Perl 6 - the future is here, just unevenly distributed

IRC log for #perl6, 2006-01-07

Perl 6 | Reference Documentation | Rakudo

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

All times shown according to UTC.

Time Nick Message
00:10 BenQ joined perl6
00:11 BenQ who speek english :))
00:11 Juerd_ Who does not.
00:13 BenQ left perl6
00:26 dduncan I just sent an email post to both dbi-users and poop-group titled "RFC: renaming SQL::Routine out of SQL"; any feedback there is appreciated
00:27 azuroth hmm. do it!
00:36 svnbot6 r8596 | putter++ | misc/JavaScript-FrontEnd/Grammar.pm - Corrected typo.  The file now parses with current pugs and parrot 0.4.0.  But being untested, it is, of course, very unlikely to actually work.
01:04 _SamB_ joined perl6
01:05 SamB is now known as _SamB_
01:09 freakman joined perl6
01:09 Cryptic_K joined perl6
01:10 freakman left perl6
01:14 svnbot6 r8597 | lwall++ | strRange thought "aa" < "z" so string ranges didn't wrap right.
01:14 svnbot6 r8596 | putter++ | misc/JavaScript-FrontEnd/Grammar.pm - Corrected typo.  The file now parses with current pugs and parrot 0.4.0.  But being untested, it is, of course, very unlikely to actually work.
01:36 putter joined perl6
01:39 putter audreyt: would you be ok with my dropping a copy of the 25MB 2000+ file mozilla/js/tests in misc/JavaScript-FrontEnd?  A bit large, but it seems like the front-end might eventually make a useful stress test for rules and oo.
01:39 putter rules now, oo later
02:08 putter joined perl6
02:08 svnbot6 r8598 | putter++ | The crude beginnings of a JS frontend test suite.  Requires mozilla_js_tests, not yet checked in.
02:11 putter audreyt/etal: Bug report: with parrot-0.4.0 and today's pugs, cd misc/JavaScript-FrontEnd; ../../pugs -I .   and use Grammar;,
02:13 putter err, never mind. :/
02:14 putter Is embedded parrot known to be broken, src/Pugs/Eval.hs:852:12-52: Irrefutable pattern failed for pattern (Pugs.AST.Internals.App (Pugs.AST.Internals.Var name'), or is my build just flakey?  sigh.
02:17 gilby joined perl6
02:19 putter hmm, pugs.kwiki.org (and www.kwiki.org) are refusing connections.
02:20 oylenshpeegul left perl6
02:30 justatheory joined perl6
02:42 putter ok, with pugs r8593, parrot 0.4.0, amd64, and an unclear gcc/glibc situation (ie, I may have updated glibc since ghc was compiled), embedded/external parrot and "a" ~~ /a/ yields segfault/ok, and /a/ ~~ "a" yields
02:43 putter pugs: src/Pugs/Eval.hs:852:12-52: Irrefutable pattern failed for pattern (Pugs.AST.Internals.App (Pugs.AST.Internals.Var name') invs' args') : rest   with both.
03:00 putter oooo... jsperl5!
03:03 putter Ok, it looks like "a" ~~ /<Foo::Bar>/ doesn't work, as it gives the same PGE error as /<nonexistentmumble>/.   So I place the JS FrontEnd back on hold.
03:06 clkao putter: what a bout jsperl5?
03:06 scrottie joined perl6
03:07 putter err, that is grammar Foo { rule Bar { x }} errors, but rule FooBar {x} works.
03:08 putter clkao: didnt exist the last time I looked.  found it on iblech's smoke list.  thought it was a nifty way to go, so glad someone (iblech?) pursued it.
03:09 putter hmm...
03:13 putter anyone: my current impression is there is little interest in my rules engine running on swi prolog within perl5.  so I've done little work on it (well, some p6 support now exists), and not even kept misc/Grammars/snapshot_of_prolog_engine.pl synced.
03:13 putter But I also thought there was little interest in js/p5 hybrids.  ;)  so if I'm also wrong about the engine, just let me know.
03:16 putter Basically its a quick path to getting _full_ p6&p5 regexp support... and good performance... and utterly undeployable (requires swi prolog compiled with non-default arguments), so it's only good for grammar development.
03:18 putter oh! clkao++ # jsperl5
03:22 putter clkao: nifty!
03:24 putter clkao: have you looked at using a pugs-embedded p5 with js?
