Perl 6 - the future is here, just unevenly distributed

IRC log for #perl6, 2010-01-08

Perl 6 | Reference Documentation | Rakudo

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

All times shown according to UTC.

Time Nick Message
00:06 Hurley hi
00:06 sjohnson hi Hurley
00:07 Hurley if i compile the perl6 developer version. Can i use perl5 modules?
00:08 sjohnson i dont believe the use v5 translator is ready yet
00:12 frettled That's right, it's not a top priority.
00:13 frettled But as I understand it, it _is_ a priority, it's just that there are Perl 6 specific issues that are considered more important to fix first.
00:13 frettled Strange that they want to finish fiddling with the actual language before adding legacy sugar, eh?  ;)
00:19 sjohnson frettled: i suppose Perl 5 is something that'll be hard to say goodbye too :)
00:19 sjohnson to*
00:21 * sjohnson hugs perl 5.10 branch
00:41 dalek book: 8fa9944 | chromatic++ | src/roles.pod:
00:41 dalek book: Minor edits to the Roles chapter; some author notes.
00:41 dalek book: review: http://github.com/perl6/book/commit/8fa994464b73131f008fa28fd2040cbd62cbca17
01:25 cotto_work joined #perl6
01:27 Exodist joined #perl6
01:56 Hurley joined #perl6
01:58 nbrown joined #perl6
01:58 colomon joined #perl6
02:06 ShaneC joined #perl6
02:17 xomas_ joined #perl6
02:20 ShaneC joined #perl6
02:33 ShaneC joined #perl6
02:44 cognominal joined #perl6
03:12 agentzh joined #perl6
03:21 eternaleye joined #perl6
03:28 cognominal joined #perl6
03:39 simcop2387_ joined #perl6
03:48 simcop23- joined #perl6
03:56 Baggiokwok joined #perl6
04:07 eternaleye joined #perl6
04:22 am0c joined #perl6
04:27 justatheory_ joined #perl6
04:47 patspam1 joined #perl6
04:56 astinus joined #perl6
04:57 patspam joined #perl6
05:13 eternaleye joined #perl6
05:45 meppl joined #perl6
05:48 xinming_ joined #perl6
05:49 mberends commute &
06:08 pugs_svn r29475 | diakopter++ | [sprixel] better memoization & a bugfix.
06:15 pugs_svn r29476 | diakopter++ | [sprixel] cleanup old adhoc tests; add empty() and star() (greedy only for now) combinators
06:17 saschi g'morning!! ^^
06:19 diakopter saschi: hi
06:19 saschi hey there! :)
06:22 dalek book: d0103a2 | duff++ | src/builtins.pod:
06:22 dalek book: [builtins] minor typos
06:22 dalek book: review: http://github.com/perl6/book/commit/d0103a290547fd6b9be52a2592aeec2711b0996e
06:22 dalek book: 6ef4d99 | duff++ | src/preface.pod:
06:22 dalek book: [preface] suggested sentence change by eternaleye++
06:22 dalek book: review: http://github.com/perl6/book/commit/6ef4d998c08bf8db09c20ebb29f6f3cc6985811e
06:22 dalek book: c8151a8 | duff++ | src/grammars.pod:
06:23 dalek book: [grammars] add =begin/=end tags around code
06:23 dalek book: review: http://github.com/perl6/book/commit/c8151a87bb889ab2fe09416c9f633e015fe90437
06:30 sjohnson hi
06:40 pugs_svn r29477 | diakopter++ | [sprixel] implement character range() (all unicode codepoints supported, but the unicode range macros/definitions aren't yet implemented).  will probably steal them from xregexp.
06:54 pugs_svn r29478 | diakopter++ | [sprixel] begin to port the support for named pattern references/resolution (i.e. grammar declaration/construction)
07:07 mberends joined #perl6
07:11 mberends ooh, Sprixel is on the boil again :-)
07:11 uniejo joined #perl6
07:12 Su-Shee joined #perl6
07:12 eternaleye joined #perl6
07:13 justatheory joined #perl6
07:15 Su-Shee good morning
07:15 mberends gm Su-Shee, do you have any truffles for us today?
07:17 Su-Shee mberends: no, but I can give you my todo list which involved mopping kitchen and bathroom. ;)
07:19 * yath hands a roomba to Su-Shee
07:19 Su-Shee doesn't the thing just sweep and not mop? ;)
07:20 yath yes, but there's a, hm
07:20 yath there's something that mops, too
07:20 * Su-Shee needs a cleaning lady.
07:20 Su-Shee plain and simple.
07:20 yath and expensive
07:21 Su-Shee not really.
07:21 Su-Shee the time I have to work to pay her is much less then the time I waste doing household on my own.
07:21 yath hehe
07:22 Su-Shee no, really. simple calculation. two technians, two technician salaries > cleaning lady once a week ;)
07:28 sjohnson Su-Shee: do you do much Javascript stuff?
07:29 xinming joined #perl6
07:29 sjohnson was just thinking about how you also got that good book about the Good and Bad parts
07:29 sjohnson i bought it as I have to use JS at work :|
07:34 pugs_svn r29479 | diakopter++ | [sprixel] implement computeRefs and the first portion of Grammar.compile
07:35 kaare joined #perl6
07:43 saschi hey Su-Shee, sjohnson, yath, mberends ^^
07:45 Su-Shee sjohnson: yes, I do. And I like it.
