Perl 6 - the future is here, just unevenly distributed

IRC log for #perl6, 2011-08-27

Perl 6 | Reference Documentation | Rakudo

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

All times shown according to UTC.

Time Nick Message
00:08 jevin joined #perl6
00:13 frhodes joined #perl6
00:26 whiteknight joined #perl6
00:27 lichtkind joined #perl6
00:28 lichtkind TimToady: was .perl your idea?
00:28 lichtkind i mean the dumper method
00:33 lichtkind allright have to sleep now
00:33 lichtkind cheers
00:43 lateau_ joined #perl6
00:45 lateau_ left #perl6
00:53 wk joined #perl6
01:11 uasi joined #perl6
01:21 jevin joined #perl6
01:24 uasi joined #perl6
02:03 Minimiscience joined #perl6
02:04 soh_cah_toa phenny: tell masak i was thinking about adding a does_ok() sub to Test. it'd be just like isa_ok() except it would check roles instead of inheritance. what do you think?
02:04 phenny soh_cah_toa: I'll pass that on when masak is around.
02:07 wolfman2000 joined #perl6
02:10 s1n joined #perl6
02:13 TimToady http://rosettacode.org/wiki/Ternary_logic#Perl_6, just for mikemol++
02:13 [Coke] joined #perl6
02:14 TimToady decommuting &
02:17 abercrombie perl is really concise compared to others..
02:21 agentzh joined #perl6
02:27 mikemol Neat.
02:27 * mikemol thanks TimToady++
02:28 mikemol Now, why use multi instead of sub?
02:28 soh_cah_toa mikemol: it allows for subs w/ the same name but different signatures and arity
02:29 soh_cah_toa let me find an example
02:30 soh_cah_toa mikemol: http://www.programmersheaven.com/2/Perl6-FAQ-Multi
02:31 f00li5h joined #perl6
02:32 soh_cah_toa mikemol: are you familiar w/ method overloading in c++? it's the same thing
02:32 soh_cah_toa well, the same concept that is
02:36 mikemol soh_cah_toa: I've written code with multis before, but none of these have the same name.
02:37 soh_cah_toa oh ok. probably missed the beginning of that conversation :)
02:37 soh_cah_toa why, did someone suggest using 'multi' over 'sub' even for subs w/ different names
02:37 mikemol soh_cah_toa: http://rosettacode.org/wiki/Ternary_logic#Perl_6
02:38 mikemol TimToady++ wrote it. I'm just trying to figure out the specific reason to use multi instead of sub.
02:39 soh_cah_toa for subs that define some type of operator, you need to use multi
02:39 soh_cah_toa notice how they all use prefix:<something>
02:39 mikemol Ah
02:40 soh_cah_toa it's the same as operator overloading which still involves multiple dispatch
02:40 jevin joined #perl6
02:40 soh_cah_toa keyword: 'multiple' ;)
02:41 soh_cah_toa oh and s/prefix/infix/ mixed that up
02:41 soh_cah_toa but you get the point :)
02:41 mikemol Well, ¬ uses prefix.
02:42 mikemol niecza: multi infix<¬> ($a, $b) {$a + $b}
02:43 p6eval niecza v8-55-g0db7bb8: OUTPUT«[31m===[0mSORRY![31m===[0m��Malformed block at /tmp/QBL7ilCxu7 line 1:�------> [32mmulti infix[33m�[31m<¬> ($a, $b) {$a + $b}[0m��Parse failed��»
02:43 soh_cah_toa yeah, but it's the only one there. i was referring to the infix's but said prefix by mistake. prefix still needs it though b/c like i said it still uses multiple dispatch
02:44 mikemol I know p6eval can return the the type signatures of objects. I'm wondering if it's possible to extract the ¬ from within &prefix:<¬>()
02:47 soh_cah_toa hm, i'm sure it's possible but i'm not too familiar w/ tinkering around w/ meta objects
02:47 soh_cah_toa perl6: prefix:<+>.^signature
02:47 p6eval niecza v8-55-g0db7bb8: OUTPUT«Unhandled exception: No value for parameter \$v in CORE prefix:<+>␤  at /home/p6eval/niecza/lib/CORE.setting line 0 (CORE prefix:<+> @ 0) ␤  at /tmp/01flJd7uhm line 1 (MAIN mainline @ 1) ␤  at /home/p6eval/niecza/lib/CORE.setting line 2047 (CORE C953_ANON @…
02:47 p6eval ..rakudo a55346: OUTPUT«too few positional arguments: 0 passed, 1 (or more) expected␤  in main program body at line 1:src/gen/core.pm␤»
02:47 p6eval ..pugs: OUTPUT«*** No compatible multi variant found: "&prefix:+"␤    at /tmp/BTXdCc6XDy line 1, column 1 - line 2, column 1␤»
02:48 soh_cah_toa yeah...not too sure about that :\
02:49 plobsing_ perl6: &prefix:<+>.^signature
02:49 p6eval rakudo a55346: OUTPUT«Method 'signature' not found for invocant of class 'ClassHOW'␤  in main program body at line 22:/tmp/_oWM9nhj8f␤»
02:49 p6eval ..pugs: OUTPUT«*** No such method in class Class: "&signature"␤    at /tmp/WzPVcLW7Vl line 1, column 1 - line 2, column 1␤»
02:49 p6eval ..niecza v8-55-g0db7bb8: OUTPUT«Unhandled exception: Unable to resolve method signature in class ClassHOW␤  at /tmp/KMBY92WAnv line 1 (MAIN mainline @ 2) ␤  at /home/p6eval/niecza/lib/CORE.setting line 2047 (CORE C953_ANON @ 2) ␤  at /home/p6eval/niecza/lib/CORE.setting line 2048 (CORE mo…
02:49 plobsing_ perl6: &prefix:<+>.signature
02:49 p6eval niecza v8-55-g0db7bb8: OUTPUT«Unhandled exception: Unable to resolve method signature in class Sub␤  at /tmp/cJS7Fj1Kn1 line 1 (MAIN mainline @ 1) ␤  at /home/p6eval/niecza/lib/CORE.setting line 2047 (CORE C953_ANON @ 2) ␤  at /home/p6eval/niecza/lib/CORE.setting line 2048 (CORE module-…
02:49 p6eval ..rakudo a55346: OUTPUT«get_attr_str() not implemented in class 'Sub'␤  in main program body at line 1:src/metamodel/RoleToInstanceApplier.nqp␤»
02:49 p6eval ..pugs:  ( no output )
02:50 plobsing_ pugs: &prefix:<+>.signature.say
02:50 p6eval pugs: OUTPUT«:($?1 is copy)␤»
02:50 soh_cah_toa access to the mop uses the .^ syntax
02:50 plobsing_ you don't need the mop
02:50 plobsing_ the sub knows its signature
02:50 soh_cah_toa to get the signature?
02:50 soh_cah_toa oh right
02:50 soh_cah_toa the capture or parcel or something right?
