»ö« Welcome to Perl 6! | perl6.org/ | evalbot usage: 'perl6: say 3;' or rakudo:, niecza:, std:, or /msg camelia perl6: ... | irclog: irc.perl6.org/ | UTF-8 is our friend!
Set by moritz on 3 May 2013.
00:00 berekuk left, ajr_ left 00:04 berekuk joined
TimToady std: .push &fork 00:04
camelia std fa59c1b: OUTPUT«===SORRY!===␤Function name &fork not allowed in infix position at /tmp/4He0L8KfIJ line 1:␤------> .push ⏏&fork␤Check failed␤FAILED 00:00 42m␤»
00:06 berekuk left 00:16 Targen left 00:18 berekuk joined 00:20 berekuk left 00:22 tgt left 00:52 autumn is now known as auty 01:12 MrMeek-afk joined 01:13 woosley1 joined 01:16 MrMeek left 01:26 FROGGS_ joined 01:30 FROGGS left 01:43 kbenson left 02:06 census left 02:19 grondilu left
diakopter PerlJam: hi 02:30
02:34 orafu left, orafu joined 02:50 PacoAir left 03:04 am0c joined 03:08 jaldhar joined 03:15 am0c left 03:30 tangentstorm left 03:32 tangentstorm joined 03:54 preflex left, preflex joined 04:08 skids left 04:10 skids joined 04:22 skids left 04:23 skids joined, DarthGandalf left 04:25 Psyche^ joined 04:28 Patterner left, Psyche^ is now known as Patterner 04:30 DarthGandalf joined 04:33 REPLeffect left 04:35 adu joined
adu FROGGS_: interesting 04:36
04:37 skids left, REPLeffect joined 04:52 DarthGandalf left 04:58 skids joined 05:05 skids left 05:09 DarthGandalf joined 05:10 thou left 05:18 thou joined 05:34 _ilbot left, _ilbot joined 05:58 adu left 06:04 dmol joined 06:10 am0c joined 06:26 dmol left, dmol joined
raiph blogs.perl.org/users/perl_6_reports...y-1st.html 06:30
moritz raiph: note that irclog.perlgeek.de is up again 06:52
sorear yay!! 06:53
wow, the new server is MUCH faster 06:54
moritz that's curious, because it only has 1/4th of the memory of the old one, and one instead of four CPUs
and is virtual, whereas the old one was "real" hardware
but maybe using a newer version of mysql helps 06:55
raiph great to hear it's up
moritz raiph++ # nice summaries
raiph it's not responding (yet) for me
moritz raiph: what IP does the domain resolve to?
might still have the old IP in DNS 06:56
should be 213.95.21.23
raiph yeah, old ip presumably (213.95.10.24) 06:57
sorear moritz: is it possible that dumping and reloading the database caused it to be stored better? 06:58
moritz raiph: until all the DNS caches are flushed, you can put an entry into your /etc/hosts, or resolve via nameserver 213.239.254.194 06:59
sorear: I have no idea. I think a RDBMS isn't supposed to suffer from that kind of problem, but then again it's mysql, so all bets are off :-) 07:00
dmol BTW, Wikipedia has switched from MySQL to MariaDB recently 07:03
moritz I guess I will when Debian makes the switch
do the mariadb folks care more about standard compliance and data validation than mysql? 07:04
07:04 pecastro joined
dmol moritz: mariadb is designed as drop-in replacement for mysql, and it is compatible with the later on the protocol level 07:06
07:06 pecastro_ left
moritz right. Which probably means "no" 07:06
dmol And I don't believe in the future of mysql under the veil of Oracle :) 07:07
tangentstorm runs off to write his own database... 07:11
moritz tangentstorm: don't, just use postgresql :-)
(unless you have *very* specific needs)
tangentstorm i was kind of trying to be funny... but i actually am writing a database. :) 07:12
moritz tangentstorm: now I'm curious. Why?
tangentstorm That is a deep question. :) 07:13
moritz has time
to read your deep answer, I mean :-) 07:14
tangentstorm Do you have any guess as to why programmers implement their own versions of things in general? 07:15
besides just for fun/learning i mean 07:16
moritz 80% "Not Invented Here"-syndrome, 20% didn't find anything that fit their needs
tangentstorm wonders which of those applies to perl6... :) 07:17
moritz both, of course :-)
tangentstorm What do you think causes NIH? 07:18
07:19 rindolf joined
moritz hubris, lack of domain knowledge which makes people believe they can do better than existing solutions 07:19
tangentstorm Hrm... Both of those are pretty good causes. 07:20
moritz there are lots of other reasons, but those tend to be most common ones
I've heard that google is very "good" at NIH 07:21
tangentstorm I've noticed a pattern where people will go and look at other implementations, really want to use them, and and come away saying all the implementations are terrible.
which i guess is your 20%
they don't really say terrible... they say bloated or messy or whatever. 07:22
moritz for the sake of propriety I should add that I don't have quantitive data to support that 80%/20% divide
tangentstorm well you hang out here. they say all the things they say about parrot.
moritz well, I don't really disagree
sorear aka, dunning & kreuger
moritz but then again when parrot was started, there weren't many VMs for dynamic languages at all
tangentstorm sorear: yes, but people who are suffering from overestimating their abilities aren't actually going to be able to make something :) 07:24
moritz tangentstorm: so, are you going to tell my why you write your own database? I'm still curious :-)
(and if the answer is NIH, I won't think negatively about you; I've done it often enough myself)
tangentstorm no it's part of solving the problem that i believe causes NIH 07:25
but also NIH :)
lizmat -OFun and -Olearning may also be involved?
tangentstorm fun and learning have definitely been involved. :) 07:26
lizmat looks back and sees tangentstorm just said that
tangentstorm learned all about b+ trees :)
new question: have you ever had the experience of looking up some obscure math term on wikipedia? 07:27
sorear yes
moritz yes
tangentstorm if not, the hodge conjecture's a good place to start :D 07:28
sorear meta.wikimedia.org/wiki/Reading_level look for landscape gardening
hodge conjecture isn't terrible obscure 07:29
tangentstorm :D
sorear I mean I have no idea what it is but it must be pretty famous and important if there's a prize for it :P
tangentstorm sure because of the prize... but every word in the definition of the problem is (to me) complete gibberish :)
even the guy on the official video for the prizes is like ... how the heck am i supposed to even give you a frame of reference for this thing?!? 07:30
so it's obscure to begin with, but the actual problem doesn't have to be obscure. 07:31
you could be looking at any densely linked site... like tvtropes or something (warning: people routinely get lost for hours in there)
basically just any time you're trying to approach understanding a complicated graph structure
sorear I understood the first sentence of that article. 07:32
bonsaikitten tangentstorm: re: "all implementations are terrible" - most code is bad, and we underestimate effort
sorear I know what algebraic varieties are, and I know what algebraic topology is... and they (superficially) have nothing to do with each other
bonsaikitten so you end up at the conclusion "it'll be faster for me to write this from scratch"
07:32 xinming_ joined
masak we know. this is #perl6 ;) 07:33
tangentstorm bonsaikitten: if you think about the things that make code bad, you'll see that wikipedia's math pages suffer from the same problems
sorear although non-singular algebraic varieties tend to be very nice topological spaces, as such things go
bonsaikitten tangentstorm: biggest problem: most coders are at the edge of their understanding
you don't just do boring stuff you know already, you try to solve new problems
sorear always(?) analytic manifolds, etc
tangentstorm bonsaikitten: agreed.