03:26 loftcity joined perl6
03:27 xinming joined perl6
03:30 putter iblech/etal: it looks like out of the box, pugs -B js doesnt find its Prelude.js, and gives an error message when offered expressions.  With -I blib6 it goes to (again) precompile it.  If a -I is required, perhaps the error message should offer it?
03:31 scook0 joined perl6
03:33 putter ah well.  I was unsuccessful in trying to run the js backend.
03:35 putter clkao: if the js backend does not as yet support eval(:lang<perl6>), one thing which might be done with jsperl is to inject that capability.  using perl5 to call pugs.
03:37 putter long ago when i tried this, iblech's modules weren't really set up for being use()ed from p5 (... extraneous output...? I don't really remember).  but if that's still a poblem, and if difficult to fix, one could now use pugs -C js.
03:38 clkao *nod*
03:38 clkao we'll see in chubei hackathon
03:38 clkao now i need to pack and move
03:38 putter There may still be some code somewhere in the pugs tree which did a zero-effort stab at it.
03:38 putter ooo, hackathon.  enjoy! :)
03:39 putter cheers &
03:39 clkao but if we call to pugs.pm
03:39 clkao how do we carry current context (variable etc)
03:40 clkao and if you care familiar with pil2js, can you look at the array/shift.t failure? i have several things blocked by that bug
03:42 putter dont immediately see the code.
03:43 clkao it's about array arg being slurpy
03:43 putter re context, didnt. ;)  let's see...
03:43 putter re array/shift.t failure, I'll take a look...
03:53 putter clkao: ping?
03:57 zuulvin_ joined perl6
04:00 joepurl joined perl6
04:03 putter clkao: the problem lies not with pil2js, but with pugs's pil output.
04:04 clkao ah.
04:04 putter given my @a = ([]); push(shift @a,'foo','bar');   you get shift called w 3 args.
04:06 putter pugs -C pil -e 'my @a = ([]); push(shift @a,23,24)'
04:07 putter hopefully there will be an easier way to get yaml-ized pil for pil2.  currently, you can get it (or at least once could) by running pugs -B js and something like eval.yaml expr at the read-eval-print loop.  in pugs-p5.pl repl, you can toggle :v, enter the expression, and scroll up.
04:09 putter given that pil1 is on it's way out, it might be easiest to wait for pil2, rather than trying to fix it.  or not...
04:13 putter someone had a one-liner to yield the pil yaml for an expression, but I cant immediately find or recreate it.
04:16 putter clkao: re context, one trick I used to pass variables to prolog was to do an aggressive scan of the regexp string, noting anything that looked like a variable, and then pass a hash of variable name to... well, that doesnt help
04:17 putter because p6 is so much more picky about predeclaration than p5.  though since the pil2js environment structure is available... ok,
04:20 putter the prolog idea was to note anything that might be a variable, pass a hash of callbacks for each, to be used if it turned out they really were one (rather than, eg, a comment or string literal).  one might play a similar game with the string to be evaled, and on
04:20 PJF joined perl6
04:21 putter the js side, prune the list to vars which are real, and wrap the eval code in an appropriately argumented function for pugs to compile.
04:22 putter not great, as the.... oh, I'm silly.
04:28 putter ok, variables arent an issue, because pugs will happily -C pil code (like -e 'say @a') with dangling vars.  the generated code gets evaled in the backend, so if that's sufficient to connect things up, you're all set.
04:35 putter but if you need prototype info, a (we dont have it yet) parser regexp modification, etc, to get the parse right, that's just not going to happen.  simple cases work ok though, as long as the pil generator continues to be tolerant.  it seems an inherent problem with an external pil generator.
04:40 putter which suggests non-hs backends using hs pugs to generate pil is more of a bootstrap strategy, a way to make backend progress, and ultimately to boot a compiler written in p6 into the non-hs backends.  rather than something which can really be correctly delegated to.
04:42 putter the (small:) price of a flexible syntax.
