Perl 6 - the future is here, just unevenly distributed

IRC log for #perl6, 2009-12-19

Perl 6 | Reference Documentation | Rakudo

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

All times shown according to UTC.

Time Nick Message
00:04 zaslon lolperl6adventhazblogged! perl6advent++ 'Day 19: Whatever': http://perl6advent.wordpress.c​om/2009/12/19/day-19-whatever/
00:07 nihiliad joined #perl6
00:12 awwaiid joined #perl6
00:37 Exodist joined #perl6
00:39 patspam joined #perl6
00:50 Wolfman2000 evening
00:52 sjohnson hi
00:56 yahooooo joined #perl6
01:42 colomon_ joined #perl6
01:51 japhb joined #perl6
02:18 patspam joined #perl6
02:20 fatface_ joined #perl6
02:29 astrojp joined #perl6
02:30 astrojp left #perl6
02:33 LionMadeOfLions joined #perl6
02:48 pointme joined #perl6
02:48 mubot joined #perl6
02:48 zaslon joined #perl6
02:50 TimToady joined #perl6
02:54 diakopter 'tis
02:54 colomon_ 'tain't
02:56 TimToady joined #perl6
02:57 wayland76 Isn't that "perl -T" ? :)
02:57 colomon_ :)
03:25 Chillance joined #perl6
03:26 * colomon_ would like to point out he just likes the word 'tain't, and has no idea what diakopter was talking about.
03:28 sjohnson i like it
04:12 meppel joined #perl6
04:15 Baggio_ joined #perl6
04:32 colomon joined #perl6
04:34 astrojp joined #perl6
04:36 diederich joined #perl6
04:36 _jaldhar joined #perl6
04:36 diederich joined #perl6
04:37 rfordinal joined #perl6
05:02 pmichaud_ joined #perl6
05:14 diakopter colomon: :)
05:16 gfx joined #perl6
05:19 Baggio_ joined #perl6
05:20 mberends joined #perl6
05:23 diakopter mberends++ | [pugs/src/perl6/vill] introducing vill, a viv+llvm mashup
05:23 diakopter mberends++ | http://perlcabal.org/svn/pugs/revision/?rev=29375
05:29 drbean_ joined #perl6
05:29 diakopter mberends++ sneaking vill into pugscode svn when no one was watching
05:29 mberends :) hoping some helpful people learn LLVM and help develop 'vill'
05:30 matsuzine joined #perl6
05:31 matsuzine left #perl6
05:31 matsuzine_ joined #perl6
05:35 matsuzine joined #perl6
05:36 matsuzine_ left #perl6
05:44 mberends the vill 'make test' is useful to prove and see what it does
05:51 dalek nqp-rx: 7009fe7 | pmichaud++ | src/Regex/Cursor-protoregex-peek.pir:
05:51 dalek nqp-rx: Switch to using getattribute for contexts instead of hash interface.
05:51 dalek nqp-rx: review: http://github.com/perl6/nqp-rx/commit/7​009fe7a5dafefb1153f0343ac67a4e7e8e16022
05:51 dalek nqp-rx: eb9c75a | pmichaud++ |  (5 files):
05:51 dalek nqp-rx: Update bootstrap, PARROT_REVISION.
05:51 dalek nqp-rx: review: http://github.com/perl6/nqp-rx/commit/e​b9c75a9b6bf144808ca6d24f31b606e9e8adba8
06:04 _stuuf joined #perl6
06:16 Jeffrey_ joined #perl6
06:18 Jeffrey_ left #perl6
06:24 * mberends skims http://unladen-swallow.googlecode.c​om/files/Unladen_Swallow_PyCon.pdf for ideas
06:29 mj41_ joined #perl6
06:35 charsbar_ joined #perl6
06:42 kst joined #perl6
06:50 nicholas_ joined #perl6
06:50 matsuzine left #perl6
06:59 Baggio_ joined #perl6
07:01 pugs_svn r29376 | mberends++ | [src/perl6/vill] tweak wordcount in Makefile to ignore .svn dirs
07:07 pugs_svn r29377 | pmurias++ | [vill] refer to "someone man known as pmichaud" by his name
07:07 mberends ;-)
07:37 pugs_svn r29378 | mberends++ | [vill] added pmurias++ as contributor
07:48 colomon_ joined #perl6
07:51 JimmyZ joined #perl6
08:01 colomon_ joined #perl6
08:33 lumi joined #perl6
08:33 kaare joined #perl6
08:40 drbean joined #perl6
08:50 meneldor joined #perl6
08:50 JimmyZ joined #perl6
08:56 Chillance joined #perl6
08:57 meneldor hello moritz_ , i forgot about rakudo on win32 and installed it successfully on my ubuntu
08:57 meneldor now i read your PDF book
08:58 meneldor but it starts with JSON tutorial?
08:58 meneldor is this the right book to start reading for perl5 to perl6 ?
08:59 moritz_ meneldor: it shouldn't start with JSON
09:00 meneldor i've opened a link from the blog
09:00 moritz_ http://cloud.github.com/downloa​ds/perl6/book/book-2009-12.pdf
09:01 lumi_ joined #perl6
09:01 meneldor http://perlgeek.de/blog-en/perl-6/w​e-write-a-perl-6-book-for-you.html
09:01 meneldor this one
09:01 moritz_ meneldor: there are newer versions than the one linked on my blog, we have monthly releases
09:01 moritz_ and that was the first one
09:01 meneldor :)
09:01 meneldor ok tnx
09:01 moritz_ maybe we should have a permanent link which always points to the latest
09:02 meneldor ah this book is different thing :) thank you moritz_
09:03 meppl joined #perl6
09:04 moritz_ meneldor: it still has a chapter about parsing JSON
09:05 meneldor i can read about it if i start to use perl6 for web later :)
09:05 meneldor now i use json module for catalyst with perl5
09:05 moritz_ you can also use the knowledge for parsing something else than JSON
09:06 Baggio_ joined #perl6
09:06 meneldor ok but how perl6 use local module libraries?
