»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! | YAPC::NA talk submissions due end of Month. Have you submitted your talk?
Set by toddr_ on 18 March 2013.
00:03 lizmat left 00:05 autumn left, sivoais left 00:08 Timbus joined, autumn joined
labster rn: '_' ~~ /<alpha>/ 00:12
p6eval rakudo ffe441, niecza v24-35-g5c06e28: ( no output )
labster rn: say '_' ~~ /<alpha>/
p6eval rakudo ffe441, niecza v24-35-g5c06e28: OUTPUT«「_」␤ alpha => 「_」␤␤»
labster Despite the fact that this is not well explained in the spec, I'm pretty sure this is wrong. 00:13
00:14 sivoais joined
rurban census: cygwin questions please @ #cygwin. And install gcc-4 as the error says, not gcc-3. 00:15
census rurban: thanks! i did post to #cygwin -- but there are not many perl folks there. i have been downloading gcc-4 for the past 2 horus now. at 95% hopefully will find out the answer soon! :) 00:16
rurban Net::SSL is in the package perl_vendor 00:17
find a faster mirror also :)
census rurban: what do you mean by mirror? 00:19
and do you mean i should be downloading perl_vendor from cygwin?
colomon labster: it might be worth looking in roast for that one, though my first inclination is you are right.
census dakopter! where ddi you go?
diakopter here 00:20
as my last msg said
:P
census OH
i was confused
i'm at 99%!
what does rurban mean?
rurban rurban is my name 00:21
diakopter *headdesk*, or, alternately, "heh"
census i meant what did you mean in what you said earlier about the mirror and perl_vendor 00:22
rurban I'm the perl maintainer for cygwin, but #perl6 has nothing to do with it
census diakopter! what is wrong?
rurban -- you are!
diakopter I was being silly
census omg! :)
awww
i got sad :(
i thought you were mad at me :(
diakopter no I was giggling at rurban's reply, which was ambiguously not tongue-in-cheek, so I replied both ways :)
labster colomon: One of the examples of class composition is: C<< <[_]+alpha+digit> >> 00:23
colomon labster: that certainly is very suggestive. ;)
00:23 sivoais left
rurban Well, I try to answer literally 00:23
diakopter census: please google what is an internet mirror
labster emotion is hard to feel though IRC
00:23 sivoais joined
census ok! :) 00:23
and what is meant by: "Net::SSL is in the package perl_vendor" ? 00:24
labster that's why we need to smile :) and raise our hands in the air \o/
diakopter labster: oh! I always thought those were legs!
rurban census: you should 1. find a faster mirror within the setup.exe mirror selection listbox, and 2. install the perl_vendor package if you need Net:SSL. Net::SSLey needs to be installed via cpan. Therefore you need to install 3. gcc-4 00:25
labster diakopter: oh, huh. Oh! Oh! can't unsee :/
census how do i know which mirror is fastest?
diakopter you just have to try different ones 00:26
rurban See what perl_vendor contains with the package search: cygwin.com/cgi-bin2/package-cat.cgi...erl_vendor
and I have to go now... :)
census oh no! 00:27
so you think now that i almost have gcc4 i should be able to install LWP::Protocol::socks, which needs Net::SSLeay ?
lue labster: S02/Names: "Underscore is always considered alphabetic." (presumably extends to regex rules too, at least S05 seems to imply it (S05/Simplified lexical parsing of patterns, see plaintext table)) 00:28
census help 00:29
It says:
"These tests *will* *fail* if you do not have network connectivity. [n]"
I think I do have network connectivity ...
lue rn: say "alpha_betic" ~ /<alpha>+/
p6eval niecza v24-35-g5c06e28: OUTPUT«alpha_beticRegex()<instance>␤»
..rakudo ffe441: OUTPUT«alpha_betic␤»
diakopter census: I suspect rurban was giving you an alternative (Net::SSL), so I don't think you need to install perl_vendor right now
so yes, try installing the same thing you did earlier 00:30
census ok thanks!
i'll try that next
i'm doing cpan install Net::SSLeay right now
the gcc4 JUST finished
labster lue: I'm sure that's going to surprise someone besides me 00:31
00:31 sivoais left
census diakopter: so how should i answer that question? y or n 00:32
the default seem n
i never got that q before!
diakopter what question
00:33 sivoais joined
census it is prompting me now 00:33
Do you want to run external tests? These tests *will* *fail* if you do not have network connectivity. [n]
lue r: say "alpha-betic?" ~~ /<alpha>+/ 00:34
p6eval rakudo ffe441: OUTPUT«「alpha」␤ alpha => 「a」␤ alpha => 「l」␤ alpha => 「p」␤ alpha => 「h」␤ alpha => 「a」␤␤»
diakopter census: y
census ok. said looks good for kit being complete :)
lue thinks hyphen should be part of the definition of <alpha> if underscore is there.
census but i don't think it worked
not sure 00:35
scsys.co.uk:8002/239036?ln=on&t...Format+it!
labster rn: say "alpha-betic" ~~ /<ident>/; 00:36
p6eval rakudo ffe441, niecza v24-35-g5c06e28: OUTPUT«「alpha」␤ ident => 「alpha」␤␤» 00:37
labster oh right, subtraction
lue ? I would think <ident> means identifier, which according the S02 section I cited is the "foo" in $foo, which allows hyphens.. 00:38
00:39 leprevost joined
census rurban: turns out that I already have perl_vendor downloaded! 00:39
00:41 sivoais left
diakopter census: do you not see the error message in the thing you nopasted? 00:41
census i do see it :(
i was in denial :(
diakopter which error message do you think you see
00:42 sivoais joined
census YAML not installed .... Failed during this command ... 00:42
SSLeay.o failed
diakopter keep trying
census cygmrf-4.dall cannot open shared object file
00:42 leprevost left
diakopter yes 00:42
00:42 Chillance left
diakopter census: did you google search that error message? 00:43
I did, and found you need to install the libmpfr library in the cygwin package manager 00:44
census www.google.com/#hl=en&q=%22cyg...mp;bih=410
diakopter it's not apparent to me why you pasted that url 00:45
census oh i'm sorry
i thought you asked me if i googled
and i wanted to show you what i googled
diakopter oh. :)
census so i need libmpfr ? 00:46
there are 8 options ....2 of hiwhc i apparently already have ...
i have libmpfr1 in libs and math
diakopter is there one with 4 in the name?
census yes!
libmpf4 in libs, libmpf4 in math
lifmpfr-devel in libs; lifmbfr-devel in math; lastly, libmpfr0 in obsolete. 7 total. 00:47
diakopter seems the missing dependency is a bug in the cygwin package 00:49
census are you implying then taht i should download the libmpf4 from the libs and from math?