04:44 putter ok, need to load shed.  good night all &
04:50 aufrank joined perl6
05:29 loftcity joined perl6
06:11 rantanplan_ joined perl6
06:26 Cryptic_K joined perl6
06:32 Aankhen`` joined perl6
06:45 svnbot6 r8599 | stevan++ | src/PIL/Native/Bootstrap/Roles/*
06:45 svnbot6 r8599 | stevan++ | - rough sketch/start to core "behavioral" Roles
06:45 svnbot6 r8599 | stevan++ | - Eq role based on Haskell Eq class
06:45 svnbot6 r8599 | stevan++ | - Ord role based on Haskell Ord class
06:45 svnbot6 r8599 | stevan++ | - Num role based on Haskell Num class & Eiffel NUMERIC
06:50 dduncan Citing a follow-up to my post here 6 hours ago, I have sent a message to several mls, including dbi-users, poop-group, and dbix-class, titled "solved the RFC on SQL::Routine; new direction planned"
06:52 Khisanth joined perl6
07:30 lisppaste3 joined perl6
07:33 Khisanth joined perl6
07:54 rashakil_ joined perl6
08:03 lao joined perl6
08:11 xern joined perl6
08:16 drbean left perl6
08:28 azuroth joined perl6
08:28 azuroth t out
08:28 azuroth err. whoops.
08:31 gaal $morning
08:33 scrottie joined perl6
08:33 audreyt yo
08:33 * audreyt wakes up
08:34 * azuroth just whacked his harddrive into an amd 64 system. awesome!
08:34 azuroth however, the n-force onboard sound doesn't seem to be working. not so awesome. :-(
08:37 khisanth_ joined perl6
08:44 audreyt gaal: let Just x = blah
08:44 azuroth bwahaha im engaged suck that space cowboys
08:46 gaal audreyt: I got GHC's brain to explode. see comment in commit :-)
08:46 svnbot6 r8600 | gaal++ | Yaml emitting:
08:46 svnbot6 r8600 | gaal++ |  - some readability improvements, need more work
08:46 svnbot6 r8600 | gaal++ |  - towards proper tagging of emitted arbitrary objects (probably need to add
08:46 svnbot6 r8600 | gaal++ |    a tag to all YamlNode types, not just YamlMap... we'll see)
08:46 audreyt gaal: oh good, I hit that in my 1st week of Hs hacking
08:46 audreyt I was _very_ afraid
08:47 gaal like I said in #haskell... I'm really happy this relationship is starting to balance out. GHC causes *my* brain to explode very often indeed.
08:48 gaal audreyt: any ideas on how to golf the v@(VRef r) case in toYaml? it's really... "profound", which is my new euphemism for "code that nests too deeply"
08:51 audreyt lookin'
08:53 audreyt wow you were writing lisp
08:54 audreyt ok, I'll golf.
08:54 gaal "))))"
08:54 gaal yeah, ifValTypeIsa kinda pulls that way I guess?
09:00 Khisanth joined perl6
09:01 integral joined perl6
09:04 audreyt done
09:04 audreyt committing
09:06 azuroth joined perl6
09:08 gaal what can I do to make openfoundry faster?
09:12 gaal audreyt: thanks.
09:12 svnbot6 r8601 | audreyt++ | * Pugs.Prim.Yaml: Slight refactoring.
09:12 gaal now, another programming pattern question:
09:13 iblechbot joined perl6
09:14 gaal in hashToYaml, and also in emitNode, there's a parameter that's passed through (d and e respectively). Is there a trick akin to the AOP "wormhole protocol" to avoid this? /me vaguely feels a monad can do this (eg. "ask"), but am not sure how to set it up.
09:14 gaal probably not worth it in these simple cases
09:14 svnbot6 r8602 | audreyt++ | * More slight refactorings.
09:15 gaal whee :)
09:18 gaal hmmm, I remember some fighting with layout last night that was only solved when I put the labmda in parentheses (yamlmap <- flip mapM assocs ...). how come it works for you? :-p
09:22 * gaal bounces in glee when, after following up on the thought "I should go study a Haskell style guide
09:23 audreyt re
09:23 audreyt so, the d/e thing
09:23 gaal ", the first thing his eye catches is "There's More Than One Way To Do It"
09:23 audreyt is among one of the very vew cases where n+k pattern is warranted
09:23 audreyt for the 'd' case that is
09:23 gaal n+k is an improvement I made on prettyVal :)
09:23 audreyt yup
09:24 audreyt classically, it's implemented as a ReaderT
09:24 gaal but if you look at emitterCallback
09:24 audreyt (`runReaderT` 0) $ do ... ask...
09:24 audreyt and then
09:24 audreyt local (+1) ...
09:24 audreyt very simple
09:24 gaal that used to be a big function which I splat to emitNode, then it passes e around w/o ever changing
09:25 audreyt oh, so "local" is not even needed
09:25 gaal so, in the case where it's not changed
09:25 gaal heh
09:25 audreyt in that case
09:25 audreyt you want implicit param
09:25 audreyt aka "env" vars in p6
09:25 audreyt aka $+e
09:25 gaal so (`runReaderT` a) sets up "a" as an askable param?