09:06 meneldor do he use own lib path?
09:07 moritz_ if you put modules into ~/.perl6/lib/ rakudo will find them
09:07 moritz_ or you can set the PERL6LIB environment variable
09:07 moritz_ or you can say BEGIN { @*INC.push: '/path/to/lib' } in your module
09:07 meneldor i see
09:08 meneldor is there a way like in perl5 to install from cpan directly to perl6 lib
09:09 meneldor like: cpan6 install Module
09:09 moritz_ there's an installer called "proto"
09:09 moritz_ http://github.com/masak/proto/
09:10 moritz_ you have to checkout the installed-modules branch
09:10 moritz_ and then you can try ./proto install json
09:10 meneldor tnx
09:16 Su-Shee joined #perl6
09:16 iblechbot joined #perl6
09:17 Su-Shee good morning
09:17 Tene hi!
09:17 mberends guten morgen
09:19 moritz_ tach!
09:20 meneldor do anyone use Padre with perl6 ?
09:21 * Su-Shee doesn't.
09:27 * moritz_ is a vim user
09:28 Su-Shee so am I. gvim with shiny tabs ;)
09:35 meneldor im trying this from the book: my @names = $file.get.split(' ');
09:35 agentzh joined #perl6
09:35 meneldor but it throws an error
09:35 meneldor Could not locate a method 'get' to invoke on class 'IO'.
09:36 meneldor there are 6-7 additional lines below this error
09:37 lisppaste3 meneldor pasted "untitled" at http://paste.lisp.org/display/92346
09:41 * frettled is a whimsical user.
09:41 meneldor can anyone help me with this error please
09:42 lumi joined #perl6
09:42 ashizawa joined #perl6
09:42 meneldor is this my fault or this code is obsolete?
09:44 frettled I'm sorry, I'm still too much of a n00b to help you with that.  :(
09:46 mberends meneldor: please paste your script file
09:47 * frettled idles again.  :)
09:47 mberends :)
09:48 mberends if you successfully open a file, get works
09:49 mikehh joined #perl6
09:50 meneldor if i only use open it works
09:50 meneldor this error appear when i use $file.get.split(' ');
09:56 moritz_ $*IN.get.split(' ').perl.say
09:57 moritz_ rakudo: $*IN.get.split(' ').perl.say
09:57 p6eval rakudo 7f1c3f: ["Land", "der", "Berge,", "Land", "am", "Strome,"]␤
09:58 meneldor moritz_: i just ran the code in the book
09:58 moritz_ meneldor: and did you create the data files first?
09:59 Baggio_ joined #perl6
09:59 meneldor yes
09:59 meneldor i wrote file named the same way with the same data inside
10:00 justin_ joined #perl6
10:01 Alias joined #perl6
10:02 justin_ left #perl6
10:02 moritz_ that's weird
10:02 moritz_ but I'm afraid I can't investigate right now, have to leave rather soon
10:03 mberends copy the source code from the book pdf give the wrong kind of quote characters
10:04 mberends *copying, gives, gah, fingers too cold
10:04 meneldor mberends: i've changed them :)
10:05 mberends Also fixing more of them here. The original git version would be more reliable for copying.
10:07 mberends works now. beware of backticks pretending to be single quotes
10:08 meneldor the line: my @names = $file.get.split(' ');
10:08 meneldor is hand written
10:09 mberends check every quote/backtick on all lines
10:14 meneldor well i did it
10:14 meneldor i tryed this: my $file = open 'scores'; say $file;
10:14 meneldor it prints: IO<0xb5fbe380>
10:15 mberends $file is an open file object, ok
10:15 meneldor but it means that OPEN works
10:15 meneldor then if i use get it throws an error
10:16 mberends pasting my copy, then you can diff them
10:16 vamped joined #perl6
10:17 meneldor your copy?
10:18 vamped left #perl6
10:20 meneldor here is my perl6 version: This is Rakudo Perl 6, revision 37414 built on parrot 0.9.1-devel
10:20 lisppaste3 mberends pasted "book example for meneldor" at http://paste.lisp.org/display/92347
10:22 meneldor mberends: the same error :(
10:22 mberends meneldor: your parrot is too old, you should get newer parrot and rakudo
10:22 meneldor but i installed it 1 hour before
10:22 mberends you downloaded from an old site
10:22 meneldor oh no
10:23 mberends parrot is now 1.8
10:23 meneldor i use http://www.rakudo.org/how-to-get-rakudo
10:23 * mberends looks
10:24 meneldor then i downloaded  rakudo-2009-12.tar.gz
10:24 meneldor then: perl Configure.pl --gen-parrot
10:24 meneldor and it check some thinks in the site i think
10:25 mberends it could be that your search path is finding an earlier install instead of the new one
10:26 meneldor nope, i start perl6 executable in the same directory i've builded
10:26 mberends try 'which perl6' and verify the directory
10:27 lumi_ joined #perl6
10:27 mberends the 0.9.1 must be removed with extreme prejudice
10:28 meneldor ill try to compile it again
10:29 mberends just remove all the new stuff first, and verify that you have *no* old version left
10:29 meneldor this is the first time i install perl6
10:30 mberends which OS is this on?
10:30 meneldor ubuntu
10:30 meneldor hmm now is different i think
10:30 meneldor Checked out revision 43076.
10:30 mberends ok, I'll try the webpage download in /tmp on another ubuntu pc
10:31 meneldor wait
10:31 meneldor let me compile it
10:33 mberends 43076 sounds good, I'm using 42558
10:34 rfordinal joined #perl6
10:35 meneldor i might had been installed from old archive
10:37 meneldor nice!