bonsaikitten so, you'll be delivering some suboptimal solution just because you don't know better (which is the exact same thing you yell at others for) 07:34
tangentstorm that's /especially/ true when you're in a situation that would prompt you to go find someone else's package to use.
well. but often your problem is pretty small 07:35
like : maybe you just want to keep a list of your favorite books or something.
so you might look at MySQL and it seems huge and bloated compared to a text file.
bonsaikitten sqlite ;)
sorear I'm only talking about complex algebraic varieties here. Algebraic varieties over finite fields are a completely different ball of wax.
07:36 xinming left
bonsaikitten mysql is a bad example because it's kinda mediocre in every way 07:36
tangentstorm okay, well sqlite then. :)
even though it's small, it's still a lot of overhead, especially if you've never used SQL before. 07:37
07:37 woolfy joined
dalek ast: bc4c9a5 | (David Warring)++ | S05-capture/caps.t:
tests separator captures RT #117831
 07:37
masak sorear: I just learned the terms homomorphism, monomorphism, etc. -- a bunch of Wikipedia articles make a ton more sense.
I can almost see the point of exact sequences. :) 07:38
tangentstorm So... you only have so much "woring ram" in your brain at any given time...
masak tangentstorm: working* ?
tangentstorm And when you're confronted with a graph structure, and you need to understand things, you'll do what masak is doing... :)
sorear warring ram. corewars?
bonsaikitten tangentstorm: head space - yeah, but once you have figured out how to talk to the DB it gets quite easy
tangentstorm depends what you're storing 07:39
bonsaikitten hmm, haven't played corewars in a long time
tangentstorm that's the other thing... you make random associations in your own mind.
sorear tangentstorm: those who do not study COBOL and CODASYL are doomed to repeat them.
tangentstorm haha... i'm not familiar with codasyl but i've been contemplating writing a tetris in cobol just for fun :) 07:40
masak "fun" :)
tangentstorm so first you have this graph of someone else's ideas in front of you ( wikimath, tv tropes, some dude's code )
sorear hey, I'm the one who's working off and on on a bootstrapping INTERCAL compiler
tangentstorm then there's all the associations you're making in your own head as you try to understand and associate what you're reading back to your own experience. 07:41
for example, for me, writing that last sentence reminds me of the official term for it : en.wikipedia.org/wiki/Transderivational_search :) 07:42
whatever "official" means
So here you are, walking this graph of ideas... some of them are in your head and some are on your screen. 07:43
Now if you don't have enough working memory, you can augment your memory by putting stuff in front of you. 07:44
like notes to yourself or whatever.
i think pretty much everyone uses their visual field to ground them... like even if you get interrupted you don't completely forget everything you were doing because your environment reminds you.
sorear reminds me of ben goertzel(sp?)'s notes on graphical language 07:45
tangentstorm but if you get interrupted and distracted and you rely on your environment to remind you...
masak leaves for some AFK activities for the rest of the day
sorear tangentstorm: explain blind people
tangentstorm and then your environment is showing you a complicated graph structure....
you are going to get lost.
lizmat Looking at S12:886: "Is equivalent to submethod BUILD (:$tail is copy ) { $!tail := $tail }" 07:46
tangentstorm sorear: sighted people then ;)
lizmat somehow the "is copy" and the binding ":=" seem to be in contradiction
sorear no
lizmat am I wrong ?
sorear $!tail binds the attribute to the freshly copied container 07:47
if there were no "is copy" the attribute would become readonly
lizmat ok, gotcha
tangentstorm anyway... i think this problem has a lot to do with our brains trying to deal with complicated, ill-structured graphs, so, as part of what i'm doing, i'm building a graph database. 07:51
but i'm also trying to write it in a way that will make it comprehensible to the next guy looking for a graph database. 07:52
... 07:54
lizmat that's always helpful: *you* might be the next guy in ~ 2 years or so 07:55
sorear graph database? so you are reinventing codasyl?
tangentstorm probably
another reason people reinvent the wheel: lack of awareness of existing wheels :) 07:56
bonsaikitten pff. all that old stuff is irrelevant ;) 07:57
(so let's reinvent it badly!)
lizmat did we discuss learning curve already?
tangentstorm ah, no... mine is a graph database but it's not necessarily a purely navigational database. 07:58
lizmat sometimes it *seems* easier to reimplement something then going through the curve of learning a system
tangentstorm it's basically a triple store, which firmly embraces the relational model when and where it makes sense. 07:59
lizmat: yes :)
bonsaikitten tangentstorm: triple stores ... oh, so hilarious
tangentstorm so my task in writing this is to make it seem easier to the next guy to use my thing than to reinvent his own. 08:00
bonsaikitten: thank you. i also do bar mitzvahs :)
lizmat .tell jnthn I'm not seeing how transposing "EXPORTHOW::<grammar> = TracedGrammarHOW;" to overriding "Mu::new" lexically would work 08:01
yoleaux lizmat: I'll pass your message to jnthn.
tadzik TimToady: ping-pong 08:03
erm
timotimo: pingpong
TimToady: sorry, tabfail
tangentstorm it's also a version control system, an extensible parser, and a tool for process management 08:04
tadzik timotimo: could you paste me Steps to Reproduce somewhere?
timotimo: and later maybe try removing %*CUSTOM_LIB<site>/panda?
08:06 census joined
tangentstorm supposes he looks like a crackpot by now. :) 08:11
sorear tangentstorm: Does it also bake pizzas?
diakopter wants pizza 08:12
tangentstorm no, because pizzas aren't trivially implemented by walking a graph.
lizmat I think it's more important that it knows what you mean
tangentstorm unless you know a recipe that uses them? :)
08:18 berekuk joined 08:19 donaldh left 08:21 donaldh joined 08:25 Chillance joined 08:30 zby_home_ joined 08:32 woosley1 left 08:47 FROGGS_ left
lizmat .tell jnthn I found S12:2516 about Custom Meta-Objects 08:47
yoleaux lizmat: I'll pass your message to jnthn.
