»ö« Welcome to Perl 6! | perl6.org/ | evalbot usage: 'perl6: say 3;' or rakudo:, niecza:, std:, or /msg p6eval perl6: ... | irclog: irc.perl6.org/ | UTF-8 is our friend!
Set by sorear on 4 February 2011.
00:06 jevin left 00:08 jevin joined 00:11 lichtkind left 00:13 frhodes joined 00:25 frhodes left 00:26 whiteknight joined 00:27 lichtkind joined
lichtkind TimToady: was .perl your idea? 00:28
i mean the dumper method
allright have to sleep now 00:33
cheers
00:33 lichtkind left 00:43 lateau_ joined 00:45 lateau_ left 00:53 wk joined 01:11 uasi joined 01:17 colomon left, uasi left 01:19 wamba left 01:20 jevin left 01:21 jevin joined 01:24 uasi joined 01:26 uasi left 01:30 whiteknight left 01:33 thou left 02:03 Minimiscience joined
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 02:10 s1n joined 02:12 [Coke] left
TimToady rosettacode.org/wiki/Ternary_logic#Perl_6, just for mikemol++ 02:13
02:13 [Coke] joined
TimToady decommuting & 02:14
02:16 wk left
abercrombie perl is really concise compared to others.. 02:17
02:18 risou is now known as risou_awy 02:21 agentzh joined 02:23 Chillance left
mikemol Neat. 02:27
mikemol thanks TimToady++
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
let me find an example
mikemol: www.programmersheaven.com/2/Perl6-FAQ-Multi 02:30
02:31 f00li5h joined, cggoebel left
soh_cah_toa mikemol: are you familiar w/ method overloading in c++? it's the same thing 02:32
well, the same concept that is
mikemol soh_cah_toa: I've written code with multis before, but none of these have the same name. 02:36
soh_cah_toa oh ok. probably missed the beginning of that conversation :) 02:37
why, did someone suggest using 'multi' over 'sub' even for subs w/ different names
mikemol soh_cah_toa: rosettacode.org/wiki/Ternary_logic#Perl_6
02:38 jevin left
mikemol TimToady++ wrote it. I'm just trying to figure out the specific reason to use multi instead of sub. 02:38
soh_cah_toa for subs that define some type of operator, you need to use multi 02:39
notice how they all use prefix:<something>
mikemol Ah
soh_cah_toa it's the same as operator overloading which still involves multiple dispatch 02:40
02:40 jevin joined
soh_cah_toa keyword: 'multiple' ;) 02:40
oh and s/prefix/infix/ mixed that up 02:41
but you get the point :)
mikemol Well, ¬ uses prefix.
02:42 wolfman2000 left
mikemol niecza: multi infix<¬> ($a, $b) {$a + $b} 02:42
02:42 Minimiscience left
p6eval niecza v8-55-g0db7bb8: OUTPUT«===SORRY!===␤␤Malformed block at /tmp/QBL7ilCxu7 line 1:␤------> multi infix⏏<¬> ($a, $b) {$a + $b}␤␤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
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:<¬>() 02:44
soh_cah_toa hm, i'm sure it's possible but i'm not too familiar w/ tinkering around w/ meta objects 02:47
perl6: prefix:<+>.^signature
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 @…
..rakudo a55346: OUTPUT«too few positional arguments: 0 passed, 1 (or more) expected␤ in main program body at line 1:src/gen/core.pm␤»
..pugs: OUTPUT«*** No compatible multi variant found: "&prefix:+"␤ at /tmp/BTXdCc6XDy line 1, column 1 - line 2, column 1␤»
soh_cah_toa yeah...not too sure about that :\ 02:48
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␤»
..pugs: OUTPUT«*** No such method in class Class: "&signature"␤ at /tmp/WzPVcLW7Vl line 1, column 1 - line 2, column 1␤»
..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…
plobsing_ perl6: &prefix:<+>.signature
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-…
..rakudo a55346: OUTPUT«get_attr_str() not implemented in class 'Sub'␤ in main program body at line 1:src/metamodel/RoleToInstanceApplier.nqp␤»
..pugs: ( no output )
plobsing_ pugs: &prefix:<+>.signature.say 02:50
p6eval pugs: OUTPUT«:($?1 is copy)␤»
soh_cah_toa access to the mop uses the .^ syntax
plobsing_ you don't need the mop
the sub knows its signature
soh_cah_toa to get the signature?
oh right
the capture or parcel or something right?