diakopter for gcc
they're the same package
one will do
census ok! thank goodness this one seems muchfaster! 00:50
00:50 sivoais left
census it's not letting me download for some reason ... 00:51
00:51 sivoais joined 00:52 lichtkind left
Juerd An idea: perl6.org/compilers/features could have a column for Perl 5. 00:56
00:57 anuby joined 01:00 sivoais left 01:01 leprevost joined, sivoais joined 01:02 d4l3k_ joined, dalek left 01:03 dbr joined, d4l3k_ is now known as dalek
diakopter Juerd: why a column? 01:05
do you mean "the p5-in-rakudo version of the corresponding p6 feature"?
labster so we could compare how awesome Perl 6 is compared to the LTA Perl 5, I think he means. 01:06
diakopter <squint> p5 has its detriments, but I'm surprised anyone thinks they outweight the current p6 implementations' detriments.. 01:08
I mean, the "ideal" p6, sure 01:09
but there's also the ideal p5
01:10 sivoais left
labster it's marketing, though. I mean, we don't have checkboxes for "not slow as hell" and "has eleventy zillion modules" there. 01:11
01:11 sivoais joined
diakopter labster: we'll see about that... 01:12
(kidding, mostly)
01:20 sivoais left 01:21 sivoais joined 01:30 sivoais left, sivoais joined 01:39 sivoais left 01:41 sivoais joined 01:46 FROGGS_ joined 01:49 sivoais left, thou left 01:50 sivoais joined, FROGGS left 01:51 hypolin left 01:52 thou joined 01:53 hypolin joined 01:55 sivoais left 01:56 sivoais joined 02:00 hypolin left 02:04 hypolin joined 02:10 drbean left 02:12 drbean joined 02:16 thou left 02:17 lustlife joined 02:18 thou joined 02:26 japhb_ left
dalek kudo/extend-dump: 8976369 | moritz++ | src/Perl6/Compiler.nqp:
report -I and -M options in --help output
02:32
kudo/extend-dump: 9b7bf0d | (Geoffrey Broadwell)++ | src/Perl6/Compiler.nqp:
Merge branch 'nom' into extend-dump
kudo/extend-dump: 63fa63e | (Geoffrey Broadwell)++ | src/core/ (6 files):
Considerable DUMP refactoring, resulting in proper automatic handling of circular data structures -- but also some uninitialized value warnings that still require squashing
02:51 daniel-s joined 02:55 census left 02:56 orafu left 02:57 orafu joined 03:00 preflex left 03:01 preflex_ joined, preflex_ is now known as preflex 03:16 census joined 03:28 census left 03:56 census joined 04:08 SamuraiJack joined 04:09 leprevost left 04:31 sftp left, sftp joined 04:32 cooper left, shachaf_ joined, shachaf_ left, shachaf_ joined, shachaf left 04:33 shachaf_ is now known as shachaf, genehack left 04:34 cooper joined, genehack joined 04:54 SHODAN joined 05:29 dmol joined 05:34 aindilis` left 05:55 grondilu joined
grondilu rn: my Bag $b .= new: <foo foo bar>; say $b.kv; 05:56
p6eval rakudo ffe441, niecza v24-35-g5c06e28: OUTPUT«0 foo 1 bar␤»
06:02 rindolf joined, dmol left 06:09 domidumont joined 06:12 domidumont left 06:33 kaleem joined 06:34 FROGGS_ left 06:40 domidumont joined 06:41 drbean left 06:43 drbean joined 06:45 drbean left 06:53 grondilu left 07:00 drbean joined 07:01 dayangkun joined 07:07 FROGGS joined 07:08 snearch joined 07:10 census left 07:11 domidumont left 07:13 domidumont joined 07:14 snearch left 07:26 frd|afk left, frdmn joined 07:32 dmol joined 07:36 rindolf left 07:37 rindolf joined 07:43 daxim joined 07:45 rindolf left 07:46 rindolf joined 07:48 frd|afk joined
FROGGS who is responsible for feather.perl6.nl/syn/ ? 07:50
ahh, there is a note at the bottom
07:52 frdmn left
moritz everybody, and noone 07:53
dalek ast: e08e40f | (Tobias Leich)++ | S04-declarations/our.t:
(hopefully) fixed pod link to S04
FROGGS moritz: can I change that file? is it on github?
07:54 drbean left
moritz sure, and sure 07:55
I think in the Mu repo
FROGGS ahh, found it 07:56
moritz mlenz@mlenz-workstation:~/p6/mu$ git ls-files|grep html|grep syn|head -n1 07:57
docs/feather/syn_index.html
gist.github.com/quchen/5280339 # how somebody tried to troll #haskell. Very amusing. 07:58
07:58 rindolf left
moritz thought the language is partially not quite SFW 07:58
07:59 rindolf joined
diakopter moritz: haha! thanks for posting 07:59
FROGGS "I assure you
we have heterosexuals in our language as well" hehe
moritz "Haskell is invariant under gender. Really!" 08:01
tadzik haha 08:02
I admire their patience 08:03
labster lol awesome *troll gives up*
08:05 fhelmberger joined, am0c left
dalek ecs: 7afa4fe | (Brent Laabs)++ | S05-regex.pod:
clarification: add existing backslash classes, note <alpha> includes underscore
08:05
08:06 kurahaupo left
labster hopefully people like that ^ 08:06
moritz the commit message sounds really good 08:07
labster++
labster it creates a new rakudobug
dalek : ab9ab11 | (Tobias Leich)++ | docs/feather/syn_index.html:
fixed links to raw pod files
moritz OH NOEZ 08:08
labster rn: /\b/
p6eval rakudo ffe441: ( no output )
..niecza v24-35-g5c06e28: OUTPUT«===SORRY!===␤␤Unsupported use of \b as word boundary; in Perl 6 please use <?wb> (or either of « or ») at /tmp/qf1esHFLeg line 1:␤------> /\b⏏/␤␤Parse failed␤␤»
FROGGS ahh, this one
08:12 rindolf left 08:13 rindolf joined 08:19 kresike joined
kresike hello all you happy perl6 people 08:19
08:23 drbean joined 08:24 rindolf left 08:25 rindolf joined
FROGGS hi kresike 08:27
08:39 frd|afk left 08:41 dakkar joined
kresike hello FROGGS o/ 08:42
08:54 sqirrel joined 08:57 drbean left 08:58 drbean joined
masak good forenoon, #perl6 09:02
FROGGS hi masak 09:05
bbkr hi #perl6
FROGGS hi bbkr 09:06
FROGGS is the new welcome manager
labster congrats FROGGS 09:08
FROGGS *g* 09:09
jnthn o/ masam 09:12
*masak
and bbkr
FROGGS hi jnthn
jnthn is teaching today and for the rest of the week, but should have tuits again after this :)
labster I promised I'd blog about being a Perl 6 user, so here it is: brentlaabs.blogspot.com/2013/04/why...erl-6.html 09:19
masak \o/
labster Of course, I had to actually start a blog to do this (I'm much more of a wiki person myself)
masak you could always get a bliki ;) 09:20
labster++ # excellent post 09:22
labster: I wish we had a hundred delurkers like you :)
you hear that, you hundred lurkers? :)
FROGGS labster++ # that makes me think of start blogging too 09:24
labster That tiny regex example really impressed census++ earlier today, so I felt like sharing. From doing all of the porting of File::Spec, I have to say that I now am now quite sick of P5-style regexes and very happy to be working with the P6 version. 09:26
FROGGS labster: that is exactly what I am thinking 09:27
a year ago it was like: damn have to do PHP while there is such a wonderful Perl 5 out there 09:28
now I feel almost the same for Perl 5 like I do for PHP :o)
(but there is still a huge gap between PHP and Perl 5)
labster no kidding 09:32
PHP, Python, and Perl are all crazy people. Perl is crazy in a eccentric, mildly but happily intoxicated way. PHP is crazy in a more manic-depressive way, where you never really know what to expect. Python is the obsessive-compulsive, where everything must be just so or it'll get upset. 09:33
09:34 kaleem left
kresike labster++ #great post 09:50
labster, in the last paragraph, you wrote 'of waiting' twice 09:51
labster fixed, thanks kresike++ 09:58
good night all
10:00 labster left 10:02 census joined
census FROGGS: good morning! 10:06
FROGGS hi census
census thank you so much for your help! (can we chat in the other room for a quick second so as to not distract here... ?) 10:07
10:11 anuby left
census (i menat the help room ...) 10:12
10:13 hypolin left
FROGGS census: currently busy at work... 10:15
census oh ok, i completely understand! my questio pretty much was just that i could not remember exactly which packages i had downloaded to get perl up and running in cygwin. that is the from the program manager ...