09:05 kaare_ joined, _ilbot left 09:08 _ilbot joined
moritz has closed the biggest April gap in the IR clogs 09:10
dalek nda/panda2: 0bbb63b | tadzik++ | lib/Panda (2 files):
Make destdir easier to override
nda/panda2: 09fc84a | tadzik++ | / (3 files):
Bugfix, and make more real-world proof
lizmat wonders if those clogs really shine in infra-red only 09:11
09:17 PacoAir joined
labster hi everyone 09:19
lizmat hi labster!
09:20 _ilbot left, _ilbot joined
labster Okay, I have a couple of nitpicky things about File::Spec I want to ask. Since I was concentrating on getting the code working, I didn't pay too much attention to naming. But now that it's just about done, the names of the classes seem more important. 09:23
09:23 sftp joined
labster What should I call the File::Spec module, now in the core? I think it should be under IO, but it's really a path specification or file specification. So, IO::Spec, IO::FileSpec, or IO::PathSpec? (IO::Path::Spec would get confusing due to looking like IO::Path::Unix). 09:23
Second, should I change the names of the OS submodules? Perl 5 uses ::Unix and ::Win32, but S16 qp// uses :posix and :win. For comparison, python uses os.posixpath and os.ntpath. 09:24
moritz how about IO::Path::Spec and then IO::Path::Spec::Unix?
09:25 spider-mario joined
lizmat was thinking more or less the same thing 09:27
labster Then IO::Path has it's own subclass IO::Path::Unix which does pathy operations, while IO::Path::Spec doing the stringy operations would have IO::Path::Spec::Unix. 09:28
lizmat brb
labster As long as IO::Path::Spec::Unix and IO::Path::Unix aren't too confusing, we're ok.
then again, we don't want anyone really using the backend, so maybe that's okay. 09:32
09:32 grondilu joined
grondilu was reading rosettacode.org/wiki/Anonymous_recu...n#PicoLisp and thought that "recurse" would be a nice keyword instead of &?ROUTINE or &?BLOCK. 09:34
dalek ast: 4ee1dab | (Brent Laabs)++ | S32-io/io-path.t:
fudge some IO::Path tests in preparation of windows support
 09:51
lizmat grondilu: maybe make this a specs issue? github.com/perl6/specs/issues 09:57
afk for a few hours at least
09:57 berekuk left 09:58 thou left, berekuk joined
grondilu lizmat: yep, why not. I should do more of this stuff. 10:07
10:11 kivutar joined 10:12 census left 10:17 kivutar left 10:30 kivutar joined 11:12 pecastro left 11:14 pecastro joined 11:20 tgt joined 11:23 pecastro_ joined, kivutar left 11:24 kivutar joined, pecastro left, pecastro_ is now known as pecastro 11:25 kivutar left 11:32 kivutar joined 11:39 cibs left, isBEKaml joined 11:49 kivutar left 11:50 dmol left 12:01 tgt left
timotimo tadzik: i had the wrong parrot-5.*.0 in my PATH, so i used the wrong panda. not your fault at all 12:02
12:17 am0c left 12:25 isBEKaml left 12:32 kivutar joined 12:47 kivutar left 12:53 orafu left 12:54 preflex left 12:55 preflex joined 13:14 MrMeek joined 13:16 MrMeek-afk left, skids joined
cognominal nqp: grammar A { token TOP { <a-a> } token a-a { a } } say(A.parse: 'a') 13:44
camelia nqp: OUTPUT«Unable to parse expression in blockoid; couldn't find final '}' at line 2, near "token TOP "␤current instr.: 'panic' pc 14721 (src/stage2/gen/NQPHLL.pir:5232) (src/stage2/gen/NQPHLL.nqp:279)␤»
cognominal nqp: grammar A { token TOP { <a-a> }; token a-a { a } }; say(A.parse: 'a')
camelia nqp: OUTPUT«a␤»
cognominal nqp: grammar A { token TOP { <a-a:<a>> }; token a-a:<a> { a } }; say(A.parse: 'a') 13:45
camelia nqp: OUTPUT«Method 'a-a' not found for invocant of class 'A'␤current instr.: 'TOP' pc 200 ((file unknown):167896626) (/tmp/oRKMwkv6Fs:1)␤»
cognominal nqp: grammar A is HLL::Parser { token TOP { <EXPR> }; token term:sym<a> { a } }; say(A.parse: 'a') 13:47
camelia nqp: OUTPUT«Could not find parent class 'HLL::Parser' at line 2, near "; say(A.p"␤current instr.: 'panic' pc 14721 (src/stage2/gen/NQPHLL.pir:5232) (src/stage2/gen/NQPHLL.nqp:279)␤»
cognominal nqp: grammar A is HLL::Grammar { token TOP { <EXPR> }; token term:sym<a> { a } }; say(A.parse: 'a')
camelia nqp: OUTPUT«a␤»
cognominal nqp: grammar A is HLL::Grammar { token TOP { <EXPR> }; token term:sym<a-a> { a } }; say(A.parse: 'a') 13:48
camelia nqp: OUTPUT«a␤»
13:49 berekuk left 13:51 census joined 14:23 pecastro_ joined 14:25 pecastro left
lee_ is the source code to dalek available somewhere? 14:33
lizmat probably somewhere on github 14:35
rn: class Foo { has $!foo; trusts Bar }; class Bar { method foo { self!Foo::foo } }
camelia niecza v24-37-gf9c8fc2: OUTPUT«===SORRY!===␤␤GLOBAL::Bar does not name any package at /tmp/lAQNPBP254 line 1:␤------> class Foo { has $!foo; trusts Bar⏏ }; class Bar { method foo { self!Foo::f␤␤Cannot call private method 'foo' on Foo because it does …
..rakudo 474102: OUTPUT«===SORRY!===␤Cannot call private method 'foo' on package Foo because it does not trust Bar␤at /tmp/HdVIzVX3E1:1␤------> ; class Bar { method foo { self!Foo::foo⏏ } }␤ expecting any of:␤ method arguments␤»
lizmat a case of "trusts" NYI ? 14:37
lee_ ah-ha! this is what i was looking for github.com/sorear/dalek-poller/blo...h.psgi#L58 14:38
for my own bot :)
14:41 pecastro_ is now known as pecastro 14:50 rking joined
cognominal nqp: grammar A is HLL::Grammar { token TOP { <EXPR> }; token term:sym<nqp-fun> { a } }; say(A.parse: 'a') 14:54
camelia nqp: OUTPUT«a␤»
TimToady lizmat: did you see irclog.perlgeek.de/perl6/2013-05-04#i_7016965 ? 15:22
it's not quite so awesome as to suggest a colon, but that turned out to be hard
hmm, maybe there's another way... 15:24
15:33 Targen joined 15:34 crab2313 joined
dalek d: 276a42f | larry++ | STD.pm6:
Warn on apparent term following baremeth
 15:41
TimToady camelia: restart std 15:43
hmm
camelia: help
camelia TimToady: Usage: <(nqp-jvm|star|pugs|nqp|std|niecza|rakudo|b|nom|npr|n|r|perl6|prn|rn|p|rnp|nrp|pnr|rpn|p6|nr)(?^::\s) $perl6_program>
TimToady hmm 15:44
geekosaur rebuild? 15:46
TimToady camelia: rebuild std 15:47
guess not...