10:37 meneldor tnx mberends
10:37 meneldor now all works
10:37 meneldor i dont know how i've installed that old version before :)
10:38 * mberends remembers the birth of 'get' at NPW in Oslo
10:38 meneldor well today im starting to migrate from perl5
10:38 meneldor cheers for the help mberends
10:39 mberends meneldor: you're very welcome
10:39 meneldor here is the old version I've installed before: http://github.com/rakudo/rakudo
10:40 meneldor when you click Download it sends you some archive rakudo-rakudo-26262626.tar.gz
10:40 meneldor then i've downloaded rakudo-2009-12.tar.gz and all is OK
10:41 mberends the site is current, but the tar file may be obsolete
10:45 mikehh joined #perl6
10:52 mjk joined #perl6
11:12 JimmyZ joined #perl6
11:29 iblechbot joined #perl6
11:44 lumi joined #perl6
11:45 astoria joined #perl6
11:54 masak joined #perl6
11:54 masak mberends++ # vill
11:54 masak mberends: I vill definitely check it out! :)
11:57 masak phenny: tell moritz that http://perlgeek.de/blog-en/per​l-6/publicity-for-perl-6.html (while otherwise great as usual) contains a HTML-related glitch in its first paragraph. you don't type in all the HTML manually, do you? :)
11:57 phenny masak: I'll pass that on when moritz is around.
11:58 masak oh dang.
11:58 masak phenny: tell moritz_ that http://perlgeek.de/blog-en/per​l-6/publicity-for-perl-6.html (while otherwise great as usual) contains a HTML-related glitch in its first paragraph. you don't type in all the HTML manually, do you? :)
11:58 phenny masak: I'll pass that on when moritz_ is around.
12:03 * masak really liked http://use.perl.org/~Alias/journal/40026
12:03 * Alias bows
12:03 colomon joined #perl6
12:04 masak I think, in all humility of course, that it's one of the areas where Perl 6 will be more adaptable than Perl 5.
12:04 Alias From my perspective, Perl 6 is worse
12:04 Alias Because the grammar is utterly intangible
12:05 Alias And modules only have a valid document structure if they also compile
12:06 Alias Does P6 have a BEGIN equivalent?
12:06 Baggio_ joined #perl6
12:06 Alias Is it possible to do code injection into the compiler?
12:06 masak oh, sure.
12:07 masak macros, defined operators, defined other pieces of syntax...
12:07 masak but it's, er, homoiconic in a way that Perl 5 isn't.
12:08 rachelBROWN what would be the best way to go about donating LSD to the most active contributors of perl6
12:09 masak rachelBROWN: are you really sure that's what's needed at this juncture? :P
12:10 masak how about we just make sure the ng branch merges instead?
12:10 rachelBROWN perl6 is supposed to do everything, right?
12:10 kidd joined #perl6
12:10 rachelBROWN I thing LSD would help us with that.
12:10 huf no, just 70% of that
12:11 masak rachelBROWN: I see your point, but... I think most people believe more in LTM around here.
12:13 rachelBROWN Large Telescope Metalanguage?
12:13 masak no, but I like the sound of that. :)
12:14 masak Longest-Token Matching.
12:14 huf hmm, my LTM mislead me about the meaning of LTM
12:15 araujo joined #perl6
12:15 masak huf: you have a telescope? :)
12:20 huf i meant long-term memory ;) no scope, only vision here.
12:23 masak oh. well, hindsight is 20-20.
12:24 huf you are correct. i have 0 eyes in the back of my head.
12:26 jnthn oh hai, #perl6
12:28 masak lolitsjnthn! \o/
12:28 Su-Shee_ joined #perl6
12:29 jnthn .oO( does LSD make meta-models more understandable? )
12:31 jnthn Anyway, the ng branch needs Hash far more. ;-)
12:33 masak if nothing else, we're pun junkies.
12:34 huf best kind of junkie.
12:34 * Su-Shee can stop punning every day, if I want to.
12:34 huf right after each pun? and until the next one? impressive.
12:35 Su-Shee just a question of will. and a backspace key in good working condition.
12:36 masak the first step is admitting that one is a punster.
12:36 Su-Shee "hello masak!" ;)
12:36 carlin Friends don't let friends use puns
12:36 masak hi.
12:36 masak my name is masak, and I pun.
12:37 huf hmm, this could lead to the punic wars if we arent careful
12:37 masak I started punning when I was 15, sneaking behind the school building and punning where nobody heard me.
12:37 Su-Shee huf: you'll have to pay for that one. ;)
12:37 carlin Good idea, make a pun jar
12:38 Su-Shee there is actually such a thing in english? I always had difficulties translating the german word.
12:38 huf one of intermediate size no longer used in the west?
12:38 huf (but still used in strife-torn bolivia as a reminder of better days)
13:25 pmurias joined #perl6
13:44 rfordinal left #perl6
13:45 Su-Shee joined #perl6
13:51 synth joined #perl6
13:51 JimmyZ joined #perl6
13:53 cb_ joined #perl6
13:58 orafu joined #perl6
14:00 cb_ joined #perl6
14:10 cb_ joined #perl6
14:15 Baggio_ joined #perl6
14:22 mikehh joined #perl6
14:42 masak quiet here today.
14:42 * masak tries vill
14:45 mberends oooh, masak!
14:46 masak mberends! \o/
14:46 mberends you'll have to install LLVM first. Since much of the work is being done at Apple, it should not be a problem :)
14:47 masak congratulations on the de-sekkritification!
14:47 mberends :D
14:47 masak mberends: I'm on it.
14:47 mberends cool!
14:47 masak reading the README in the meantime :)
14:49 mberends just planning more README items, to pre-answer FAQs
14:50 mberends so if *you* ask, it's ok, it's not RTFM yet ;)
14:51 mberends it's nice to see that llvm-config is a Perl 5 script
14:52 pugs_svn r29379 | pmurias++ | [mildew-js] strings and ints are not wrapped
14:52 masak http://llvm.org/docs/GettingStarted.html "1. Read the documentation.\n2. Read the documentation.\n3. Remember that you were warned twice about reading the documentation." :)
14:52 mberends heehee
15:02 hicx174 joined #perl6
15:03 patspam joined #perl6
15:07 masak mberends: any particular reason the 'hello world' example in the README is 'print "howdy!"' and not 'say "howdy!"'?