02:50 plobsing_ using the .^ would ask class Sub for a signature
02:51 plobsing_ pugs: &prefix:<+>.WHAT.say
02:51 p6eval pugs: OUTPUT«Sub␤»
02:55 soh_cah_toa that's odd. on my installation that code gives me 'Code()'
02:55 soh_cah_toa rakudo: &prefix:<+>.WHAT.say
02:55 p6eval rakudo a55346: OUTPUT«Code()␤»
02:56 plobsing_ I suspect that's a parrot leak, which makes some methods give parrot errors
02:56 soh_cah_toa ok
02:56 plobsing_ rakudo, pugs: &prefix:<+>.Str
02:57 plobsing_ pugs: &prefix:<+>.Str
02:57 p6eval pugs: OUTPUT«*** No such method in class Sub: "&Str"␤    at /tmp/RyURqwUNRH line 1, column 1 - line 2, column 1␤»
02:57 plobsing_ rakudo: &prefix:<+>.Str
02:57 p6eval rakudo a55346: OUTPUT«get_attr_str() not implemented in class 'Sub'␤  in main program body at line 1:src/metamodel/RoleToInstanceApplier.nqp␤»
02:57 soh_cah_toa you're right. now it's class 'Sub'
03:00 soh_cah_toa mikemol: anyway, i digress. there are quite a few p6 super ninjas that come around a little later that can help you w/ that a lot more than i can. just stick around :)
03:01 plobsing_ in attempting to answer the original question (getting just the 'x' out of &fixity:<x>), I don't see a method on the sub to do that
03:01 plobsing_ rakudo: &infix:<+>.^methods(:local)
03:01 p6eval rakudo a55346:  ( no output )
03:01 plobsing_ rakudo: &infix:<+>.^methods(:local).say
03:01 p6eval rakudo a55346: OUTPUT«candidatesmultiname␤»
03:02 plobsing_ rakudo: say for &infix:<+>.^methods(:local)
03:02 p6eval rakudo a55346: OUTPUT«␤␤␤»
03:02 plobsing_ rakudo: .say for &infix:<+>.^methods(:local)
03:02 p6eval rakudo a55346: OUTPUT«candidates␤multi␤name␤»
03:02 plobsing_ my copy must be old, I get a different result
03:02 soh_cah_toa plobsing_: that reminds me, what does the :local adverb indicate for the .^methods() method? the phrase "local method" doesn't make sense to me
03:03 plobsing_ local means not to look up the inheritance heirarchy (IIRC)
03:03 plobsing_ it prunes out a lot of irrelevant methods
03:04 soh_cah_toa oh, just the methods it defines. i see
03:06 ruz grammar tracer and debugger, suspect these are only useful with master branch only?
03:07 TimToady perl6: say ~&prefix:<+>
03:07 p6eval niecza v8-55-g0db7bb8: OUTPUT«Sub()<instance>␤»
03:07 p6eval ..rakudo a55346: OUTPUT«&prefix:<+>␤»
03:07 p6eval ..pugs: OUTPUT«␤*** No compatible multi variant found: "&prefix:+"␤    at /tmp/BMW1VCzYnP line 1, column 1 - line 2, column 1␤»
03:10 TimToady perl6: say &prefix:<+>.name
03:10 p6eval pugs: OUTPUT«&::␤»
03:10 p6eval ..rakudo a55346: OUTPUT«&prefix:<+>␤»
03:10 p6eval ..niecza v8-55-g0db7bb8: OUTPUT«Unhandled exception: Unable to resolve method name in class Sub␤  at /tmp/YnBxOF4lUe line 1 (MAIN mainline @ 1) ␤  at /home/p6eval/niecza/lib/CORE.setting line 2047 (CORE C953_ANON @ 2) ␤  at /home/p6eval/niecza/lib/CORE.setting line 2048 (CORE module-CORE …
03:12 soh_cah_toa rakudo: &prefix:<+>.name.comb[9]
03:12 p6eval rakudo a55346:  ( no output )
03:12 soh_cah_toa rakudo: &prefix:<+>.name.comb[9].say
03:13 p6eval rakudo a55346: OUTPUT«+␤»
03:13 soh_cah_toa i win :)
03:13 soh_cah_toa although, that's a really horrible way to do it
03:15 soh_cah_toa rakudo: &prefix:<+>.name.comb(/<punct>/)[3].say
03:15 p6eval rakudo a55346: OUTPUT«+␤»
03:16 soh_cah_toa timtowtdi baby ;)
03:18 cexsum joined #perl6
03:19 mikemol Does that work with infix, too?
03:19 soh_cah_toa you'd need a different index number though
03:20 soh_cah_toa i totally suggest not using that though. i was just goofing around
03:20 TimToady in any case, niecza doesn't support .name yet
03:21 TimToady and the code in question doesn't run under rakudo for some reason
03:23 Su-Shee_ joined #perl6
03:30 colomon joined #perl6
03:41 uvtc joined #perl6
04:00 uvtc left #perl6
04:19 kaare_ joined #perl6
04:43 moritz good morning
04:43 soh_cah_toa good night ;)
04:53 kaare_ joined #perl6
05:04 birdwindupbird joined #perl6
05:19 REPLeffect joined #perl6
05:29 Chillance joined #perl6
05:37 mj41 joined #perl6
05:37 Chillance joined #perl6
05:48 xinming joined #perl6
06:20 sili joined #perl6
06:29 ZaphrodZenovka left #perl6
06:31 ZaphrodZenovka joined #perl6
06:46 daniel-s joined #perl6
06:49 kst joined #perl6
07:04 woosley joined #perl6
07:39 yath joined #perl6
07:48 envi joined #perl6
08:06 molaf joined #perl6
08:23 wk joined #perl6
08:34 wk joined #perl6
09:06 wamba joined #perl6
09:10 orafu joined #perl6
09:14 whiteknight joined #perl6
09:24 masak joined #perl6
09:24 masak morning, #perl6
09:24 phenny masak: 26 Aug 23:24Z <TimToady> tell masak "receptible" and "pund", unless they were punned.
09:24 phenny masak: 02:04Z <soh_cah_toa> tell masak i was thinking about adding a does_ok() sub to Test. it'd be just like isa_ok() except it would check roles instead of inheritance. what do you think?
09:28 im2ee joined #perl6
09:30 masak TimToady: found and fixed s/pund/pound/. I think I might not see what's wrong with "receptible". :/
09:35 mberends ahoj masak: wow. your last blog post was a long read! It must have been a long write :)
09:36 masak mberends: I almost felt like Stevey there for a while.
09:42 masak and yes, it took all of last night :)
09:42 mberends the "do tests" == wrong idea did seem a bit extreme to me, but I get your point. Ad-hoc testing is very natural - before you buy a car, marry someone etc.
09:43 mberends It seemed the ideal opportunity to mention tote as well.
09:45 masak ooh
09:45 masak I'm not saying manual testing is wrong. I'm saying my language sense about the word has been warped.
09:46 wullikam joined #perl6
09:47 mberends ah. warped. very reflective :)
09:48 masak adding mention about ad-hoc tests being beneficial and about tote. mberends++
09:48 masak mberends: ooc, is there a situation where you could see yourself making an ad-hoc test, it triggers a failure, and you'd *not* add it as a unit test?
09:49 mberends masak: no. on the contrary, those produce some of the best tests.
09:50 masak then we are in accord.
09:50 mberends violently
09:50 masak ;)
09:51 masak some day I'll write a blog post/talk about turning against your own application, adopting the "exploit" mindset towards it in order to make it more robust.
09:53 mberends hmm, nice idea. The creators are often less able to "turn against" than some users from hell, though.
09:53 masak yes.
09:53 masak that was a (commonly known) problem that I experienced when writing the adventure game.
09:53 masak I kept running along the "happy path".
09:54 mberends :)
09:54 masak it was beta testers who kept finding bugs along various other paths.
09:54 masak in fact, I plan to re-write the whole game using tests.
09:54 masak maybe not this weekend, but sometime soon.
10:00 mberends excellent - that's the approach you also used with the module installation manager (post-proto). It worked very well.
10:00 masak yes.
10:00 masak and that now, IIUC, forms the basis of Pies.
10:04 mberends today I'll be writing some docs for Niecza - a kind of developer's guide. Testing forms a big part of the procedure.
10:05 masak mberends: Knuth has a nice story in some book about the creation of TeX, about how he kept around an "exploit" file for the TeX engine, in which he had tried to be as evil and out-of-the-box as possible, just to crash-test the program for robustness. ISTR he described that file as one of the most enjoyable things he had written.
10:05 masak mberends++ # docs for Niecza
10:10 pernatiy joined #perl6
10:10 masak pernatiy: hi!
10:24 cognominal nom: grammar A is HLL::Grammar
10:24 p6eval nom 2742f6: OUTPUT«===SORRY!===␤No applicable candidates found to dispatch to for 'trait_mod:<is>'. Available candidates are:␤:(Attribute $attr, Any $rw)␤:(Attribute $attr, Any $readonly)␤:(Routine $r, Any $rw)␤:(Parameter $param, Any $readonly)␤:(Parameter $param, Any $rw)␤:(Parameter …
10:25 cognominal is this problem known? that makes grammars not very useful.