plobsing_ using the .^ would ask class Sub for a signature 02:51
pugs: &prefix:<+>.WHAT.say
p6eval pugs: OUTPUT«Sub␤»
soh_cah_toa that's odd. on my installation that code gives me 'Code()' 02:55
rakudo: &prefix:<+>.WHAT.say
p6eval rakudo a55346: OUTPUT«Code()␤»
plobsing_ I suspect that's a parrot leak, which makes some methods give parrot errors 02:56
soh_cah_toa ok
plobsing_ rakudo, pugs: &prefix:<+>.Str
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␤»
plobsing_ rakudo: &prefix:<+>.Str
p6eval rakudo a55346: OUTPUT«get_attr_str() not implemented in class 'Sub'␤ in main program body at line 1:src/metamodel/RoleToInstanceApplier.nqp␤»
soh_cah_toa you're right. now it's class 'Sub'
02:58 cexsum left
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 :) 03:00
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
rakudo: &infix:<+>.^methods(:local)
p6eval rakudo a55346: ( no output )
plobsing_ rakudo: &infix:<+>.^methods(:local).say
p6eval rakudo a55346: OUTPUT«candidatesmultiname␤»
plobsing_ rakudo: say for &infix:<+>.^methods(:local) 03:02
p6eval rakudo a55346: OUTPUT«␤␤␤»
plobsing_ rakudo: .say for &infix:<+>.^methods(:local)
p6eval rakudo a55346: OUTPUT«candidates␤multi␤name␤»
plobsing_ my copy must be old, I get a different result
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)
it prunes out a lot of irrelevant methods
soh_cah_toa oh, just the methods it defines. i see 03:04
ruz grammar tracer and debugger, suspect these are only useful with master branch only? 03:06
TimToady perl6: say ~&prefix:<+> 03:07
p6eval niecza v8-55-g0db7bb8: OUTPUT«Sub()<instance>␤»
..rakudo a55346: OUTPUT«&prefix:<+>␤»
..pugs: OUTPUT«␤*** No compatible multi variant found: "&prefix:+"␤ at /tmp/BMW1VCzYnP line 1, column 1 - line 2, column 1␤»
TimToady perl6: say &prefix:<+>.name 03:10
p6eval pugs: OUTPUT«&::␤»
..rakudo a55346: OUTPUT«&prefix:<+>␤»
..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 …
soh_cah_toa rakudo: &prefix:<+>.name.comb[9] 03:12
p6eval rakudo a55346: ( no output )
soh_cah_toa rakudo: &prefix:<+>.name.comb[9].say
p6eval rakudo a55346: OUTPUT«+␤» 03:13
soh_cah_toa i win :)
although, that's a really horrible way to do it
rakudo: &prefix:<+>.name.comb(/<punct>/)[3].say 03:15
p6eval rakudo a55346: OUTPUT«+␤»
soh_cah_toa timtowtdi baby ;) 03:16
03:18 cexsum joined
mikemol Does that work with infix, too? 03:19
soh_cah_toa you'd need a different index number though
i totally suggest not using that though. i was just goofing around 03:20
TimToady in any case, niecza doesn't support .name yet
and the code in question doesn't run under rakudo for some reason 03:21
03:23 Su-Shee_ joined 03:26 Su-Shee left 03:30 colomon joined 03:41 uvtc joined 03:42 abercrombie left 04:01 uvtc left 04:19 kaare_ joined 04:30 kaare_ left
moritz good morning 04:43
soh_cah_toa good night ;)
04:44 soh_cah_toa left 04:53 kaare_ joined 05:04 birdwindupbird joined 05:05 REPLeffect left 05:19 REPLeffect joined 05:29 Chillance joined 05:37 mj41 joined, Chillance left, Chillance joined 05:41 xinming left 05:49 xinming joined 05:56 mj41 left 06:15 sili left 06:21 sili joined 06:29 ZaphrodZenovka left 06:31 ZaphrodZenovka joined 06:46 daniel-s joined 06:48 kst left 06:49 kst joined 07:04 woosley joined 07:39 yath joined 07:43 envi left 07:48 envi joined 07:51 birdwindupbird left 07:56 dual left 08:06 molaf joined 08:19 Su-Shee_ is now known as Su-Shee 08:23 wk joined 08:25 wk left 08:34 wk joined 09:06 wamba joined 09:08 molaf left 09:10 orafu left, orafu joined 09:14 whiteknight joined 09:24 masak joined
masak morning, #perl6 09:24
phenny masak: 26 Aug 23:24Z <TimToady> tell masak "receptible" and "pund", unless they were punned.
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
masak TimToady: found and fixed s/pund/pound/. I think I might not see what's wrong with "receptible". :/ 09:30
mberends ahoj masak: wow. your last blog post was a long read! It must have been a long write :) 09:35
masak mberends: I almost felt like Stevey there for a while. 09:36
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.
It seemed the ideal opportunity to mention tote as well. 09:43
masak ooh 09:45
I'm not saying manual testing is wrong. I'm saying my language sense about the word has been warped.
09:46 wullikam joined
mberends ah. warped. very reflective :) 09:47
masak adding mention about ad-hoc tests being beneficial and about tote. mberends++ 09:48
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?
mberends masak: no. on the contrary, those produce some of the best tests. 09:49
masak then we are in accord. 09:50
mberends violently
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. 09:51
mberends hmm, nice idea. The creators are often less able to "turn against" than some users from hell, though. 09:53
masak yes.
that was a (commonly known) problem that I experienced when writing the adventure game.
I kept running along the "happy path".
mberends :) 09:54
masak it was beta testers who kept finding bugs along various other paths.
09:54 wullikam left
masak in fact, I plan to re-write the whole game using tests. 09:54
maybe not this weekend, but sometime soon.
mberends excellent - that's the approach you also used with the module installation manager (post-proto). It worked very well. 10:00
masak yes.
and that now, IIUC, forms the basis of Pies.
mberends today I'll be writing some docs for Niecza - a kind of developer's guide. Testing forms a big part of the procedure. 10:04
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
mberends++ # docs for Niecza
10:10 pernatiy joined
masak pernatiy: hi! 10:10
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 …
cognominal is this problem known? that makes grammars not very useful. 10:25
10:25 kst left 10:26 kst joined
mberends cognominal: probably just NYI, such things should work 10:26
cognominal I think every language created with create_language.pl depends on it
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␤␤»
cognominal HLL::Grammar is not on the path, I checked
next is QRegex
mberends I investigated a little when you previously asked. There are similar examples that work in earlier Rakudo branches, so it *almost* works.
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, 10:31
these are probably some pbc that should be loaded. 10:32
mberends nom: say "%s and %s".sprintf(<salt pepper>); 10:33
p6eval nom 2742f6: OUTPUT«salt and pepper␤»
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?
cognominal I have trouble to understand the bootstrap process so I am not sure what can be loaded at this final stage.
where are the pbc loaded anyway? 10:34
masak mberends: +1 for de-spec. 10:37
10:37 MayDaniel joined
masak mberends: sending a message to a format string feels... weird. 10:37
mberends: a bit like python's '|'.join(stuff)
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... 10:38
masak mberends: we already have $value.fmt($format) 10:39
mberends true. I'd like to see a few more opinions before committing
(not bikeshed ones tho') 10:40
10:42 wamba left 10:45 szabgab left
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 10:47
mberends masak: sprintf is a predecessor of variable interpolation, so Perl seldom needs it but C does. Likewise strftime(). 10:49
masak well, I'm not disputing that. 10:52
mberends anyhow, where do you think printf() gets its hard work done?