15:07 masak the latter gives nicer output, and is more of a moral boost, since it's a Perl6ism.
15:08 masak s/moral/morale/
15:11 mberends masak: there was a trivial reason.  I thought I would leave 'say' as a piece of LHF to tempt some hacker to implement it in terms of the existing 'print' ;)
15:12 masak ahah. :)
15:12 mberends so please do
15:12 arnsholt I like the idea of leaving stuff to trick new people into contributing
15:12 masak still building LLVM... :)
15:12 mberends "trick" ? are you insinuating ?
15:13 arnsholt I'd never insinuate anything O=)
15:14 arnsholt But more seriously, it sounds like a good idea. Having some stuff that people not intimately familiar with the codebase can do might be a good way of attracting more contributors
15:14 mberends innocently put, it's optimizing -Ofun for the maximum number of people
15:15 mberends diakopter++ got me into Sprixel that, intentionally or not
15:16 mberends *like that # fingers too cold
15:19 masak is your IRC client outside?
15:20 mberends the shop that sells inkjet cartridges was a short walk away :)
15:22 masak mberends: this only creates more questions. :)
15:23 masak does your IRC client need ink?
15:24 masak will the inkjet cartridge salesmen be the first against the wall when the revolution comes?
15:24 masak etc.
15:24 mberends this FAQ will need to become bigger than planned :)
15:26 masak by the way, compiling LLVM takes a long time.
15:27 mberends hmm. did you read about not building for *all* targets?
15:28 mberends the default LLVM can cross compile to numerous CPUs
15:28 masak is that so?
15:28 masak no, I just followed the build instructions in the documentation.
15:30 mberends ./configure --prefix=/usr --enable-optimized --enable-targets=host-only ?
15:30 mberends (from the vill README)
15:31 masak oh.
15:31 masak I'll abort and do like that then.
15:31 masak thanks.
15:31 mberends without that, it adds about 10 CPU backends
15:32 mberends it still took me 25 45 minutes
15:32 mberends 25 - 45 minutes system dependent
15:34 Psyche^ joined #perl6
15:39 masak mberends: the README is extensive and very enjoyable.
15:45 colomon_ joined #perl6
15:47 mberends masak: you know I value docs more than most people do
15:53 masak mberends: it's a positive trait. kudos.
15:55 mberends masak: you're pretty good yourself too.
15:56 mberends oh, the mutual appreciation society meeting is just beginning
16:01 moritz_ masak: thanks
16:01 phenny moritz_: 11:58Z <masak> tell moritz_ that http://perlgeek.de/blog-en/per​l-6/publicity-for-perl-6.html (while otherwise great as usual) contains a HTML-related glitch in its first paragraph. you don't type in all the HTML manually, do you? :)
16:02 masak mberends: every community is a mutual appreciation society to some extent. :)
16:02 mberends masak++ # just because
16:02 masak :)
16:03 masak I wish I took the time to write more documentation and architectural comments, though.
16:03 masak but I am sorta proud of the GGE README, even though it's much smaller than vill's :) http://github.com/masak/gge
16:06 diakopter Alias: could PPI do "keyhole" grammar mutation in the same way Devel::Declare does?
16:06 masak mberends: are you sure about the v5.8.0 minimal version requirement. ISTR that STD.pm runs on > v5.10 only.
16:08 diakopter Alias: (though, if you did that, it'd become as expressive as the languages STD.pm can handle...)
16:08 diakopter mberends: yes, v5.10
16:10 holli joined #perl6
16:11 holli hi. i'm toying with todays calendar entry.
16:11 holli this line: say @x.map: * ~ 'A';
16:11 mberends thanks masak and diakopter, will update with other edits
16:11 holli doesnt work in rakudo (No applicable candidates found to dispatch to for 'map' in Main)
16:12 masak rakudo: my @x = <a b c>; say @x.map: * ~ 'A';
16:12 p6eval rakudo 7f1c3f: No applicable candidates found to dispatch to for 'map'␤in Main (file <unknown>, line <unknown>)␤
16:12 masak holli: correct.
16:12 masak rakudo: say (* ~ 'A").WHAT
16:12 p6eval rakudo 7f1c3f: Confused at line 2, near "~ 'A\").WHA"␤in Main (file <unknown>, line <unknown>)␤
16:12 masak rakudo: say (* ~ 'A').WHAT
16:12 p6eval rakudo 7f1c3f: Str()␤
16:12 masak holli: there you go.
16:13 diakopter what?
16:13 masak 's a bug.
16:13 lumi_ joined #perl6
16:14 diakopter rakudo: my @x = <a b c>; say @x.map: { * ~ 'A' }
16:14 p6eval rakudo 7f1c3f: AAA␤
16:16 masak rakudo: my @x = <a b c>; say @x.map: { $_ ~ 'A' }
16:16 p6eval rakudo 7f1c3f: aAbAcA␤
16:16 masak diakopter: both * and {} is tautological.
16:17 diakopter *and* repetitively redundant :P
16:17 masak and unnecessary :)
16:17 diakopter and extraneous
16:17 masak here we go, autopunning again.
16:17 diakopter in addition to superfluous
16:17 holli rakudo: my @x = <a b c>; say @x.map: { * }
16:17 p6eval rakudo 7f1c3f: !whatever_closure!whatever​_closure!whatever_closure␤
16:17 holli lol
16:18 diakopter lolhi indeed
16:18 masak we didn't mention repetitive, did we? and repetitive.
16:18 carlin rakudo: my @x = <a b c>; say @x.map: *;
16:18 p6eval rakudo 7f1c3f: get_number() not implemented in class 'Sub'␤in Main (file <unknown>, line <unknown>)␤
16:19 holli and then they say people should write modules for Perl6
16:19 masak holli: you think that's bad? :)
16:20 masak holli: try implementing a functioning wiki engine 18 months ago!