07:47 sjohnson Su-Shee: did you find a good book that kind of gives a good tutorial of JS other than the "Definitive Guide"?
07:48 sjohnson i found it to be a bit lacking, even though it's a huge book
07:48 sjohnson saschi: hi
07:48 mberends hi saschi!
07:48 Su-Shee sjohnson: how do you mean exactly? there's javascript, the language and "handling DOM with javascript"
07:49 Su-Shee (and programming firefox, but that's really specific.)
07:50 sjohnson Su-Shee: something like a... camel book, or something close to it
07:51 sjohnson i keep hearing about jQuery maybe i should try it out
07:51 diakopter sjohnson: http://www.amazon.com/dp/047022780X/
07:51 Su-Shee well tutorial-wise, you usally go by subject you're interested in and use something from the web; js is really extremely well discussed. as an overview, a short reference should do. and there's a bunch of cookbook style books (I think, ORA has one)
07:52 Su-Shee sjohnson: that's a framework doing web widget, ajax and such for you.
07:52 rachelBROWN joined #perl6
07:52 diakopter sjohnson: the one I linked above is written by the author I respect the most when it comes to JavaScript.
07:53 sjohnson thanks u 2 will look into these
07:53 Su-Shee hasn't john resig written some book?
07:53 diakopter http://www.amazon.com/dp/193398869X/
07:54 diakopter http://www.amazon.com/dp/1590597273/
07:54 diakopter afk&
07:55 Su-Shee sjohnson: ^^^ there. that's the guy who has written jquery and implemented processing in js.
07:56 sjohnson thanks
07:56 sjohnson !
07:58 cognominal joined #perl6
07:59 cognominal joined #perl6
08:03 bryan[c1] joined #perl6
08:14 eternaleye joined #perl6
08:15 eternaleye joined #perl6
08:16 elkng joined #perl6
08:33 bcat joined #perl6
08:36 soupdragon joined #perl6
08:37 iblechbot joined #perl6
08:37 bbkr joined #perl6
08:42 partisan joined #perl6
08:59 JohnDoe365 joined #perl6
08:59 JohnDoe365 left #perl6
09:04 saschi i hope in webgl ^-^
09:20 dakkar joined #perl6
09:28 ihrd left #perl6
09:30 bryan[c1] joined #perl6
09:40 ashizawa joined #perl6
10:01 cognominal joined #perl6
10:11 payload joined #perl6
10:26 moritz_ good localtime() fellow #perl6 inhabitants
10:26 szbalint hai2u
10:38 Mantis joined #perl6
10:38 lmc joined #perl6
10:39 frettled rakudo: say localtime();
10:39 p6eval rakudo 3867ff: Could not find non-existent sub localtime␤in Main (file src/gen_setting.pm, line 324)␤
10:39 frettled :(
10:39 frettled ;)
10:53 moritz_ rakudo: say time
10:53 p6eval rakudo 3867ff: 1262947337.07978␤
11:36 patspam joined #perl6
11:38 rhuxton joined #perl6
11:53 lisppaste3 rhuxton pasted "Sorting a hash and splitting key/value at the same time" at http://paste.lisp.org/display/93141
11:53 rhuxton Hi all - just spent my first hour playing with rakudo and it feels like Christmas already :-)  Is there a "perlish" way to sort a hash and split key/value at the same time?
11:54 SmokeMachine joined #perl6
11:54 moritz_ rakudo: say (a => 3).kv.perl
11:54 p6eval rakudo 3867ff: ["a", 3]␤
11:55 moritz_ rhuxton: how do you want to sort?
11:55 moritz_ rakudo: my %h = (c => 1, b => 2, c => 3); say %h.sort».kv
11:55 p6eval rakudo 3867ff:  ( no output )
11:56 moritz_ don't know why p6eval doesn't like that
11:56 moritz_ rebuilding now, I think it should work
11:56 moritz_ $ perl6 -e 'my %h = (c => 1, b => 2, a => 3); say %h.sort».kv.perl'
11:56 moritz_ [["a", 3], ["b", 2], ["c", 1]]
11:57 moritz_ $ perl6 -e 'my %h = (c => 1, b => 2, a => 3); say %h.sort.map({ @(.kv)}).perl'
11:57 moritz_ ["a", 3, "b", 2, "c", 1]
11:57 hejki erm
11:57 hejki my rakudo does the first one right
11:57 moritz_ might be a problem with the evalbot :(
11:57 hejki oh
11:57 hejki :>
11:58 hejki btw.. >> > » :)
11:58 moritz_ anyway, that sorts the pairs in the hash, and then turns the pairs into lists
12:01 rhuxton Thanks moritz_ - I'm getting a "StopIteration" from rakudo after the first two pairs, but that's probably a buglet in the precise version of rakudo I grabbed. Off to read up about ">>" now :-)
12:01 moritz_ @array>>.method just calls method on every array element
12:02 moritz_ StopIteration is the less-than-awesome error message you often get when you iterate with two variables over an odd number of of elements
12:02 moritz_ rakudo: for 1, 2, 3 -> $a, $b { say "$a|$b" }
12:02 p6eval rakudo 3867ff: 1|2␤StopIteration␤in Main (file <unknown>, line <unknown>)␤
12:05 rhuxton Ah, then it's a bug in the precise version of rakudo I grabbed then (since the other ways of hash-sorting work fine). I'll do a "git pull" (if that's the equivalent of svn update) and try again later.
