid(*this)) uniquely identifies the "next" method
00:02 sorear Perl 6 doesn't have that, so it seems like I'd have to allocate storage for the nextsame iterator on every method call
00:10 Aleph_ is now known as Dunearhp
00:20 qw3rty joined #perl6
00:37 risou joined #perl6
00:40 mfollett Not that it really matters, but out of curiosity why is '.?' and such an operator instead of '.' being an operator and '?' being a metaoperator that operates on '.'?
00:45 qw3rty left #perl6
00:45 redicaps joined #perl6
00:46 redicaps left #perl6
01:02 pythonian4000 is now known as pythonian4000afk
01:06 QinGW joined #perl6
01:16 meppl left #perl6
01:25 masonkramer left #perl6
01:25 masonkramer joined #perl6
01:38 ggoebel left #perl6
01:44 ggoebel joined #perl6
01:46 ggoebel left #perl6
01:52 mfollett left #perl6
01:52 mfollett joined #perl6
01:55 silug joined #perl6
01:55 mfollett Has try.rakudo.org always had the chapters?
02:01 Eevee joined #perl6
02:09 meppl joined #perl6
02:11 dukeleto mfollett: probably for performance reasons, concerning .?
02:11 sorear mfollett: because . isn't a binary operator
02:12 sorear mfollett: a binary operator takes two arguments, both of which can logically stand on their own
02:12 sorear in $foo.methodname, $foo can stand on its own, but methodname can't
02:12 mfollett sorear:  I didn't know that a metaoperator could only operate on binary operators
02:12 sorear . is syntactically a postfix operator, because only the left side can stand on its own
02:13 mfollett sorear:  hyperoperator is a meta operator and can operate on the method invocation operator
02:13 sorear so metaoperators always attach like $foo>>.method
02:13 sorear never like $foo.>>method
02:13 mfollett sorear:  oh, I see
02:14 sorear as for why $foo*.method isn't used, I can only ask - what would $foo*++ mean?
02:14 sorear or any non-method operator
02:16 mfollett sorear:  I guess it would depend on how many ways $foo could do postfix<++>($x)
02:18 mfollett sorear:  I thought $foo++ was the same as $foo.++
02:18 mfollett sorear:  or, wait, I may have just confused mysel
02:18 mfollett f
02:31 aesop left #perl6
02:35 aesop joined #perl6
02:37 risou left #perl6
02:37 risou_ joined #perl6
02:38 risou_ left #perl6
02:39 risou joined #perl6
02:40 pythonian4000afk is now known as pythonian4000
02:42 pythonian4000 is now known as pythonian4000afk
02:44 agentzh joined #perl6
02:58 eternaleye joined #perl6
02:58 Atle joined #perl6
02:58 eternaleye left #perl6
02:58 eternaleye joined #perl6
03:26 pythonian4000afk is now known as pythonian4000
03:35 meppel joined #perl6
03:36 meppel left #perl6
03:37 risou_ joined #perl6
03:39 meppl left #perl6
03:39 risou left #perl6
03:42 meppl joined #perl6
03:53 tadzik left #perl6
03:54 Atle left #perl6
03:54 satyavvd joined #perl6
04:08 rainy_day joined #perl6
04:21 f00li5h is now known as f00li5h[HireMe]
04:25 revdiablo left #perl6
04:25 revdiablo joined #perl6
04:25 revdiablo left #perl6
04:25 revdiablo joined #perl6
04:27 QinGW1 joined #perl6
04:29 QinGW left #perl6
04:36 ruoso left #perl6
04:37 risou joined #perl6
04:39 risou_ left #perl6
04:54 plobsing left #perl6
05:00 justatheory joined #perl6
05:02 dalek niecza/master: f65d43c | sorear++ | lib/Kernel.cs:
05:02 dalek niecza/master: Keep the full dispatch list during method calls
05:02 dalek niecza/master: review: http://github.com/sorear/niecza/commit/f65d43c4d1f8b339efd8a69e7551e12a399bb0c6
05:02 dalek niecza/master: c6bd89e | sorear++ | / (2 files):
05:02 dalek niecza/master: Implement nextsame
05:02 dalek niecza/master: review: http://github.com/sorear/niecza/commit/c6bd89e5da855f112acf50b156a725d9495258c6
05:15 Chillance left #perl6
05:22 Mowah joined #perl6
05:23 justatheory left #perl6
05:34 envi joined #perl6
05:36 sorear How does automatic generation of proto methods work?
05:38 kaare joined #perl6
05:39 kaare is now known as Guest85408
05:39 rainy_day left #perl6
05:39 Guest85408 is now known as kaare_
05:40 rainy_day joined #perl6
05:40 rainy_day left #perl6
05:49 dju left #perl6
05:51 sftp left #perl6
05:52 sftp joined #perl6
05:53 envi left #perl6
05:56 envi joined #perl6
06:04 mfollett left #perl6
06:08 sftp left #perl6
06:09 Raynes left #perl6
06:13 risou left #perl6
06:13 risou_ joined #perl6
06:14 uniejo joined #perl6
06:20 sftp joined #perl6
06:23 jjore left #perl6
06:28 sftp_ joined #perl6
06:29 sftp left #perl6
06:36 cosimo left #perl6
06:38 jjore joined #perl6
06:38 risou joined #perl6
06:40 nymacro left #perl6
06:40 risou_ left #perl6
06:43 jjore left #perl6
06:45 jjore joined #perl6
06:49 nymacro joined #perl6
06:54 mathw phenny: tell masak your presentation looks awesome, but what software did you use to actually present it? OS X's Preview?
06:54 phenny mathw: I'll pass that on when masak is around.