16:20 masak mberends: slight vill make failure: http://gist.github.com/260133
16:21 * mberends thinks
16:22 mberends ok, gcc on OS X may have a different set of C/C++ lib defaults and cannot statically link malloc()
16:22 masak sounds plausible.
16:23 mberends you're the first non linux customer (or maybe the first customer, period)
16:24 mberends certainly the first customer with a complaint ;)
16:24 masak help me solve this, and we'll see about that low-hanging 'say' fruit! :)
16:25 mberends :)
16:25 holli how complete is file io?
16:26 masak holli: not 100%, but usable.
16:26 masak anything in particular?
16:26 colomon__ joined #perl6
16:26 holli nah, just letting my thoughts flowing =)
16:27 masak that's encouraged. :) but don't hesitate to make specific requests.
16:27 nihiliad joined #perl6
16:33 holli to be honest, i have absolutely no perl6 experience. was just thinking i try at latest rakudo (given all the talk about rakudo* lately). but when i see that a simple
16:33 holli example from the advent calendar doesnt run i'm effectively discouraged.
16:34 holli i mean it's map not working, map!
16:35 masak holli: there are a couple of known issues with the Whatever * and operators.
16:35 mberends masak: finally reading every word in your error gist, the make has not reached the library linking stage yet. The C++ compiler does not have the include dir that contains malloc.h in its search path. This is a typical reason why projects have to run a configure before make, to patch the Makefile. I was hoping to avoid having to do that :(
16:35 masak holli: I agree that it's unfortunate that one such example crept into the Advent Calendar.
16:36 masak holli: but I won't lie to you -- you're bound to run into similar omissions here and there. it's possible to write elaborate things on top of Rakudo, but one has to work and know which deficiencies to avoid.
16:36 masak mberends: ah, so it's more of your fault than mine? ;)
16:37 mberends yes, ok, blame me, Richard Stallman and Steve Jobs
16:39 * holli blames Merlyn
16:39 masak together, you'll have a reality distortion field to make objects materialize!
16:39 subichan joined #perl6
16:41 * mberends struggles to avoid adding a pre-make configure
16:41 holli rakudo: say "foo { 1 + 2 } bar"
16:41 p6eval rakudo 7f1c3f: foo 3 bar␤
16:41 holli neat
16:42 holli rakudo: my $foo = 'foo { 1 + 2 } bar'; say $foo.eval
16:42 p6eval rakudo 7f1c3f: Method 'eval' not found for invocant of class 'Str'␤in Main (file src/gen_setting.pm, line 324)␤
16:42 masak eval is a function.
16:43 holli why?
16:44 masak holli: most things are. 'say' and a few others are the exception.
16:45 colomon I'm with holli.  Why the heck isn't there a .eval method on Str?  Makes perfectly good sense to me...
16:45 masak it does.
16:45 mberends holli++
16:45 masak let's add it while no-one's looking :)
16:45 colomon Okay, give me sec.  :)
16:45 holli i contributed to perl6 =) =) =)
16:45 masak if someone else does the spec, I'll do Rakudo.
16:45 colomon or you get the spec and I'll get ng.
16:46 masak I was thinking master.
16:46 masak holli++ indeed.
16:47 colomon Should it be on Any or just Str?
16:47 mberends eval expects Str
16:48 mj41 joined #perl6
16:48 masak I added it. running spectests.
16:48 colomon really?
16:48 colomon rakudo: say eval 10;
16:48 p6eval rakudo 7f1c3f: Parameter type check failed on call to 'eval'.␤in Main (file <unknown>, line <unknown>)␤
16:48 masak colomon: I'm not pushing until the spectests show green. :)
16:48 colomon masak: did you do Any.eval or Str.eval?
16:49 colomon masak: that's why it's easier to do it in ng.  :)
16:49 masak I put it in Any-str,pm, but I coerce to Str.
16:49 colomon eval ~self  ?
16:49 masak yes.
16:49 colomon that's exactly what I wrote here.  :)
16:49 masak not much room for variation :)
16:50 colomon could have been eval self.Str, I suppose.
16:50 masak actually, I used parentheses, because Damian told me so.
16:50 masak (and I agree)
16:50 colomon really?
16:51 abra joined #perl6
16:52 masak yes. parentheses save lives.
16:54 colomon ng spectest passes (no parens)
16:55 colomon seems like Code should have a method version as well?
16:55 colomon Callable?
16:55 * holli fights to surpress a xkcd reference
16:57 holli a new quoting style for filenames? neat.
16:57 masak colomon: how do you figure that? why not just .() the Callable?
16:57 masak colomon: there's no corresponding subroutine eval for Callable.
16:57 holli is there a Perl6 book worth reading?
16:57 colomon ah
16:57 carlin holli: Now I'm curious, which xkcd? :-)
16:57 masak holli: it's not done yet, but yes. definitely.
16:58 masak holli: it's written for people like you.
16:58 holli http://xkcd.com/297/ of course
16:59 mberends not 224 for a change
16:59 carlin Ah, I was thinking 224
16:59 carlin mberends++ # heh
17:00 colomon ng spectest passes (parens)
17:00 * mberends likes 297 as well
17:01 carlin xkcd really is a great, so we'll forgive him for 353 ;-)
17:02 masak carlin: nothing to forgive, really. :)
17:02 TimToady I have more of a hard time forgiving xkcd for not realizing you can end a Perl program with a right parenthesis
17:02 masak just because Python has some positive qualities doesn't mean Perl is bad, or one shouldn't use Perl...
17:03 holli masak: i'm a good proofreader =)
17:03 masak holli: excellent.
17:04 mberends holli: right paren, on cue
17:04 carlin masak: the alt text
17:05 holli mberends: huh?