12:06 hejki rakudo: for 1..4 -> $a, $b { if 1 % ($a|$b) { say "foo" } }
12:06 p6eval rakudo 3867ff: foo␤foo␤
12:06 hejki dang.. typo :P
12:06 hejki rakudo: for 1..4 -> $a, $b { if 2 % ($a|$b) { say "foo" } }
12:06 p6eval rakudo 3867ff: foo␤
12:06 hejki najs
12:06 hejki really najs :)
12:08 pmurias joined #perl6
12:09 pmurias diakopter: hi
12:10 rhuxton [cue light-bulb going on sound-effect] Ah, the kv method is on Pair, as well as Hash.
12:10 moritz_ right
12:10 moritz_ there's even a .kv method on List
12:10 moritz_ rakudo: say <a b c>.kv.perl
12:10 p6eval rakudo 3867ff: [0, "a", 1, "b", 2, "c"]␤
12:11 moritz_ arrays have keys also
12:11 moritz_ usually called index
12:11 rhuxton OK, Next question: how do I find what methods Pair (or any other class) support? Happy enough using ack to rummage through source if necessary.
12:11 moritz_ ack is good :-)
12:11 moritz_ rakudo: say Pair.^methods».name.join(' ')
12:11 p6eval rakudo 3867ff: perl ACCEPTS get_string Scalar fmt key value pairs kv atan ord acos :l eval chop tanh int atanh cosec :s chars roots uc reverse keys isa ucfirst tan log10 substr cos bytes pick fmt evalfile acosec join sinh asinh cosech trim asec chr acotanh floor sech asech rand truncate sort round
12:11 p6eval ..rinde…
12:12 moritz_ rakudo: say Pair.^methods(:local)».name.join(' ')
12:12 p6eval rakudo 3867ff: value pairs kv perl ACCEPTS get_string Scalar fmt key␤
12:16 rhuxton Aha! Introspection. Good :) What's the caret mean there? I know of its use in regexps, closure-variable-thingummies and junctions.
12:17 moritz_ it means "call a method on the meta class"
12:21 rhuxton Because "Pair" itself returns an instance of class Pair rather than an object representing the class, presumably.
12:22 moritz_ rhuxton: no, because introspection is handled by the meta class
12:22 moritz_ rakudo: say ~(a => 3).^methods(:local)».name
12:22 p6eval rakudo 3867ff: get_string Scalar fmt key value pairs kv perl ACCEPTS␤
12:22 moritz_ that's with a Pair object, and it looks exactly the same
12:24 rhuxton Hmm - time to restart the reading / playing cycle I think :-)
12:24 rhuxton moritz_: thanks very much - most helpful.
12:24 moritz_ rhuxton: you're welcome. And feel free to continue asking
12:26 ruoso joined #perl6
12:34 payload joined #perl6
12:50 mariano__ joined #perl6
12:53 Su-Shee if I role-d something into a class with a role and I inherit this class, do I get the roled in stuff with it or would I have to role in again?
12:53 moritz_ you get the composed methods an atributes
12:54 moritz_ composing a role is a lot like writint these methods and attributes in the class directly
12:56 Su-Shee assume, I did a role somethere in class A, inherited it and now I'm including the same role in Class F (which already inherited everything) - are role methods and attributes overwritten? can they be overloaded?
12:57 Su-Shee will I get some error like "Role XX already included" or something?
13:01 rodi joined #perl6
13:03 moritz_ the spec says something along the lines "a role may never conflict with itself"
13:05 ignacio_ joined #perl6
13:11 Baggiokwok joined #perl6
13:14 takadonet morning all
13:26 cognominal joined #perl6
13:27 cognominal joined #perl6
13:34 fridim_ joined #perl6
13:40 mathw heh
13:40 mathw I just wrote some C++ code using the >> operator
13:40 mathw keep thinking 'that's not right hyper syntax'
13:42 takadonet hehe
13:43 mathw almost wrote a hyper method call at one point
13:45 takadonet I have not written any Perl 6 code in my Perl 5 yet....
13:46 takadonet Sometime I notice that a few places that a junction would be great right about now in my code
13:47 mathw :)
13:47 mathw I do that too
13:47 mathw once you've used any() a couple of times in an if condition...
13:52 payload joined #perl6
13:55 pmurias joined #perl6
14:08 rodi use Quantum::Superpositions # any() in Perl5, inspiration for any() in Perl6 IIRC.
14:09 soupdragon joined #perl6
14:09 mathw of course I could write myself an any() in C++
14:13 elmex joined #perl6
14:17 rhuxton left #perl6
14:18 PerlJam rodi: you mean "use Perl6::Junctions" these days Q::S is old
14:19 flip214 joined #perl6
14:19 flip214 Hello, short question again ...
14:21 rodi PerlJam++: mm, shiny :)  s/Junctions/Junction/ .
14:21 flip214 Is there a way to make a function signature that allows some parameter with a specified class (sub x(myClass obj))
14:21 flip214 which then accepts *only* this class, and not any derived?
14:21 flip214 But maybe that doesn't make sense ... or does it?
14:22 flip214 If an inherited class is passed as parameter, is it "dumbed down" to the lower-level class object before being passed?