06:56 baux joined #perl6
06:58 cosimo joined #perl6
07:04 cosimo left #perl6
07:12 nym joined #perl6
07:14 araujo joined #perl6
07:14 araujo left #perl6
07:14 araujo joined #perl6
07:14 Axius joined #perl6
07:15 nymacro left #perl6
07:16 cosimo joined #perl6
07:20 dalek niecza/master: 087263f | sorear++ | lib/Kernel.cs:
07:20 dalek niecza/master: Some low-level multi support
07:20 dalek niecza/master: review: http://github.com/sorear/niecza/commit/087263f7228e8429e7f392401e895dd3731e7b52
07:20 dalek niecza/master: 6393299 | sorear++ | / (3 files):
07:20 dalek niecza/master: Implement slurpy hashes
07:20 dalek niecza/master: review: http://github.com/sorear/niecza/commit/639329998b31decad70c8782ad2d92ba48a03fd4
07:21 cosimo left #perl6
07:21 cosimo joined #perl6
07:26 Axius left #perl6
07:43 daxim joined #perl6
07:45 QinGW joined #perl6
07:46 QinGW1 left #perl6
07:50 timbunce joined #perl6
08:04 risou left #perl6
08:04 risou_ joined #perl6
08:04 pythonian4000 is now known as pythonian4000afk
08:31 dakkar joined #perl6
08:36 xinming_ is now known as xinming
08:43 phenny left #perl6
08:43 IllvilJa joined #perl6
08:44 sorear seen tylercurtis
08:44 aloha tylercurtis was last seen in #perl6 2 days 19 hours ago joining the channel.
08:45 phenny joined #perl6
08:52 araujo left #perl6
08:56 jlaire left #perl6
08:58 jlaire joined #perl6
08:58 daxim left #perl6
09:00 dalek niecza/master: a668022 | sorear++ | / (3 files):
09:00 dalek niecza/master: Start on the documentation rewrite
09:01 dalek niecza/master: review: http://github.com/sorear/niecza/commit/a66802273d9100135dc57d82ad1e70a0364001e8
09:02 daxim joined #perl6
09:28 araujo joined #perl6
09:47 ggoebel joined #perl6
10:05 QinGW left #perl6
10:47 wamba joined #perl6
11:01 zby_home_ joined #perl6
11:08 ggoebel left #perl6
11:08 Italian_Plumber joined #perl6
11:09 ggoebel joined #perl6
11:35 satyavvd left #perl6
11:39 flussence whoa, who made my test run 30 minutes shorter today?
11:39 moritz_ flussence: I've removed one test in which a dies_ok only succeeded by running into the recursion limit
11:42 flussence oh, the arrays thing?
11:42 moritz_ typed arrays, yes
11:53 nadim left #perl6
11:55 moritz_ Use of uninitialized value in numeric context  in 'MapIter::reify' at line 1
11:55 moritz_ from the spectest output
11:55 moritz_ S03-metaops/hyper.rakudo
11:55 moritz_ seems to point to a rakudobug, if you ask me
11:55 moritz_ (even if the warning is genuine, it shouldn't come from line 1)
11:56 mathw whoa, haskell interpreter written in Javascript
12:16 masak joined #perl6
12:16 masak \o/
12:16 phenny masak: 06:54Z <mathw> tell masak your presentation looks awesome, but what software did you use to actually present it? OS X's Preview?
12:18 MayDaniel joined #perl6
12:18 masak mathw: yes. with 'open *.pdf', Preview easily allows me to jump between individual slides without needing to merge them into one PDF.
12:19 masak and thanks. :) all the nice feed back on the presentation system makes me want to use it again, soon.
12:19 * masak backlogs
12:26 nadim joined #perl6
12:30 smash left #perl6
12:33 nym left #perl6
12:36 MayDaniel left #perl6
12:38 timbunce left #perl6
12:41 risou_ left #perl6
12:42 masak rakudo: my @*a; subset A of Int where { @*a.push("A") }; subset B of A where { @*a.push("B") }; 42 ~~ B; say @*a # discovered and RT'd yesterday
12:42 p6eval rakudo 5bdbd4: OUTPUT«BB␤»
12:43 masak jnthn described it as "a place where some cheating catches up with us".
12:44 moritz_ well, I think the YOU_ARE_HERE thing in rakudo needs some work
12:44 moritz_ masak: btw I'm nearly sure that the '&say not found in where-block' is a dupe
12:45 moritz_ masak: there's alread a bug report for not being able to use scalar lexicals in where-blocks
12:46 masak moritz_: thanks. I'll have a look andthen merge. :)
12:48 risou joined #perl6
12:50 masak indeed. this seems to be the new ticket's old twin: http://rt.perl.org/rt3/Ticket/Display.html?id=77356
12:50 * masak merges
12:51 masak moritz_++
12:51 moritz_ masak++ # finding the ticket
12:52 masak RT isn't so bad once one bends with it rather than against it. :)
12:53 moritz_ masak: re RT #78322, is the order actually specced?
12:53 masak moritz_: don't think so; but I convinced jnthn that only one order makes sense.
12:53 masak narrower subset types "presuppose" the constraints of wider ones.