10:26 kst joined #perl6
10:26 mberends cognominal: probably just NYI, such things should work
10:26 cognominal I think every language created with create_language.pl depends on it
10:29 masak nom: say HLL::Grammar ~~ Mu
10:29 p6eval nom 2742f6: OUTPUT«Could not find symbol 'HLL::&Grammar'␤  in mu <anon> at /tmp/uf9g0A3fDr:1␤  in mu <anon> at /tmp/uf9g0A3fDr:1␤␤»
10:29 cognominal HLL::Grammar is not on the path, I checked
10:29 cognominal next is QRegex
10:29 mberends I investigated a little when you previously asked. There are similar examples that work in earlier Rakudo branches, so it *almost* works.
10:31 cognominal the                     $*ST.apply_trait('&trait_mod:<is>', $declarand, $trait);  in Actions.pm is not executed as it should becausse HLL::Grammar is not defined,
10:32 cognominal these are probably some pbc that should be loaded.
10:33 mberends nom: say "%s and %s".sprintf(<salt pepper>);
10:33 p6eval nom 2742f6: OUTPUT«salt and pepper␤»
10:33 mberends S32-str/sprintf.t has tests for the sub form of sprintf(), but not the method form, although the method is specced in S32::Str. I have never seen sprintf() used as a method and it looks unintuitive. Should we de-spec the method form or add tests?
10:33 cognominal I have trouble to understand the bootstrap process so I am not sure what can be loaded at this final stage.
10:34 cognominal where are the pbc loaded anyway?
10:37 masak mberends: +1 for de-spec.
10:37 MayDaniel joined #perl6
10:37 masak mberends: sending a message to a format string feels... weird.
10:37 masak mberends: a bit like python's '|'.join(stuff)
10:38 mberends masak: agreed. otoh, there is a general architecture for these kind of functions, and de-speccing would break the order. It  can be argued both ways...
10:39 masak mberends: we already have $value.fmt($format)
10:39 mberends true. I'd like to see a few more opinions before committing
10:40 mberends (not bikeshed ones tho')
10:47 masak mberends: over the years, I've used .substr, .index, .rindex, .chars, .chomp, .split, .comb, .lc, .ucfirst, .ord, .flip, .trim, .lines more times than I can count. but never .sprintf
10:49 mberends masak: sprintf is a predecessor of variable interpolation, so Perl seldom needs it but C does. Likewise strftime().
10:52 masak well, I'm not disputing that.
10:52 mberends anyhow, where do you think printf() gets its hard work done?
10:53 masak rather, I'm suggesting that if .sprintf has a method form at all, the invocant shouldn't be the format string, but the list of values.
10:53 masak just like with .join
10:53 masak the format string isn't the main agent here, the list of values to be substituted in are.
10:54 mberends agreed, but since the main goal is less surprise for existing sprintf users, let's aim to keep only the sub form.
10:54 masak agreed.
10:54 masak that's also the conservative route, that can later be changed.
10:55 masak nom: say ("a", "b", "c").WHAT
10:55 p6eval nom 2742f6: OUTPUT«Parcel()␤»
10:56 * mberends feels the "forgiveness" itch overriding the "permission" twitch
10:56 masak nom: augment class Parcel { method sprintf($format) { sprintf $format, |self } }; say ("Martin", "Berends").sprintf("I trust %s %s with a spec commit bit.")
10:56 p6eval nom 2742f6: OUTPUT«===SORRY!===␤augment not allowed without 'use MONKEY_TYPING' at line 1, near "{ method s"␤»
10:57 masak nom: use MONKEY_TYPING; augment class Parcel { method sprintf($format) { sprintf $format, |self } }; say ("Martin", "Berends").sprintf("I trust %s %s with a spec commit bit.")
10:57 p6eval nom 2742f6: OUTPUT«I trust Martin Berends with a spec commit bit.␤»
10:57 masak \o/
10:57 mberends phew
10:57 masak time to make lunch.
10:58 mberends gorgonzola: mouldy by design
10:58 masak nom: say Parcel ~~ List
10:58 p6eval nom 2742f6: OUTPUT«Bool::False␤»
10:58 masak nom: say Parcel.^parents
10:58 p6eval nom 2742f6: OUTPUT«Cool() Any() Mu()␤»
10:59 masak boy, I'm starting to really like those spaces. :)
11:01 masak nom: say Parcel.^parents(:local)
11:01 p6eval nom 2742f6: OUTPUT«Cool()␤»
11:02 masak arguably a slight overloading of both a method and the term "parent", but alright.
11:02 masak I can see the type of consistency that was chose, with .^methods
11:02 masak chosen*
11:10 mberends "was chose" sounds perfectly good in some states ;)
11:11 cognominal it seems that HLL::Grammar depends on protoregex that are NYI :(
11:13 szabgab joined #perl6
11:15 mberends o/ szabgab!
11:23 flussence .oO( oh dear, my urxvt's forgotten how to draw uppercase "m". )
11:31 masak mberends: agreed. I thought the same thing when I saw the error. choose, chose, chose. looks OK. :)
11:32 ilogger2 joined #perl6
11:35 zby_home_ joined #perl6
11:36 wamba joined #perl6
11:44 thelazydeveloper joined #perl6
11:46 masak flussence: software is hard!
11:48 agentzh joined #perl6
11:56 ilogger2 joined #perl6
12:01 jedai_ joined #perl6
12:06 thelazydeveloper left #perl6
12:08 jrockway joined #perl6
12:25 masak I just want to ++ sorear on general principle.
12:25 masak sorear: thank you for Niecza.
12:25 masak sorear++
12:32 mkramer joined #perl6
12:33 * mberends seconds that. sorear++ :)
12:42 masak having Rakudo only, we're a cyclops. with Niecza and Rakudo, we get stereo vision.
12:42 masak depth.
12:44 mberends Niecza makes http://perl6.org/compilers/features much more interesting.
12:44 lichtkind joined #perl6
12:45 woosley joined #perl6
12:46 masak someone came to me and jnthn and pmichaud during the YAPC::EU hallway track.
12:46 masak he said "shall we revive Pugs?". pmichaud said "why not?". I said "oh YESS!"
12:47 masak he said he'd proposed it here on channel but got Warnock'd.
12:47 masak we all told him he was just unlucky and should come back and try again.
12:47 masak here's how you revive Pugs, by the way:
12:47 masak 1. get together a dedicated group of about three people.
12:48 masak 2. form a Haskell study group, open to everyone, meeting each week for 15 minutes or so.
12:48 masak 3. ???
12:48 masak 4. Pugs!
12:49 * masak .oO( maybe 2012 *will* be the end of the world because that's the year Perl 6 will be released )
12:49 mberends I'd like to join, but am daunted by the time probably required to master Haskell
12:50 mberends I suggest offering a grant to someone dedicated enough to revive Pugs
12:50 flussence Haskell doesn't look that hard... :)
12:50 flussence (and I never would've said that before learning p6...)
12:50 wamba joined #perl6
12:51 arnsholt mberends: I think the main challenge with Haskell is getting into the whole mindset of it all
12:52 arnsholt (I speak pidgin Haskell from configuring my XMonad)
12:52 lichtkind if @@ gone?
12:52 arnsholt ?
12:52 mberends what was @@ ?
12:53 lichtkind lol context op
12:56 lichtkind so i think that || is also gone
12:58 lichtkind wait im talking crap @@ went to **@ long before
13:06 masak lichtkind: || replaced @@ replaced **@
13:06 masak prefix:<||> is still there, AFAIK.
13:06 masak arnsholt: I also speak pidgin Haskell, from writing some small things on my own, and from reading Pugs source code.
13:07 pochi joined #perl6
13:08 lichtkind masak: wiht || i mean positional flattenign of captures which should be still valid
13:09 masak that's what I mean, too.
13:09 lichtkind its no longer ||?
13:11 masak no, I think it still is.
13:11 masak (how did you get the impression from what I said...? oh, forget it...) :P
13:17 [Coke] tl;dr
13:18 lichtkind masak: before i forgot it i wated to say that "lichtkind: || replaced @@ replaced **@" confused me a bit
13:19 masak lichtkind: it's no more confusing than 3 < 4 < 5 IMO.