09:25 Khisanth joined perl6
09:26 audreyt yup
09:26 audreyt only it's spelled as ?e in GHC
09:26 audreyt I'll check in a demo
09:26 gaal "?e"? I thought "e <- ask"
09:26 gaal you mean there's even golfy syntax?
09:26 * gaal 's brain explodes with delight
09:28 audreyt yeah. there is special syntax for the reader monad
09:28 audreyt but it's not really a reader monad per se
09:28 audreyt because you can jump in and out of it using mplicit param
09:28 audreyt I'll check in the code in 1min
09:29 gaal oooohhh asks is so clever in its simplicity! I never realized it works because it accepts a function on a, and a constructor is of course a function!
09:30 lao /leave #perl6
09:31 audreyt gaal: yeah. arrow/monad/functor/sequence/etc has that kind of sparkling elegance
09:33 audreyt committed.
09:34 audreyt gaal: so, this is not golfy syntax for reader monad -- it's something else, but amounts to the same effect as reader monad.
09:34 svnbot6 r8603 | audreyt++ | * Refactoring Yaml code to use implicit parameters for gaal++.
09:34 audreyt I've applied it to both the d case and e case
09:35 gaal "it's something else, but amounts to the same effect as reader monad.
09:35 gaal " -> typical :)
09:35 audreyt :)
09:36 audreyt now, if you want to eliminate the
09:36 audreyt let ?d = pred ?d in do
09:36 audreyt call, and have each recursion automatically subsctract ?d
09:36 audreyt you'd want linear implicits (%d) instead of ?d
09:36 audreyt but in this case, the explicitness doesn't really hurt
09:36 audreyt and arguably makes things more apparent
09:37 audreyt env $d = $+d - 1; # that line is equiv to this
09:37 audreyt or rather, :=
09:37 * gaal nods
09:37 gaal cool stuff
09:38 audreyt the good thing is that, unlike Reader
09:38 audreyt you can call other function without (?d) in its context normally
09:38 audreyt without lifting
09:38 audreyt also you can have the context (?a, ?b, ?c)
09:38 gaal does this work on arbitrarily long call chains?
09:38 audreyt and call something with (?a, ?c)
09:38 audreyt yes.
09:38 audreyt and efficiently too
09:38 gaal ah, you're answering my question :)
09:39 gaal this is VERY cool then.
09:39 audreyt :)
09:39 audreyt yup
09:39 audreyt env vars is going to rock in p6.
09:40 audreyt so, the 'e' case has no recursion
09:40 audreyt so ?e doesn't really save you anything
09:40 gaal Perl's greatest fear now is that someone comes up with a p->Hs emitter and then everyone will just pull CPAN and code in Hs.
09:40 audreyt so it's a matter of style
09:40 audreyt nah... Hs only works if you know what you are doing.
09:40 audreyt half of the time I don't.
09:40 gaal audreyt: I rememer a quote from you on that topic... :)
09:41 audreyt oh? :)
09:41 buu Hs is scary =[
09:41 buu I don't even know what it means and it scares me!
09:41 gaal * autrijus stares at type Eval x = forall r. ContT r (ReaderT x IO) (ReaderT x IO x) and feels very lost
09:42 gaal <shapr> Didn't you write that code?
09:42 gaal <autrijus> yeah. and it works
09:42 gaal <autrijus> I just don't know what it means.
09:42 gaal -- http://www.haskell.org/hawiki/QuotesPage
09:43 audreyt yeah :)
09:43 audreyt and that, well, doesn't really work.
09:44 audreyt I had to put another EvalT on top of ContT and ReaderT.
09:44 audreyt so it supports my point :)
09:44 gaal the "e" case doesn't save me, except that from a refactoring viewpoint, it was annoying to break up a function because I had to introduce paramaters
09:44 audreyt right
09:44 audreyt implicit vars does the \lifting for you
09:44 audreyt at the cost of a sigil
09:45 audreyt anyway... I'll go back to perl6 land and finish repr types for stevan :)
09:45 gaal well, suppose I was very naive. (not far from the case). I'd say hey, this function is long but I don't want the helper functions to need long useless args. I know! I'll put them in a global!
09:46 gaal this is the same thing, except that it doesn't suck.
09:47 audreyt but in Hs, globals takes 60 chars to declare.