12:54 masak very much like binary boolean ops && || allows one to "presuppose" their lhs (or their negation, as it were)
12:55 masak s/as it were/depending/ :)
12:55 moritz_ fwiw, Eiffel has two 'and' operators, one short circuiting, one where the order of execution is undefined
12:56 masak a bit like ?&, then :)
12:56 masak (no pun intended)
12:58 PerlPilot is now known as PerlJam
13:08 Trashlord left #perl6
13:11 Trashlord joined #perl6
13:13 Axius joined #perl6
13:16 nymacro joined #perl6
13:18 snearch joined #perl6
13:19 Patterner left #perl6
13:20 Psyche^ joined #perl6
13:20 Psyche^ is now known as Patterner
13:20 masak yesterday I told jnthn about my plans to put Yapsi on an STD.pm6 foundation, and how I still haven't solved the problem of getting the STD.pm6 output into Perl 6 rapidly and reliably. he suggested that I use Rakudo's Perl 6 grammar instead. now I can't remember what I thought the advantage of using STD.pm6 would be. :)
13:21 moritz_ better error messages
13:21 moritz_ parsing more things than rakudo (like heredocs)
13:21 moritz_ closer to the spec :-)
13:22 nadim left #perl6
13:22 nadim joined #perl6
13:22 moritz_ rakudo: say Perl6::Grammar.parse('1').WHAT
13:22 p6eval rakudo 5bdbd4: OUTPUT«Regex::Match()␤»
13:22 wamba left #perl6
13:22 wamba joined #perl6
13:23 mathw masak: What I really liked about your presentation system is that it has this wonderful aesthetic idea to it. I didn't have time to look at it properly though - what's the source file? Inkscape SVG marked up with metadata?
13:23 moritz_ I don't know how to get Match objects out of it, and dealing with Regex::Match is a pain
13:24 masak mathw: source file is in the github repo, named canvas.svg. almost all the "metadata" happens to be objects in the SVG. I only added a few .class attributes manually.
13:24 mathw nice
13:24 mathw best to keep these things as simple as possible
13:24 masak moritz_: oh. there you go, then. thanks for reminding me. :)
13:25 masak mathw: I could make the whole presentation from within Inkscape. that felt liberating.
13:25 moritz_ masak: hasn't ingy written a yaml loader? I think STD exports YAML....
13:25 mathw and it looked gorgeous
13:25 mathw so fluid
13:25 masonkramer left #perl6
13:25 masak moritz_: he's started, but I haven't seen a finished product.
13:26 masak moritz_: do you think using Blizkost might work?
13:26 mathw although it's possible do that kind of thing in Powerpoint or Impress (although those transitions probably aren't possible), people tend to get stuck in the default slide layouts. Treating it all as one giant SVG breaks you away from that.
13:26 masak mathw: exactly.
13:26 lamstyle left #perl6
13:27 masak mathw: I expect to learn to exploit the spatial aspect more as I make more presentations.
13:27 mathw I must see if I can learn some lessons from it for my next presentation - but since that's a work thing, it's going to be in Powerpoint again. Still, I can do something a bit more fun perhaps :)
13:27 masak best of luck.
13:27 mathw in the mean time... I have software to break
13:27 mathw and a compiler to befuddle
13:29 justatheory joined #perl6
13:31 reqamst joined #perl6
13:33 kaare_ left #perl6
13:34 justatheory left #perl6
13:34 plobsing joined #perl6
13:39 Axius left #perl6
13:40 risou left #perl6
13:43 risou joined #perl6
13:54 ruoso joined #perl6
13:56 timbunce joined #perl6
13:57 masak rakudo: sub foo($x, $y) { 2 * $x + $y }; say foo(5, 32)
13:57 p6eval rakudo 5bdbd4: OUTPUT«42␤»
13:57 masak rakudo: sub infix:<o>(&f, &g) { -> $x { f g $x } }; my &foo = -> $x { -> $y { $x + $y } } o -> $z { 2 * $z }; say foo(5)(32)
13:57 p6eval rakudo 5bdbd4: OUTPUT«42␤»
13:57 masak inspired by http://blog.plover.com/prog/haskell/pointfree.html :)
14:01 alester left #perl6
14:01 araujo left #perl6
14:08 masak time to go catch a plane.
14:08 masak left #perl6
14:09 uniejo left #perl6
14:14 snearch left #perl6
14:17 lamstyle joined #perl6
14:22 MayDaniel joined #perl6
14:25 moritz_ rakudo: die 'foo'
14:25 p6eval rakudo 5bdbd4: OUTPUT«foo␤  in main program body at line 22:/tmp/Y7b64i5XMs␤»
14:25 moritz_ rakudo: my $x; my $x
14:25 p6eval rakudo 5bdbd4: OUTPUT«===SORRY!===␤Redeclaration of symbol $x at line 22, near ""␤»
14:25 moritz_ missing filename
14:25 lamstyle left #perl6
14:27 agentzh left #perl6
14:30 MayDaniel left #perl6
14:39 justatheory joined #perl6
14:40 eternaleye left #perl6
14:42 risou left #perl6
14:42 silug left #perl6
14:46 timbunce left #perl6
14:52 risou joined #perl6
14:58 thepler joined #perl6
15:06 timbunce joined #perl6
15:08 timbunce left #perl6
15:22 timbunce_ joined #perl6
15:24 timbunce_ left #perl6
15:26 araujo joined #perl6
15:26 araujo left #perl6
15:26 araujo joined #perl6
15:26 MayDaniel joined #perl6
15:28 timbunce_ joined #perl6
15:31 timbunce_ is now known as timbunce
15:37 tadzik joined #perl6
15:39 tadzik hello!