13:19 * masak sees http://www.mdk.me.uk/fun/raptor/vader/index.html again and laughs
13:24 abercrombie joined #perl6
13:30 masak mberends: re the "exploit" stance: https://twitter.com/carlmas​ak/status/11820728127066112 :)
13:34 mberends masak: indeed. like calling a class 'is' :)
13:39 donri joined #perl6
13:39 masak I sure fell for that one :)
13:39 mberends "Click Download to download Download"
13:43 masak "Click to download Click"
13:44 mberends There is the reverse problem choosing unique names in programming. I seem to collide increasingly with reserved words (eg next, date, list).
13:45 masak ooh, an ancient autopun by quietfanatic: https://twitter.com/carlmasak/status/8715995394
13:46 masak mberends: sigils are nice that way.
13:47 jevin joined #perl6
13:59 birdwindupbird joined #perl6
14:12 dorlamm joined #perl6
14:12 wamba joined #perl6
14:18 hirschnase joined #perl6
14:26 * masak realizes djanatyn hasn't been around in a while
14:26 masak hope he comes back sometime.
14:37 woosley left #perl6
14:43 donri how you know djanatyn? :p
14:43 JimmyZ joined #perl6
14:47 masak he came in here last month.
14:50 JimmyZ good evening, #perl6
14:51 masak 晚上好, JimmyZ
14:51 donri i want to be missed too, i should leave
14:51 lichtkind haha
14:51 donri nah no one would notice
14:52 JimmyZ 下午好, masak
14:52 * donri commences arm cutting instead
14:52 lichtkind donri: love always comes for free
14:52 lichtkind cant force it
14:52 donri can't by me loooove love
14:53 donri buy also
14:53 masak donri: you just reminded me of http://en.wikipedia.org/wiki​/Parable_of_the_Prodigal_Son
14:54 donri :D
14:54 masak i.e. there's more joy in someone coming back than in someone being around all the time. :)
14:56 donri yay parrots http://www.picrandom.com/images/6cim.jpg
14:56 masak :)
14:57 donri dat glass
14:57 jevin joined #perl6
15:17 isBEKaml joined #perl6
15:32 araujo joined #perl6
15:35 ilogger2 joined #perl6
15:35 birdwindupbird joined #perl6
15:36 masak nom: say "windup".join("bird" xx 2)
15:36 p6eval nom 2742f6: OUTPUT«windup␤»
15:36 masak oh, right. not Python :)
15:37 masak nom: say join "windup", "bird" xx 2
15:37 p6eval nom 2742f6: OUTPUT«birdwindupbird␤»
15:38 mberends nom: "%swindup%s".sprintf("bird" xx 2).say # :P
15:38 p6eval nom 2742f6: OUTPUT«birdwindupbird␤»
15:43 JimmyZ nom: dir
15:43 p6eval nom 2742f6:  ( no output )
15:44 JimmyZ nom: say dir
15:44 p6eval nom 2742f6: OUTPUT«.vim std_mine sprixel Perlito .ccache old_perl5 niecza mono-2.10.1.tar.bz2 .cache partcl-nqp log .pugs_history .bash_history nqp std_hilited 6model _sprixel01 p1 rakudo-star-2011.04.tar.gz .lesshst .aptitude .bashrc .cpanm nom-inst2 rakudo-star-2010.09 .ghc perl5 eval…
15:44 bluescreen10 joined #perl6
15:45 JimmyZ left #perl6
15:45 JimmyZ joined #perl6
15:45 moritz o/
15:45 mberends o/
15:45 lichtkind mberends: cheers
15:46 birdwindupbird joined #perl6
15:46 masak moritz: \o
15:55 mberends lichtkind: hi, how does Kephra get on with Perl 6?
15:56 lichtkind mberends: was pretty much busy with articles
15:56 mberends lichtkind: yes, it was good to see all that activity :)
15:57 hirschnase joined #perl6
15:59 lichtkind and the is much more in planing
16:03 abercrombie Do we still have a rakudo August release despite of the memory leaking issue?
16:03 moritz there'll be one, yes
16:04 abercrombie Oh, so the leaking will get resolved in Sept release?
16:04 lichtkind due leaking
16:06 moritz abercrombie: or maybe the Aug release will be after the fix
16:06 lichtkind i con a different problem with my rakudo build; PARROT VM: Could not load bytecode; Incompatible versions of `perl6_ops' oplib, possibly due to loading bytecode generated by an old version of Parrot. Found 3.6.0 but loaded 3.7.0
16:06 lichtkind any clue what i could do?
16:06 moritz lichtkind: remove your installation directory
16:06 moritz lichtkind: which verb did you mean to write instead of 'con'?
16:06 lichtkind had
16:07 mberends enCOuNtered
16:07 moritz mberends++
16:07 lichtkind haha
16:07 lichtkind mberends++
16:07 mberends moritz: everybody knows that! ;)
16:07 lichtkind moritz: you mean rease the dir or change a var?
16:08 lichtkind mberends++
16:08 mberends lichtkind: rease the directory, or erase it of that doesn't work.
16:08 mberends *if
16:10 lichtkind mberends: maybe the pronlem is that i have 2 verson of rakudo compiled
16:10 lichtkind and the one set some env vars
16:12 mberends lichtkind: in that case changing PERL6LIB might help
16:14 lichtkind to late just reased my dir :)
16:14 mberends lichtkind: also check that the other perl6 is not in $PATH
16:15 lichtkind yeah i had the path to the other parrot in $PATH
16:16 [Coke] ni haA
16:16 [Coke] O_o?
16:16 masak [Coke]: 你好
16:17 [Coke] ni hǎo
16:18 [Coke] (has to be an easier way to type that on os x.)
16:19 [Coke] (than opening up the character input widget and clicking on the combining diacritical.
16:19 masak [Coke]: there are great input modes for Mac OS X.
16:20 [Coke] 你好
16:21 [Coke] for hanzi, yah.
16:22 * [Coke] will play around.
16:22 daniel-s joined #perl6
16:22 [Coke] masak: are you the one that recommend colloquial chinese?
16:22 [Coke] *ed
16:22 masak [Coke]: 是
16:23 JimmyZ dà jiā hǎo
16:25 Khisanth joined #perl6
16:25 masak [Coke]: how do you like it so far?
16:26 whiteknight joined #perl6
16:27 jlaire joined #perl6
16:30 [Coke] I need someone to speak out loud with, or the book's useless. ;)
16:31 [Coke] thankfully, I work with someone who's a native speaker. (but she doesn't know pinyan, so she needs to see the english first .;)
16:33 masak Pinyin.
16:33 JimmyZ pinyin?
16:35 JimmyZ good ningt
16:35 * JimmyZ sleeps
16:36 masak 'night, JimmyZ.
16:40 TimToady std: any<me expoits grammerz>
16:40 p6eval std e3c970e: OUTPUT«ok 00:01 118m␤»
16:40 TimToady hmm, not updated
16:44 TimToady evalbot rebuild std
16:44 p6eval OK (started asynchronously)
16:47 lichtkind TimToady: sorry for repeating my question but the .perl dumper method was your idea?
16:48 TimToady calling it .perl was my idea
16:49 TimToady in Python it's .repr, and I said to myself, but what if I want the representation in APL, or COBOL?
16:49 TimToady so I thought .repr was too general
16:50 masak seems the Apocalypses don't mention .perl
16:50 TimToady and .perl is almost a type conversion
16:50 TimToady I don't remember when I decided that
16:50 TimToady but .perl is more of a practical thing than an apocalyptic thing
16:51 TimToady std: any<me expoits grammerz>
16:51 p6eval std e3c970e: OUTPUT«ok 00:01 118m␤»
16:52 masak std: say any<one for peanuts?>
16:52 p6eval std e3c970e: OUTPUT«ok 00:01 119m␤»
16:52 TimToady strange
16:52 masak that STD likes peanuts? :P
16:52 TimToady std: say;
16:52 p6eval std e3c970e: OUTPUT«Potential difficulties:â�¤  Unsupported use of bare 'say'; in Perl 6 please use .say if you meant $_, or use an explicit invocant or argument at /tmp/qSSaVXDSDp line 1:â�¤------> [32msay[33mâ��[31m;[0mâ�¤ok 00:01 117mâ�¤Â»
16:53 TimToady std: any;
16:53 p6eval std e3c970e: OUTPUT«ok 00:01 117m␤»
16:53 TimToady bizarre, that's the version I pushed, and it complains about any here, but not as a  bot
16:54 * TimToady wonders if it's reporting the new version but running the old one somehow
16:55 masak perhaps it only thought it updated?