10:27 rafl joined 10:33 skids left 10:35 skids joined 10:40 census left 10:50 lustlife left 10:55 dayangkun left
masak labster++ # love your categorization of P-language craziness 11:00
11:05 kaleem joined 11:14 brrt joined
moritz labster++ # blogging, crazy p-languages 11:24
timotimo i lay awake in bed at night thinking about how to implement a maze generator in perl6. open up rosetta code in the morning and sure enough ... 11:32
daxim it puts the indentation in the code or else it gets the hose again! 11:39
masak timotimo: strangelyconsistent.org/blog/an-int...ate-perl-6 11:40
11:41 drbean left
masak timotimo: if you don't know about union-find, then learn it. it's great stuff. 11:41
11:41 drbean joined
timotimo i know about it 11:42
in fact, i even knew it was the "perfect" way to make a maze
oh, actually i think i read that blog post before 11:43
masak :)
if any algorithm deserves the epithet "perfect", it's one with time complexity O( inverse of Ackermann function ) :) 11:44
11:46 arnsholt left
timotimo indeed 11:47
11:52 sqirrel left 11:54 sqirrel joined 11:57 gcole joined 12:07 kivutar joined
Woodi_ hallo 12:12
FROGGS hallo Woodi_
Woodi_ I cannot find * 2013.03... there wasn't one ?
FROGGS Woodi_: there is none yet as far as I know 12:18
timotimo that's true
Woodi_ just wanted to confirm, no problem for me
masak Woodi_! \o/ 12:19
Woodi_ btw. is there any TeamSpeak3 server for perl6 ? :)
yo masak !
(ts3 is free for 32 persons, 512 for non-profit organizations)
12:21 skids left
daxim proprietary software begone 12:21
masak IRC is my teamspeak. 12:23
Woodi_ you see, irc is used by software devs and gamers+devs. but gamers (and their devs) use voice coms too
mumble is gpl :0
I would love to lurk on perl6 voice coms :) 12:24
timotimo i prefer mumble over ts3, but i do have to admit ts3 has some neat features.
Woodi_ but more seriously: there is a 'cache' inside perl6 implementations... can it be used as 'module' or like hash ? 12:25
daxim modules.perl6.org/#Cache::Memcached
Woodi_ daxim: it's bindings library ? 12:26
daxim I haven't looked at the code, have you?
Woodi_ yes (in a sec) :) 12:28
timotimo that anchor doesn't seem to work :o 12:29
Woodi_ it's library for Memcache :)
but I was thinking about reusing internals of implementations... you know, hashes sometimes are used internally, so maybe caches and other thing can be maked available for general use... 12:30
12:31 tgt joined 12:33 census joined 12:35 jerome left 12:47 arnsholt joined
arnsholt labster++ # "I'm one of those mythical beasts that you've been hearing about for years but have never actually seen -- I am a Perl 6 user." 12:47
masak only unicorns use Perl 6. 12:48
12:48 sqirrel left
arnsholt And their programs are inscribed into slabs of pure unobtainium 12:48
13:04 anant joined 13:07 daniel-s left, daniel-s joined 13:12 shinobicl joined 13:15 kaleem left 13:16 SamuraiJack left
pmichaud labster++ # great post 13:24
good morning, #perl6 13:25
13:25 bluescreen10 joined, colomon left
moritz good am, pm 13:28
13:30 kaleem joined
masak pmichaud! \o/ 13:31
13:32 kaleem left 13:33 kaare_ joined 13:35 eternaleye left 13:38 PacoAir joined 13:40 FROGGS left 13:45 grondilu joined
grondilu created rosettacode.org/wiki/FASTA_format because he's having trouble figuring out what's the best way to parse this format. 13:46
[Coke] labster++ 13:50
pmichaud: 'allo 'allo. 13:51
13:51 skids joined
arnsholt [Coke]: Is this when I say "Good moaning"? =) 13:51
[Coke] Good Show! 13:52
masak r: say "{.[0]}: {.[1].comb(/ \N+ /).join}" for ">Rosetta_Example_1\nTHERECANBENOSPACE\n>Rosetta_Example_2\nTHERECANBESEVERAL\nLINESBUTTHEYALLMUST\nBECONCATENATED\n".comb(/ '>' (\N+)\n (<!before '>' >\N+\n?)+ /, :match) 13:54
p6eval rakudo ffe441: OUTPUT«Rosetta_Example_1: THERECANBENOSPACE␤Rosetta_Example_2: THERECANBESEVERAL LINESBUTTHEYALLMUST BECONCATENATED␤»
masak grondilu: ^
PerlJam gutenMorgen #perl6!