14:22 flip214 I feel confused ...
14:22 mathw standard OO polymorphism
14:22 mathw it'll accept an instance of any derived type
14:22 mathw you could use a where clause to do an exact check, I think
14:22 PerlJam flip214: sub foo (my Thingy $x where { .isa(Thingy) })  { ... }
14:22 mathw but it should be very rare to need such a thing
14:23 mathw as it's an indicator of a dodgy OO design
14:23 PerlJam flip214: what mathw said.
14:23 flip214 thanks
14:23 flip214 perljam: but wouldn't that check the lower-level object already?
14:24 flip214 So that sub foo($x where { .isa(myClass) }) would work? Avoids the repetition, too.
14:25 PerlJam yeah, that would work.
14:26 PerlJam not sure about the order of the checks though; specifying the nominal type may avoid the constraint check entirely when you accidently pass some non-MyClass object.  But whatever ... that's probably a premature optimization  :)
14:28 PerlJam Ha!  from #perl ...
14:28 PerlJam < squeeky> perl6 actually exists. "cloud computing" is exactly what a cloud is. vapor.
14:29 flip214 Well, was just some thought. Thank you very much!
14:29 flip214 left #perl6
14:29 PerlJam Clearly we are making progress when people start saying "perl6 actually exists"  :)
14:29 mathw yes
14:29 mathw although without context it's impossible to tell if it was sarcastic
14:30 mathw I'm assuming it wasn't
14:30 PerlJam It wasn't I don't think.  The text leading up to the statement had me all set to defend Perl 6 as not being vapor, but then he went in the unexpected direction to claim that it was cloud computing that was vapor
14:30 colomon_ joined #perl6
14:31 takadonet morning colomon
14:31 mathw cool
14:32 mathw For someone to say that in #perl who isn't one of us lot is definitely a step forward
14:32 soupdragon one of us! one of us!
14:32 rgrau_ joined #perl6
14:34 payload joined #perl6
14:36 rodi left #perl6
14:51 douglashunter joined #perl6
15:08 soupdragon joined #perl6
15:12 kaare_ joined #perl6
15:12 payload joined #perl6
15:14 soupdragon joined #perl6
15:21 mariano__ joined #perl6
15:21 soupdragon joined #perl6
15:34 perlsyntax joined #perl6
15:35 perlsyntax Does anyone know when perl 6 comeing out?
15:36 jferrero joined #perl6
15:36 Gothmog_ They usually say it comes christmas.
15:37 perlsyntax that long i thought it was this christms.
15:37 perlsyntax thanks
15:39 perlsyntax left #perl6
15:39 iblechbot joined #perl6
15:39 PerlJam blah
15:39 Baggiokwok joined #perl6
15:41 ash_ joined #perl6
15:45 Psyche^ joined #perl6
15:48 justatheory joined #perl6
15:51 pugs_svn r29480 | masak++ | [S12-objects] various consistencifications
15:51 pugs_svn r29480 |
15:51 pugs_svn r29480 | - one case of 'enum type' which should be 'enumeration type'
15:51 pugs_svn r29480 | - the C<.true> method is now called C<.so>, according to S32
15:51 pugs_svn r29480 | - so is the function
15:51 pugs_svn r29480 | - s/an an/an/
15:51 pugs_svn r29480 |
15:51 pugs_svn r29480 | [S32] removed comment about musings about .true and .defined
15:51 pugs_svn r29480 |
15:51 pugs_svn r29480 | The musings are actually addressed in some of the above corrected
15:51 pugs_svn r29480 | paragraphs in S12.
16:00 payload joined #perl6
16:01 ignacio_ joined #perl6
16:11 cognominal joined #perl6
16:12 Levex joined #perl6
16:15 Cyclops3590 joined #perl6
16:21 hercynium joined #perl6
16:25 ignacio_ joined #perl6
16:28 pmurias joined #perl6
16:37 TimToady PerlJam: um, .isa is not going to an exact type check either, if that's what flip214 really wanted...
16:37 Levex left #perl6
16:38 PerlJam oh, I guess not.
16:38 PerlJam that a was pre-caffienated PerlJam though.  He's crazy.
16:42 ash_ to do an exact match, you could check the object's WHAT
16:43 moritz_ right
16:43 moritz_ $obj.WHAT === ::Type
16:44 pugs_svn r29481 | lwall++ | [S12] extra tweak to masak++'s tweak, now that "so" is less confusable as a value.
16:46 TimToady std: m:5x(42)//
16:46 p6eval std 29480: [31m===[0mSORRY![31m===[0m␤Missing term at /tmp/uc8bkaITvF line 1 (EOF):␤------> [32mm:5x(42)//[33m⏏[31m<EOL>[0m␤    expecting any of:␤      prefix or term␤   standard stopper␤ term␤   terminator␤       whitespace␤FAILED 00:02 110m␤
16:48 pmurias joined #perl6
16:53 pmurias TimToady: should things in STD_P5.pm hava a p5 prefix? like p5term?
16:57 TimToady pmurias: currently termish renames p5term to term; probably so that method EXPR doesn't get confused
16:58 TimToady in general, it depends on how much we want to reuse the p6 code
16:59 pmurias TimToady: -e '1' doesn't work with STD_P5, is it expected?
17:01 TimToady STD_P5 is only partially translated from a p6 parser to a p5 parser.  I wouldn't expect anything to work yet...