geekosaur evalbot rebuild std 15:49
camelia OK (started asynchronously)
geekosaur grep ftw
TimToady evalbot help 15:51
LTA...
std: .push 42 15:52
camelia std fa59c1b: OUTPUT«===SORRY!===␤Two terms in a row (method call with args requires colon or parens without whitespace) at /tmp/ETUPi0iMes line 1:␤------> .push⏏ 42␤ expecting any of:␤ infix or meta-infix␤ infixed function␤ statement modifier
..loop…
timotimo TimToady: are you going to spec that & + no-space + something is forbidden? 15:53
TimToady well, runs off the end
timotimo maybe &, no space, bareword/identifier?
TimToady std: .push &i
camelia std fa59c1b: OUTPUT«ok 00:00 42m␤»
timotimo r: say i 15:54
camelia rakudo 474102: OUTPUT«0+1i␤»
timotimo ah, that's why it works
TimToady wanted something defined but not a function
anyway, not done rebuilding, apparently
std: .push &i 15:55
camelia std fa59c1b: OUTPUT«ok 00:00 42m␤»
TimToady it would appear that the rebuild failed or stalled 16:00
16:16 dmol joined 16:28 woolfy left, thou joined 16:34 thou left 16:38 crab2313 left
GlitchMr huh? The IRC logs page works, but it doesn't have anything? 16:50
TimToady temporarily missing between March 1 and a day or so ago
std: .push &i 16:51
camelia std 276a42f: OUTPUT«Potential difficulties:␤ Colon (or parens) required for method name with arguments at /tmp/6j5aE3w3bz line 1:␤------> .push⏏ &i␤ok 00:00 42m␤»
TimToady std: .push &fork
camelia std 276a42f: OUTPUT«===SORRY!===␤Function name &fork not allowed in infix position at /tmp/avWm_rac4L line 1:␤------> .push ⏏&fork␤Potential difficulties:␤ Colon (or parens) required for method name with arguments at /tmp/avWm_rac4L line 1:␤------> …
TimToady timotimo: most of the warnings in STD are not specced like that; in that sense, STD is serving as the spec 16:52
moritz GlitchMr: there are two name servers for irclog.perlgeek.de, and one of them serving the old IP, which has an old database
(the old host is up again, but it doesn't have the new data) 16:53
GlitchMr Confusing... 16:54
moritz like all eventually consistent systems are sometimes
GlitchMr At least my private logs seem to work realtime...
I'm going to wait until my system notices IP had changed. 16:55
I'm still goint to have logs, even if I don't think it's needed. Just in case. 16:58
moritz a bit of redundant redundancy never hurts 16:59
lizmat TimToady: cool! 17:08
moritz TimToady++ 17:09
lizmat++ # nagging
lizmat TimToady++ indeed
lizmat is not sure whether nagging is one of her vices or virtues 17:10
moritz maybe both. 17:11
17:11 FROGGS joined
arnsholt It can be used for good and evil 17:14
With great power comes great responsibility =)
flussence
.oO( With great power comes great power bills... )
 17:16
moritz with grate responsitibilty comes much nagging from others :-)
*great 17:17
17:22 berekuk joined
lizmat :-) so what am I doing wrong here: 17:24
r: class Foo { has $!foo; trusts Bar }; class Bar { method foo { self!Foo::foo } }
camelia rakudo 474102: OUTPUT«===SORRY!===␤Cannot call private method 'foo' on package Foo because it does not trust Bar␤at /tmp/E4zFVVR6lK:1␤------> ; class Bar { method foo { self!Foo::foo⏏ } }␤ expecting any of:␤ method arguments␤»
timotimo :o 17:25
you don't need to predeclare Bar?
r: class Bar { ... }; class Foo { has $!foo; trusts Bar }; class Bar { method foo { self!Foo::foo } }
camelia rakudo 474102: ( no output )
moritz you probably need to
timotimo LTA error, i'd claim
dalek ast: 33746c6 | larry++ | S05-modifier/ignorecase.t:
[M] is not <[M]>
moritz aye
timotimo who feels like making a ticket? 17:26
moritz masak!
lizmat submits rakudobug
17:30 kaare_ left
dalek d: 291c7f0 | larry++ | CORE.setting:
add some missing function declarations
 17:41
ast: cd184fd | larry++ | S01-perl-5-integration/context.t:
use '=== Nil', not '~~ Nil'
 17:42
17:45 domidumont joined 17:53 kivutar joined 17:55 domidumont left 17:56 domidumont joined
dalek ast: 3c14ada | larry++ | S02-types/declare.t:
avoid ~~ Nil
 17:58
ast: 1655bb5 | larry++ | S02-types/nil.t:
avoid ~~ Nil
p: e3a67e4 | (Alessandro Ghedini)++ | 3rdparty/dyncall/dyncallback/dyncall_ (2 files):
Fix build failure on PowerPC (closes #62)
 18:12
p: 5a319af | (Alessandro Ghedini)++ | 3rdparty/dyncall/dyncall/dyncall_callvm_sparc.c:
Fix build failure on SPARC
p: 249b9cf | moritz++ | 3rdparty/dyncall/dyncallback/dyncall_ (2 files):
Merge branch 'ftbfs-ppc' of github.com/Debian/nqp
p: 8d9d7be | moritz++ | 3rdparty/dyncall/dyncall/dyncall_callvm_sparc.c:
Merge branch 'ftbfs-sparc' of github.com/Debian/nqp
ast: 220bf16 | larry++ | S05-capture/named.t:
STD warning suppression
 18:34
ast: cb011d5 | larry++ | S12-attributes/undeclared.t:
class Foo does Role{} needs ws before {}
ast: cda8a0d | larry++ | S12-methods/parallel-dispatch.t:
don't smartmatch against Nil
ast: 86125fc | larry++ | S13-overloading/typecasting-short.t:
|$capture --> |capture
ast: 151f725 | larry++ | S32-str/capitalize.t:
capitalize -> wordcase
ast: 3b4611b | larry++ | integration/class-name-and-attribute-conflict.t:
$.foo form not allowed in BUILD
18:36 ajr joined, ajr is now known as Guest87011 18:37 Guest87011 left
dalek d: b33d8e0 | larry++ | CORE.setting:
more missing names
 18:38
18:43 japhb_ left
dalek p: 4934e64 | (Stéphane Payrard)++ | / (2 files):
support parameters with dash

Conflicts:
src/NQP/Grammar.nqp
Signed-off-by: Moritz Lenz [email@hidden.address]		 18:43
moritz masak, GlitchMr: both nameservers now return the right IP for irclog.perlgeek.de, so once the DNS caches (typically max. 1 day) expire, your problems with the IR clogs should go away 18:50
18:50 Chat8721 joined
diakopter \o/ 18:50
lue has a guess as to where IR clogs came from... 18:51
Chat8721 C:
18:51 Chat8721 left
diakopter c drive indeed 18:51
fj 18:52
ergh
dalek p: a24051c | (Stéphane Payrard)++ | src/QAST/Operations.nqp:
Remove stray blanks
 18:53
p: 582df75 | (Stéphane Payrard)++ | src/QAST/Operations.nqp:
Rolled QAST::Operations.add_hll_unbox calls into loop
p: a3d59ba | (Stéphane Payrard)++ | / (349 files):