masak PerlJam! \o/
PerlJam masak! How are you today? 13:55
masak PerlJam: I can't complain. I literally can't. :P
grondilu: I whipped up this solution while at $work. feel free to add it to the page. it's a bit brief, and could be expanded with nice varaible names and such. 13:56
r: say "{.[0]}: {.[1].lines.join}" for ">Rosetta_Example_1\nTHERECANBENOSPACE\n>Rosetta_Example_2\nTHERECANBESEVERAL\nLINESBUTTHEYALLMUST\nBECONCATENATED\n".comb(/ '>' (\N+)\n (<!before '>' >\N+\n?)+ /, :match)
p6eval rakudo ffe441: OUTPUT«Method 'lines' not found for invocant of class 'Parcel'␤current instr.: '' pc 229 ((file unknown):171240541) (/tmp/puJomJBEe7:1)␤called from Sub '' pc 1411 ((file unknown):171240964) (/tmp/puJomJBEe7:1)␤called from Sub '' pc 274014 (src/gen/CORE.setting.pir:120247)…
masak ooh
r: say "{.[0]}: {(~.[1]).lines.join}" for ">Rosetta_Example_1\nTHERECANBENOSPACE\n>Rosetta_Example_2\nTHERECANBESEVERAL\nLINESBUTTHEYALLMUST\nBECONCATENATED\n".comb(/ '>' (\N+)\n (<!before '>' >\N+\n?)+ /, :match)
p6eval rakudo ffe441: OUTPUT«Rosetta_Example_1: THERECANBENOSPACE␤Rosetta_Example_2: THERECANBESEVERAL LINESBUTTHEYALLMUST BECONCATENATED␤»
masak that's better. &lines already means .comb(/ \N+ /) 13:57
arnsholt *sigh* The globbing feature in shell is nice, but it has terrible semantics (or lack thereof, perhaps) when nothing matches
PerlJam labster++ (reading scroll back and now your post :) 13:59
moritz ok, what I *really* want for the successor of planetsix.perl.org: a blog aggregator, and a few administrators can manually add single posts from elsewhere 14:01
14:01 am0c joined
PerlJam labster++ for the first sentence 14:02
and labster++ again for the last two sentences.
[Coke] moritz: yes, that would be nice. I will ask if that feature could be added.
grondilu masak++
kresike bye folks
14:02 kresike left
PerlJam actually ... has anyone started a Perl 6 quotes file? 14:03
pmichaud PerlJam: close might be pages on perl6.org/fun/ 14:04
masak PerlJam: rindolf has some quotes on Perl 6. but they're a bit dated, and mostly talk about Perl 6 in a Duke Nukem Forever kind of way ;) 14:08
arnsholt grondilu: Isn't there someone in here who's been dealing with bioinformatics stuff? Or that's you, perhaps?
rindolf PerlJam: you can use XML-Grammar-Fortune for that. 14:09
masak: I have some more recent stuff.
duckduckgo.com/?q=chuck+norris+per...mifish.org
grondilu masak: how could I get rid of spaces in the second example?
[Coke] moritz: oh. here's a thought - we could have a single feed that only had curated articles in it, and add that feed to the planet. Then the special processing is in the one feed, not the planet. 14:10
masak grondilu: oh, oops.
r: say "{.[0]}: {[~] (~.[1]).lines}" for ">Rosetta_Example_1\nTHERECANBENOSPACE\n>Rosetta_Example_2\nTHERECANBESEVERAL\nLINESBUTTHEYALLMUST\nBECONCATENATED\n".comb(/ '>' (\N+)\n (<!before '>' >\N+\n?)+ /, :match) 14:11
p6eval rakudo ffe441: OUTPUT«Rosetta_Example_1: THERECANBENOSPACE␤Rosetta_Example_2: THERECANBESEVERAL LINESBUTTHEYALLMUST BECONCATENATED␤»
masak hrm.
r: say "{.[0]}: {.[1]>>.lines.join}" for ">Rosetta_Example_1\nTHERECANBENOSPACE\n>Rosetta_Example_2\nTHERECANBESEVERAL\nLINESBUTTHEYALLMUST\nBECONCATENATED\n".comb(/ '>' (\N+)\n (<!before '>' >\N+\n?)+ /, :match)
p6eval rakudo ffe441: OUTPUT«Method 'lines' not found for invocant of class 'Match'␤current instr.: '' pc 166558 (src/gen/CORE.setting.pir:72841) (src/gen/CORE.setting:1054)␤called from Sub 'hyper' pc 150036 (src/gen/CORE.setting.pir:65526) (src/gen/CORE.setting:13255)␤called from Sub 'hyper' …
masak should Match have a .lines method? 14:12
rn: say Match ~~ Cool
p6eval rakudo ffe441, niecza v24-35-g5c06e28: OUTPUT«True␤»
masak r: say "{.[0]}: {.[1]>>.comb(/\N+/).join}" for ">Rosetta_Example_1\nTHERECANBENOSPACE\n>Rosetta_Example_2\nTHERECANBESEVERAL\nLINESBUTTHEYALLMUST\nBECONCATENATED\n".comb(/ '>' (\N+)\n (<!before '>' >\N+\n?)+ /, :match)
p6eval rakudo ffe441: OUTPUT«Rosetta_Example_1: THERECANBENOSPACE␤Rosetta_Example_2: THERECANBESEVERALLINESBUTTHEYALLMUSTBECONCATENATED␤»
masak grondilu: there we go. 14:13
14:13 xilo left
masak grondilu: .lines should work too, methinks. 14:13
arnsholt Hah! Good luck future self: $(DATADIRS): data/130314_parsereksperiment.tar.gz; DIR=$@ tar zxf $< -C data --strip-components=1 130314filer_til_arne/$${DIR#data/} 14:15
huf hope nobody ever puts spaces in paths :) 14:17
arnsholt I don't wanna talk about it =p 14:18
grondilu masak: I added your solution to RC but I mentioned that a more elegant way to do it would be welcome.
masak submits Match.lines rakudobug
grondilu++
14:18 census left
arnsholt huf: I'm the only one working on this (it's for my research), so that's not gonna happen. But yeah. A great big hack in the Unix tradition =) 14:19
masak r: "1\n2\n3" ~~ /.*/; say $/.lines
p6eval rakudo ffe441: OUTPUT«No such method 'lines' for invocant of type 'Match'␤ in block at /tmp/u_Sd9GI3H5:1␤␤»
14:19 telex left
grondilu would like to use a grammar to do these kind of stuff properly 14:19
masak my solution can easily be transformed into a grammar.