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
just like with .join
the format string isn't the main agent here, the list of values to be substituted in are.
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.
that's also the conservative route, that can later be changed.
nom: say ("a", "b", "c").WHAT 10:55
p6eval nom 2742f6: OUTPUT«Parcel()␤»
mberends feels the "forgiveness" itch overriding the "permission" twitch 10:56
masak nom: augment class Parcel { method sprintf($format) { sprintf $format, |self } }; say ("Martin", "Berends").sprintf("I trust %s %s with a spec commit bit.")
p6eval nom 2742f6: OUTPUT«===SORRY!===␤augment not allowed without 'use MONKEY_TYPING' at line 1, near "{ method s"␤»
masak nom: use MONKEY_TYPING; augment class Parcel { method sprintf($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.␤»
masak \o/
mberends phew
masak time to make lunch.
mberends gorgonzola: mouldy by design 10:58
masak nom: say Parcel ~~ List
p6eval nom 2742f6: OUTPUT«Bool::False␤»
masak nom: say Parcel.^parents
p6eval nom 2742f6: OUTPUT«Cool() Any() Mu()␤»
masak boy, I'm starting to really like those spaces. :) 10:59
nom: say Parcel.^parents(:local) 11:01
p6eval nom 2742f6: OUTPUT«Cool()␤»
masak arguably a slight overloading of both a method and the term "parent", but alright. 11:02
I can see the type of consistency that was chose, with .^methods
chosen*
mberends "was chose" sounds perfectly good in some states ;) 11:10
cognominal it seems that HLL::Grammar depends on protoregex that are NYI :( 11:11
11:13 szabgab joined
mberends o/ szabgab! 11:15
11:22 risou_awy is now known as risou
flussence
.oO( oh dear, my urxvt's forgotten how to draw uppercase "m". )
11:23
11:29 agentzh left 11:32 ilogger2 joined, ChanServ sets mode: +v ilogger2 11:35 zby_home_ joined 11:36 wamba joined 11:44 thelazydeveloper joined
masak flussence: software is hard! 11:46
11:48 agentzh joined 11:52 wamba left 11:56 ilogger2 joined, ChanServ sets mode: +v ilogger2 12:01 jedai_ joined 12:08 jrockway joined
masak I just want to ++ sorear on general principle. 12:25
sorear: thank you for Niecza.
sorear++
12:32 mkramer joined
mberends seconds that. sorear++ :) 12:33
masak having Rakudo only, we're a cyclops. with Niecza and Rakudo, we get stereo vision. 12:42
depth.
mberends Niecza makes perl6.org/compilers/features much more interesting. 12:44
12:44 lichtkind joined 12:45 woosley joined
masak someone came to me and jnthn and pmichaud during the YAPC::EU hallway track. 12:46
he said "shall we revive Pugs?". pmichaud said "why not?". I said "oh YESS!"
he said he'd proposed it here on channel but got Warnock'd. 12:47
we all told him he was just unlucky and should come back and try again.
here's how you revive Pugs, by the way:
1. get together a dedicated group of about three people.
2. form a Haskell study group, open to everyone, meeting each week for 15 minutes or so. 12:48
3. ???
4. Pugs!
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
I suggest offering a grant to someone dedicated enough to revive Pugs 12:50
flussence Haskell doesn't look that hard... :)
(and I never would've said that before learning p6...)
12:50 wamba joined
arnsholt mberends: I think the main challenge with Haskell is getting into the whole mindset of it all 12:51
(I speak pidgin Haskell from configuring my XMonad) 12:52
lichtkind if @@ gone?
arnsholt ?
mberends what was @@ ?
lichtkind lol context op 12:53
so i think that || is also gone 12:56
wait im talking crap @@ went to **@ long before 12:58
masak lichtkind: || replaced @@ replaced **@ 13:06
prefix:<||> is still there, AFAIK.
arnsholt: I also speak pidgin Haskell, from writing some small things on my own, and from reading Pugs source code.
13:07 pochi joined
lichtkind masak: wiht || i mean positional flattenign of captures which should be still valid 13:08
masak that's what I mean, too. 13:09
lichtkind its no longer ||?
masak no, I think it still is. 13:11
(how did you get the impression from what I said...? oh, forget it...) :P
[Coke] tl;dr 13:17
lichtkind masak: before i forgot it i wated to say that "lichtkind: || replaced @@ replaced **@" confused me a bit 13:18
masak lichtkind: it's no more confusing than 3 < 4 < 5 IMO. 13:19
masak sees www.mdk.me.uk/fun/raptor/vader/index.html again and laughs
13:24 abercrombie joined 13:28 wamba left
masak mberends: re the "exploit" stance: twitter.com/carlmasak/status/11820728127066112 :) 13:30
mberends masak: indeed. like calling a class 'is' :) 13:34
13:39 donri joined
masak I sure fell for that one :) 13:39
mberends "Click Download to download Download"
masak "Click to download Click" 13:43
mberends There is the reverse problem choosing unique names in programming. I seem to collide increasingly with reserved words (eg next, date, list). 13:44
masak ooh, an ancient autopun by quietfanatic: twitter.com/carlmasak/status/8715995394 13:45
mberends: sigils are nice that way. 13:46
13:47 jevin joined 13:59 birdwindupbird joined 14:12 dorlamm joined, wamba joined 14:18 hirschnase joined 14:21 hirschnase left 14:25 birdwindupbird left
masak realizes djanatyn hasn't been around in a while 14:26
hope he comes back sometime.