09:47 audreyt instead of in p6, where it takes one
09:47 audreyt (the "*" in "$*foo")
09:48 audreyt the 60 chars are "{-# NOINLINE #-}", ":: IORef ()", "unsafePerformIO (newIORef (...))"
09:49 gaal both solutions are similar in that in their respective languages, they can't be golfed further by much.
09:49 audreyt gaal++
09:49 audreyt but global vars are so wrong, I'm not sure leaving ::* open is a great idea.
09:50 gaal one liners.
09:50 audreyt (as now we have envs, and @*ARGS etc can reasonably be @+ARGS so they can be overridden lexically instead of using "temp")
09:50 r0nny joined perl6
09:50 audreyt one liners have "no strict" by default
09:50 audreyt and as such can be lax
09:52 Khisanth joined perl6
09:52 gaal so $x means (our) ::Main::<$x> yes?
09:54 audreyt in 'no strict'
09:54 audreyt yeah
09:54 gaal probably good enough in one liners :)
09:55 audreyt I think so too
09:55 * audreyt still doesn't see a reasonable case for an open$*
09:55 audreyt I mean, ::* is conceptually the outermost lexical scope
09:55 audreyt to any compilation unit
09:55 audreyt but in any other lexical scope, you can't introduce symbols
09:55 audreyt i.e. adding to ::OUTER
09:56 * gaal nods
09:56 audreyt but somehow you can add to ::* _and_ it affects other people importing your cod
09:56 audreyt e
09:56 audreyt so it's even worse than NOINLINE foo
09:56 audreyt because you still have to import foo explicitly
09:56 audreyt or export it
09:56 audreyt but ::* does its action from an distance without chance of detection
09:57 audreyt I don't know, but I'm very uncomfortable with huffmanizing it this heavy
09:57 * audreyt goes back to piln haxxoring
09:58 * gaal eyes rubyext.c some more
10:16 kaktus joined perl6
10:17 nothingmuch joined perl6
10:20 r0nny re
10:21 r0nny audreyt: is it usable again ?
10:21 nothingmuch hola
10:22 audreyt r0nny: by the end of tomorrow, you should be able to write p6 class/role code and have it run on piln runcore
10:23 gaal BTW, to pointfree this:
10:23 gaal emitterCallback e vp = let ?e = e in emitNode =<< thawNode vp
10:23 gaal why wasn't it enough to do
10:23 gaal emitterCallback e = = let ?e = e in emitNode =<< thawNode
10:23 gaal ?
10:24 gaal s/= =/=/
10:24 audreyt change =<< to +
10:24 gaal +!?
10:24 audreyt foo x y = x + f y
10:24 audreyt you can't point free it to
10:24 audreyt foo x = x + f
10:24 audreyt could you?
10:24 gaal ah.
10:25 gaal hrmpfh.
10:25 * nothingmuch goes for a hike
10:25 nothingmuch ciao!
10:25 gaal nothingmuch: it just started raining here :)
10:26 audreyt gaal: so the pf it
10:26 audreyt foo x = (x+) . f
10:26 audreyt and similarily
10:26 audreyt emitterCallback e = let ?e = e in (emitNode =<<) . thawNode
10:27 gaal yes. apart from being completely unreadable, this is beautiful.
10:27 nothingmuch ooh
10:27 nothingmuch rain good
10:27 nothingmuch i'll take ponchos
10:27 audreyt gaal: well, you can define an infix # (as some has done)
10:27 audreyt and write it as
10:27 audreyt emitterCallback e = let ?e = e in emitNode # thawNode
10:27 gaal ah, that's the reverse $ right?
10:28 audreyt er, some people define that to that as well
10:28 audreyt different purpose
10:28 audreyt or you can define infix =<<.
10:28 audreyt which is self explanatory
10:28 gaal # :: ??
10:28 audreyt emitNode =<<. thawNode
10:28 audreyt that may even be a good idea.
10:33 gaal nothingmuch: if by any chance you're areound tlv tonight, I'm serving pizza
10:34 gaal hmmm, the tagging I was doing *is* the right way to tag an object (no need for syck_emit_tag, just give a tag in syck_emit_{map,seq,str}). but there is some initialization apparently needed to make it actually work. tracing...
10:35 larsen joined perl6
10:36 nothingmuch gaal: crap, was there yesterday
10:39 gaal come again! frequent driver miles!