14:20 census joined 14:26 FROGGS joined, Patterner left 14:28 Psyche^ joined, Psyche^ is now known as Patterner
timotimo i wonder if libtcod (the roguelike development library) is a good target to be NativeCall'd from perl6 14:28
arnsholt googles 14:29
timotimo ooooh, 7 days ago the tcod dev bumped the version number to 1.6.0 14:33
arnsholt timotimo: From a skim of some of the docs, doesn't look entirely unreasonable
As long as it doesn't need callbacks. 'Cause those appear to be broken =(
timotimo i'll just avoid callbacks if they are not required and stall the project if they are 14:34
i believe in the perl6 community's ability to get this problem out of the way :)
or maybe the parrot people. who knows!
masak callbaks are like the holy grail of NCI, it seems.
arnsholt For smallish stuff I think callbacks will still work
Something seems to be broken once Parrot does a GC, so as long as you have enough RAM to avoid that, you should be good 14:35
pmichaud r: say (12.34).lines
p6eval rakudo ffe441: OUTPUT«No such method 'lines' for invocant of type 'Rat'␤ in block at /tmp/6SZMjjMahp:1␤␤»
pmichaud wonders if .lines belongs on Cool 14:36
masak r: say Rat ~~ Cool
p6eval rakudo ffe441: OUTPUT«True␤»
arnsholt masak: For the time being I'm assuming it's my code doing something (more or less) stupid. Parrot is innocent until proven guilty =)
masak pmichaud: maybe not. that was my intuition about it, though. just like .comb or .split
arnsholt I think the general layout of the code handling the callbacks is sound. Just something breaking Parrot
masak *nod* 14:37
14:39 xilo joined
timotimo masak: got a time estimate for the next p6cc results? :) 14:40
masak timotimo: this coming weekend. 14:41
timotimo nice :)
masak my $work curses me repeatedly by having me live in interesting times. :) 14:42
so side activities go slowly at present. 14:43
timotimo oh my 14:44
jnthn The trick is to make it sufficiently painful for $work to make your times interesting... :P
timotimo jnthn: were you able to look at EXPORT_hash yet? 14:45
oh, interesting. my first post to perl.perl6.users did come through. although delayed by about one day 14:46
no software can be that slow o_O
jnthn timotimo: Not yet...mixture of jetlag/other stuff recovery plus needing to teach this week :)
timotimo oh,i actually read that you'll be teaching this week. i forgot it :|
jnthn I meant to look last night, but felt exhausted earlier than expected. 14:47
I have a horrible feeling the nasty sick people sat near me on the train yesterday may have shared their infections too :/
masak it's time to make a stand. boot sick people from crowded, non-ventilated public spaces. 14:51
14:54 shinobicl left, kivutar left 14:57 jerome joined
timotimo hm, i don't really see how i could properly TDD the bindings for the most part :( 14:58
jnthn hotel & 15:04
15:07 Guest61242 left 15:14 spider-mario joined 15:16 kaleem joined
timotimo hum. what operator do i override for equivalence testing? (in this case the color class) 15:17
15:17 tgt left 15:19 telex joined 15:21 Guest61242 joined
moritz what kind of equivalence testing? infix:<eqv>? 15:23
timotimo it would dispatch to TCOD_color_equals 15:25
so i suppose eqv is the right one?
15:26 tgt joined
timotimo oh, yikes. the vim color scheme thinks $s in a parameter list starts a substitution 15:29
15:32 brrt left
moritz I don't know which kind of equivalence testing you want to override; that's what I asked you :-) 15:32
timotimo i'm going with eqv, it seems to be the right one in this case 15:33
15:37 tgt left
timotimo if there's a struct i want to bind, do i just have to say "is native 'TCOD_color_t'" and put a has int8 $.r and $.g and $.b in there if that's what the struct looks like in c? 15:38
15:38 pmichaud left, pmichaud joined
moritz not quite 15:38
you write class TCOD_color_t is repr('cstruct') { has int8 $.r; has int8 $.r; has int8 $.g; } 15:39
timotimo ah, ok
moritz and then you use the new type as a type constraint
timotimo can i give it a different name and use is symbol('TCOD_color_t')?
15:39 sqirrel joined
moritz sub set_color(OpaquePointer, TCOD_color_t) returns int is native() {*} 15:40
timotimo also, if i have to stub the class, i don't have to put all the is in there at the beginning, right?
moritz timotimo: it's a Perl 6 level type. You can name it anything you want.
timotimo ah,good 15:41
15:41 am0c left
timotimo and if a function expects a pointer to TCOD_color_t, i'll just use CArray[TCOD_color_t], correct? 15:42
moritz correct
timotimo what's that about "is native()", don't i have to give the library name?
maybe only once? 15:43
moritz yes, I forgot that
FROGGS timotimo: no ,not just once 15:45
timotimo got it
FROGGS and you only can omit it if you mean libc6 15:46
timotimo good to know
15:47 rindolf left 15:49 atta_ left 15:53 atta_ joined
timotimo huh. "Nominal type check failed for parameter ''; expected Any but got Color instead" 15:56
apparently it's in the ===infix from the core setting coming from nativecall's PARAMETERIZE_TYPE
15:59 crab2313 joined
timotimo the offending line seems to be "our sub TCOD_color_set_HSV(CArray[Color] $c, num32 $h, num32 $s, num32 $v) is native('libtcod.so') { * }; 16:00
probably something wrong with my CArray type thingie?
woohoo, i segfault! 16:08
16:09 benabik joined 16:12 domidumont left
dalek kudo/extend-dump: ba5e04f | coke++ | t/spectest.data:
run test we already pass
16:12
kudo/extend-dump: 2441b01 | moritz++ | src/core/Pod.pm:
fix signature of Pod::Block.gist
kudo/extend-dump: ffe441f | (Tobias Leich)++ | / (2 files):
install our-scoped variables in the nextmost package
kudo/extend-dump: 5f169a3 | (Geoffrey Broadwell)++ | / (3 files):
Merge branch 'nom' into extend-dump
16:19 colomon joined 16:25 grondilu left, grondilu joined 16:34 kaleem left 16:38 alester joined
masak hey, why was '$:foo' changed to '$!foo' many years ago? 16:44
flussence looks too much like :$foo? 16:45
16:46 domidumont joined
masak hm, it does, doesn't it? 16:46
perigrin it would be easy to typo when typing at speed 16:47
moritz now we have $:foo again :-)
but with a different meaning
16:47 kivutar joined
moritz (implicit named parameter) 16:47
perigrin it might be easy to typo when typing at speed 16:48
16:54 eternaleye joined
grondilu just laughed at this meme: www.quickmeme.com/meme/3tnkv6/ 16:55
masak heh. 16:56
I think I would stop and ask myself what the heck I was doing not writing tests, somewhere around line 55.
PerlJam masak: why do you remember that $!foo was once $:foo ? :) 16:57
masak PerlJam: parallel discussion over at #moe :) 16:59
16:59 japhb_ joined
japhb_ Is Rakudo::Debugger known not-panda-installable right now, or is it just me? 16:59
masak PerlJam: also, I remember the Apocalypses quite well. A12 was even released shortly before I ventured into #perl6 for the first time.