17:05 masak carlin: ah. that might be something that needs forgiveness. :)
17:06 colomon BTW, Any.eval pushed to ng.
17:06 mberends holli: TimToady asks for right parentheses and you type one before (probably) reading his remark
17:07 dalek rakudo/ng: 9d5018c | (Solomon Foster)++ | src/core/Any-str.pm:
17:07 dalek rakudo/ng: Add Any.eval with obvious definition.
17:07 dalek rakudo/ng: review: http://github.com/rakudo/rakudo/commit/9​d5018c94f854af23cf33b9846cb119184da8fb5
17:07 holli noh, i didnt even realize him talking
17:08 mberends synchronicity++
17:08 TimToady asynchronicity++ too
17:08 mberends fate, even
17:09 holli speaking of it. is there builtin multicore support in Perl6?
17:09 TimToady in the language, yes
17:09 holli like a parallellize opeartor
17:09 colomon holli: spec'd but not yet implemented
17:10 TimToady rakudo: say 1,2,3 »+« 4,5,6
17:10 p6eval rakudo 7f1c3f: 12756␤
17:10 TimToady hmm, precedence failure
17:10 TimToady rakudo: say (1,2,3) »+« (4,5,6)
17:10 p6eval rakudo 7f1c3f: 579␤
17:11 TimToady rakudo's parser still doesn't believe in list infixes, I guess
17:11 masak TimToady: Rakudo has that one right, right?
17:11 masak or not.
17:11 masak I thought all hyperops had the same prec as their non-hyper counterparts.
17:12 masak but perhaps that's just [] ops, not >> << ops.
17:12 TimToady hmm, you may be right, lemme look
17:12 colomon masak: seems to me like that would be the other way around?
17:12 colomon how can you stack [] ops?
17:12 masak colomon: stack?
17:12 TimToady rakudo: say 1,2,3 X 4,5,6
17:12 p6eval rakudo 7f1c3f: 141516242526343536␤
17:13 TimToady that one is right
17:13 colomon masak: have more than one of them in an expression.
17:13 colomon masak: in such a fashion that precedence would be an issue.
17:13 TimToady so you're right, rakudo does know about list infix, but >>+<< isn't one
17:13 TimToady (yet)
17:14 masak colomon: [+] 1, 2, [*] 3, 4
17:14 masak colomon: but no, precedence is not an issue there.
17:14 holli what was the name again of this language with all this math symbols as operators? the one that was never widely used because there were no keyboards for it
17:14 masak it all falls out naturally, because they're listops.
17:14 colomon whereas @a >>+<< @b >>*<< @c is perfectly sensible...
17:14 TimToady colomon: actually, pugs did implement the multithreaded hypers at one point
17:14 masak holli: APL?
17:14 masak holli: and there were keyboards for it. :)
17:15 colomon TimToady: really?  wonder if that still works?
17:16 TimToady holli: also, we have feed operators ==> and <== that work like pipelines for cell architectures
17:17 TimToady but those aren't implemented yet either
17:17 * Juerd wonders if they'll be used much
17:17 holli masak: yeah, that's it.
17:17 diakopter programs get hungry
17:18 TimToady but basically any lazy lists could be constructed by a separate processor, as long as there aren't bad dependencies (side effects)
17:18 masak pipes are another place where I have the feeling that the spec will mutate a bit before settling...
17:18 mberends masak: could you confirm that your OS X does *not* have /usr/include/malloc.h or /usr/local/include/malloc.h ?
17:19 colomon mine has /usr/include/malloc/malloc.h
17:19 diakopter /usr/include/sys/malloc.h
17:19 colomon but neither one you mention.
17:20 mberends ouch, that's the problem
17:20 colomon and what diakopter says, too.
17:20 diakopter but stdlib.h should cover it
17:20 mberends ah, good suggestion
17:21 diakopter otherwise, what colomon said
17:22 holli TimToady: why limiting yourself to a single machine? think cloud =)
17:23 holli btw cloud. i wnat perl6 on to be runnable in the google app engine =)
17:23 abra joined #perl6
17:23 masak holli: that's great! thanks for volunteering!
17:24 diakopter holli: do you mean hosted in one of the supported gae languages? or as its own supported langage?
17:24 TimToady holli: sure, but lazy lists have to keep ordering, even if calculated out of order; clouds tend to be better for map/reducey things
17:24 holli the latter preferably
17:24 mberends holli's head is in the clouds
17:25 diakopter holli: ask again when they support RealPerl
17:25 holli i think that will never happen.
17:25 diakopter I mean, Perl 5... oops, did I leak the new name?
17:25 masak "RealPerl"? is that the new nickname for Perl 5? :)
17:26 masak strangely, I find myself amused rather than offended.
17:26 TimToady as opposed to ComplexPerl? :)
17:26 diakopter it was meant to amuse, so.. .whew
17:27 TimToady ImaginaryPerl is a bit long to type
17:27 holli so that becomes iPerl
17:27 TimToady but I'd like both Perl 5 and Perl 6 to be real, so maybe NumPerl vs RatPerl
17:27 hydra35 joined #perl6
17:28 diakopter or RatPerl vs IrratPerl
17:28 * TimToady contemplates jokes on Rat.pack
17:29 diakopter Frat.pack; Brat.pack; Borat.pack
17:30 diakopter or
17:30 IllvilJa joined #perl6
17:30 holli how the fuck do i update the rakudo repository?
17:30 diakopter HerePerl vs LagPerl
17:30 mberends masak: to avoid a potentially futile edit/commit/revert, could you edit src/run.cpp, s/malloc/stdlib/ on the first occurrence (line 3) and see if 'make' compiles one file more? It works ok on Linux.
17:31 diakopter holli: not like that
17:31 masak mberends: sure thing. hold on.
17:31 colomon holli: did you get it using git?
17:31 holli yeah, but git --help tells me there is no update (<-- svn user)
17:32 colomon git pull
17:32 masak mberends: that works. next failure is in save.cpp. the same cause.