Merge remote-tracking branch 'upstream/master'
p: da25ee1 | moritz++ | src/vm/parrot/QAST/Operations.nqp:
Merge branch 'master' of github.com/cognominal/nqp
moritz question for the OS X users: is github.com/drKreso/nqp/commit/86a3...3edc4348a5 (the assumption that $^X eq 'darwin' implies 64bit) good? 18:55
geekosaur no 18:59
flussence OS X was 32-bit userland on 64-bit kernel until recently, IIRC 19:00
geekosaur other way around, only recently the 64 bit kernel became default
64-bit userland on 32 bit has existed since 10.5 and was default in 10.6
flussence that... confuses me 19:01
geekosaur confuses a lot of people :)
anyway, the assumption is, as the comment says, imperfect but largely true for 10.6+ and most machines using Apple's perl5 (the earliest Apple Intel hardware did not support 64bit extensions) 19:03
and finding out the truth is absurdly difficult :/ 19:04
so anyway, that assumption is false on: PPC, 10.5.x, early Core1-based Intel hardware. do we care? 19:05
and may be false on 32-bit builds of perl from other sources, although in many cases those will only be 32 bit in the case where 64 bit won't run anyway 19:07
moritz $ ./perl6 -Ilib t/spec/S32-str/capitalize.rakudo 19:08
1..15
Cannot get character of empty string
that's a parrot error; I'm sure there's a rakudobug to be tracked down and submitted in there
geekosaur: can you think of a better heuristic?
geekosaur not really, see 'absurdly difficult 19:11
'
labster the file /System/Library/CoreServices/SystemVersion.plist has a OS Version number. You could check that. 19:12
moritz if *I* did that, it would be largely guesswork, because I don't have OS X anywhere and can't test it 19:13
geekosaur and just to make things more annoying, there is no difference between perl -V output on this machine with `arch -32` vs. `arch -64`
moritz geekosaur: does 'uname -a' differ?
geekosaur apple went to a lot of trouble to hide as much as possible the difference between 32 and 64 bit on OS X 19:14
I don't have any 32-bit-only machines to check, sorry
moritz does uname -a mention 64 anywhere? 19:15
on the machine(s) you do have available? 19:16
geekosaur yes; the problem is it also mentioned 64 bit everywhere else unles syou have one of the things I mentioned earlier (10.5, Core 1 processor, etc.)
labster Well, uname -r will check that you don't have 10.5 or earlier. You just need Darwin version > 10 19:17
well, >= 10 19:18
19:18 kaare_ joined
moritz what does uname -r print on those systems? 19:20
labster osxdaily.com/2009/09/07/how-to-tell...w-leopard/
my google-fu is strong
geekosaur note that kernel is not relevant
as I said earlier, from 10.6 on it is normal to have a 64 bit userspace on a 32 bit kernel
19:20 domidumont left
moritz will just merge that pull request, and wait for complaints. 19:21
labster *sigh*. go for it, moritz.
geekosaur yes, this is moderately insane
dalek p: 86a3b20 | drKreso++ | Configure.pl:
OSX 64 bit target for NativeCall
 19:22
p: a78780c | moritz++ | Configure.pl:
Merge branch 'master' of github.com/drKreso/nqp
moritz labster: speaking of pull requests, I'm spectesting yours right now
labster It's passed on my and FROGGS' machines. With the exception of io-path, which I'm going to have to rewrite for different OS versions anyway. 19:24
timotimo moritz: did you have time to take a look at my branch for Package::EXPOTR?
moritz timotimo: no
dalek p: cae6f61 | (Gerhard R)++ | 3rdparty/dyncall/buildsys/gmake/tool/gcc.gmake:
[dyncall] next shot at fixing the build on Strawberry Perl

Should work universally, but testing on OSX would be appreciated.
p: af021f4 | moritz++ | 3rdparty/dyncall/buildsys/gmake/tool/gcc.gmake:
Merge branch 'master' of github.com/gerdr/nqp
labster The good news is that by stea... borrowing tests from Perl 5, I should have literally hundreds of tests to make sure things paths are working. 19:27
colomon labster++ 19:28
geekosaur pulls and builds on his OS X desktop 19:32
although I don't think I've done any NativeCall stuff before, so not sure how I'd test any of this 19:34
(well, ideally, it's already got tests and I can just run them) 19:35
dalek kudo/nom: 5b53365 | (Brent Laabs)++ | / (7 files):
added IO::Spec, started work on IO::Path
 19:37
kudo/nom: 22397c5 | (Brent Laabs)++ | src/core/IO.pm:
add support for IO::Path::<os subclasses>
kudo/nom: beab87b | (Brent Laabs)++ | src/core/IO (5 files):
rename `file-name-is-absolute` to `is-absolute` in IO::Spec
kudo/nom: a2505bc | (Brent Laabs)++ | src/core/ (3 files):
IO::Spec.tmpdir now returns `.` on failure, trying $*TMPDIR out
kudo/nom: ca77ce3 | (Brent Laabs)++ | / (5 files):
move IO to later in the build: lib.pm now works, regex works
kudo/nom: 87edfc0 | (Brent Laabs)++ | src/core/IO/Spec/ (3 files):
cleanup tmpdir, remove unused and already commented out code
kudo/nom: 02c7fbb | (Brent Laabs)++ | src/core/IO (3 files):
fix dir(), panda now runs \o/. Also change splitpath to use :nofile param
kudo/nom: 5838e93 | moritz++ | / (8 files):
Merge branch 'iospec' of github.com/labster/rakudo into nom
pan style="color: #395be5">perl6-examples: 95cc309 | (David Warring)++ | parsers/CSSGrammar.pm:
removed null operator fudge
pan style="color: #395be5">perl6-examples: 165e633 | (David Warring)++ | rosalind/itwv-grondilu.pl:
Merge branch 'master' of github.com:perl6/perl6-examples
19:39 dmol left, dmol1 joined 19:41 kivutar left
dalek kudo/nom: 454c190 | moritz++ | docs/ChangeLog:
update ChangeLog
 19:42
timotimo ah great, iospec 19:45
dalek pan style="color: #395be5">perl6-examples: b71f739 | (David Warring)++ | parsers/CSSGrammar.pm:
removed null combinator fudge
 19:47
19:51 berekuk left 19:53 kivutar joined 19:56 woolfy joined 19:57 berekuk joined
labster cool. That means it's time for me to add some tests. 19:58
moritz labster++ # great work in rakudo 20:00
labster: have you signed a CLA for rakudo? it's a prerequisite for getting a commit bit 20:01
labster Yeah, I sent it in last month.