17:00 dakkar left
PerlJam Ah. 17:00
masak , Perl 6 historian 17:01
japhb_ .seen jnthn 17:04
yoleaux I saw jnthn 15:04Z in #perl6: <jnthn> hotel &
17:04 SamuraiJack joined
japhb_ jnthn, is Rakudo::Debugger supposed to be installable with panda right now? 17:05
17:05 daxim left
colomon jnthn: that reminds me, I found Grammar::Debugger completely non-functional a few days ago, some sort of internal error whenever I tried to enter a command. 17:06
jnthn: I was talking with somebody about it here, but I don't know if it was ever properly noted or sorted.
arnsholt I think it's known to be bitrotted, unfortunately 17:11
Dunno what it'll take to unbreak it
17:16 colomon left 17:17 grondilu_ joined
japhb_ starts placing 'say Backtrace.new.full' in strategic places out of frustration in tracking down his problem 17:19
17:19 grondilu left 17:21 fhelmberger left
FROGGS ó.ò 17:23
poor japhb_ 17:24
japhb_: I believe I've installed rakudo::debugger yesterday
japhb hmm 17:26
FROGGS retrying currently
17:27 colomon joined
FROGGS ==> Successfully installed Rakudo::Debugger 17:27
japhb_ hmm 17:28
FROGGS japhb: that was `pandabrew rebootstrap`though
dalek rl6-roast-data: 390dc70 | coke++ | / (3 files):
today (automated commit)
japhb_ has not used pandabrew
Clearly I missed something important in the last 3 weeks.
[Coke] adds a barbershop pole to his bash prompt. 17:29
17:34 Chillance joined
FROGGS japhb: btw, if you would paste the error you get, we maybe can help 17:34
japhb_ For the Rakudo::Debugger install, I get this: 17:36
$ panda install Rakudo::Debugger
==> Fetching Rakudo::Debugger
==> Building Rakudo::Debugger
resolve stage failed for Rakudo::Debugger: Could not find Perl6::BOOTSTRAP in any of: /home/geoff/git/rakudo/install/lib/parrot/4.10.0-devel/languages/perl6/site/lib, /home/geoff/git/rakudo/install/lib/parrot/4.10.0-devel/languages/perl6/vendor/lib, /home/geoff/git/rakudo/install/lib/parrot/4.10.0-devel/languages/perl6/lib, /home/geoff/.perl6/2013.03-35-g5f169a3/lib, .
FROGGS hmmm, do you have several nqp's? 17:38
japhb Lots. But I generally build with --gen-parrot, so I get a local NQP 17:39
FROGGS and there is no other nqp in path?
that's just a guess, but that is treally something that causes trouble
bbiab
17:39 thou left
jnthn japhb_: Last I knew, Rakudo::Debugger was pandaable 17:39
japhb_: Haven't tried it in a while 17:40
japhb_ jnthn, FROGGS just had success using pandabrew (which I've not seen before)
japhb_ wonders if there was another NQP in path the last time he did a *full* Rakudo rebuild 17:41
17:43 census left
japhb_ Well, I'll try a full flatten-and-rebuild of Rakudo, and see what happens. 17:44
17:45 grondilu_ left
jnthn FROGGS: I don't understand what ffe441f is doing 17:46
Just because a scope happens to have a $?PACKAGE in it doesn't mean that scope corresponds to the package itself... 17:48
17:53 SamuraiJack left 17:54 SamuraiJack joined 18:00 census joined 18:03 dmol left 18:05 dmol joined 18:11 dmol left 18:12 dmol joined 18:15 benabik left
japhb Hmmm, Rakudo flatten and rebuild did not help install Rakudo::Debugger. 18:19
18:19 SamuraiJack left 18:20 thou joined
japhb wonders if the fail is that 'which nqp' shows the symlinked location of nqp. 18:20
18:22 kaleem joined, kaleem left
tadzik japhb: pandabrew is just a convenience wrapper over panda 18:25
so it automatically detects when you've updated rakudo and need to rebootstrap panda
japhb_ Looks like a lib path mismatch. Perl6::BOOTSTRAP exists in ./install/lib/parrot/4.10.0-devel/languages/nqp/lib/Perl6/BOOTSTRAP.pbc , but panda is looking in .../install/lib/parrot/4.10.0-devel/languages/perl6/lib 18:26
japhb_ wonders why the mismatch between his rakudo and FROGGS'
18:26 colomon left
tadzik FROGGS may be using his own hacked Rakudo :) 18:27
tadzik tries locally
japhb_ tadzik, Ah. I have a bash script that does the rakudo rebuild and panda rebootstrap, which takes care to save backup copies of the state file in case Rakudo can't rebuild, or rebuilds but only gets partway through the panda installs before failing, or the machine crashes partway through the process 18:29
tadzik I want pandabrew to do all that
I just installed Rakudo::Debugger with success locally
japhb_ dangitall
r: .say for @*INC; 18:32
p6eval rakudo ffe441: OUTPUT«/home/p6eval/nom-inst/lib/parrot/4.10.0-devel/languages/perl6/site/lib␤/home/p6eval/nom-inst/lib/parrot/4.10.0-devel/languages/perl6/vendor/lib␤/home/p6eval/nom-inst/lib/parrot/4.10.0-devel/languages/perl6/lib␤/home/p6eval/.perl6/2013.03-12-gffe441f/lib␤»
japhb_ tadzik, could you do the same? 18:33
18:34 tgt joined
tadzik japhb_: gist.github.com/tadzik/5303877 18:34
japhb_ tadzik, what about `find . -name 'BOOTSTRAP*'`? 18:36
tadzik that's going to take a lot of time, even on this ssd :P 18:42
japhb_: gist.github.com/tadzik/5303984
18:43 thou_ joined, thou left
japhb_ Wait, there's a difference -- you're using a separate NQP build, not doing --gen-nqp (or --gen-parrot) in Rakudo 18:44
.oO( But why does that matter ...?)
18:45
18:46 woolfy joined
japhb_ Gah, merely forcing that nqp directory into RAKUDOLIB just gives me a different problem 18:49
tadzik yes
japhb_ tadzik, what do you get for: nqp --version; perl6 --version 18:51
tadzik This is nqp version 2013.03-1-g5e85657 built on parrot 4.10.0 revision RELEASE_4_10_0 18:58
This is perl6 version 2013.03-9-g8976369 built on parrot 4.10.0 revision RELEASE_4_10_0
FROGGS jnthn: I thought that only packagish things have $?PACKAGE set 18:59
jnthn: like packages, classes, roles ...