15:39 bluescreen left #perl6
15:41 risou_ joined #perl6
15:42 bluescreen joined #perl6
15:42 risou left #perl6
15:43 moritz_ hi tadzik
15:47 plobsing left #perl6
15:50 estrabd_ joined #perl6
15:50 BinGOs left #perl6
15:50 estrabd left #perl6
15:50 BinGOs joined #perl6
15:51 BinGOs left #perl6
15:51 BinGOs joined #perl6
15:54 lamstyle joined #perl6
16:00 lamstyle left #perl6
16:02 timbunce_ joined #perl6
16:03 pyrimidine joined #perl6
16:04 timbunce_ left #perl6
16:05 timbunce left #perl6
16:05 justatheory left #perl6
16:08 Entonian joined #perl6
16:16 _kaare joined #perl6
16:16 baux left #perl6
16:17 ggoebel left #perl6
16:19 bbkr rakudo: say ("a".."f",0..9).pick(56, :replace)
16:19 p6eval rakudo 5bdbd4: OUTPUT«136bcda70258ef49␤»
16:20 bbkr hmm, bug? 56 chars are expected with "replace", right?
16:21 jhuni left #perl6
16:22 tadzik rakudo: say ("a".."f",0..9).roll(56)
16:22 p6eval rakudo 5bdbd4: OUTPUT«7eeee0ed00f7cc86c8f4a3c83b75ac1678da4163d4de690c82e2a781␤»
16:22 bbkr tadzik++
16:22 tadzik but it looks weird indeed. I mean, pick :replace
16:23 bbkr maybe pick with replace disappeared from spec when "roll" was implemented? checking...
16:23 tadzik looks like pick takes no :replace, but why doesn't it complain then?
16:23 tadzik rakudo: say ("a".."f",0..9).pick(56, :pbrrr)
16:23 p6eval rakudo 5bdbd4: OUTPUT«7def0a92bc518346␤»
16:23 tadzik neh
16:23 baest older versions of rakudo returns 56 chars with above pick
16:24 bbkr yep, spec change
16:24 bbkr "pick takes a list or array of values and returns a random selection of elements from the list (without replacement; see roll  for dice roll aka replacement semantics)"
16:24 tadzik rakudo: sub foo() {}; foo(:asasasa); say 'alive'
16:24 p6eval rakudo 5bdbd4: OUTPUT«Unexpected named parameter 'asasasa' passed␤  in 'foo' at line 1:/tmp/OpiyRuTf2C␤  in main program body at line 22:/tmp/OpiyRuTf2C␤»
16:24 tadzik weird
16:25 bbkr clear now - spec has changed. thanks for help :)
16:26 tadzik bbkr++ # bugfinding
16:28 Entonian left #perl6
16:28 moritz_ not a bug
16:28 moritz_ subs != methods
16:28 moritz_ rakudo: say pick(<a b c>, 2, :foobar)
16:28 p6eval rakudo 5bdbd4: OUTPUT«Unexpected named parameter 'foobar' passed␤  in 'pick' at line 1764:CORE.setting␤  in main program body at line 22:/tmp/sBV37nhTMr␤»
16:29 tadzik why do methods always allow named params?
16:30 bbkr who maintains perl6advent.wordpress.com ? in http://perl6advent.wordpress.com/2009/12/15/day-15-pick-your-game/ pick(Int, :replace) can be replaced by roll() to keep up with spec. Simple change that prevents confusion.
16:30 tadzik I can change that, just lemee get some pizza
16:31 bbkr awesome :)
16:32 moritz_ bbkr: I can also give you edit access, if you have (or get) a wordpress.com account and tell me your email address
16:33 alester joined #perl6
16:37 bbkr moritz_: i don't have one, currently i'm too busy at work to spare some time for Rakudo :(
16:37 moritz_ that's OK too
16:38 mberends joined #perl6
16:42 risou joined #perl6
16:43 timbunce joined #perl6
16:43 risou_ left #perl6
16:46 Entonian joined #perl6
16:50 silug joined #perl6
16:55 bbkr how can I convert Str to printable hex values? I'm looking for something that will work as 'print unpack("H*", "abc")' P5 code or mysql HEX() function.
16:55 ggoebel joined #perl6
16:56 moritz_ rakudo: say 'abc'.comb».char».fmt('%x')
16:56 p6eval rakudo 5bdbd4: OUTPUT«Method 'char' not found for invocant of class 'Str'␤  in main program body at line 22:/tmp/BjKMUSaljI␤»
16:56 moritz_ rakudo: say 'abc'.comb».chr».fmt('%x')
16:56 masonkramer joined #perl6
16:57 p6eval rakudo 5bdbd4: OUTPUT«000␤»
16:57 moritz_ that was less than impressive :-)
16:57 moritz_ rakudo: say 123.fmt('%x')
16:57 p6eval rakudo 5bdbd4: OUTPUT«7b␤»
16:57 moritz_ rakudo: say 123.fmt('%02x')
16:57 p6eval rakudo 5bdbd4: OUTPUT«7b␤»
16:58 moritz_ rakudo: say 'abc'.comb».ord».fmt('%02x')
16:58 p6eval rakudo 5bdbd4: OUTPUT«616263␤»
16:58 bbkr \o/
16:58 bbkr moritz++
16:59 diakopter rakudo: say 'abc'.ord».fmt('%02x')
16:59 p6eval rakudo 5bdbd4: OUTPUT«616263␤»
17:00 x3nU left #perl6
17:00 moritz_ diakopter++
17:08 silug left #perl6
17:09 bbkr P6 is awesome for "helper oneliners" - code that does useful stuff but doesn't go to production machines itself. We use PERL in company for 10 years and I have to say that P6 can be considered as "reliable" for many side-tasks now. That's huge progress :)
17:12 estrabd_ is now known as estrabd
17:13 szbalint Yes, I'm using P6 for production side tasks aswell (git hooks, reporting and monitoring scripts)
17:13 moritz_ \o/
17:18 dakkar left #perl6
17:25 ggoebel left #perl6
17:26 PerlJam bbkr, szbalint: that sounds like a great topic for a short article or a whole series of blog posts.