It should be there by now, the address is going about an hour's drive away from me. 20:02
moritz ok, great
20:02 adu joined
lizmat moritz: do you need one for nqp as well? 20:03
moritz lizmat: no 20:04
lizmat but you do need one for rakudo, so where do I get the CLA ? 20:07
moritz lizmat: www.perlfoundation.org/contributor_..._agreement 20:08
FROGGS ola 20:11
lizmat FROGGS! 20:12
moritz ola FROGGS, que tal?
FROGGS err, fine thanks 20:13
moritz retires for resting 20:14
dalek kudo/nom: 5ad46fd | moritz++ | src/Perl6/Grammar.nqp:
start to switch **0..1 back to ?
FROGGS gnight moritz
lizmat night moritz! 20:16
20:18 rindolf left 20:26 dmol1 left 20:28 dmol joined, zby_home_ left 20:31 kaare_ left
timotimo something's going wrong with my EXPORT sub :| 20:35
in my EXPORT sub I can't access the "create_adt" symbol. it's defined as "our sub create_adt(...) is export" 20:38
why can't i access it as ADT::create_adt?
FROGGS can you paste the code?
timotimo sure
adu FROGGS: thanks for the tips 20:40
FROGGS adu: you're welcome
timotimo github.com/timo/ADT/tree/EXPORT_sub - this branch here 20:41
FROGGS hmmm, if the export sub is outside of the module, dont you need to import that module? 20:42
timotimo i ... do?
i don't know how to perl :)
FROGGS r: module M { sub s is export { } }; say M::s
camelia rakudo 454c19: OUTPUT«Could not find symbol '&s'␤ in method <anon> at src/gen/CORE.setting:10027␤ in any at src/gen/Metamodel.nqp:2504␤ in any find_method_fallback at src/gen/Metamodel.nqp:2492␤ in any find_method at src/gen/Metamodel.nqp:939␤ in block at /tmp/wU0zQjkOTT:1␤␤»…
FROGGS r: module M { sub s is export { } }; import M; say M::s 20:43
camelia rakudo 454c19: OUTPUT«Could not find symbol '&s'␤ in method <anon> at src/gen/CORE.setting:10027␤ in any at src/gen/Metamodel.nqp:2504␤ in any find_method_fallback at src/gen/Metamodel.nqp:2492␤ in any find_method at src/gen/Metamodel.nqp:939␤ in block at /tmp/HSbL8v0Gly:1␤␤»…
FROGGS r: module M { our sub s is export { } }; import M; say M::s
camelia rakudo 454c19: OUTPUT«Nil␤»
FROGGS r: module M { our sub s is export { } }; say M::s
camelia rakudo 454c19: OUTPUT«Nil␤»
FROGGS hmmm
looks like you dont need to import
timotimo i'm currently building a newer rakudo 20:44
can you get my code to work, though?
there's already test coverage for the export sub
lizmat well, import is needed so you can say "s"
timotimo good point
geekosaur hm, I think my build failed to find icu
FROGGS geekosaur: your parrot build? 20:45
geekosaur: what OS arfe you on?
geekosaur OS X
which means getting icu support is a bit interesting
FROGGS you have the dev package installed? (I guess there is something on osx)
geekosaur (and keeping it from colliding with macports even more so)
lizmat rn: module M { our sub s is export { "foo" } }; import M; say s 20:46
camelia rakudo 454c19: OUTPUT«===SORRY!===␤Null regex not allowed␤at /tmp/IElptOgDLc:1␤------> s is export { "foo" } }; import M; say s⏏<EOL>␤ expecting any of:␤ argument list␤ prefix or term␤ prefix or meta-prefix␤»
..niecza v24-37-gf9c8fc2: OUTPUT«===SORRY!===␤␤Action method statement_control:import not yet implemented at /tmp/3QCPWrBZLx line 1:␤------> ur sub s is export { "foo" } }; import M⏏; say s␤␤No delimiter found at /tmp/3QCPWrBZLx line 1 (EOF):␤------> …
lizmat rn: module M { our sub s is export { "foo" } }; import M; say s()
camelia rakudo 454c19: OUTPUT«foo␤»
..niecza v24-37-gf9c8fc2: OUTPUT«===SORRY!===␤␤Action method statement_control:import not yet implemented at /tmp/UKiyvPBiq9 line 1:␤------> ur sub s is export { "foo" } }; import M⏏; say s()␤␤Unhandled exception: Unable to resolve method statement_lev…
lizmat I wonder whether the need for () after "s" is a bug or not
FROGGS lizmat: it is, and there is already a ticket 20:47
s without () is treated as substitution
lizmat hehe
r: module M { our sub a is export { "foo" } }; import M; say a
camelia rakudo 454c19: OUTPUT«foo␤»
lizmat so, the import allows you to not have to specify fully qualified names 20:48
FROGGS ya, I see
but then I wonder what is wrong with timotimo's code...
will clone it
lizmat maybe his rakudo is out of date ?
I think import didn't work until maybe sometime last week ? 20:49
timotimo well, i'm already compiling the newest nom :)
FROGGS already done so :P
timotimo but i think my previous nom was pretty recent, too
lizmat we'll see :-)
FROGGS timotimo: I dont get an error for: PERL6LIB=lib perl6 -MADT -e '1'
timotimo yeah, the t/01-tree.t works no-problem 20:50
but t/02-EXPORT.t doesn't
(did you check out the correct branch?)