18:59 fgomez left 19:06 crab2313 left 19:09 fgomez joined
FROGGS jnthn: okay, I see what you mean: 19:12
r: class A; { my $?PACKAGE; our $thing = 7; }; say $thing
p6eval rakudo ffe441: OUTPUT«===SORRY!===␤Variable '$thing' is not declared␤at /tmp/rN0vruWBRB:1␤------> $?PACKAGE; our $thing = 7; }; say $thing⏏<EOL>␤ expecting any of:␤ postfix␤»
19:18 colomon joined 19:20 dmol left
census hi! does anybody know the channel in which people might be familiar with Tor? 19:22
moritz have you tried #tor? 19:23
that would be my very first try :-)
census just now! and i'm there alone :(
moritz maybe ##tor
nwc10 they seem to be on a different IRC network: www.torproject.org/about/contact.html.en#irc 19:24
census i find nottor ... not sure if that's the same thing!
nwc10: does different irc network mean that i cannot join # room name from here? 19:25
thank you for that link! :)
masak census: yes, it means that.
nwc10 correct, you can't join it from here just using /join
masak census: your IRC client joins different networks. this is the freenode network.
census oh ... is that the thing with xchat? (although my xchat expired ...) 19:26
masak census: if you haven't joined the network where the Tor channel is, you can't join the channel either.
census: think of it like countries and cities. to go to the right city, you first have to be in the right country :)
there's no Stockholm in France.
census yes ... and i don't even know how to read the map .. or where to buy one :( 19:27
taxicab -- take me to stockhol!! where ever that is! hahhah
sorear hmm, freenode web irc 19:29
you probably can't connect to OFTC with that.
census what does that mean? requested CTCP VERSION from census ?
19:29 kivutar left
census thank y'all for your help! :) 19:30
sorear www.oftc.net/oftc/WebChat
i went to the tor page masak linked, clicked on the "OFTC" link, then read the "Connecting" section 19:31
census so i can log in from: webchat.oftc.net/ ? 19:34
masak yeah.
then you'll be on the right network. 19:35
masak wouldn't recommend going from France to Stockholm by taxi
flussence It probably took longer to ask that question than to just try it.
Tene It probably took less time to recommend against going from France to Stockholm by taxi than it would have to just try it. 19:36
19:36 Guest61242 left
masak :P 19:37
FROGGS jnthn: github.com/rakudo/rakudo/commit/ff...nt-2938579
masak flussence: it's easy to look down on someone who hasn't learned to learn properly.
FROGGS are there other block declarations other than classes/packages/... ?
Tene looking down on people doesn't have a very steep learning curve.
flussence
.oO( before you look down on someone, walk uphill a mile in their shoes first )
19:39
masak besides, people have been trying to teach census how to fish for a while now. he seems more amenable to be given fishes. :)
FROGGS flussence: are they allowed to put them off first?
19:43 anant left 19:44 zby_home joined
census awww ... census just gets overwhelmed by computers so easily. 19:46
but appreciates your pateience :)
masak .oO( like a fisherman out of the water ) 19:51
census awww i'm so sorry :(
you guy are the best! :)
FROGGS \o/ 19:52
masak flattery will get you anywhere.
census welli tis sincere flattery!! for real!
19:53 LlamaRider joined
PerlJam census: it's because we are fishers of men ;-) 19:53
19:54 fgomez left
huf are you then? 19:54
stay the _hell_ away from seders
census huf: what is your question? 19:55
masak oh, so *that's* why there's only twelve of us!
census seders? as in passover?
huf as in i made a joke about the last supper 19:56
PerlJam we just to figure out who denies Larry 3 times so that we know who to call pope 19:57
census i missed you huf! 19:58
PerlJam census: aim better next time!
census aim for what?
huf a better brain. 19:59
or level up in "pun understanding"
masak dang, I think I've denied TimToady more times than three... :/
huf (or whatever the hell kind of wordplay this one was>)
PerlJam #perl6 -- the RPG 20:00
masak .oO( punderstanding )
huf all hail pope masak
gtodd anyone know anything about vmkit (llvm project) ?
masak no, I'm too denying to be pope.
I deny being pope.
huf you deny yourself three times
?
that's it. you're superpope for life.
masak easily.
dang.
moritz gtodd: I've heard the name, and have a vague idea what it does. Does that qualify?
oh, it's April again. There should be a new HPMoR progress note. 20:02
masak moritz: don't get your hopes up :/
moritz czechs
masak mopes
huf :)
nwc10 vmkit's mailing list doesn't seem to be massively active: lists.cs.uiuc.edu/pipermail/vmkit-commits/ 20:03
some months, no traffic
20:04 kaare_ left 20:06 fgomez joined
PerlJam masak++ (for nudging #moe in good directions) 20:07
20:08 domidumont left
moritz nwc10: mailing list activity isn't always a good indicator 20:08
20:08 zby_home left 20:09 araujo left
[Coke] moritz: really? our mailing list indicates we're mostly dead! 20:13
gtodd moritz: heheh guess so ... reading about it more it seems to be a tool for vm building using llvm clang etc. but possibly defined/designed in a way that specifically would not allow it to work with dynamic interpreted languages 20:15
masak PerlJam: I guess I care because I think stevan_++ is doing a good thing, technically and socially. 20:17
stevan_ auuuh, thanks guys 20:18
flussence (more toys to play with)++
masak (community bridges)++ 20:19
stevan_ it was only a matter of time before I realized that an "ultra modern Perl 5" would basically turn into a "slightly stunted Perl 6"
PerlJam (all you guys)++ :-)
masak stevan_: well from our perspective, that's a really good realization. "I ended up trying to make a better Perl 5, and I kept reinventing Perl 6!" :) 20:20
sorear o/ stevan_
stevan_ heya sorear 20:21
masak: s/reinventing/referring to/
masak aye.
20:22 sqirrel left
stevan_ WWP6D 20:22
stevan_ is gonna make bumper stickers
20:22 jrockway left, jrockway joined
nwc10 moritz: yes, but that list seemed also to be the commit list 20:23
20:23 rindolf joined
nwc10 so sometimes several months gaps in commit activity 20:23
PerlJam The community bridge comment is very important too. I noticed that there were some people on #moe who were not necessarily anti-perl6, but they weren't *anything-perl6* and now they're reading and learning about it. 20:24
s/were/are/ maybe
masak yeah.
20:26 kurahaupo joined
PerlJam If moe can bridge the laziness gap between p5 and p6, *that* would be major accomplishment IMHO 20:27
All the other stuff seems easy to me by comparison
20:28 spider-mario left
moritz laziness gap? 20:28
masak the reluctance to switch, I guess. 20:29
PerlJam p5 has mostly eager evaluations, while p6 has mostly lazy evaluations.
Getting people used to this is the problem.
masak oh! 20:30
it's only lists that are lazy, though.
PerlJam and lists are used everywhere. 20:32
maybe I'm making a mountain out of a molehill though.