14:37 woosley left
donri how you know djanatyn? :p 14:43
14:43 JimmyZ joined
masak he came in here last month. 14:47
JimmyZ good evening, #perl6 14:51
14:51 wamba left
masak 晚上好, JimmyZ 14:51
donri i want to be missed too, i should leave
lichtkind haha
donri nah no one would notice
JimmyZ 下午好, masak 14:52
donri commences arm cutting instead
lichtkind donri: love always comes for free
cant force it
donri can't by me loooove love 14:53
14:53 jevin left
donri buy also 14:53
masak donri: you just reminded me of en.wikipedia.org/wiki/Parable_of_th...odigal_Son
donri :D 14:54
masak i.e. there's more joy in someone coming back than in someone being around all the time. :)
donri yay parrots www.picrandom.com/images/6cim.jpg 14:56
masak :)
donri dat glass 14:57
14:57 jevin joined 15:17 isBEKaml joined 15:30 isBEKaml left 15:32 araujo joined 15:35 ilogger2 joined, ChanServ sets mode: +v ilogger2, birdwindupbird joined
masak nom: say "windup".join("bird" xx 2) 15:36
p6eval nom 2742f6: OUTPUT«windup␤»
masak oh, right. not Python :)
nom: say join "windup", "bird" xx 2 15:37
p6eval nom 2742f6: OUTPUT«birdwindupbird␤»
mberends nom: "%swindup%s".sprintf("bird" xx 2).say # :P 15:38
p6eval nom 2742f6: OUTPUT«birdwindupbird␤»
JimmyZ nom: dir 15:43
p6eval nom 2742f6: ( no output )
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 15:45 JimmyZ joined
moritz o/ 15:45
15:45 birdwindupbird left
mberends o/ 15:45
lichtkind mberends: cheers
15:46 birdwindupbird joined
masak moritz: \o 15:46
mberends lichtkind: hi, how does Kephra get on with Perl 6? 15:55
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
lichtkind and the is much more in planing 15:59
abercrombie Do we still have a rakudo August release despite of the memory leaking issue? 16:03
moritz there'll be one, yes
abercrombie Oh, so the leaking will get resolved in Sept release? 16:04
lichtkind due leaking
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
any clue what i could do?
moritz lichtkind: remove your installation directory
lichtkind: which verb did you mean to write instead of 'con'?
lichtkind had
mberends enCOuNtered 16:07
moritz mberends++
lichtkind haha
mberends++
mberends moritz: everybody knows that! ;)
lichtkind moritz: you mean rease the dir or change a var?
mberends++ 16:08
mberends lichtkind: rease the directory, or erase it of that doesn't work.
*if
lichtkind mberends: maybe the pronlem is that i have 2 verson of rakudo compiled 16:10
and the one set some env vars
mberends lichtkind: in that case changing PERL6LIB might help 16:12
lichtkind to late just reased my dir :) 16:14
mberends lichtkind: also check that the other perl6 is not in $PATH
lichtkind yeah i had the path to the other parrot in $PATH 16:15
[Coke] ni haA 16:16
O_o?
masak [Coke]: 你好
[Coke] ni ha􏿽xCC􏿽x8Co 16:17
(has to be an easier way to type that on os x.) 16:18
(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.
[Coke] 你好 16:20
for hanzi, yah. 16:21
[Coke] will play around. 16:22
16:22 daniel-s joined
[Coke] masak: are you the one that recommend colloquial chinese? 16:22
*ed
masak [Coke]: 是
JimmyZ dà jiā hǎo 16:23
16:25 Khisanth joined
masak [Coke]: how do you like it so far? 16:25
16:26 whiteknight joined 16:27 jlaire joined
[Coke] I need someone to speak out loud with, or the book's useless. ;) 16:30
thankfully, I work with someone who's a native speaker. (but she doesn't know pinyan, so she needs to see the english first .;) 16:31
masak Pinyin. 16:33
JimmyZ pinyin?
good ningt 16:35
JimmyZ sleeps
16:35 JimmyZ left
masak 'night, JimmyZ. 16:36
TimToady std: any<me expoits grammerz> 16:40
p6eval std e3c970e: OUTPUT«ok 00:01 118m␤»
TimToady hmm, not updated
evalbot rebuild std 16:44
p6eval OK (started asynchronously)
lichtkind TimToady: sorry for repeating my question but the .perl dumper method was your idea? 16:47
TimToady calling it .perl was my idea 16:48
in Python it's .repr, and I said to myself, but what if I want the representation in APL, or COBOL? 16:49
so I thought .repr was too general
masak seems the Apocalypses don't mention .perl 16:50
TimToady and .perl is almost a type conversion
I don't remember when I decided that
but .perl is more of a practical thing than an apocalyptic thing
std: any<me expoits grammerz> 16:51
p6eval std e3c970e: OUTPUT«ok 00:01 118m␤»
masak std: say any<one for peanuts?> 16:52
p6eval std e3c970e: OUTPUT«ok 00:01 119m␤»
TimToady strange
masak that STD likes peanuts? :P
TimToady std: say;
p6eval std e3c970e: OUTPUT«Potential difficulties:␤ 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:␤------> say⏏;␤ok 00:01 117m␤» 16:53
TimToady std: any;
p6eval std e3c970e: OUTPUT«ok 00:01 117m␤»
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
masak perhaps it only thought it updated? 16:55
right.