16:55 masak right.
16:55 TimToady std: WHICH;
16:55 p6eval std e3c970e: OUTPUT«Potential difficulties:â�¤  Unsupported use of bare 'WHICH'; in Perl 6 please use .WHICH if you meant $_, or use an explicit invocant or argument at /tmp/H9GUkjJW_z line 1:â�¤------> [32mWHICH[33mâ��[31m;[0mâ�¤ok 00:01 117mâ�¤Â»
16:55 TimToady that's the old version
16:55 TimToady because WHICH switch over to any's message
16:56 TimToady *ed
16:56 frhodes joined #perl6
16:56 benabik joined #perl6
16:57 * TimToady should rejoice for having exploits-moded ur evalbot, but doesn't
16:59 masak not unless you did it deliberately.
17:05 lichtkind TimToady: thank you
17:07 lichtkind TimToady: a seq() is a flat fully reified list?
17:09 cognominal tough to go between coffeescript and perl6.    coffeescript does not need parentheses on method call with arguments but requires them on method call without argument (to avoid confusion with attribute access).
17:09 cognominal on Perl 6 this is the opposite.
17:12 masak yes.
17:13 masak and Perl 6 *wants* you to confuse method calls with (public) attribute access. :)
17:17 TimToady lichtkind: something like that, though we're still a bit unsure whether we'll keep it
17:22 masak as with the Seq type as such, I guess.
17:25 * lichtkind understand each time reading WHICH just Witch
17:28 ccc_ joined #perl6
17:37 * lichtkind likes it also how * in  $*OUT  works like a placeholder like in F*** Y** when meaning STDOUT
17:37 masak I read it more like "...for now."
17:41 cognominal I thought that "confusion" was a design feature. As a user of a class I don't want to know if I call a non accessor method or call an accessor method. Attributes are just a implementtion detail.
17:42 cognominal ho, TimToady answered lichtkind.
17:43 cognominal like the compiler, I am esaily "confused"
17:45 wamba joined #perl6
17:45 cognominal btw, why the need for parentheses method calls for arguments?
17:46 moritz cognominal: the parser must always know whether to expect a term or an infix
17:48 cognominal I am not sure where it comes to play here.
17:48 moritz cognominal: if $a.foo  allowed parameters after it, but optionally, it could be both an infix (for the no argument case) or a term (for the argument case)
17:48 moritz s/paramters/arguments/
17:49 Mowah_ joined #perl6
17:50 TimToady sorear: does niecza ignore 'is equiv'?
17:53 TimToady niecza: multi infix:<∧> is equiv(&infix:<*>) ($a,$b) {say 'and'; $a min $b }; multi infix:<∨> is equiv(&infix:<+>) ($a,$b) {say 'or'; $a max $b }; say True ∨ True ∧ False
17:53 p6eval niecza v8-55-g0db7bb8: OUTPUT«or␤and␤Bool::False␤»
17:53 TimToady this should be and or True
17:53 TimToady or am I doing something rong?
17:59 masak maybe 'is equiv' is NYI.
18:09 TimToady usually niecza++ is good about carping when NYI
18:10 masak niecza: multi infix:<∧> is a-little-horse("niegh") ($a, $b) {}; say "alive"
18:10 p6eval niecza v8-55-g0db7bb8: OUTPUT«[31m===[0mSORRY![31m===[0mâ�¤â�¤Unhandled trait a-little-horse for this context at /tmp/nG5jMk0_gA line 1:â�¤------> [32mti infix:<∧> is a-little-horse("niegh") [33mâ��[31m($a, $b) {}; say "alive"[0mâ�¤â�¤Potential difficulties:â�¤  $a is declared but not used at…
18:10 TimToady niecza: multi infix:<∧> is tighter(&infix:<*>) ($a,$b) {say 'and'; $a min $b }; multi infix:<∨> is looser(&infix:<+>) ($a,$b) {say 'or'; $a max $b }; say True ∨ True ∧ False
18:10 p6eval niecza v8-55-g0db7bb8: OUTPUT«or␤and␤Bool::False␤»
18:12 TimToady std: any<me expoits grammerz>
18:12 p6eval std e3c970e: OUTPUT«ok 00:01 118m␤»
18:12 TimToady moritz: any kloo why std says it has the right version but doesn't?
18:14 moritz TimToady: does 'make snap' work for you, and reports the correct revision?
18:15 TimToady yes
18:16 TimToady and the any fails with: The 'any' listop may not be called without arguments (please use () or whitespace to clarify) at (eval) line 1
18:17 moritz I can git clean -xdf and see if that improves things
18:17 TimToady thanks
18:18 moritz Can't locate YAML/XS.pm in @INC
18:19 mberends moritz: perlbrew version change?
18:19 moritz mberends: most likely, yes
18:20 moritz so it seems that 'make snap' updates the revision file even for failed builds :(
18:20 dalek roast: 1aa4587 | Coke++ | S02-names_and_variables/names.t:
18:20 dalek roast: nom fudge
18:20 dalek roast: review: https://github.com/perl6/roast/commit/1aa4587a48
18:21 frhodes joined #perl6
18:21 dalek rakudo/nom: 0adc742 | Coke++ | t/spectest.data:
18:21 dalek rakudo/nom: run this fudged test
18:21 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/0adc742612
18:24 moritz std: any
18:24 p6eval std e3c970e: OUTPUT«ok 00:01 117m␤»
18:24 moritz std: any<foo>
18:24 p6eval std e3c970e: OUTPUT«ok 00:01 118m␤»
18:25 moritz that's a new, clean, successful build
18:26 TimToady ./viv -e 'any<foo>' in the snap directory should fail
18:26 moritz fwiw p6eval runs tryfile, not viv'
18:26 moritz not sure if it matters
18:27 moritz evalbot control restart
18:28 p6eval joined #perl6
18:29 TimToady tryfile fails for me here
18:29 moritz std: any<f>
18:29 TimToady I mean, succeeds in failing :)
18:29 p6eval std e3c970e: OUTPUT«ok 00:01 118m␤»
18:29 TimToady très étrange
18:32 TimToady 奇特
18:33 jevin joined #perl6
18:38 stepnem joined #perl6
18:41 flussence I just spent 3 hours on a ferry with no wifi and only a netbook - still managed to get some Pod::To::HTML stuff done though
18:42 Minimiscience joined #perl6
18:43 flussence it takes 5.5 minutes to generate s26.html, 3 of which are before pod2html even gets called :(
18:45 flussence oh wait, that was with all the debug output. 4m43 without...
18:48 kst joined #perl6
18:49 lichtkind what is the new name of /parrot_install/bin ?
18:49 flussence s/parrot_//
18:49 lichtkind thank you
18:50 lichtkind flussence: you mean with underscore?
18:50 flussence it's just install/bin now
18:51 lichtkind i see
18:51 lichtkind had some brain latency
18:51 lichtkind it was a regex
18:52 djanatyn joined #perl6
18:52 djanatyn Hi!
18:52 djanatyn masak: hello :D
18:54 lichtkind nom doest understand =begin END
18:54 lichtkind master did
18:55 donri masak: ^ look who i brought
18:55 masak djanatyn! \o/
18:55 masak djanatyn: how's things?
18:56 flussence tadzik: ping - seems like the pod parser doesn't do nested <>s inside FormattingCodes
18:57 flussence ("C<B<>>" in the spec is ending up as "<code>B&lt;</code>&gt;")
19:00 donri djanatyn: now that you're here, you'll have to write some perl6
19:01 flussence .oO( am I a bad person for using <mark>/<em> where the spec suggests <em>/<u>? :)
19:04 djanatyn masak: Doing great!. :)
19:04 djanatyn donri: I've written perl6 code before ;)
19:04 donri Show me.
19:04 djanatyn masak: I took a little bit of time off to take a dip in python.