FROGGS yeah, 01 passes, 02 missing 20:51
what branch ooc?
ahh, okay, got it
adu why is it still { <…> } in NQP and {*} in Perl6? 20:52
lizmat because NQP is "Not Quite" ?
adu I thought they were changing it to {*} 20:53
lizmat I'm not aware of that, but that doesn't mean much :-)
adu hmm 20:54
FROGGS timotimo: the Perl6::Grammar was fixed to support arglist properly?
timotimo er ... wha? 20:55
FROGGS it wasnt working a week ago or so
timotimo what is arglist in this case? 20:56
FROGGS use Module "arglist"
making it an our-sub changes its behaviour though
timotimo oh 20:57
FROGGS Nominal type check failed for parameter '@attrs'; expected Positional but got Any instead
so $_<types> is Any at line 78 20:58
timotimo let me bring everything up to date with my new rakudo 20:59
FROGGS hmmm, there is no <types> in your grammar 21:00
ahh, there, I see
timotimo: $adt is a failed match for "foo bar" 21:01
timotimo oh, duh 21:02
m)
but it gets executed on your end
didn't on mine. i got "cannot find &create_adt" no my end
FROGGS ya, that was before I made it an our-sub
lizmat night #perl6 21:05
timotimo huh, wasn't it an our sub before as well?
night lizmat
FROGGS hmmm, maybe it was, but since I was on the wrong branch first...
timotimo: no, the diff says it wasnt our 21:06
timotimo strange. i'll investigate :) 21:07
indeed!
21:07 berekuk left
timotimo # Error: Type check failed in assignment to '$a'; expected 'Tree' but got 'Tree' 21:12
that's just great :(
flussence
.oO( maybe it wanted a larch instead of a data structure )
 21:13
geekosaur elmost right... 21:15
21:21 japhb_ joined
gtodd ᑲᐦᑲᑯᐤ 21:35
FROGGS ? 21:36
diakopter r: say ??!!q!??!??q!??!!!q??; 21:42
camelia rakudo 5ad46f: OUTPUT«??␤»
diakopter n: say ??!!q!??!??q!??!!!q??; 21:43
camelia niecza v24-37-gf9c8fc2: OUTPUT«===SORRY!===␤␤Expecting a term, but found either infix ?? or redundant prefix ?␤ (to suppress this message, please use space between ? ?) at /tmp/GTY_vyhv2d line 1:␤------> say ??⏏!!q!??!??q!??!!!q??;␤␤Parse failed␤␤»…
moritz std: say ??!!q!??!??q!??!!!q??;
camelia std b33d8e0: OUTPUT«===SORRY!===␤Expecting a term, but found either infix ?? or redundant prefix ?␤ (to suppress this message, please use space between ? ?) at /tmp/GljRSEOi2Z line 1:␤------> say ??⏏!!q!??!??q!??!!!q??;␤Parse failed␤FAILED 00:00 41m…
labster rn: ??!!'foo' 21:45
camelia rakudo 5ad46f: ( no output )
..niecza v24-37-gf9c8fc2: OUTPUT«===SORRY!===␤␤Expecting a term, but found either infix ?? or redundant prefix ?␤ (to suppress this message, please use space between ? ?) at /tmp/UxqB0_s8Uj line 1:␤------> ??⏏!!'foo'␤␤Parse failed␤␤»
21:45 spider-mario left
FROGGS rn: my $a; $a = ??!!42; say $a 21:46
camelia niecza v24-37-gf9c8fc2: OUTPUT«===SORRY!===␤␤Expecting a term, but found either infix ?? or redundant prefix ?␤ (to suppress this message, please use space between ? ?) at /tmp/qxgJmqaSgn line 1:␤------> my $a; $a = ??⏏!!42; say $a␤␤Parse failed␤␤»…
..rakudo 5ad46f: OUTPUT«True␤»
FROGGS ahh, sure
moritz couldn't sleep, so he implemented 'recurse': github.com/perl6/specs/issues/37#i...t-17442034 21:47
grondilu lol that was fast. moritz++ 21:48
r: sub postfix:<!>($n) { $n > 1 ?? $n * recurse($n-1) !! 1 }; say 4! 21:49
camelia rakudo 5ad46f: OUTPUT«===SORRY!===␤Undeclared routine:␤ recurse used at line 1. Did you mean '&reverse'?␤␤»
moritz grondilu: not as a built-in, as a user-supplied subroutine
grondilu oh yeah
moritz I also had an idea for a new project (again) :-) 21:51
FROGGS listens
moritz my working name is donor-o-meter
grondilu moritz wrote: "How often do you actually recurse into anonymous functions in real-world code?"
^ it's not just about anonymous functions. 21:52
moritz the idea is that I consider asking for donations for running the IRC logs
but I want some kind of coordination, so that I get neither flooded with donations, nor get none at all
grondilu I don't like to explicitely use the name of the function for recursion, because if I change the name later, I'll have to do it several times. It's not good imho
FROGGS moritz: ahh, otherwise I'd said you just make a paypal-donate button on that page 21:53
moritz so I want to display some message saying "your donations pay for this service until 2013-08-01" or so
along with a graphical bar that illustrates how far into the future or past that is
FROGGS and it will display a huge pic of the donator for the given time?
moritz and if it's maybe 2 months or more into the future, I won't even show the "donate" button
FROGGS k
sounds cool 21:54
moritz FROGGS: nah, just an acknowledgement and a link
grondilu: I'd argue that it's safer to use the name for recursion (more)
labster wow, that's amazing how github just mails out responses right away, unlike RT.
moritz grondilu: otherwise if your sub grows, and you refactor it by splitting stuff into a nother subroutine, you accidentally recurse into the wrong function 21:55
labster Agreed, it's probably safer to call itself by name when possible.
moritz grondilu: whereas when you rename a function, you typcially just grep for all occurrences, so you won't mess the recursion
*miss
grondilu oh I see. valid point. I trust your judgment anyway. So I'm fine with no "recurse". 21:56
and actually I wanted "recurse" instead of &?ROUTINE mainly for aesthetic reasons, so it's not a big deal. 21:57
(and also because "recurse" is easier to type than &?ROUTINE) 21:58
geekosaur labster, rt should be sending them immediately as well; it sounds like there are severe mail server issues with respect to rt.perl.org or whatever relaying server it might use
22:02 rindolf joined
sorear a large part of rt.perl.org's delay problem is caused by the fact that it uses manual moderation for spam control 22:06
22:10 dmol left 22:14 thou joined
moritz found lwn.net/SubscriberLink/548560/26d15e832d21a483/ nice 22:15
lue moritz: "nice" literally or sarcastically? 22:18
moritz lue: literally 22:22
.oO( REALLY!!!! )
 22:23
22:24 berekuk joined
pmichaud good afternoon, #perl6 22:32
diakopter hi Pm
22:33 pecastro left
adu good afternoon 22:33
labster Good PM, Pm.