17:09 colomon_ joined #perl6
17:10 uniejo joined #perl6
17:14 uniejo joined #perl6
17:17 colomon_ takadonet: apologies for not returning your hello this morning.  I started to and then I got disconnected from the server and my software didn't want to reconnect.  And then I took a long nap because I am sick.
17:17 takadonet colomon_: np man
17:17 takadonet was only mad for a few hours :P
17:17 colomon_ :p
17:18 jackyf joined #perl6
17:19 diakopter pmurias: hi
17:20 pmurias diakopter: rehi
17:20 diakopter what's up
17:20 rodi joined #perl6
17:21 pmurias diakopter: what would i need to make mildew-js use sprixel for rules?
17:22 diakopter heh
17:22 Tene I tried commenting on a post on blogs.perl.org
17:22 Tene I got: Your comment submission failed for the following reasons: Text entered was wrong. Try again.
17:22 Tene That's... great.
17:23 TimToady MLTA
17:23 diakopter captcha?
17:23 diakopter pmurias: not much, probably; but *I* would have to do a lot (still) ;)
17:23 Tene http://blogs.perl.org/users/ovid/2010/01/cool-things-in-perl-6.html -- I don't see a captcha...
17:24 moritz_ Tene: do you have javascript activated?
17:24 Tene moritz_: yes.
17:25 diakopter wow, that's utterly broken.
17:25 Tene I was going to comment on not needing a closure, as 'where :f' is sufficient.
17:25 diakopter I click "Sign in", and it loads a new page, but the back button is grayed out.
17:26 diakopter and there's no "cancel" button/link on the sign-in page.
17:26 Tene I just tried to comment anonymously.
17:28 takadonet I love that the members of the community always want to post useful comments on anything that relates to Perl 6
17:28 PerlJam Hmm. I wonder if subset Filename of Str where :f or die;  # works.   I would guess that it does not.
17:28 moritz_ being logged in it seems that I can comment
17:28 PerlJam (or rather it "works" but doesn't do what I intend :)
17:29 Tene where is :f documented anyway?
17:29 moritz_ in S32/IO somewhere, probably
17:29 moritz_ but it's outdated anyway
17:29 uniejo_ joined #perl6
17:30 moritz_ it should .IO.f or where .IO ~~ :f or so
17:30 * takadonet waiting for someone to post a comment...
17:30 moritz_ rakudo: subset Filename of Str where :f; my Filename "/etc/password/";
17:30 p6eval rakudo 3867ff: Malformed declaration at line 2, near "Filename \""␤in Main (file <unknown>, line <unknown>)␤
17:30 Tene the ~~ :foo still seems a bit sketchy to me.  Is that Pair.ACCEPTS ?
17:30 TimToady moritz_: did you ever see the comment on your Perl 6 in 2009 blogging?
17:30 moritz_ TimToady: yes
17:31 moritz_ Tene: it calls the 'foo' method on $_
17:31 Tene moritz_: no, that's .foo, not :foo
17:31 Tene yes?
17:31 moritz_ ... and does something with the return value
17:32 moritz_ XOR or so
17:32 moritz_ probably S03, smartmatch table
17:32 moritz_ $a ~~ :b($c)   is something like   $a.b() !xor $c
17:34 cdarroch joined #perl6
17:34 ash_ PerlJam: I think someone on the mailing list was tlaking about Filename as a subset of Str like that, but they were using fail i think, not die
17:34 PerlJam ash_: that was me :)
17:35 ash_ oops, my bad
17:35 moritz_ rakudo: subset Filename of Str where :f; my Filename $x =  "/etc/password/";
17:35 p6eval rakudo 3867ff:  ( no output )
17:35 moritz_ rakudo: subset Filename of Str where :f; my Filename $x =  "/etc/asdfasödf";
17:36 p6eval rakudo 3867ff:  ( no output )
17:36 moritz_ so it doesn't seem to work
17:36 * moritz_ aborts commenting
17:37 Tene I've worried a little bit a few times about the length of the smartmatch table contradicting the stated goal of p6 to avoid memorizing arbitrary lists.
17:37 TimToady the point is to put all the memorizing into one spot
17:38 PerlJam and most of that table is "obvious"
17:38 PerlJam the thing that it does is the thing you'd expect given the two sides of ~~
17:39 TimToady and given that you don't expect symmetry, *cough*
17:39 PerlJam well, that's part of the non-obvious :)
17:40 TimToady and given that you don't expect lists to automatically any-fy themselves
17:40 Tene Any ~~ Pair isn't quite obvious to me, I guess.  I'd expect to do method smartmatches with *.foo, not :foo
17:40 Tene So I'm not quite convinced of the utility, but maybe I should think on it more.
17:41 TimToady the :foo form is more easily negated, and arguably looks simpler in a junction
17:41 * Tene nods.
17:42 TimToady though a junction of closures oughta work, I suppose
17:42 Tene Not a real objection at all, just thinking out loud.
17:42 TimToady and p5 programmers are expecting a special form for filetests anyway
17:43 TimToady it's easy to translate -f to :f in your head
17:45 Tene :foo is :foo(1) or :foo(Bool::True) ?