TimToady std: WHICH;
p6eval std e3c970e: OUTPUT«Potential difficulties:␤ 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:␤------> WHICH⏏;␤ok 00:01 117m␤»
TimToady that's the old version
16:55 daniel-s left
TimToady because WHICH switch over to any's message 16:55
*ed 16:56
16:56 frhodes joined, benabik joined
TimToady should rejoice for having exploits-moded ur evalbot, but doesn't 16:57
masak not unless you did it deliberately. 16:59
lichtkind TimToady: thank you 17:05
TimToady: a seq() is a flat fully reified list? 17:07
17:08 hirschnase left
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
on Perl 6 this is the opposite.
masak yes. 17:12
and Perl 6 *wants* you to confuse method calls with (public) attribute access. :) 17:13
TimToady lichtkind: something like that, though we're still a bit unsure whether we'll keep it 17:17
17:18 frhodes left
masak as with the Seq type as such, I guess. 17:22
lichtkind understand each time reading WHICH just Witch 17:25
17:28 ccc_ joined
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."
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. 17:41
ho, TimToady answered lichtkind. 17:42
like the compiler, I am esaily "confused" 17:43
17:45 wamba joined
cognominal btw, why the need for parentheses method calls for arguments? 17:45
moritz cognominal: the parser must always know whether to expect a term or an infix 17:46
cognominal I am not sure where it comes to play here. 17:48
moritz cognominal: if $a.foo allowed parameters after it, but optionally, it could be both an infix (for the no argument case) or a term (for the argument case)
s/paramters/arguments/
17:49 Mowah_ joined
TimToady sorear: does niecza ignore 'is equiv'? 17:50
niecza: multi infix:<∧> is equiv(&infix:<*>) ($a,$b) {say 'and'; $a min $b }; multi infix:<∨> is equiv(&infix:<+>) ($a,$b) {say 'or'; $a max $b }; say True ∨ True ∧ False 17:53
p6eval niecza v8-55-g0db7bb8: OUTPUT«or␤and␤Bool::False␤»
TimToady this should be and or True
or am I doing something rong?
masak maybe 'is equiv' is NYI. 17:59
TimToady usually niecza++ is good about carping when NYI 18:09
masak niecza: multi infix:<∧> is a-little-horse("niegh") ($a, $b) {}; say "alive" 18:10
p6eval niecza v8-55-g0db7bb8: OUTPUT«===SORRY!===␤␤Unhandled trait a-little-horse for this context at /tmp/nG5jMk0_gA line 1:␤------> ti infix:<∧> is a-little-horse("niegh") ⏏($a, $b) {}; say "alive"␤␤Potential difficulties:␤ $a is declared but not used at…
TimToady niecza: multi infix:<∧> is tighter(&infix:<*>) ($a,$b) {say 'and'; $a min $b }; multi infix:<∨> is looser(&infix:<+>) ($a,$b) {say 'or'; $a max $b }; say True ∨ True ∧ False
p6eval niecza v8-55-g0db7bb8: OUTPUT«or␤and␤Bool::False␤»
TimToady std: any<me expoits grammerz> 18:12
p6eval std e3c970e: OUTPUT«ok 00:01 118m␤»
TimToady moritz: any kloo why std says it has the right version but doesn't?
moritz TimToady: does 'make snap' work for you, and reports the correct revision? 18:14
TimToady yes 18:15
and the any fails with: The 'any' listop may not be called without arguments (please use () or whitespace to clarify) at (eval) line 1 18:16
moritz I can git clean -xdf and see if that improves things 18:17
TimToady thanks
moritz Can't locate YAML/XS.pm in @INC 18:18
mberends moritz: perlbrew version change? 18:19
moritz mberends: most likely, yes
so it seems that 'make snap' updates the revision file even for failed builds :( 18:20
dalek ast: 1aa4587 | Coke++ | S02-names_and_variables/names.t:
nom fudge
18:21 frhodes joined
dalek kudo/nom: 0adc742 | Coke++ | t/spectest.data:
run this fudged test
18:21
moritz std: any 18:24
p6eval std e3c970e: OUTPUT«ok 00:01 117m␤»
moritz std: any<foo>
p6eval std e3c970e: OUTPUT«ok 00:01 118m␤»
moritz that's a new, clean, successful build 18:25
TimToady ./viv -e 'any<foo>' in the snap directory should fail 18:26
moritz fwiw p6eval runs tryfile, not viv'
not sure if it matters
evalbot control restart 18:27
18:28 p6eval joined, ChanServ sets mode: +v p6eval
TimToady tryfile fails for me here 18:29
moritz std: any<f>
TimToady I mean, succeeds in failing :)
p6eval std e3c970e: OUTPUT«ok 00:01 118m␤»
TimToady très étrange 18:30
奇特 18:32
18:33 jevin joined 18:38 stepnem joined
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 18:41
18:42 Minimiscience joined
flussence it takes 5.5 minutes to generate s26.html, 3 of which are before pod2html even gets called :( 18:43
oh wait, that was with all the debug output. 4m43 without... 18:45
18:46 Minimiscience left 18:48 kst joined
lichtkind what is the new name of /parrot_install/bin ? 18:49
flussence s/parrot_//
lichtkind thank you
flussence: you mean with underscore? 18:50
flussence it's just install/bin now
lichtkind i see 18:51
had some brain latency
it was a regex
18:52 djanatyn joined
djanatyn Hi! 18:52
masak: hello :D
18:53 wamba left
lichtkind nom doest understand =begin END 18:54
master did
donri masak: ^ look who i brought 18:55
masak djanatyn! \o/
djanatyn: how's things?
flussence tadzik: ping - seems like the pod parser doesn't do nested <>s inside FormattingCodes 18:56
("C<B<>>" in the spec is ending up as "<code>B&lt;</code>&gt;") 18:57
donri djanatyn: now that you're here, you'll have to write some perl6 19:00
flussence
.oO( am I a bad person for using <mark>/<em> where the spec suggests <em>/<u>? :)
19:01
djanatyn masak: Doing great!. :) 19:04
donri: I've written perl6 code before ;)
donri Show me.
djanatyn masak: I took a little bit of time off to take a dip in python.