10:39 audreyt is there such a thing? :p
10:40 audreyt it's like "frequent piloting miles"
10:40 gaal no, but the pizza's for real
10:40 gaal I got tagging to (almost!) work
10:40 gaal - !pugs/object/Foo
10:41 gaal instead of !pugs:object/Foo
10:41 gaal getting warm :)
10:41 * audreyt feels the warm glow of the Amulet
10:41 audreyt the portal is near!
10:42 gaal I just hope it isn't made of plastic
10:42 audreyt plastic amulets don't glow :)
10:42 audreyt s/glow/warm up/
10:42 audreyt (as the glowing one is life-saving)
10:42 gaal that's because all you ever found was the cheap plastic imitations
10:42 gaal i'm talking about the expensive plastic imitations
10:43 audreyt ahh. _expensive plastic imitations_
10:43 audreyt gaal++ # rofl
10:43 verve_ joined perl6
10:43 audreyt I assume they are produced by the same vendor as the camera
10:44 gaal Acme? :)
10:45 gaal they have orthogonal use (blinding your enemies, keeping warm)
10:45 gaal brb
10:46 audreyt "warming your friends and blinding your enemies"
10:56 gaal fixed. committing.
10:57 PJF Mmmm.... Amulet of Yendor.  It shall be mine.
10:58 * gaal cannot but help noticing that "monadic lifting" is a kind of.. "ascention"
10:58 * PJF tries not to choke on his dinner in laughter.
10:58 verve_ joined perl6
11:00 svnbot6 r8604 | gaal++ | * .yaml emits objects.
11:00 svnbot6 r8604 | gaal++ | TODO:
11:00 svnbot6 r8604 | gaal++ | - make it parse them, too
11:00 svnbot6 r8604 | gaal++ | - serialize rules, native types, etc.
11:03 gaal ?eval class Foo; has $.x; Foo.new(:x<42>).yaml
11:03 evalbot_8604 "--- !pugs/object:Foo \nx: 42\n"
11:03 gaal whee!
11:13 lao left perl6
11:16 qmole joined perl6
11:17 chris2 joined perl6
11:17 elmex joined perl6
11:30 Aankh|Clone joined perl6
11:31 Khisanth joined perl6
11:33 xinming joined perl6
11:35 gaal why's this a parse error?
11:35 gaal data YamlNode = MkYamlNode { Id :: !SYMID , Tag :: !YamlTag , Anchor :: !YamlAnchor , Elem :: !YamlElem } deriving (Show, Ord, Eq)
11:35 gaal ("parse error on input `Id'")
11:36 audreyt gaal: label names aint data constructors.
11:36 audreyt so must be lower case.
11:40 gaal rrrr.
11:40 gaal thanks :)
11:41 gaal I can imagine a more helpful error message here...
11:46 audreyt patches welcome
11:49 gaal :)
11:49 gaal working on some golf in Syck,hsc, next patch up is the YAML::Syck port
11:56 G2 joined perl6
11:57 xern joined perl6
12:00 GeJ joined perl6
12:01 wolverian wait, wasn't parrot on the alioth shootout page?
12:02 wolverian it's disappeared!
12:04 azuroth sandbox
12:04 azuroth http://shootout.alioth.debian.org/sandbox/
12:05 wolverian ah. thanks
12:10 blm joined perl6
12:12 svnbot6 r8605 | scook0++ | * s/it's/its/ in a few places
12:20 kane_ joined perl6
12:21 Limbic_Region joined perl6
12:23 xinming joined perl6
12:27 borkened joined perl6
12:37 bsb joined perl6
12:38 Khisanth joined perl6
12:43 iblechbot_ joined perl6
12:44 PJF left perl6
12:44 ingy audreyt: ping
12:45 audreyt pong
12:48 ingy audreyt: what is 'Cannot determine perl version info from lib/YAML.pm'
12:50 svnbot6 r8606 | gaal++ | .yaml:
12:50 svnbot6 r8606 | gaal++ |   - make YamlNode resemble SyckNode a little better in structure. This
12:50 svnbot6 r8606 | gaal++ |     will come useful when eventually we need to serialize "undef but" etc.
12:50 svnbot6 r8606 | gaal++ |   - golfage (I hope it's sane, or in the right form of insanity at least)
12:50 audreyt ingy: put a "use 5.005" in it
12:51 ingy use 5.006_001; # ?