17:45 PerlJam yes  :)
17:45 Tene :P
17:45 moritz_ True
17:45 moritz_ not 1
17:46 moritz_ .oO( but not 1 is False, so that answer wasn't very helpful :-)
17:46 PerlJam I'm probably going to write :foo(0) before I write :foo(False) though
17:46 PerlJam (see? I just did it)
17:46 Tene PerlJam: !:foo is standard for negated, iirc.
17:47 PerlJam Tene: right.  I was just commenting on the :foo() form.
17:49 TimToady :!foo, not !:foo
17:49 TimToady then there's :0foo
17:50 PerlJam but that's :0fun
17:50 PerlJam :)
17:51 FCO joined #perl6
17:51 Tene http://www.reddit.com/r/programming/comments/an337/cool_things_in_perl_6/c0ifk1y?context=1
17:51 TimToady heh, the specs are still full of :by
17:52 xinming_ joined #perl6
17:53 PerlJam :fun  :nofun  ...  could we get some magic there?  (that'll probably be one of the grammar mods soon after people really start using Perl 6 if not)
17:54 ignacio_ joined #perl6
17:55 TimToady what's wrong with :!fun ?
17:55 diakopter :frap
17:55 TimToady and do you really want people to get confused by things like :none
17:55 TimToady "don't go northeast?"  O_o
17:55 PerlJam nothing at all, but someone always seems to introduce a "no" prefix to negate something
17:56 TimToady someone is always wrong
17:56 TimToady negation should not be inband with identifiers
17:56 PerlJam inventory?
17:57 PerlJam you have: no tea
17:57 PerlJam It's all douglas adams' fault  :)
17:57 moritz_ should have used a Bag
17:57 TimToady I'd rather have a Tea Set
17:58 tylerni7 joined #perl6
17:59 TimToady actually, at the moment I'd rather have a Hash
17:59 takadonet need more people upvoting Tene comment on reddit!
17:59 diakopter I was going to say, Tea Pot and Tea Hash seem redundant
17:59 TimToady well, why didn't you?
17:59 diakopter I guess we'll never know
17:59 huf it's better if you blend the two, so no, not redundant
18:00 huf but the tea bit is confusing me
18:00 cognominal I am using the parrot/tools/dev/create_language.pl but want to use nqp instead of perl 6. I tweaked the generated Makefile but parrot does not want to compile the perl6 generated pir file :  /Users/stef/git/rakudo/parrot_install/bin/parrot -o psal.pbc src/Psal.pir
18:00 cognominal error:imcc:syntax error, unexpected PREG, expecting '(' ('$P40')
18:00 cognominal any idea what I am doing wrong
18:00 diakopter probably something in what you're typing
18:01 moritz_ cognominal: uhm, I thought create_language.pl defaulted to NQP?
18:01 cognominal moritz_, that's my point
18:01 moritz_ cognominal: so you want perl6 instead of nqp??
18:01 moritz_ s/'??'/?/ :-)
18:02 cognominal but why a pir file generated by perl6 would not be acceptable to parrot?
18:02 moritz_ because rakudo needs some dynops
18:02 moritz_ that need to be loaded
18:03 cognominal moritz_, hum, why that would explain a syntax error?
18:03 lmc joined #perl6
18:04 Tene cognominal: Parrot questions are more appropriate in #parrot on irc.perl.org
18:04 cognominal I want perl6 because I can't remember the used subset. Also I am just refactoring code that worked before using the create_language tool chain
18:05 moritz_ cognominal: IMCC knows about operators. If it doesn't know about dynops => syntax error
18:05 Tene cognominal: you need to make sure that rakudo is loaded before the generated PIR is parsed.
18:05 Tene put "load_language 'perl6'" in a :immediate sub at the start of the file.
18:06 cognominal thx, I will try it
18:07 cognominal #perl6 and #haskell are probably both the most helpful channels on freenode :)
18:08 cognominal btw I am dabbling with some pir preprocessor  :)
18:11 diakopter don't burn your navel into your retinas
18:11 diakopter cognominal: sry; that wasn't directed at you
18:12 cognominal is that a pun?  :)
18:12 sirrobert joined #perl6
18:12 diakopter a quip
18:13 rachelBROWN joined #perl6
18:15 * Su-Shee burns indian spices for dinner and reads snow catastrophy warnings.
18:15 Su-Shee good evening all.
18:15 takadonet Su-Shee: evening
18:18 rodi In S02, what should I infer about the inclusion of "Widget" on the list of Undefined types?
18:18 stephenlb joined #perl6
18:18 moritz_ no idea
18:19 moritz_ probably "it was Cool at that time" SCNR :-)
18:20 rodi LOL
18:20 rodi Cool is on the list right before it, though :)
18:28 PerlJam rodi: Just that you can get a typed but undefined value for any class in that manner I think
18:29 PerlJam rodi: (no, there is no official Widget class  :)
18:30 cognominal I asked TimToady :  <cognominal>TimToady, btw, S02 mentions the class Widget twice without ever saying what it is
18:30 cognominal his answer: widgets: http://www.perl.com/2003/07/16/graphics/slide72.jpg
18:30 Tene That's such a useful answer.
18:30 rodi hahahahaha, thanks cognomial
18:30 PerlJam escher++
18:31 Chillance joined #perl6
18:32 cognominal your master will answer clearly when you will have progressed in the path on enlightment. But then, you will need no master.
18:35 cognominal http://www.perl.com/pub/a/2003/07/16/soto2003.html?page=7  # for more information...