I already have, but sure.
donri: github.com/djanatyn/array-of-doom/...er/game.pl
masak djanatyn: nice! how's python doing? 19:05
djanatyn masak: Great.
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.
donri djanatyn: you can use "given" to assign $_
djanatyn However, I was having some issues and I eventually switched to python + pygame.
donri: *huh*. cool.
donri given prompt { when ... } 19:06
masak djanatyn: heh, I guessed you'd gravitate to pygame :)
djanatyn: I've been glancing at it many times too.
djanatyn Yeah. I actually got a pretty decent game released and did better than I expected.
Learning python quickly was fun. :)
19:06 frhodes left
masak \o/ 19:06
19:06 frhodes joined, frhodes left
djanatyn Since the most substantial work I've done with OOP has been in perl 5, it was quite a paradigm shift 19:06
masak why? 19:07
djanatyn Since I haven't even really used Moose very much.
masak people say Python's OO and Perl 5's are pretty similar.
djanatyn I had always considered it a little bit of work to write object oriented perl.
Like, writing a new method, and things like that.
masak without Moose, sure! :)
djanatyn Yeah, I meant without Moose :D 19:08
masak and people put a lot of cargo-cult crap in their 'new' methods :P
djanatyn I found it pretty fun to use. 19:09
And of course. Perl 6's OOP is awesome, though I don't have much experience with it yet
masak all in good time ;)
nom: class OOP {}; class awesome {}; role Perl6[OOP] is awesome {}; say "yay!" 19:10
p6eval nom 0adc74: OUTPUT«yay!␤»
donri nom: say .WHAT given * 19:11
p6eval nom 0adc74: OUTPUT«Whatever()␤»
masak nom: say .WHAT given .WHAT
19:11 mj41 joined
p6eval nom 0adc74: OUTPUT«Any()␤» 19:11
19:11 frhodes joined
donri perl 6 will not have competitions in obfuscation but in code-as-natural-language 19:12
masak oh, it can do either :)
nom: my $dna = "ttaagg"; sub translate($dna) { "FFLLSSSSYY!!CC!WLLLLPPPPHHQQRRRRIIIMTTTTNNKKSSRRVVVVAAAADDEEGGGG".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␤␤»
masak rakudo: my $dna = "ttaagg"; sub translate($dna) { "FFLLSSSSYY!!CC!WLLLLPPPPHHQQRRRRIIIMTTTTNNKKSSRRVVVVAAAADDEEGGGG".comb[map { :4($_) }, $dna.trans("tcag" => "0123").comb(/.../)] }; say translate($dna)
p6eval rakudo a55346: OUTPUT«LR␤»
masak nom doesn't do .trans yet?
why wouldn't that be a showstopper to merging to master. .trans is everywhere. 19:14
19:14 Limbic_Region joined
masak s/\./?/ 19:14
dalek ecza: a7575a1 | diakopter++ | / (8 files):
make Kernel.cs compile under visual studio

add stub for sprintf add Printf.cs to Makefile
19:15
masak djanatyn: the above one-liner is from this blog post: strangelyconsistent.org/blog/the-ge...-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
masak djanatyn: I needed to translate TimToady's adieu from DNA bases to amino acids :)
flussence: you're telling me. :) I implemented it in master.
well, re-implemented it. 19:17
dalek ecza: a89a9a6 | diakopter++ | lib/Printf.cs:
add Printf.cs
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.
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.
kthakore djanatyn: how is it going?
masak I already asked that. :P 19:20
kthakore oh
:\
djanatyn: what were your issues? How can we make it better?
masak kthakore++ # double-checking
ooh, I wanted to ask that one :)
TimToady so maybe we don't write .trans just yet, but .transcomp and transrun
kthakore masak: thanks
donri issues with what?
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
donri: we are still youngish ...
donri oh
diakopter nom: say sprintf( "%d%d%d%d%d", 0...*)
p6eval nom 0adc74: OUTPUT«...␤»
djanatyn kthakore: Well...^_^
I guess it was more personal issues. 19:22
kthakore masak: is 1year some months stil youngish?
djanatyn: that is fine your experience will help us
donri we also need a .transvestite method for polymorphism
djanatyn I found the pygame documentation easier to navigate, and their abstraction was a bit higher
kthakore djanatyn: ok
masak kthakore: for a project? I guess...
djanatyn I liked how you could set the framerate without having to add handlers in pygame
kthakore ok 19:23
djanatyn I could structure my code the way I wanted easily without having $app->run();
kthakore djanatyn: want to move this to pm btw?
djanatyn kthakore: uhh, sure
donri trying to keep the rest of us out eh
kthakore oh ... I was afraid of going of topic
this is perl5 module
sorry
donri ;)
kthakore masak++ thanks 19:24
apologies again
masak oh, come on :) 19:26
I should apologise to *you* for being so silly.
me being silly, I mean. not you. 19:27
also, there's no ban on talking about Perl 5 here. as long as you don't go overboard about it.
diakopter because then we'd have to throw you a lifeline or float ring 19:28
19:29 mj41 left
masak :P 19:31
"Perl 6: like being saved from drowning." 19:32
TimToady
.oO(Hazardous Waters! Enter At Your Own Perl!)
19:42
kthakore heh 19:44
19:46 orphu joined 19:51 rlb3 joined 19:52 kst left 19:53 kst joined, orphu left
diakopter TimToady: S32::Str sprintf grammar doesn't mention the ability to specify which argument supplies the vector override separator string 19:56
but perlfunc sprintf mentions it 19:57
masak diakopter: consider that grammar more a guideline than solid spec.