17:26 isBEKaml joined #perl6
17:26 isBEKaml hi,6folks!
17:31 dukeleto isBEKaml: good *
17:31 moritz_ o/
17:32 tadzik hi isBEKaml!
17:32 szbalint PerlJam: indeed.
17:32 moritz_ 'make spectest_smolder' fails at the submission step
17:32 moritz_ curl: (26) failed creating formpost data
17:32 moritz_ because rakudo_test_run.tar.gz is missing
17:35 bbkr PerlJam: we recently started work on publishing Dynamic Content grammar in Perl6 (that's our language that allows to personalize email correspondence - http://dev.getresponse.com/dc-doc/ ). Once it is finished I'll ask management if I can write real-life-perl6-usage-testimonial to publish on blog.
17:35 isBEKaml I was chomping on last night's chomp benchmarks, and I got a question. How does captures compare against substr  especially when trying to get the last character out of any arbitrary length string?
17:35 PerlJam bbkr++ awesome!
17:36 isBEKaml (it sounds like a good benchmark exercise - but I'm looking around for ideas)
17:38 isBEKaml dukeleto: good *. :) is String.reverse making into parrot again anytime soon?
17:39 dalek rakudo: e2740b8 | moritz++ | build/Makefile.in:
17:39 dalek rakudo: add smoke archive to list "clean" make target
17:39 dalek rakudo: review: http://github.com/rakudo/rakudo/commit/e2740b8dfee3d6a593123a7ce650b47cbb072d98
17:43 daxim left #perl6
17:43 risou left #perl6
17:46 risou joined #perl6
17:46 dukeleto isBEKaml: i don't know about "again" because I don't know if it was ever there. But I will try to make String.reverse happen
17:46 araujo left #perl6
17:47 moritz_ it was there.
17:47 isBEKaml dukeleto: iirc, it was there earlier and vanished . :/
17:47 dukeleto moritz_: really? I am baffled as to why it is no longer there. Was it removed as a deprecation? Or did it vanish?
17:48 dukeleto moritz_: because if it vanished, that is a serious breach of our deprecation policy
17:48 dukeleto moritz_: perhaps a svn merge ate it. But I don't see any docs for it either, which makes me think someone did it on purpose
17:50 moritz_ dukeleto: somebody pasted a link to an example or benchmark in PIR that commented that .reverse being gone, it had to iterate over the characters
17:50 dukeleto Author: darbelo <darbelo@d31e2699-5ff4-0310-a27c-f18f2fbe73fe>
17:50 dukeleto Date:   Tue May 4 15:17:00 2010
17:50 dukeleto Obliterate the 'reverse' method on String PMC's.
17:50 dukeleto that was Parrot r46290
17:52 isBEKaml dukeleto: there was an example PIR that clued me in on the reverse removed from the api - but I was confused when I saw that it's specced/defined in PDD28
17:54 dukeleto isBEKaml: i think it was removed because it modifies it's argument
17:56 moritz_ why not have it return a reversed copy?
17:57 dukeleto moritz_: that is what I would imagine it would be changed to. I don't see a deprecation notice, so I am thinking that perhaps it was removed incorrectly
17:57 isBEKaml I think what dukeleto was trying to say was - it was removed because it didn't conform to spec.
17:57 dukeleto isBEKaml: kind of.
17:58 dukeleto isBEKaml: things that modify their arguments are hard to optimize, so parrot has been trying to minimize ops/vtables that modify their arguments
17:59 dukeleto they are so-called "immutable strings" now.
17:59 dukeleto I think we will add String.reverse back in, that does not modify it's argument. But I need to check with parrot-dev first
18:00 isBEKaml dukeleto: I see. Last night, I was benchmarking chomp against substr. One of the ways thrown around was using flip, which just performed horribly (pure agony)
18:01 isBEKaml if reverse made it to parrot, I'd happily take that around for flip.
18:01 moritz_ isBEKaml: flip is now implemented as split.reverse.join
18:01 moritz_ and split('') iterates and uses substr
18:01 isBEKaml moritz_: I saw that too. Perhaps, that was why it performs poorly.
18:02 moritz_ s/Perhaps/Surely/
18:02 isBEKaml 3 costly calls wrapped into a single method.
18:03 dukeleto Parrot has a StringIterator PMC that can be traversed in a "forward" or "reverse" directions.
18:03 dukeleto But I still think String PMCs should have a reverse method.
18:04 timbunce left #perl6
18:06 isBEKaml dukeleto++
18:07 risou left #perl6
18:07 risou_ joined #perl6
18:08 tadzik left #perl6
18:09 dual left #perl6
18:09 ggoebel joined #perl6
18:10 timbunce joined #perl6
18:10 b0m1ro joined #perl6
18:10 fridim joined #perl6
18:13 b0m1ro left #perl6
18:14 dual joined #perl6
18:26 dalek roast: fe777cc | moritz++ | S02-builtin_data_types/instants-and-durations.t:
18:26 dalek roast: [instants-and-durations.t] fix a test description
18:26 dalek roast: review: http://github.com/perl6/roast/commit/fe777cc3a965131adce2bf520976b20614d8effc
18:26 Kodi joined #perl6
18:29 am0c^ left #perl6
18:29 zby_home_ left #perl6
18:31 Raynes joined #perl6
18:34 sorear good * #perl6
18:34 hudnix joined #perl6
18:36 moritz_ good now, sorear
18:38 rgrau_ joined #perl6
18:38 M_o_C joined #perl6
18:42 sftp_ left #perl6
18:42 sftp_ joined #perl6
18:43 sftp_ left #perl6
18:44 sftp joined #perl6
18:45 isBEKaml moritz_:  http://perlcabal.org/syn/S02.html  - why are the spectest examples pointing to azawawi's folder on the server(files which may not be there) ? Why not point to roast git repo?