FROGGS std: use v5; open( FH, '>', 'input.txt' ); print FH "hello"; close(FH) 22:36
camelia std b33d8e0: OUTPUT«ok 00:01 54m␤»
FROGGS std: use v5; open( my $fh, '>', 'input.txt' ); print $fh "hello"; close($fh) 22:37
camelia std b33d8e0: OUTPUT«===SORRY!===␤Confused at /tmp/lWNbujt8WQ line 1:␤------> ( my $fh, '>', 'input.txt' ); print $fh ⏏"hello"; close($fh)␤ expecting any of:␤ p5infix␤ statement modifier␤Parse failed␤FAILED 00:01 53m␤»
22:38 sftp left 22:39 kivutar left 22:43 sftp joined 22:44 pecastro joined
sorear yo pm. 22:45
22:52 kurahaupo joined
jnthn Good evening, #perl6 22:59
yoleaux 08:01Z <lizmat> jnthn: I'm not seeing how transposing "EXPORTHOW::<grammar> = TracedGrammarHOW;" to overriding "Mu::new" lexically would work
08:47Z <lizmat> jnthn: I found S12:2516 about Custom Meta-Objects
jnthn (for just a moment)
FROGGS o/
jnthn yoleaux: tell lizmat you want to subclass ClassHOW, I expect. You then write a role with the new functionality you want in, and then arrange for it to be mixed in to every class automatically (easiest is probably method compose($obj) { self.add_role($obj, YourRolewithNewIn); nextsame; } 23:01
FROGGS nr: my \FH
camelia niecza v24-37-gf9c8fc2: OUTPUT«===SORRY!===␤␤Term definition requires an initializer at /tmp/rjpOS1QkXm line 1 (EOF):␤------> my \FH⏏<EOL>␤␤Unhandled exception: Unable to resolve method statement_level in type Str␤ at /home/p6eval/niecza/src/NieczaA…
..rakudo 5ad46f: OUTPUT«===SORRY!===␤Method 'ast' not found for invocant of class 'NQPMu'␤»
jnthn ...so, that's not the syntax for yoleaux?
FROGGS nr: my \FH = 42
camelia rakudo 5ad46f, niecza v24-37-gf9c8fc2: ( no output )
FROGGS jnthn: .tell somebody test
23:02 kurahaupo left
FROGGS .tell jnthn YOLO 23:02
yoleaux FROGGS: I'll pass your message to jnthn.
jnthn .tell lizmat you want to subclass ClassHOW, I expect. You then write a role with the new functionality you want in, and then arrange for it to be mixed in to every class automatically (easiest is probably method compose($obj) { self.add_role($obj, YourRolewithNewIn); nextsame; } 23:03
yoleaux 23:02Z <FROGGS> jnthn: YOLO
jnthn: I'll pass your message to lizmat.
23:04 rindolf left
TimToady FROGGS: STD_P5 does not yet attempt to parse indirect objects :) 23:21
it's...complicated...
census what is YOLO again? 23:24
diakopter a justification for license to do anything regardless of any other consequences 23:25
geekosaur 'you only live once'?
labster Oh, like starting a sentence "With all due respect" allows you to say anything no matter how disrespectful it is. 23:29
FROGGS TimToady: well, thanks :P 23:31
TimToady: the problem is that I need to support that some day (these days basically) 23:32
timotimo it seems like something's still wrong with the EXPORT sub 23:39
(or possibly with the way i'm using ClassHOW in my ADT module) 23:40
the symptom is that i get error messages like Error: Type check failed in assignment to '$a'; expected 'Tree' but got 'Tree'
23:40 adu left
diakopter I wish that error message would include the .WHICH output of each of those 23:43
23:43 thou left
timotimo how would you feel if i granted that wish? 23:44
would it actually be helpful? beyond seeing that the WHICH is actually different?
diakopter yeah that I suppose 23:45
jnthn Sleep; 'night & 23:47
census thanks for the yolo explanation :) i knew i haerd it around this channel before :) 23:51
timotimo oh, huh?! 23:52
i added if $.expected.^name eq $.expected.^name { return $msg ~ " ($.expected.WHICH() and $.got.WHICH() respectively)"; } 23:53
but that doesn't trigger apparently
ah, ok, ::Assignment has its own message method
labster r: say IO::Spec.splitpath("dir", :nofile).perl; my $a = ("file", :nofile).item; say IO::Spec.splitpath( |$a ).perl; 23:55
camelia rakudo 5ad46f: OUTPUT«("", "dir", "")␤Too many positional parameters passed; got 3 but expected 2␤ in method splitpath at src/gen/CORE.setting:10471␤ in method splitpath at src/gen/CORE.setting:10897␤ in block at /tmp/qvdFInYXtl:1␤␤»
labster okay, how am I using method(|$a) wrong here? 23:56
timotimo r: say IO::Spec.splitpath("dir", :nofile).perl; my $a = ("file", :nofile); say IO::Spec.splitpath( |$a ).perl; 23:57
camelia rakudo 5ad46f: OUTPUT«("", "dir", "")␤Too many positional parameters passed; got 3 but expected 2␤ in method splitpath at src/gen/CORE.setting:10471␤ in method splitpath at src/gen/CORE.setting:10897␤ in block at /tmp/Eagr6M6Efo:1␤␤»
timotimo r: my $a = :("file", :nofile); say IO::Spec.splitpath( |$a ).perl;
camelia rakudo 5ad46f: OUTPUT«===SORRY!===␤Malformed parameter␤at /tmp/CMksMof7yu:1␤------> my $a = :("file", :⏏nofile); say IO::Spec.splitpath( |$a ).␤ expecting any of:␤ colon pair␤ signature␤ double quotes␤ constraint␤ …
timotimo r: my $a = \("file", :nofile); say IO::Spec.splitpath( |$a ).perl;
camelia rakudo 5ad46f: OUTPUT«("", "file", "")␤»
timotimo that's why i guess
labster r: my $a = \("file", :nofile); say $a.WHAT: 23:58
camelia rakudo 5ad46f: OUTPUT«===SORRY!===␤Two terms in a row␤at /tmp/ZxJA4Nmova:1␤------> my $a = \("file", :nofile); say $a.WHAT⏏:␤ expecting any of:␤ method arguments␤ postfix␤ infix stopper␤ infix or meta-infix␤ stat…
labster r: my $a = \("file", :nofile); say $a.WHAT;
camelia rakudo 5ad46f: OUTPUT«(Capture)␤»
timotimo no, i do not know why you can't just use the $a you had before and no, i do not know why it claimed to be 3 elements :|
FROGGS maybe |$a flattens into positionals only? 23:59
timotimo that doesn't explain how 2 become 3

Powered by Cro and the Raku® Programming Language.

Please report any issues / comments / feature requests as an issue on App::Raku::Log.

Thank you!