12:54 ingy soon to be `use 5.008_003;`
12:54 ingy as punishment to the unfaithful
12:55 ingy Brian the Protector has become Ingy the Punisher
12:55 gaal ingy: make it "v6"
12:56 ingy gaal: mmmm 6
12:56 Limbic_Region rough for people in shops who don't have the luxury of upgrading - though I suppose they can continue to use older module versions too
12:56 ingy use 7.007_007;
12:56 ingy Limbic_Region: exactly
12:57 ingy rough modules for rough perls
13:02 audreyt ingy: "use 5.006001"
13:03 kanru joined perl6
13:03 * gaal needs to prep the house for pizza party later. audreyt, I'll try to add tag support in between mozarella attacks :)
13:04 * gaal waves &
13:04 gaal s//to YAML::Syck/
13:05 audreyt woot
13:06 * Limbic_Region kicks gaal for mentioning pizza
13:06 Limbic_Region it is bad enough that I am on a diet, but I am currently on a forced fast too for bloodwork later this morning
13:06 azuroth where's my pizza?
13:14 ingy !
13:14 ingy A
13:14 ingy Z
13:14 ingy Z
13:14 ingy I
13:14 ingy P
13:15 azuroth hmm. thanks, but I'm not arabic
13:17 audreyt    P
13:17 audreyt   I I
13:17 audreyt  ZZZZZ
13:17 audreyt Z Z Z Z
13:17 audreyt AAAAAAAAA
13:20 Limbic_Region ?eval say `which pizza`;
13:20 evalbot_8606 Error:  unexpected "`" expecting block construct, ":", identifier, "(", term, term postfix, operator, postfix conditional, postfix loop, postfix iteration, ";" or end of input
13:35 bsb_ joined perl6
13:36 khisanth_ joined perl6
13:37 xinming joined perl6
13:56 Juerd_ http://www.cafepress.com/perl6.37792693
13:57 audreyt ooooh
13:58 Juerd_ Damn, s/onliner/oneliner/
14:01 alexe_home joined perl6
14:02 khisanth_ joined perl6
14:02 drbean joined perl6
14:30 khisanth_ joined perl6
14:38 gaal It's not a well-known fact, but anyone with a pugs committer bit is automatically invited to any pizza party thrown at foo's diner.
14:42 azuroth that's awesome. do they get free travel?
14:42 gaal Juerd_: shouldn't that url be perl6.28318530717959?
14:42 gaal travel and batteries regrettably not included in the bit.
14:43 gaal out of olive oil. cornerstore &
14:52 audreyt foo's dinner?
14:55 qmole joined perl6
14:56 azuroth my brother moved a long way away, and within a half hour of getting there he's doing the mixing for a jazz band? how does that work?
14:59 gaal audreyt: diner. it's the name of a place.
15:00 azuroth foo's diner is a real place?
15:01 gaal sure
15:01 gaal it's located cospatially with roo's room.
15:01 gaal here's another service it provides apart from pizza: http://gaal.livejournal.com/2002/03/14/
15:08 azuroth gee. libraries can cost over there?
15:09 gaal some do.
15:10 gaal the national library doesn't, and purportedly carries a copy of every book published here
15:10 khisanth_ joined perl6
15:11 gaal the university library in my city is open to academics, but others need to pay a fee. I don't know if the municipal library costs anything.