18:38 cognominal I realize that my pir preprocessor in perl6 is just that :  http://www.perl.com/2003/07/16/graphics/slide65.jpg
18:39 PerlJam That's just the normal bootstrap picture. :)
18:39 mj41 joined #perl6
18:41 PerlJam people need to be reminded of this one today: http://www.perl.com/2003/07/16/graphics/slide61.jpg
18:42 ShaneC joined #perl6
18:45 cognominal "But what you really want to know about is the future state of Perl. That's nice. I don't know much about the future of Perl. Nobody does. That's part of the design of Perl 6. Since we're designing it to be a mutable language, it will probably mutate."
18:45 cognominal Only change is for sure. Let's explain that to people who want a stable perl 6
18:47 rodi I sure like the semantics of KeySet.
18:48 diakopter tbf, that's a poor excuse. while *want*ing isn't a bad thing (demanding is), holding accountable to [poorly set/managed] expectations isn't always bad. just because one might want to adjust the expectations now doesn't mean they weren't mis-aligned in the past.
18:48 cognominal tbf?
18:48 diakopter to be frank
18:49 diakopter "you are not allowed to have expectations. any expectations you may have had were misguided. nothing is certain. stop asking until we say it's okay"
18:51 goksie joined #perl6
18:51 rodi "the design process will be spiral, cooperative, and convergent. The rate of convergence is an emergent property, and cannot be forced, only encouraged."
18:52 cognominal that's S01?
18:52 diakopter those are very different messages
18:58 diakopter "Only change is for sure. Let's explain that to people who want a stable perl 6" is caustic and communicates nothing but derision, futility, scorn, apathy, but most of all, _impunity_.
19:00 PerlJam Hmm.
19:01 PerlJam I think your projecting a tad much of yourself into those sentences.
19:01 PerlJam s/your/you're/
19:02 diakopter if that's true, it buttresses "it takes one to know one" :/
19:03 PerlJam (or, put another way, I don't see derision, scorn, or apathy at all)
19:03 diakopter oh
19:03 diakopter I wasn't describing the denotation. I was describing the possible connotations.
19:04 cognominal designing for change is just typical perl hubris...
19:04 PerlJam Well, futility is something that most humans are going to need to learn anyway  :)
19:05 IllvilJa joined #perl6
19:06 PerlJam diakopter: Sure, those are all possibilities ... it all depends on what baggage the reader brings with them as to how they interpret the text.
19:07 diakopter I agree. I think it's important to consider the reader's/hearer's baggage.
19:09 diakopter inasmuch as it communicates futility & impunity, it can sting with derision/scorn/apathy.
19:11 rgrau_ joined #perl6
19:12 PerlJam It's a slippery fish no matter how you look at it.   You should consider the reader, but you can't be responsible for their take on your text.  They have to own up to their interpretation and you have to own up to possibly not being clear.  So then you dialog.
19:13 cognominal now I crash parrot :(
19:13 cognominal at least load_language was the right thing
19:13 diakopter I'm not claiming that a communicator is entirely responsible for the interpretation, but certainly largely.
19:14 diakopter but "Only change is for sure." is just plain wrong
19:14 cognominal good thing is that it is in load_language...
19:14 cognominal diakopter:  plain wrong? I thought it as a tautology :)
19:14 diakopter "only change is for sure" was true only prior to the first Apocalypse.
19:15 nihiliad joined #perl6
19:16 PerlJam depends on what you're talking about :)
19:17 diakopter what are you talking about?
19:17 cognominal what I was talking about? :)
19:18 diakopter which was?
19:18 cognominal forgot the incantation to say to gdb wher to search for the sources :(
19:21 PerlJam If you look at the world through the lens of Perl 6 development, "only change is for sure" was true well after the first Apocalypse.
19:26 mberends joined #perl6
19:41 rodi Another spec question: In S32/Containers, what can I infer from the "junction" heading being lowercase, when the rest the classes in that list are C<ucfirst>?
19:42 PerlJam rodi: a typo?
19:42 * PerlJam looks
19:43 * diakopter something about a trout's lap
19:43 PerlJam rodi: where exactly?
19:44 rodi PerlJam: http://perlcabal.org/syn/S32/Containers.html#junction
19:45 PerlJam yeah, I'd go with typo
19:46 rodi Okay, TimToady usually fixes those, I guess?
19:46 PerlJam rodi: do you have a commit bit for the pugs repo?
19:46 rodi Yeah.
19:47 rodi Or at least, I used to.
19:47 PerlJam then you can fix it.
19:47 PerlJam If it's wrong, someone will fix your fix :)
19:47 rodi :)
19:50 ash_ isn't that normally reverting a change?
19:50 ash_ fixing a fix sounds recursive
19:58 rachelBROWN joined #perl6
19:58 pugs_svn r29482 | rodi++ | C<ucfirst(junction)> with apologies if it was that way for a good reason.
20:03 ignacio_ joined #perl6
20:03 rodi In another spec question, I am surfing between S02 and S32/Containers.  KeySet and KeyBag are on a list of Mutable types in S02, defined in terms of KeyHash.  KeyHash is not on that list, but has a very nice description in S32/Containers (as do KeySet and KeyBag).  Is KeyHash left off the S02 list by intent?