17:33 mberends masak: thanks, I'll patch 'em all. diakopter++
17:33 diakopter AGAIN WIN
17:33 masak mberends++ # great customer service
17:33 diakopter that's, like, twice in one year
17:34 holli grrmpff. thanks
17:34 * diakopter used to own a mac
17:35 diakopter then windows rehab ended, and I fell off the wagon
17:36 masak mberends: belated confirmation: I have /usr/include/malloc/malloc.h
17:38 diakopter std: ((())()((())()((((()))))))()()(()())(​(())()()(()()))((())(((((())()())))))
17:38 p6eval std 29379: ok 00:01 104m␤
17:38 diakopter I like that language
17:38 carlin I didn't know std could parse lisp
17:39 diakopter perl6: ((())()((())()((((()))))))()()(()())(​(())()()(()()))((())(((((())()())))))
17:39 p6eval pugs: *** Cannot cast from VList [] to Pugs.AST.Types.VCode (VCode)␤    at /tmp/g25doSyV16 line 1, column 2-26␤
17:39 p6eval ..elf 29379: Use of uninitialized value in subroutine entry at (eval 138) line 3.␤Can't use string ("") as a subroutine ref while "strict refs" in use at (eval 138) line 3.␤ at ./elf_h line 5881␤
17:39 p6eval ..rakudo 7f1c3f: invoke() not implemented in class 'Undef'␤in Main (file <unknown>, line <unknown>)␤
17:40 diakopter mildew: ((())()((())()((((()))))))()()(()())(​(())()()(()()))((())(((((())()())))))
17:40 p6eval mildew: Can't open perl script "mildew": No such file or directory␤
17:40 diakopter mildew-js: ((())()((())()((((()))))))()()(()())(​(())()()(()()))((())(((((())()())))))
17:40 p6eval mildew-js: Can't open perl script "mildew": No such file or directory␤
17:40 diakopter o
17:41 diakopter p6eval asciirot
17:42 rfordinal3643 joined #perl6
17:44 pmurias diakopter: i don't have access to the evalbot box anymore so i can fix that
17:46 skyheights joined #perl6
17:46 pointme joined #perl6
17:46 mubot joined #perl6
17:46 zaslon joined #perl6
17:46 TimToady botrot
17:54 holli rakudo: say qp{/usr}
17:54 p6eval rakudo 7f1c3f: Confused at line 2, near "{/usr}"␤in Main (file <unknown>, line <unknown>)␤
17:54 lumi joined #perl6
17:55 holli that should work, shouldnt it?
17:57 masak yes.
17:57 masak not implemented yet in Rakudo.
17:57 masak std: say qp{/usr}
17:57 p6eval std 29379: ok 00:01 106m␤
17:58 holli so how do i read a directory in rakudo?
17:58 masak you mean list all the files in it?
17:58 holli for example
17:58 masak I'd use something like qx{ ls } right now.
17:59 holli iiek!
18:00 masak 'ilek!'? was that an expression of approbation or distaste? :)
18:00 masak can't keep up with the slang nowadays...
18:00 holli more an expression of panic =)
18:00 masak don't panic :)
18:05 holli yeah, it's mostly harmless
18:07 masak 'The Guide is definitive. Reality is frequently inaccurate.'
18:08 justatheory joined #perl6
18:10 holli you gotta love that book.
18:11 holli i like the idea of this sprixel thing. beeing able to use perl6 for both, server and client side, would be definitly sexy.
18:12 holli combined with a neat widget library,,,
18:18 pugs_svn r29380 | mberends++ | [vill] portability change malloc.h to stdlib.h, masak++ diakopter++
18:18 masak \o/
18:18 * masak tries new, improved version of vill
18:19 * mberends crosses fingers and legs
18:20 masak awww. :(
18:20 masak it got to linking this time...
18:20 masak http://gist.github.com/260181
18:22 mberends ok, that's *very* different, not seen such before
18:23 mberends and yet it looks like one detail causes all 151 lines
18:24 masak that sounds like slightly good news...
18:24 mberends the LLVM config is not pointing to a lib directory properly
18:26 mberends that probably would need rebuilding LLVM with a different ./configure parameter
18:27 mberends .oO( where would OS X C++ expect to find lib dirs? )
18:28 masak no idea.
18:28 mberends the --prefix option must be different for OS X
18:28 TimToady I thought llvm didn't really support C++ yet...
18:29 mberends TimToady: it *uses* C++, but is only partly compiling C++
18:31 mberends their greatest achievement so far is a C compiler that adds -O4, optimization across linked code modules
18:32 dalek rakudo/master: 8dc1895 | masak++ | src/setting/Any-str.pm:
18:32 dalek rakudo/master: [Any-str.pm] added method form of eval
18:32 dalek rakudo/master: review: http://github.com/rakudo/rakudo/commit/8​dc18950efcde1b9af442709846859490a99ca4b
18:32 masak holli: there you go. if you hang around for half an hour or so, the rakudo evalbot will update, and you can have your method .eval :)
18:34 masak and now, I need to go home and fill up on nutrients. running low on blood sugar...
18:34 masak o/
18:35 cotto joined #perl6
18:36 diakopter pmurias: it's in pugscode svn
18:37 diakopter it, meaning p6eval botdna
18:39 diakopter pmurias: http://svn.pugscode.org/pu​gs/misc/evalbot/evalbot.pl
18:55 cottoo joined #perl6
18:56 holli seriously, it really takes a linguist to breed the idea of a "logical non negation" as an operator =D
18:56 holli masak: tx
19:00 holli rakudo: my $foo = 'foo { 1 + 2 } bar'; say $foo.eval
19:00 p6eval rakudo 7f1c3f: Method 'eval' not found for invocant of class 'Str'␤in Main (file src/gen_setting.pm, line 324)␤
19:00 holli p6eval sprixel my $foo = 'foo { 1 + 2 } bar'; say $foo.eval
19:04 mberends holli: try sprixel: ...