18:47 moritz_ isBEKaml: hysterical reasons, they had markup...
18:47 moritz_ isBEKaml: but probably not updated anymore... so patches for changing that are very welcome
18:48 dalek specs: dffc82f | KodiB++ | S (2 files):
18:48 dalek specs: Specified a little more support for the local time zone, so you don't have to say 'DateTime.now.timezone'.
18:48 dalek specs: review: http://github.com/perl6/specs/commit/dffc82fa6b3ed622fb8eeb787419915e6b9b26b7
18:54 dukeleto moritz_, isBEKaml: i sent an email to parrot-dev about String.reverse, if y'all have further comments, please add them to that thread
18:57 dalek roast: 866d5ab | KodiB++ | S32-temporal/ (2 files):
18:57 dalek roast: Added tests for the local time zone.
18:57 dalek roast: review: http://github.com/perl6/roast/commit/866d5ab56e12a47d51244e33b9c402ca198c6d74
18:58 M_o_C left #perl6
19:00 M_o_C joined #perl6
19:03 isBEKaml moritz_: if the script that generates html versions of the spec is on feather, can I have a look at it?
19:04 isBEKaml moritz_: it should be on pugs repo, but I don't find it.
19:04 isBEKaml dukeleto: thanks!
19:11 zby_home joined #perl6
19:11 M_o_C left #perl6
19:15 M_o_C joined #perl6
19:16 silug joined #perl6
19:18 moritz_ isBEKaml: it's in the mu repo, in util/smartlinks/
19:18 moritz_ erm, smartlinks.pl
19:18 moritz_ and util/Text-SmartLinks/*
19:22 zby_home left #perl6
19:22 isBEKaml right, tracked down an older svn co'd pugs repo on my machine and went through it. Don't find anything odd at first glance, will look at the newer files.
19:24 ggoebel left #perl6
19:26 jferrero joined #perl6
19:30 moritz_ isBEKaml: it's in until/Text-SmartLinks/lib/Text/SmartLinks.pm
19:30 isBEKaml moritz_: I was just thinking if it'd be a simple fix as removing the prefixed ~/azawawi when you pinged in. :)
19:32 envi left #perl6
19:35 moritz_ isBEKaml: I tried something... let's see if it works out
19:36 isBEKaml moritz_: I'm tesitng it out now... :)
19:36 Kodi left #perl6
19:37 moritz_ uhm, seems I removed too much :(
19:39 dalek mu: 7249b4b | moritz++ | util/Text-SmartLinks/lib/Text/SmartLinks.pm:
19:39 dalek mu: [smartlinks] remove broken HTML links
19:39 dalek mu: review: http://github.com/perl6/mu/commit/7249b4b522142ceb4273a7b66a07a906b80381ea
19:39 dalek mu: 2da197f | moritz++ | util/Text-SmartLinks/lib/Text/SmartLinks.pm:
19:39 dalek mu: Revert "[smartlinks] remove broken HTML links"
19:39 dalek mu:
19:39 dalek mu: This reverts commit 7249b4b522142ceb4273a7b66a07a906b80381ea.
19:39 dalek mu:
19:39 dalek mu: It removed too much... let's try again later
19:39 dalek mu: review: http://github.com/perl6/mu/commit/2da197f13dfc664b094ec46079691bb16b054770
19:41 isBEKaml moritz_: there's a way to test it. It's outlined in Text-SmartLinks/script/smartlinks.pl
19:42 isBEKaml moritz_: damn, I'd now have to resolve external deps that I don't have currently...
19:42 moritz_ isBEKaml: there's a way to test it... push the commits, and see what perlcabal.org/syn/ makes of it :-)
19:43 isBEKaml moritz_: ah, that reminds me. I don't have commit bit since git migration. :(
19:43 moritz_ isBEKaml: what's your github ID?
19:43 isBEKaml svatsan
19:44 moritz_ isBEKaml: fixed. Welcome back to mu, roast, specs and all the rest.
19:44 dalek mu: 40d2add | moritz++ | util/Text-SmartLinks/lib/Text/SmartLinks.pm:
19:44 dalek mu: [smartlinks] another attempt to remove outdated links; also try to fix the link text
19:44 dalek mu: review: http://github.com/perl6/mu/commit/40d2add37e558931d32539b0674ba32a313add85
19:45 isBEKaml moritz_: you're committing them in and firing the gen script?
19:45 moritz_ isBEKaml: yes
19:46 isBEKaml I was thinking it could be something so simple as removing the prefixed ~/azawawi/html/ from the two link locations pointed in there.
19:46 isBEKaml line 663
19:47 isBEKaml after all, looking at the error page I see, that's what it seems to be.
19:47 moritz_ isBEKaml: it could be as simple now, but since azawawi doesn't seem to maintain the whole stuff, I'd rather delete the links
19:48 isBEKaml let's just remove the prefix and see what happens.
19:48 moritz_ is ~/azawawi/html/ up-to-date?
19:49 dalek mu: 5e58136 | moritz++ | util/Text-SmartLinks/lib/Text/SmartLinks.pm:
19:49 dalek mu: [smartlinks] another attempt to fix path
19:49 dalek mu: review: http://github.com/perl6/mu/commit/5e58136a9195f51f538e515e35e7056a80e7491d
19:49 isBEKaml it doesn't seem to hold anything of what we are looking at, it seems. click on summary or full test link
19:52 isBEKaml git still barfs at http://github.com clones.. should that happen?
19:52 isBEKaml I've switched to git://github.com clones..