20:06 rodi Oh, and did I mention I like the semantics of KeySet? :)
20:07 cognominal joined #perl6
20:09 somebody_ joined #perl6
20:09 rodi I like it so much, I wish I could have a KeySet of KeySet.
20:09 saschi hi
20:09 mberends hi
20:09 soupdragon joined #perl6
20:10 PerlJam rodi: I think it was an error of omission WRT KeyHash in S02
20:22 patspam joined #perl6
20:23 hercynium joined #perl6
20:24 tarski joined #perl6
20:26 avar joined #perl6
20:26 literal joined #perl6
20:27 tarski left #perl6
20:33 pugs_svn r29483 | mberends++ | [vill] added and updated many files, but still not generating code properly
20:50 dugg mberends: does using viv + STD mean that vill will get STD's warnings and error messages by default?
20:51 mberends dugg: yes, they are emitted to stderr, no matter what you do with the viv output.
20:52 dugg mberends++;  # sweet
20:57 bbkr joined #perl6
21:05 cj_ joined #perl6
21:12 perimosocordiae joined #perl6
21:12 cj_ left #perl6
21:12 |JackYF| joined #perl6
21:14 perimosocordiae joined #perl6
21:15 tylerni8 joined #perl6
21:18 * dugg suspects his ridiculously old laptop might take a minute to build llvm 2.6
21:21 ash_ the llvm is pretty big
21:22 mberends dugg: building llvm can take over 30 minutes on a young laptop, especially when configuring without the --enable-targets and --enable-optimized options, see file:///usr/share/doc/llvm-doc/html/GettingStarted.html#config
21:24 mberends --enable-targets=host-only is the biggest time saver
21:24 Tene http://www.reddit.com/r/programming/comments/an337/cool_things_in_perl_6/c0ifzfu
21:24 Tene -- "anybody knows a good, short introductory text on Perl 6 for Perl 5 programmers?"
21:25 Su-Shee Tene: are you just quoting or asking?
21:26 mberends http://perlgeek.de/en/article/5-to-6 by moritz_++
21:26 Tene Su-Shee: quoting.
21:27 Su-Shee ah, ok.
21:27 Tene mberends: thanks
21:27 Su-Shee otherwise I would have said the same as mberends :)
21:27 saschi need tomate soup
21:28 saschi eh, tomato >_<
21:30 xinming joined #perl6
21:33 rodi std: my KeyHash of KeyHash %foo; #what does this mean?
21:33 p6eval std 29483: ok 00:01 108m␤
21:33 rodi std my KeySet of KeySet %foo; # I meant this, not the previous.
21:34 rodi std: my KeySet of KeySet %foo; # I meant this, not the previous.
21:34 p6eval std 29483: ok 00:01 106m␤
21:34 rodi specifically, is this something which parses, but is nonsensical, or something that has semantics that I just can't grok?
21:37 ive joined #perl6
21:38 dugg mberends++: yeah, took that from your nice README
21:38 mberends :) glad somebody reads docs
21:44 ash_ docs, what are those? i am not sick
21:44 ash_ </sarcasm>
21:44 mberends it's where jnthn++ hides his passwords ;-)
21:56 dbrock joined #perl6
22:06 japhb joined #perl6
22:13 dugg compiled llvm 2.6 in 45 min 36 sec.  *whew*
22:15 ash_ sweet, did you build all the bindings? i like how it comes with C/C++ and OCaml bindings, i know very few people that know ocaml, seems odd
22:17 dugg ash_: I didn't, it was a basic fulfillment of a vill requirement, and I followed the "build as fast as you can" style configure options for it from the vill/README.
22:18 ash_ ah, well, as long as you have what you need then, i just like playing with programming languages, ocaml is interesting... syntax is odd though
22:18 dugg Totally!  I'll blame you when no yaks in my area have hair.
22:19 dugg mberends:  In vill/Makefile I needed the make clean target to have:
22:19 dugg -       @find . -executable -type f -exec rm {} ';'  # er, executables?
22:19 dugg +       @find . -perm +x -type f -exec rm {} ';'  # er, executables?
22:20 dugg at line 363 or so.
22:20 dugg My `find` didn't like -executable.
22:20 mberends dugg++: the testing has all been Linux based, this feedback is useful :)
22:21 dugg ash_: They will have been shaved with Ocaml's razor [grin]
22:22 ascent joined #perl6
22:26 pugs_svn r29484 | mberends++ | [vill] 'make clean' tweak from dugg++, another TODO in link_codegen.cpp
22:27 ascent_ joined #perl6
22:27 pmurias joined #perl6
22:29 constant joined #perl6
22:30 colomon_ joined #perl6
22:42 wayland76 I'm thinking about parallelism, and wondering if we've considered http://en.wikipedia.org/wiki/OpenCL
22:45 TimToady not directly, but the reason we have hypers is to support such coprocessing easily; likewise feed operators are for cell processors really
22:46 TimToady hypers are a natural for GPUs, though
22:46 TimToady especially once we get compact arrays of native types
22:49 pugs_svn r29485 | lwall++ | [STD] disallow :!foo(0) and :5bar[42] as suggested by masak++
23:10 dukeleto joined #perl6
23:28 mariano__ joined #perl6
23:40 orafu joined #perl6
23:56 pugs_svn r29486 | lwall++ | [S02,S09,S32] get rid of :by fossils

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

Perl 6 | Reference Documentation | Rakudo