19:04 djanatyn I already have, but sure.
19:04 djanatyn donri: https://github.com/djanatyn/arr​ay-of-doom/blob/master/game.pl
19:05 masak djanatyn: nice! how's python doing?
19:05 djanatyn masak: Great.
19:05 djanatyn I had originally planned to use Perl and SDL to create a game for a 48 hour game-making competition I participated in, ludum dare.
19:05 donri djanatyn: you can use "given" to assign $_
19:05 djanatyn However, I was having some issues and I eventually switched to python + pygame.
19:05 djanatyn donri: *huh*. cool.
19:05 donri given prompt { when ... }
19:06 masak djanatyn: heh, I guessed you'd gravitate to pygame :)
19:06 masak djanatyn: I've been glancing at it many times too.
19:06 djanatyn Yeah. I actually got a pretty decent game released and did better than I expected.
19:06 djanatyn Learning python quickly was fun. :)
19:06 masak \o/
19:06 frhodes joined #perl6
19:06 djanatyn Since the most substantial work I've done with OOP has been in perl 5, it was quite a paradigm shift
19:07 masak why?
19:07 djanatyn Since I haven't even really used Moose very much.
19:07 masak people say Python's OO and Perl 5's are pretty similar.
19:07 djanatyn I had always considered it a little bit of work to write object oriented perl.
19:07 djanatyn Like, writing a new method, and things like that.
19:07 masak without Moose, sure! :)
19:08 djanatyn Yeah, I meant without Moose :D
19:08 masak and people put a lot of cargo-cult crap in their 'new' methods :P
19:09 djanatyn I found it pretty fun to use.
19:09 djanatyn And of course. Perl 6's OOP is awesome, though I don't have much experience with it yet
19:09 masak all in good time ;)
19:10 masak nom: class OOP {}; class awesome {}; role Perl6[OOP] is awesome {}; say "yay!"
19:10 p6eval nom 0adc74: OUTPUT«yay!␤»
19:11 donri nom: say .WHAT given *
19:11 p6eval nom 0adc74: OUTPUT«Whatever()␤»
19:11 masak nom: say .WHAT given .WHAT
19:11 mj41 joined #perl6
19:11 p6eval nom 0adc74: OUTPUT«Any()␤»
19:11 frhodes joined #perl6
19:12 donri perl 6 will not have competitions in obfuscation but in code-as-natural-language
19:12 masak oh, it can do either :)
19:13 masak nom: my $dna = "ttaagg"; sub translate($dna) { "FFLLSSSSYY!!CC!WLLLLPPPPHHQQRRRRIIIMT​TTTNNKKSSRRVVVVAAAADDEEGGGG".comb[map { :4($_) }, $dna.trans("tcag" => "0123").comb(/.../)] }; say translate($dna)
19:13 p6eval nom 0adc74: OUTPUT«Method 'trans' not found for invocant of class 'Str'␤  in sub translate at /tmp/AYXOVlGxhu:1␤  in mu <anon> at /tmp/AYXOVlGxhu:1␤  in mu <anon> at /tmp/AYXOVlGxhu:1␤␤»
19:13 masak rakudo: my $dna = "ttaagg"; sub translate($dna) { "FFLLSSSSYY!!CC!WLLLLPPPPHHQQRRRRIIIMT​TTTNNKKSSRRVVVVAAAADDEEGGGG".comb[map { :4($_) }, $dna.trans("tcag" => "0123").comb(/.../)] }; say translate($dna)
19:13 p6eval rakudo a55346: OUTPUT«LR␤»
19:13 masak nom doesn't do .trans yet?
19:14 masak why wouldn't that be a showstopper to merging to master. .trans is everywhere.
19:14 Limbic_Region joined #perl6
19:14 masak s/\./?/
19:15 dalek niecza: a7575a1 | diakopter++ | / (8 files):
19:15 dalek niecza: make Kernel.cs compile under visual studio
19:15 dalek niecza: make Builtins a partial class
19:15 dalek niecza: add stub for sprintf
19:15 dalek niecza: add Printf.cs to Makefile
19:15 dalek niecza: review: https://github.com/sorear/niecza/commit/a7575a1d22
19:16 masak djanatyn: the above one-liner is from this blog post: http://strangelyconsistent.org/blog/​the-genetic-code-oneliner-in-perl-6
19:16 flussence masak: I think tadzik++ was trying to implement it for inside this Text::Escape module, turns out it's surprisingly complicated
19:16 masak djanatyn: I needed to translate TimToady's adieu from DNA bases to amino acids :)
19:16 masak flussence: you're telling me. :) I implemented it in master.
19:17 masak well, re-implemented it.
19:17 dalek niecza: a89a9a6 | diakopter++ | lib/Printf.cs:
19:17 dalek niecza: add Printf.cs
19:17 dalek niecza: review: https://github.com/sorear/niecza/commit/a89a9a623c
19:18 TimToady trans should probably turn into a little grammar that does LTM
19:18 masak guess it's not that hard to build grammars using the MOP.
19:19 TimToady but recognizing immutable arguments and not recreating the grammar will be a trick
19:19 masak TimToady: but when you put it like that, it feels like a caching solu... right.
19:19 kthakore djanatyn: how is it going?
19:20 masak I already asked that. :P
19:20 kthakore oh
19:20 kthakore :\
19:20 kthakore djanatyn: what were your issues? How can we make it better?
19:20 masak kthakore++ # double-checking
19:20 masak ooh, I wanted to ask that one :)
19:20 TimToady so maybe we don't write .trans just yet, but .transcomp and transrun
19:20 kthakore masak: thanks
19:20 donri issues with what?
19:21 kthakore donri =>  I had originally planned to use Perl and SDL to create a game for a 48 hour game-making competition I participated in, ludum dare. ... but I had some issues and I switched to Python and Pygame
19:21 kthakore donri: we are still youngish ...
19:21 donri oh
19:21 diakopter nom: say sprintf( "%d%d%d%d%d", 0...*)
19:21 p6eval nom 0adc74: OUTPUT«...␤»
19:21 djanatyn kthakore: Well...^_^
19:22 djanatyn I guess it was more personal issues.
19:22 kthakore masak: is 1year some months stil youngish?
19:22 kthakore djanatyn: that is fine your experience will help us
19:22 donri we also need a .transvestite method for polymorphism
19:22 djanatyn I found the pygame documentation easier to navigate, and their abstraction was a bit higher
19:22 kthakore djanatyn: ok
19:22 masak kthakore: for a project? I guess...
19:22 djanatyn I liked how you could set the framerate without having to add handlers in pygame
19:23 kthakore ok
19:23 djanatyn I could structure my code the way I wanted easily without having $app->run();
19:23 kthakore djanatyn: want to move this to pm btw?
19:23 djanatyn kthakore: uhh, sure
19:23 donri trying to keep the rest of us out eh
19:23 kthakore oh ... I was afraid of going of topic
19:23 kthakore this is perl5 module
19:23 kthakore sorry
19:23 donri ;)
19:24 kthakore masak++ thanks
19:24 kthakore apologies again
19:26 masak oh, come on :)
19:26 masak I should apologise to *you* for being so silly.
19:27 masak me being silly, I mean. not you.
19:27 masak also, there's no ban on talking about Perl 5 here. as long as you don't go overboard about it.
19:28 diakopter because then we'd have to throw you a lifeline or float ring
19:31 masak :P
19:32 masak "Perl 6: like being saved from drowning."
19:42 TimToady .oO(Hazardous Waters! Enter At Your Own Perl!)
19:44 kthakore heh
19:46 orphu joined #perl6
19:51 rlb3 joined #perl6
19:53 kst joined #perl6
19:56 diakopter TimToady: S32::Str sprintf grammar doesn't mention the ability to specify which argument supplies the vector override separator string
19:57 diakopter but perlfunc sprintf mentions it
19:57 masak diakopter: consider that grammar more a guideline than solid spec.
19:57 masak (I didn't write it, but I've been studying it)
19:57 * diakopter is working on sprintf for niecza
20:00 diakopter I'll assume the 5.14.1 docs are correcter
20:01 lichtkind diakopter: yes lots of docs effort lately
20:01 diakopter ?