19:52 moritz_ no, but git:// clones usually work for me
19:52 orafu left #perl6
19:53 dukeleto isBEKaml: are you appending .git to the http:// clone urls ?
19:53 isBEKaml yes.
19:53 jferrero left #perl6
19:53 dukeleto isBEKaml: what error do you get?
19:53 jferrero joined #perl6
19:54 isBEKaml dukeleto: zsh: segmentation fault  git clone http://github.com/perl6/specs.git
19:54 isBEKaml . It goes some way getting parent nodes and the like before hitting this.
19:54 dalek mu: 2966fd7 | moritz++ | util/Text-SmartLinks/lib/Text/SmartLinks.pm:
19:54 dalek mu: [smartlinks] put back a line which I accidentally deleted
19:54 dalek mu: review: http://github.com/perl6/mu/commit/2966fd7d31d277c054c0fe1424fa4c26c5ee5857
19:55 * moritz_ now much happier with the result
19:55 isBEKaml no images, mucho bettero.
19:56 moritz_ one more idea
19:56 moritz_ currently the spectests are inlined
19:56 moritz_ which is a lot of data
19:56 dukeleto isBEKaml: wow, fun. What version of git? You are actually getting a coredump in Git? You must be lucky.
19:56 moritz_ and makes the .html files slow to load
19:56 moritz_ and means that if you use the search feature of your browser, it'll take a minute before telling you that a substring was not found
19:57 moritz_ what do you think about just linking to line numbers on github instead?
19:57 Italian_Plumber left #perl6
19:58 isBEKaml dukeleto: no, I don't see any coredump file. I'm equally baffled at this behaviour. btw, git version is v1.6.4
19:59 plobsing joined #perl6
19:59 dukeleto isBEKaml: ulimit -c is probably set to 0
19:59 isBEKaml moritz_: yeah, I was wondering about that too. I've seen dialog boxes asking me whether to stop script from loading or continue on my FF window.
19:59 dukeleto isBEKaml: which prevents coredumps from being created
19:59 isBEKaml dukeleto: right.
20:00 dukeleto isBEKaml: http clone urls are very inefficient, so it is best not to use them unless you are firewalled. But that is still a strange error.
20:03 isBEKaml dukeleto: I have done this quite a lot of times before (my git fu isn't that strong or as sufficiently competent as my svn fu). and never faced this. Let me thinko about raising an issue with git folks.
20:03 moritz_ \o/  success
20:05 moritz_ much faster now to search the specs
20:05 dalek mu: 9a57c5b | moritz++ | util/Text-SmartLinks/lib/Text/SmartLinks.pm:
20:05 dalek mu: [smartlinks] try to reduce page load times by non-inlining test scripts; link to them instead
20:05 dalek mu: review: http://github.com/perl6/mu/commit/9a57c5bc7a1b28b47d9d09adb7b83dbee411ef1e
20:06 isBEKaml dukeleto: I just noticed another weird thing. http clone works fine on another account here that uses bash shell. Could it be related to the $SHELL ?
20:07 isBEKaml my guess is, there's some env variable that's misbehaving - leading to coredump.
20:08 dukeleto isBEKaml: yeah, if you are using git completion or a git-ified PS1, that could be possible
20:08 isBEKaml moritz_: awesome. Today I learned that github highlight works by including the line numbers in the url...
20:08 dukeleto isBEKaml: yes, i love that feature
20:08 isBEKaml dukeleto: none of that.
20:10 dalek mu: b3c0507 | moritz++ | util/Text-SmartLinks/lib/Text/SmartLinks.pm:
20:10 dalek mu: [smartlinks] code cleanup
20:10 dalek mu: review: http://github.com/perl6/mu/commit/b3c0507e42170dbd2d0de12a503144c835930e94
20:11 dukeleto isBEKaml: interesting. Perhaps zsh is having a bad day. You can always backtrace the coredump if you are really interested in it ....
20:11 isBEKaml moritz_: I'm happy running yak shaving sessions with rakudo folks! :D
20:12 * dukeleto makes a large wig from freshly shorn yak hair
20:13 * isBEKaml watches in _no_ wonder as it falls off with a huge THUD
20:15 isBEKaml dukeleto: some other day. It's 2 am here and I'm in no mood to see another segfault. :)
20:15 * sorear has started porting STD.pm6 in earnest
20:20 timbunce left #perl6
20:24 jferrero left #perl6
20:27 timbunce joined #perl6
20:27 * jnthn made it home from trip to .nl and OSDC.fr :-)
20:27 phenny jnthn: 10 Oct 16:57Z <sorear> ask jnthn Where in S14 does it talk about body run time of parametric roles?
20:28 jnthn sorear: Hmm. Well, wherever I decide to put it when I patch it, I guess. :-)
20:28 * jnthn thought it was already in there
20:28 jnthn Will look tomorrow or so. Mostly interested in being unconcious at the moment. ;-)
20:35 diakopter jnthn: g'nite
20:36 diakopter sorear: porting STD.pm6 ...?
20:36 diakopter to niecza's subset of p6?
20:38 Bucciarati joined #perl6
20:44 Mowah left #perl6
20:45 sorear diakopter: I prefer to think of it as porting STD.pm6 to Perl 6.0
20:45 sorear viv has a very idiosyncratic feature set
20:46 PhatEddy joined #perl6
20:46 diakopter ahh... good point.
20:48 orafu joined #perl6
20:52 Kodi joined #perl6
20:54 sorear also I'm collecting a list of small things Niecza needs
20:54 sorear like &substr
20:54 sorear (I already have the method form, so the sub is really quite easy)
20:55 PhatEddy phenny: tell moritz_ his missing rakudo_test_run.tar.gz file could be caused by not having TAP::Harness::Archive installed.  Without the module installed the tests are still run but you get the LTA error message.