15:12 gaal the most impressive public library I'd ever seen (and the one that impressed upon me that a non-free library is in fact a grave thing) was the one in the pompidou center in paris
15:15 SamB you can't copy the books?</stupid-joke>
15:18 joepurl joined perl6
15:30 lisppaste3 joined perl6
15:37 justatheory joined perl6
15:37 bernhard joined perl6
15:40 khisanth_ joined perl6
15:50 alexe_home left perl6
16:07 khisanth_ joined perl6
16:14 nothingmuch joined perl6
16:28 elmex_ joined perl6
16:31 xinming joined perl6
16:43 cabbage is now known as integral
16:44 ntgrl is now known as cabbage
16:48 azuroth ?eval [2,5,1][*]
16:48 evalbot_8606 Error:  unexpected "[" expecting term postfix, operator, postfix conditional, postfix loop, postfix iteration, ";" or end of input
16:48 azuroth ?eval [2,5,1](*)
16:48 evalbot_8606 Error:  unexpected "(" expecting term postfix, operator, postfix conditional, postfix loop, postfix iteration, ";" or end of input
16:48 azuroth ?eval [*][2,5,1]
16:48 evalbot_8606 [2, 5, 1]
16:48 azuroth ah... ha
16:48 azuroth ?eval say ([*][2,5,1])
16:48 evalbot_8606 OUTPUT[2 5 1 ] bool::true
16:48 azuroth ?eval say (*[2,5,1])
16:48 evalbot_8606 OUTPUT[2 5 1 ] bool::true
16:48 azuroth ?eval say ((*)[2,5,1])
16:48 evalbot_8606 Error:  unexpected "(" expecting block construct, ":", term, term postfix, operator, postfix conditional, postfix loop, postfix iteration, ";" or end of input
16:49 azuroth ?eval say (+[2,5,1])
16:49 evalbot_8606 OUTPUT[3 ] bool::true
16:49 azuroth ?eval say ([+][2,5,1])
16:49 evalbot_8606 OUTPUT[2 5 1 ] bool::true
16:49 anatolyv joined perl6
16:50 azuroth ?eval say [+]([2,5,1])
16:50 evalbot_8606 OUTPUT[2 5 1 ] bool::true
16:50 azuroth ?eval my @a = [2,1,2]; my $v = [+]@a; say $v;
16:50 evalbot_8606 OUTPUT[2 1 2 ] bool::true
16:51 azuroth ?eval my @a = [2,1,2]; my $v = [*]@a; say $v;
16:51 evalbot_8606 OUTPUT[2 1 2 ] bool::true
16:51 azuroth ?eval my @a = [2,1,2]; my $v = [-]@a; say $v;
16:51 evalbot_8606 OUTPUT[2 1 2 ] bool::true
16:51 azuroth ?eval my @a = [2,1,2]; my $v = @a; say $v;
16:51 evalbot_8606 OUTPUT[2 1 2 ] bool::true
16:51 azuroth ?eval my @a = [2,1,2]; my $v = @a[*]; say $v;
16:51 evalbot_8606 Error:  unexpected "[" expecting word character, "::", term postfix, operator, ";" or end of input
16:51 azuroth I thought I saw audreyt use the hyper reduce earlier
16:52 azuroth ?eval my @a = [2,1,2]; my $v = [*]>>@a; say $v;
16:52 evalbot_8606 Error:  unexpected ">" expecting "\171", "<<", ".", ":", "(", block construct, identifier, term, term postfix, operator, ";" or end of input
16:53 azuroth ?eval my @a = [2,1,2]; my $v = [*]<<@a; say $v;
16:53 evalbot_8606 OUTPUT[4 ] bool::true
16:53 azuroth ?eval my @a = [2,1,2]; my $v = [+]<<@a; say $v;
16:53 evalbot_8606 OUTPUT[5 ] bool::true
16:53 azuroth yayy
17:05 xinming joined perl6
17:16 penk joined perl6
17:26 penk joined perl6
17:30 buu joined perl6
17:32 rafl ingy: How's YAML development going? Can we expect a release in the next time?
17:36 xinming joined perl6
17:41 khisanth_ joined perl6
17:42 dduncan joined perl6
17:52 Juerd_ gaal: No :)
17:53 wilx joined perl6
18:06 pen1 joined perl6
18:08 elmex joined perl6
18:19 khisanth_ joined perl6
18:49 khisanth_ joined perl6
18:52 Khisanth is now known as khisanth_
19:10 xinming joined perl6
19:26 khisanth_ joined perl6
19:27 Khisanth is now known as khisanth_
19:28 integral is now known as ntgrl
19:31 rantanplan_ joined perl6
19:47 stevan audreyt: ping
19:55 stevan oh well probably middle of the night there
19:55 * stevan wanders off again &
20:10 nothingmuch http://www.pandora.com/
20:20 mjl69 joined perl6
20:43 nnunley joined perl6
21:00 frederico joined perl6
21:26 gaal hola
21:32 nothingmuch hola
21:39 gaal no more pizza :)
21:43 nothingmuch good thing or bad thing?
21:44 gaal a force of nature
21:45 vel joined perl6
22:03 mjl69_ joined perl6
22:03 cognominal joined perl6
22:28 Shachaf joined perl6
22:30 chris2 joined perl6
22:46 nnunley Hrm.  Don't grok PArrow/Haskell enough to do the hack that I want to do.  Just trying to figure out what's necessary to convert the IO Monad stuff into what's required (String) for use with Text.Parser.Rule.
23:00 PJF joined perl6
23:02 borkened joined perl6
23:30 lisppaste3 joined perl6
23:52 stennie joined perl6
23:54 bsb_ left perl6

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

Perl 6 | Reference Documentation | Rakudo