(I didn't write it, but I've been studying it)
diakopter is working on sprintf for niecza
I'll assume the 5.14.1 docs are correcter 20:00
lichtkind diakopter: yes lots of docs effort lately 20:01
diakopter ?
flussence sprintf in rakudo is still pretty lacking, iirc.
djanatyn Well, I've got rakudo-star installed on my Fedora box.
flussence nom: printf('%d %1$x', 45);
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␤␤»
masak djanatyn: great!
lichtkind diakopter: fram reading p5p i know the docs fot improved lately , even from 5.14 to 5.14.1 20:02
s/fot/got/
masak TimToady: did you see the first observation in gist.github.com/1148915 ? do you agree that S06 needs a fix for that example? 20:07
TimToady: also, I'm concerned about irclog.perlgeek.de/perl6/2011-08-18#i_4293141 20:08
(is writing on a macro grant proposal, so I got to thinking about those)
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?)
masak I noticed that too. 20:10
I went to edit it to fix it, but in the edit view it looked right. :/
flussence yeah, seems like it's on their end...
lue hello planet! o/
masak hi lue! 20:12
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
lue [I heard he doesn't like IRC. He should visit sometime] 20:14
masak lue: don't immanentize the eschaton! o.O 20:16
TheDamian and IRC... it would be like matter and antimatter...
lue
.oO(I'd think it'd only be immanentizing if I were to send him a letter about coming on)
20:17
20:17 birdwindupbird left
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?
masak the chances of that are slim. 20:19
from a pure entropy perspective, we'd be much more likely to end up with PHP 7. :/
flussence AAAAAAAAAaaa!
masak exactly.
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?)
sbp this is the scariest conversation I have so far seen in here
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.
sbp nope, I'm conserving my conversation
lue
.oO(great. I just got *another* idea for a silly little project)
20:22
20:22 Mowah_ left
masak lue: how'd the game go? 20:22
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
lue: by the way, have you been feeling like submitting an AI for Little Animal Farm lately? :) 20:24
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!] 20:25
masak :) 20:26
lichtkind the speakers training in riga with thedamian was good
20:26 Mowah_ joined
lichtkind currently practicing the learned 20:26
lue github.com/lue/Frotzlexity --- I don't trust my instincts about the rules, have a look. 20:28
[Of course, implementation will be a challenge, but any programming that deals with English is. Perhaps an esperanto version first...] 20:29
20:29 Mowah_ left 20:30 huf joined 20:31 Mowah_ joined
masak looks 20:32
sbp or lojban
20:33 frhodes left
masak lue: actually, analyzing the word for correct usage doesn't seem that hard. 20:35
it's analyzing the rest of the sentence that's tricky.
the game looks nice. 20:36
20:37 mj41 left
djanatyn sorear: Did someone say lojban? 20:37
ermm, sbp
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.
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...
masak realizes that mathw has been kinda silent lately 20:38
lue: I could think of a couple of ways. but it's your game :)
lue thinks about a more all(fun,challenging,interesting) game... 20:39
At least I didn't start implementing anything... 20:40
djanatyn kthakore: paste.pocoo.org/show/465458/
masak lue++ # releasing early and often
djanatyn that's how the clock object works in pygame
20:41 pernatiy joined
masak nom: say <per>.roll, <naity>.roll 20:41
p6eval nom 0adc74: OUTPUT«pernaity␤»
sbp loaded dice!
masak oops.
nom: say <per>.roll, <natiy>.roll
p6eval nom 0adc74: OUTPUT«pernatiy␤»
masak \o/ 20:42
sbp: not loaded, just d1 dice :P
sbp and loading is a d6 to d1 transformation?
djanatyn paste.pocoo.org/show/465461/ <-- sample output of the first 50 iterations on my machine
not sure how I'd expect it in perl 20:43
but I like pygame's approach to it
lue nom: say ("a".."z").roll(6).join(''); 20:48
p6eval nom 0adc74: OUTPUT«pnhatr␤»
masak pn hatrs will pn hat. 20:50
kthakore djanatyn: ok 20:51
lue
.oO(it could also perceived as the first step of a backronym generator)
sbp nom has a mind of its own 20:53
pernatiy masak: what so special in my nick? ;)
masak pernatiy: oh, hi :) sorry, just being Saturday-evening silly. 20:54
pernatiy: how are you liking Perl 6 (and #perl6) so far, by the way?
(besides the silliness, I mean) 20:55
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
pernatiy not sure if I like it
to be honest I not sure if I like any programming language
20:56 uvtc joined
masak pernatiy: :) 20:56
pernatiy =)
masak pernatiy: so you're the Steve Yegge type, then.
pernatiy their all are just tools
*they 20:57
masak yes. they're just substrates for programs.
uvtc kthakore, hi. Just curious: what is the status of SDL bindings for Perl 6? I poked around in github's PerlGameDev/SDL6,
kthakore, but am not sure how far along it is. 20:58
masak pernatiy: what do you like about Perl 5? (if anything)
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
masak of course.
pernatiy well, I like perl5 for it's XS capaility
masak well, we don't have that :) 21:00
ingy Pegex is self hosting!!
pernatiy Inline::C package is one of the best =)
masak ingy: \o/
ingy does a little dance
pernatiy also OO approach in perl5 is funny
masak pernatiy: as in "ridiculous"? :P 21:01
pernatiy not much people understand it but it is brilliant from some point of view
TimToady masak: yes, the circumfix macro is wrong
ingy does rakudo have access to a pcre/p5 re engine yet?
masak TimToady: I suggest either you fix it, or you suggest a fix and I fix it.
21:01 rlb3 left
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"
masak ingy: maybe through blizkost. I don't know that anyone has tried.