20:55 phenny PhatEddy: I'll pass that on when moritz_ is around.
21:07 M_o_C left #perl6
21:10 chitragupt left #perl6
21:11 sorear TimToady: ping
21:11 masak joined #perl6
21:11 masak jnthn: seems you beat me to the front door :)
21:11 sorear o/ masak
21:11 masak sorear: \o
21:12 Italian_Plumber joined #perl6
21:16 Italian_Plumber left #perl6
21:16 chitragupt joined #perl6
21:17 Italian_Plumber joined #perl6
21:17 Italian_Plumber left #perl6
21:18 dalek rakudo: db221e3 | KodiB++ | src/c (2 files):
21:18 dalek rakudo: Implemented $*TZ.
21:18 dalek rakudo: review: http://github.com/rakudo/rakudo/commit/db221e3d2b14da597d4b71b9525380a5fa7ca60a
21:18 dalek rakudo: 3494676 | KodiB++ | src/core/Temporal.pm:
21:18 dalek rakudo: [Temporal] Ensured .new(Str) works correctly in derived classes.
21:18 dalek rakudo: review: http://github.com/rakudo/rakudo/commit/3494676c7dac88fad9b71ce8065fdd85bad7b695
21:18 dalek rakudo: a1cd99b | KodiB++ | src/ops/perl6.ops:
21:18 dalek rakudo: [ops] Added encodelocaltime.
21:18 dalek rakudo: review: http://github.com/rakudo/rakudo/commit/a1cd99b175f4fa150aa1f8f40b20df4ebf92f36b
21:18 dalek rakudo: 5d23e30 | KodiB++ | t/spectest.data:
21:18 dalek rakudo: Run S32-temporal/local.t.
21:18 dalek rakudo: review: http://github.com/rakudo/rakudo/commit/5d23e3082ce89a5dae607f45846059c64a0e921d
21:18 dalek rakudo: 8711e05 | KodiB++ | docs/ChangeLog:
21:18 dalek rakudo: Updated ChangeLog.
21:18 dalek rakudo: review: http://github.com/rakudo/rakudo/commit/8711e0578705b85f2fb8503bb7c29ef3f1dd1f5e
21:18 dalek rakudo: e6aa633 | KodiB++ | src/core/Temporal.pm:
21:18 dalek rakudo: [Temporal] Don't clone the formatter, so .perl can always recognize the default.
21:18 dalek rakudo: review: http://github.com/rakudo/rakudo/commit/e6aa633e3b99b9165776954b53311f88c785db29
21:18 masak Kodi++
21:19 leprevost joined #perl6
21:22 masak wow. lots of sensible discussion about the actual matter at hand here: http://www.reddit.com/r/programming/comments/dpqh7/longest_palindrome_by_regex/
21:22 masak reddit does surprise sometimes.
21:23 nymacro left #perl6
21:24 justatheory joined #perl6
21:25 _kaare left #perl6
21:25 leprevost left #perl6
21:25 colomon err... they do recognize that 1.5 seconds was in Rakudo, right?
21:26 masak no idea :)
21:27 Kodi left #perl6
21:27 colomon they don't seem to.
21:27 colomon I don't think I can comment there...
21:28 * masak glances longingly at the bed
21:28 isBEKaml left #perl6
21:28 masak bed: oh, ok. you've convinced me. :)
21:28 masak #perl6: good night. see you tomorrow.
21:28 masak left #perl6
21:29 chitragupt left #perl6
21:29 chitragupt joined #perl6
21:30 colomon (some of the people do realize it is, later on.)
21:30 pyrimidine left #perl6
21:32 MayDaniel left #perl6
21:34 Eevee left #perl6
21:40 f00li5h[HireMe] is now known as f00li5h
21:44 sftp left #perl6
21:50 nymacro joined #perl6
21:50 timbunce left #perl6
22:04 ggoebel joined #perl6
22:04 TomDLux joined #perl6
22:36 justatheory left #perl6
22:37 risou_ left #perl6
22:41 fridim left #perl6
22:43 PhatEddy left #perl6
22:47 wamba left #perl6
23:11 thundergnat joined #perl6
23:12 thundergnat rakudo: say 8.sqrt;
23:12 p6eval rakudo 5d23e3: OUTPUT«2.82842712474619␤»
23:12 thundergnat rakudo: say (-8).sqrt
23:12 p6eval rakudo 5d23e3: OUTPUT«NaN␤»
23:12 thundergnat phooey
23:12 thundergnat rakudo: say (-8).Complex.sqrt
23:12 p6eval rakudo 5d23e3: OUTPUT«0 + 2.82842712474619i␤»
23:13 thundergnat Just an observation, but it seems to me that taking a square root of a negative number should automatically coerce to a Complex.
23:24 thundergnat Added a new perl6 implementation to rosettacode: http://rosettacode.org/wiki/Roots_of_a_quadratic_function#Perl6
23:40 araujo joined #perl6
23:48 thundergnat left #perl6
23:51 Raynes left #perl6
23:52 araujo left #perl6
23:53 masonkramer thundergnat: I don't necessarily agree with that - it could hide a flaw in the program
23:54 pythonian4000afk is now known as pythonian4000
23:55 diakopter I think it should depend on whether that "negative number" was a Complex
23:56 masonkramer rakudo: say 5.Complex.sqrt
23:56 p6eval rakudo 5d23e3: OUTPUT«2.23606797749979 + 0i␤»
23:56 masonkramer ahh, right, that's what he's got
23:56 masonkramer yeah, that's what's happening now

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

Perl 6 | Reference Documentation | Rakudo