19:04 holli sprixel: my $foo = 'foo { 1 + 2 } bar'; say $foo.eval
19:05 p6eval sprixel 29380: NYI: term__S_scope_declarator␤
19:05 mberends sprixel's like that :(
19:05 holli makes me wonder, are the error messages spec'ed?
19:05 TimToady the ones in STD.pm are, more or less
19:06 TimToady to the extent that STD.pm is spec..
19:06 TimToady sprixel: $_ = 'foo { 1 + 2 } bar'; say $_.eval
19:06 p6eval sprixel 29380: NYI: term__S_variable␤
19:07 holli so the same syntax error is supposed to throw the same error in every implementation?
19:07 TimToady sprixel: say 'foo { 1 + 2 } bar'
19:07 holli s/same error/same message/
19:07 p6eval sprixel 29380: NYI: term__S_identifier␤
19:08 TimToady 'is supposed to' is not, strictly speaking, a defined phrase in Standardese... :)
19:08 holli btw, TimToady if you ever happen to be in Freiburg and need a couch, be my guest =)
19:08 TimToady but yeah, more or less :)
19:09 TimToady thanks...I think...
19:09 TimToady I'll have to figure out someplace to put the couch, though...
19:10 holli no need for scepticism. i just need someone to teach me some emacs tricks ^
19:10 holli ^^
19:10 colomon joined #perl6
19:11 subichan TimToady: my proposal for perl6: multi tape finite state transducers
19:12 TimToady I like the ^X^Cvi emacs command, myself  :P
19:14 subichan it would be neat and you know it ;)
19:14 * TimToady has a Pooh-brain that melts easily.
19:15 subichan i know i may be late for proposals but i tried
19:16 subichan it*
19:16 TimToady in what way would this not be possible in a Perl 6 module?
19:16 holli multi tape finite state transducer? sounds like a device from Star Trek (tm)
19:18 subichan everything is possible, but somebody's gotta code it ! as a computational linguist i loved perl, but I longed a lot for two-tape transducers. had to resort to other toolkits for that. now that two-tape is kinda old and perl6 is coming out then i thought that would be a nice thing to add with its own operator or something..
19:18 subichan .. as a multi-tape implementation
19:19 subichan that's been done only with prolog afaik.
19:22 rfordinal3643 left #perl6
19:22 subichan perhaps i'm just being too lazy and egoistical
19:25 holli i thought it's possible to invent your own operators in perl6.
19:26 subichan yeah but i wanted it to be official.. you know, like s/// and tr//
19:26 colomon holli: it certainly is possible.
19:28 holli ahhh, good old hybris ;-D
19:32 holli i'm just reading the wikipedia artictle about this tranducer thingie. my head aches.
19:32 holli too much symbols.
19:32 zloyrusskiy joined #perl6
19:33 TimToady subichan: speaking of official, what's the official spelling of your name? 子美チャン、かなあ。。。
19:34 TimToady or is it some other exotic language, like English?
19:35 holli TimToady: you should have called this Mu object Poo. so one could say "everything comes down to poo"
19:36 Su-Shee Mu is cool. I already hooked at least 3 japanologists on Perl 6 because of it.
19:37 holli .oO( writing perl6 code in kanji)
19:38 TimToady std: say 無
19:38 p6eval std 29380: Undeclared routine:␤  '無' used at line 1␤ok 00:01 106m␤
19:38 TimToady hmm
19:39 subichan it's a nickname lol so i wouldn't know how you should really spell it in japanese.. chan is the japanese suffix but the rest is really a nick.. so katakana i guess
19:41 TimToady well, generally 'bi' comes out 美 'beauty' in names
19:41 TimToady and 'su' could be any of various things
19:41 subichan haha, no this evolved out of something else
19:48 Psyche^ joined #perl6
20:01 iblechbot joined #perl6
20:26 vamped joined #perl6
20:35 vamped is there a preferential file suffix: .pl or .p6 ?
20:37 mberends vamped: .pl because the contents determine 5 or 6
20:37 vamped that's what I thought, but I'm getting my p5 and p6 code mixed up, so I thought maybe .p6
20:38 vamped (until perl6 will run perl5 code)
20:39 mberends use .p6 if you prefer, you'll just encounter minor friction occasionally
20:39 mberends rename when the interoperability lands
20:40 vamped well, maybe I had better conform to the one_true_way ;)
20:40 mberends lots of files in Unix have no extension, and rely on #!/something
20:41 vamped yes. that's another (GOOD) possibility. I think I will do it that way
20:41 nihiliad joined #perl6
20:43 y3llow_ joined #perl6
20:49 PerlJam for me .pl is still either "prolog" or "perl library"  (ergo, I tend to use .p6 if I use anything at all)
20:50 mberends phenny, tell masak an expert at #llvm said the --prefix=/usr *should* work on OS X as well. "probably something is expecting 32-bit code and using 64-bit or vice versa"
20:50 phenny mberends: I'll pass that on when masak is around.
20:51 mberends are macs schizophrenic?
21:03 mberends diakopter: ping
21:04 xomas joined #perl6
21:06 xomas joined #perl6
21:07 pugs_svn r29381 | lwall++ | [S02] rough in idea of moritz_++ to have separate type for convenience loopbacks
21:07 pugs_svn r29381 | also conjecturally put all "last-resort" multis there to see if jnthn++ freaks
21:38 colomon joined #perl6
21:57 meppl joined #perl6
22:06 astoria joined #perl6
22:19 justatheory joined #perl6
22:27 astrojp joined #perl6
22:43 MikeD joined #perl6
22:45 subichan joined #perl6
22:46 moritz_ TimToady++
23:03 lumi_ joined #perl6
23:41 colomon joined #perl6
23:48 meteorjay joined #perl6
23:59 * holli choses chrome as the one true browser

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

Perl 6 | Reference Documentation | Rakudo