20:01 flussence sprintf in rakudo is still pretty lacking, iirc.
20:01 djanatyn Well, I've got rakudo-star installed on my Fedora box.
20:01 flussence nom: printf('%d %1$x', 45);
20:01 p6eval nom 0adc74: OUTPUT«'$' is not a valid sprintf format␤  in sub sprintf at src/gen/CORE.setting:1508␤  in sub printf at src/gen/CORE.setting:1513␤  in mu <anon> at /tmp/W4k1fryc16:1␤  in mu <anon> at /tmp/W4k1fryc16:1␤␤»
20:01 masak djanatyn: great!
20:02 lichtkind diakopter: fram reading p5p i know the docs fot improved lately , even from 5.14 to 5.14.1
20:02 lichtkind s/fot/got/
20:07 masak TimToady: did you see the first observation in https://gist.github.com/1148915 ? do you agree that S06 needs a fix for that example?
20:08 masak TimToady: also, I'm concerned about http://irclog.perlgeek.de/​perl6/2011-08-18#i_4293141
20:08 masak (is writing on a macro grant proposal, so I got to thinking about those)
20:09 masak s/is/am/ # g'ah
20:09 flussence (is it just me or is github screwing up utf-8 code in a number of places?)
20:10 masak I noticed that too.
20:10 masak I went to edit it to fix it, but in the edit view it looked right. :/
20:10 flussence yeah, seems like it's on their end...
20:10 lue hello planet! o/
20:12 masak hi lue!
20:12 masak heh. it's so recently that I met TheDamian, that now when I read the Exegeses, they're all in Australian! :D
20:14 mj41 joined #perl6
20:14 lue [I heard he doesn't like IRC. He should visit sometime]
20:15 masak lue: don't immanentize the eschaton! o.O
20:16 masak TheDamian and IRC... it would be like matter and antimatter...
20:17 lue .oO(I'd think it'd only be immanentizing if I were to send him a letter about coming on)
20:18 masak lue: true, true. but you can never be too sure.
20:18 lue masak: would the resulting annihilation create and leave Perl 7 in the debris?
20:19 masak the chances of that are slim.
20:19 masak from a pure entropy perspective, we'd be much more likely to end up with PHP 7. :/
20:19 flussence AAAAAAAAAaaa!
20:19 masak exactly.
20:20 lue .oO(perhaps we can isolate #perl6 from the rest of the Internet just before annihilation to avoid PHP influences)
20:20 flussence (does that mean PHP will cause the heat death of the universe?)
20:20 sbp this is the scariest conversation I have so far seen in here
20:21 Juerd sbp: Were you having any conversation here at all, just now?
20:21 masak sbp: anything involving TheDamian takes on an edge of the scary.
20:21 sbp nope, I'm conserving my conversation
20:22 lue .oO(great. I just got *another* idea for a silly little project)
20:22 masak lue: how'd the game go?
20:23 lue I'm uploading the repo now. [ Just the rules so far, want to make sure they're OK before implementing :) ]
20:23 * masak is expectant
20:24 masak lue: by the way, have you been feeling like submitting an AI for Little Animal Farm lately? :)
20:25 lue It looked intereseting, have to take a crack at it [also interesting is a P6 IF game, rewriting my Pod parser..... gah! so many fun projects! I hath been converted!]
20:26 masak :)
20:26 lichtkind the speakers training in riga with thedamian was good
20:26 Mowah_ joined #perl6
20:26 * lichtkind currently practicing the learned
20:28 lue https://github.com/lue/Frotzlexity --- I don't trust my instincts about the rules, have a look.
20:29 lue [Of course, implementation will be a challenge, but any programming that deals with English is. Perhaps an esperanto version first...]
20:30 huf joined #perl6
20:31 Mowah_ joined #perl6
20:32 * masak looks
20:32 sbp or lojban
20:35 masak lue: actually, analyzing the word for correct usage doesn't seem that hard.
20:35 masak it's analyzing the rest of the sentence that's tricky.
20:36 masak the game looks nice.
20:37 djanatyn sorear: Did someone say lojban?
20:37 djanatyn ermm, sbp
20:37 masak lue: if I may offer an observation: it looks like even with made-up words, most people would be able to use them correctly in a sentence, purely based on their endings. there doesn't appear to be much of a challenge in it.
20:37 lue Indeed. It's the first way I thought of making the "frotzlexity dice" more than just dice-rolling. There's likely a better way to turn it into a game...
20:38 * masak realizes that mathw has been kinda silent lately
20:38 masak lue: I could think of a couple of ways. but it's your game :)
20:39 * lue thinks about a more all(fun,challenging,interesting) game...
20:40 lue At least I didn't start implementing anything...
20:40 djanatyn kthakore: http://paste.pocoo.org/show/465458/
20:40 masak lue++ # releasing early and often
20:40 djanatyn that's how the clock object works in pygame
20:41 pernatiy joined #perl6
20:41 masak nom: say <per>.roll, <naity>.roll
20:41 p6eval nom 0adc74: OUTPUT«pernaity␤»
20:41 sbp loaded dice!
20:41 masak oops.
20:41 masak nom: say <per>.roll, <natiy>.roll
20:41 p6eval nom 0adc74: OUTPUT«pernatiy␤»
20:41 masak \o/
20:42 masak sbp: not loaded, just d1 dice :P
20:42 sbp and loading is a d6 to d1 transformation?
20:42 djanatyn http://paste.pocoo.org/show/465461/ <-- sample output of the first 50 iterations on my machine
20:43 djanatyn not sure how I'd expect it in perl
20:43 djanatyn but I like pygame's approach to it
20:48 lue nom: say ("a".."z").roll(6).join('');
20:48 p6eval nom 0adc74: OUTPUT«pnhatr␤»
20:50 masak pn hatrs will pn hat.
20:51 kthakore djanatyn: ok
20:51 lue .oO(it could also perceived as the first step of a backronym generator)
20:53 sbp nom has a mind of its own
20:53 pernatiy masak: what so special in my nick? ;)
20:54 masak pernatiy: oh, hi :) sorry, just being Saturday-evening silly.
20:54 masak pernatiy: how are you liking Perl 6 (and #perl6) so far, by the way?
20:55 masak (besides the silliness, I mean)
20:56 pernatiy currently I am working with perl5, but I interested in perl6
20:56 sbp you have like the duodecimal of nicks. it makes perfect sense, but hardly anybody knows it yet
20:56 pernatiy not sure if I like it
20:56 pernatiy to be honest I not sure if I like any programming language
20:56 uvtc joined #perl6
20:56 masak pernatiy: :)
20:56 pernatiy =)
20:56 masak pernatiy: so you're the Steve Yegge type, then.
20:56 pernatiy their all are just tools
20:57 pernatiy *they
20:57 masak yes. they're just substrates for programs.
20:57 uvtc kthakore, hi. Just curious: what is the status of SDL bindings for Perl 6? I poked around in github's PerlGameDev/SDL6,
20:58 uvtc kthakore, but am not sure how far along it is.
20:58 masak pernatiy: what do you like about Perl 5? (if anything)
20:59 TimToady or contrariwise, what do you dislike the least? :)
20:59 pernatiy perl6 have some features which I miss in perl5, but I need to do some work on perl6 to make complete opinion
20:59 masak of course.
20:59 pernatiy well, I like perl5 for it's XS capaility
21:00 masak well, we don't have that :)
21:00 ingy Pegex is self hosting!!
21:00 pernatiy Inline::C package is one of the best =)
21:00 masak ingy: \o/
21:00 * ingy does a little dance
21:00 pernatiy also OO approach in perl5 is funny
21:01 masak pernatiy: as in "ridiculous"? :P
21:01 pernatiy not much people understand it but it is brilliant from some point of view
21:01 TimToady masak: yes, the circumfix macro is wrong
21:01 ingy does rakudo have access to a pcre/p5 re engine yet?
21:01 masak TimToady: I suggest either you fix it, or you suggest a fix and I fix it.
21:02 pernatiy the way how it done allow you to use some similar technic in not OO programming
21:02 TimToady well, we'd need a syntactic category "transparent"
21:02 masak ingy: maybe through blizkost. I don't know that anyone has tried.