TimToady or maybe it belons in ws somewhere
ingy should start writing a p6 grammar in pegex so he can parse p6 in python
pernatiy so you can mix OO style with procedural, in quantity which you need
flussence or some `is native<libpcre>` magic
masak TimToady: isn't there a category "comment"? 21:03
TimToady I guess there is
masak why "transparent" and not "comment"?
TimToady was just thinking of something more general, but it doesn't really work
unless you can have a macro that matches anywhere 21:04
which is not really what anyone wants
masak especially not the poor schmuck who promised to implement them :/
Tene masak: let me know how your macros work goes; I'm very interested in that.
masak Tene: thanks! I will.
Tene I've come pretty close to working on it myself half a dozen times over the past few years.
21:04 natureboy joined
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.
masak Tene: at least the AST-based macros.
21:06 envi joined 21:07 Fuad joined
Fuad hi 21:07
masak Fuad! \o/
Fuad masak: Hello bro!
masak wow, this is like a family gathering!
Fuad :) 21:08
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.
TimToady: that might actually solve it in that particular case.
21:10 daemon joined 21:11 soh_cah_toa joined
masak (then again, maybe not) 21:13
TimToady almost certainly not... 21:17
diakopter TimToady: I had a question about S32::Str sprintf ^^ 21:18
TimToady after all these years, I still don't know if I like that feature of sprintf; it seems wrong somehow ^H 21:19
we should be doing that with named args or some such 21:20
diakopter whimpers
masak what's the last/latest release of Pugs? 21:21
diakopter hackage.haskell.org/package/Pugs
masak diakopter: thanks.
diakopter she updates it everytime a new ghc is released I think
masak yes, that's what I guessed too. 21:22
21:22 Fuad left
masak 6.2.13, it seems. 21:22
which is between 6.2 "Basic IO and control flow elements; mutable variables; assignment" and 6.28 "Classes and traits"
even though Pugs has classes, at least. 21:23
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?
that's not how I remember it... 21:25
(which is not to say that it didn't happen that way)
donri is pugs doomed to oblivion? 21:26
flussence small random unrelated question - is there an obvious way to do (?$string.looks-like-an-int-literal) that I don't know about? 21:27
Tene flussence: match against the int rule in the perl6 grammar 21:28
masak donri: "Cannot predict now"
or maybe "Outlook not so good"
donri i recall hearing something about perl6 made the pugs design difficult
TimToady perl6 makes every implementation difficult 21:29
donri its use of parsec maybe
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.
TimToady well, it doesn't need to use parsec anymore
donri kinda sad :(
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. 21:30
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.
TimToady or maybe audreyt++ was going to hack it in 21:34
masak would like to meet stevan some day
TimToady anyway, it crashed
masak well, we got a MOP for Rakudo :)
which is not entirely disjunct from Moose.
21:40 Mowah_ left 21:45 explorer__ joined 21:49 wamba joined
silug i really should catch up on that one of these years. 21:57
oops, wrong channel. :) 21:58
masak thought you meant Moose and Rakudo's MOP ;)
silug well, sure, that too. :)
why modern window managers don't have focus-follows-eyes yet, i'll never understand. 22:00
22:00 ccc_ left
masak TimToady: how are method-like macros such as .WHAT declared? 22:01
silug: ugh, and I thought focus-follows-mouse was bad enough :/ 22:02
silug masak: i don't tend to accidentally type in the wrong window because my mouse has moved on though... :) 22:03
masak the few times I've had focus-follows-mouse, I've done that. 22:04
YMMV, I guess.
silug no, no, focus-follows-mouse is an abomination. i agree. :) 22:05
TimToady masak: I presume they could just be postfix:<WHAT> 22:10
but maybe not
might have to be postfix:<.WHAT>
and disallow . form on that as we do on \w-ish postfixes 22:11
flussence
.oO( I'd prefer mouse-follows-eyes )
masak TimToady: that oughta work, yes.
TimToady left-eye blink is left button, right-eye is right button, both eyes, is center :)
huf nah, that's uncomfortable 22:12
we need mindreading
TimToady well, we could probably tell intentional eyeblink from unintentional even now
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?
masak ok, reasonable. 22:14
center-button drags are uncommon anyway.
uvtc first you think, you wink, you do a double blink, ... close your eyes, and jump
TimToady maybe clicks should do tongue recognition 22:15
and snakes could click or drag two things at once
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
TimToady std: any<foo> 22:18
lue for drag-and-drop, clamp your mouth shut.
p6eval std e3c970e: OUTPUT«ok 00:01 118m␤»
TimToady sigh
huf the other solution is genetic engineering 22:19
i'd like having 4 arms :)
just imagine, a world where you can eat chips while ircing WITHOUT getting your keyboard sticky...
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...
TimToady ruz: nom is precisely where it will work, since it uses 6model
masak yeah, won't ever work in master. 22:21
22:21 uvtc left
masak ruz: nom has support for grammars, just not full support. yet. 22:21
ruz yep, so I'm still wating for nom to catch up with grammars 22:22
masak: spectest rate suggests that it's really limitted
going to check if nom compiles grammar I have 22:23
lue afk 22:24
TimToady going to check if a nap is about to occur
22:28 abercrombie joined 22:41 drbean joined 22:47 Minimiscience joined 22:56 cexsum joined 22:58 ZaphrodZenovka joined 23:21 Minimiscience left
ruz cool, with a few changes grammar is parsed and traced 23:26
masak \o/
23:30 am0c joined 23:37 abercrombie left
masak heh. didn't have time for a blog post tonight. will have time tomorrow for at least one, though. 23:39
'night, #perl6.
23:39 sili joined 23:40 Psyche^ joined, Psyche^ is now known as Patterner 23:41 Khisanth left 23:42 kfo joined 23:55 Khisanth joined, Limbic_Region left
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?