21:02 TimToady or maybe it belons in ws somewhere
21:02 * ingy should start writing a p6 grammar in pegex so he can parse p6 in python
21:02 pernatiy so you can mix OO style with procedural, in quantity which you need
21:02 flussence or some `is native<libpcre>` magic
21:02 masak TimToady: isn't there a category "comment"?
21:03 TimToady I guess there is
21:03 masak why "transparent" and not "comment"?
21:03 TimToady was just thinking of something more general, but it doesn't really work
21:04 TimToady unless you can have a macro that matches anywhere
21:04 TimToady which is not really what anyone wants
21:04 masak especially not the poor schmuck who promised to implement them :/
21:04 Tene masak: let me know how your macros work goes; I'm very interested in that.
21:04 masak Tene: thanks! I will.
21:04 Tene I've come pretty close to working on it myself half a dozen times over the past few years.
21:04 natureboy joined #perl6
21:05 masak Tene: I was really surprised when it fell into place for me how regular and simple they really are.
21:05 Tene I should have published the prototypes I wrote; I can't find them anymore.
21:05 masak Tene: at least the AST-based macros.
21:06 envi joined #perl6
21:07 Fuad joined #perl6
21:07 Fuad hi
21:07 masak Fuad! \o/
21:07 Fuad masak: Hello bro!
21:07 masak wow, this is like a family gathering!
21:08 Fuad :)
21:09 TimToady masak: and yes, the RC macro is bogus, but let's be sure that an AST can stringify back to its original code somehow, eh?
21:09 masak TimToady: sure.
21:09 masak TimToady: that might actually solve it in that particular case.
21:10 daemon joined #perl6
21:11 soh_cah_toa joined #perl6
21:13 masak (then again, maybe not)
21:17 TimToady almost certainly not...
21:18 diakopter TimToady: I had a question about S32::Str sprintf ^^
21:19 TimToady after all these years, I still don't know if I like that feature of sprintf; it seems wrong somehow 
21:20 TimToady we should be doing that with named args or some such
21:20 * diakopter whimpers
21:21 masak what's the last/latest release of Pugs?
21:21 diakopter http://hackage.haskell.org/package/Pugs
21:21 masak diakopter: thanks.
21:21 diakopter she updates it everytime a new ghc is released I think
21:22 masak yes, that's what I guessed too.
21:22 masak 6.2.13, it seems.
21:22 masak which is between 6.2 "Basic IO and control flow elements; mutable variables; assignment" and 6.28 "Classes and traits"
21:23 masak even though Pugs has classes, at least.
21:24 TimToady the MOP never dropped into it, but turned into Moose instead
21:24 masak oh? was there a point when the MOP was targeted for Pugs, before it was Moose?
21:25 masak that's not how I remember it...
21:25 masak (which is not to say that it didn't happen that way)
21:26 donri is pugs doomed to oblivion?
21:27 flussence small random unrelated question - is there an obvious way to do (?$string.looks-like-an-int-literal) that I don't know about?
21:28 Tene flussence: match against the int rule in the perl6 grammar
21:28 masak donri: "Cannot predict now"
21:28 masak or maybe "Outlook not so good"
21:28 donri i recall hearing something about perl6 made the pugs design difficult
21:29 TimToady perl6 makes every implementation difficult
21:29 donri its use of parsec maybe
21:29 Tene donri: right now, nobody has made public plans to adopt and maintain and improve pugs.  Unless someone starts working on it, nobody will work on it.
21:29 TimToady well, it doesn't need to use parsec anymore
21:29 donri kinda sad :(
21:30 Tene donri: it's entirely possible that someone could come in here tomorrow and say "Hey, I'm working on Pugs, and I was wondering about ...", but it seems unlikely.
21:33 TimToady masak: iirc, stevan++ was designing the MOP at least in part for Perl 6 at the Toronto hackathon, but I don't know if he was intending to hack it into pugs himself or delegate that to someone like nothingmuch++, but anyway, it didn't happen.
21:33 masak ok.
21:34 TimToady or maybe audreyt++ was going to hack it in
21:34 * masak would like to meet stevan some day
21:34 TimToady anyway, it crashed
21:34 masak well, we got a MOP for Rakudo :)
21:34 masak which is not entirely disjunct from Moose.
21:45 explorer__ joined #perl6
21:49 wamba joined #perl6
21:57 silug i really should catch up on that one of these years.
21:58 silug oops, wrong channel.  :)
21:58 masak thought you meant Moose and Rakudo's MOP ;)
21:58 silug well, sure, that too.  :)
22:00 silug why modern window managers don't have focus-follows-eyes yet, i'll never understand.
22:01 masak TimToady: how are method-like macros such as .WHAT declared?
22:02 masak silug: ugh, and I thought focus-follows-mouse was bad enough :/
22:03 silug masak: i don't tend to accidentally type in the wrong window because my mouse has moved on though...  :)
22:04 masak the few times I've had focus-follows-mouse, I've done that.
22:04 masak YMMV, I guess.
22:05 silug no, no, focus-follows-mouse is an abomination.  i agree.  :)
22:10 TimToady masak: I presume they could just be postfix:<WHAT>
22:10 TimToady but maybe not
22:10 TimToady might have to be postfix:<.WHAT>
22:11 TimToady and disallow . form on that as we do on \w-ish postfixes
22:11 flussence .oO( I'd prefer mouse-follows-eyes )
22:11 masak TimToady: that oughta work, yes.
22:11 TimToady left-eye blink is left button, right-eye is right button, both eyes, is center :)
22:12 huf nah, that's uncomfortable
22:12 huf we need mindreading
22:12 TimToady well, we could probably tell intentional eyeblink from unintentional even now
22:13 masak but what if I want to drag-and-drop?
22:13 TimToady drag it with your other eye, assuming you don't use lasers?
22:14 masak ok, reasonable.
22:14 masak center-button drags are uncommon anyway.
22:14 uvtc first you think, you wink, you do a double blink, ... close your eyes, and jump
22:15 TimToady maybe clicks should do tongue recognition
22:15 TimToady and snakes could click or drag two things at once
22:15 huf you dont need double blinks or mouse buttons actually, if you're not using the mouse, you have your hands on the kb anyway
22:18 TimToady std: any<foo>
22:18 lue for drag-and-drop, clamp your mouth shut.
22:18 p6eval std e3c970e: OUTPUT«ok 00:01 118m␤»
22:18 TimToady sigh
22:19 huf the other solution is genetic engineering
22:19 huf i'd like having 4 arms :)
22:19 huf just imagine, a world where you can eat chips while ircing WITHOUT getting your keyboard sticky...
22:20 ruz who tried jnthn's grammar::tracer? master complains when I try to use the module. didn't try nom as I think it still lacks support for grammars :(
22:20 lue just imagine, a world where caffeine is a vital nutrient...
22:20 TimToady ruz: nom is precisely where it will work, since it uses 6model
22:21 masak yeah, won't ever work in master.
22:21 masak ruz: nom has support for grammars, just not full support. yet.
22:22 ruz yep, so I'm still wating for nom to catch up with grammars
22:22 ruz masak: spectest rate suggests that it's really limitted
22:23 ruz going to check if nom compiles grammar I have
22:24 lue afk
22:24 TimToady going to check if a nap is about to occur
22:28 abercrombie joined #perl6
22:41 drbean joined #perl6
22:47 Minimiscience joined #perl6
22:56 cexsum joined #perl6
22:58 ZaphrodZenovka joined #perl6
23:26 ruz cool, with a few changes grammar is parsed and traced
23:26 masak \o/
23:30 am0c joined #perl6
23:39 masak heh. didn't have time for a blog post tonight. will have time tomorrow for at least one, though.
23:39 masak 'night, #perl6.
23:39 sili joined #perl6
23:39 Psyche^ joined #perl6
23:42 kfo joined #perl6
23:55 Khisanth joined #perl6
23:57 flussence woo, got L<#Section> things working
23:57 ruz how hard it would be to say in a grammar that here may go indented block with <this> inside?

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

Perl 6 | Reference Documentation | Rakudo