20:37 bluescreen10 left
masak and maybe not :) 20:37
sometimes it certainly surprises me when I go back to Perl 5 :)
20:37 bluescreen10 joined
PerlJam To me, it's one of those underlying currents throughout the language. Like when you've only spoken a language that only had present-tense and then you learn about a language with both past and future tenses too. Your thinking has to adapt to these new concepts for you to effectively use the language. 20:40
(okay, I might be getting a little extreme in my comparison ;)
anyway, it's a difference that's more "within the programmer" than "within the language" IMHO 20:42
masak ...which is part of why we language people might not see that aspect very well. PerlJam++ 20:43
20:45 rindolf left 20:49 lizmat joined
lizmat thinks the topic is out of date 20:50
FROGGS well, there is always an end-of-month :o) 20:51
20:52 skids left
lizmat afaik the YAPC::NA talk submission window is closed 20:53
felher is surprised that <alpha> is not equivalent to <+alpha-[_]> :) 20:54
FROGGS felher: why should it? 20:57
masak felher: that has surprised me too sometimes. 20:58
felher: also, that _ is considered alphanumeric ;)
PerlJam well, _ is *clearly* a number! ;)
masak rn: say 1_000_000_00 20:59
p6eval rakudo ffe441, niecza v24-35-g5c06e28: OUTPUT«100000000␤»
LlamaRider felher: for me the bigger source of confusion both in P5 and P6 is why would \w+ match "vice_president" and not "vice-president" :)
masak PerlJam: you seem to be right :P
PerlJam of course I'm right! (I jam at perl! ;)
LlamaRider PerlJam: by your logic, isn't -1 a number? Then - should also be in there ;) 21:03
FROGGS and 'e'
err, nvm
-.-
PerlJam LlamaRider: no, 1 is the number, - is the adjective describing the number. 21:05
21:06 snearch joined
LlamaRider PerlJam: that's a very positive take on things ;) 21:07
lizmat rn: say 100\ 000\ 000
p6eval niecza v24-35-g5c06e28: OUTPUT«===SORRY!===␤␤Two terms in a row at /tmp/g_AIEBX3mM line 1:␤------> say 100\ ⏏000\ 000␤␤Parse failed␤␤»
..rakudo ffe441: OUTPUT«===SORRY!===␤Confused␤at /tmp/YbI0EuDstU:1␤------> say 100\ ⏏000\ 000␤ expecting any of:␤ unspace␤ horizontal whitespace␤ postfix␤»
masak lizmat: \ ends the term
lizmat expecting any of: unspace
but there is a an unspace there ! 21:08
:-)
masak yeah, but after that :P
rn: "OH HAI"\ \ .say
p6eval rakudo ffe441: OUTPUT«===SORRY!===␤Confused␤at /tmp/GLc7SahIm9:1␤------> "OH HAI"\ ⏏\ .say␤ expecting any of:␤ unspace␤ horizontal whitespace␤ postfix␤»
..niecza v24-35-g5c06e28: OUTPUT«===SORRY!===␤␤Method call found where infix expected (change whitespace?) at /tmp/yjAEiL5E1L line 1:␤------> "OH HAI"\ \ ⏏.say␤␤Parse failed␤␤»
masak *that*, however...
masak submits rakuodbug 21:09
rakudo*
lizmat I was just thinking the same thing
rn: say 100\ \ 000\ 000
p6eval niecza v24-35-g5c06e28: OUTPUT«===SORRY!===␤␤Two terms in a row at /tmp/fpQzbtkR06 line 1:␤------> say 100\ \ ⏏000\ 000␤␤Parse failed␤␤»
..rakudo ffe441: OUTPUT«===SORRY!===␤Confused␤at /tmp/u0R2Xl1yad:1␤------> say 100\ ⏏\ 000\ 000␤ expecting any of:␤ unspace␤ horizontal whitespace␤ postfix␤»
felher FROGGS: well, I didn't consider '_' to be alphanumeric. 21:13
LlamaRider: yeah, good point :)
I didn't even know that in p5, \w matches '_' :) 21:15
Which may have been a reason for me being surprised :)
FROGGS I always considered _ as alnum, but I can't explain why 21:17
LlamaRider FROGGS: That's a common bias coming from programming languages, since most langs allow _ in var names 21:19
at least that's how I always thought of it
FROGGS troo 21:20
timotimo in soviet timo's apartment, nap takes you
21:20 araujo joined
timotimo :S 21:20
21:22 tgt left, am0c joined 21:23 tgt joined
lue std: say 100\ 000 21:26
p6eval std 86b102f: OUTPUT«===SORRY!===␤Two terms in a row at /tmp/winx5Sgqq9 line 1:␤------> say 100\ ⏏000␤ expecting any of:␤ horizontal whitespace␤ infix or meta-infix␤ infixed function␤ postcircumfix␤ postfix␤
..postfix_prefix_meta_operator␤ postop␤ s…
lizmat std: say 100 000 21:27
p6eval std 86b102f: OUTPUT«===SORRY!===␤Two terms in a row at /tmp/TTsWjNFpxs line 1:␤------> say 100 ⏏000␤ expecting any of:␤ infix or meta-infix␤ infixed function␤ statement modifier loop␤Parse failed␤FAILED 00:00 42m␤»
lue
.oO(now I see why that doesn't work)
lizmat wonders whether it would make sense to simply concat numerics in this case 21:28
PerlJam and strings too!
std: say "foo" " bar";
p6eval std 86b102f: OUTPUT«===SORRY!===␤Two terms in a row at /tmp/VvK5512npg line 1:␤------> say "foo" ⏏" bar";␤ expecting any of:␤ infix or meta-infix␤ infixed function␤ statement modifier loop␤Parse failed␤FAILED 00:00 42m␤»
lue that's probably not a good idea (if you want visual space in numbers, use an underscore)
lizmat well, *that* would probably be hiding errors
moritz lizmat: forbidding two terms in a row is crucial for getting somewhat sensible parsing error messages 21:29
masak lizmat: I don't think so. the unspace indicates that we have already left "term" mode.
er, "term mode"*
lizmat stops wondering
21:29 grondilu joined
timotimo is this "spelled" correctly for zavolaj? "our sub TCOD_color_set_HSV(CArray[Color] $c, num32 $h, num32 $s, num32 $v) is native('libtcod.so') { * };" 21:29
Color is my class that is repr('CStruct') 21:30
PerlJam I'm not even sure zavolaj groks num32.
masak r: say num32 21:31
p6eval rakudo ffe441: OUTPUT«(num32)␤»
timotimo i looked at the sources to find out and saw it there
it says num32 is how to spell float
[Coke] r: say num128 21:32
p6eval rakudo ffe441: OUTPUT«===SORRY!===␤Undeclared routine:␤ num128 used at line 1␤␤»
FROGGS timotimo: I had problems using types at all 21:33
timotimo oh, i think i recognize the type of error!
FROGGS so I'd try to pass ($c, $h, $s, $v ..
timotimo last time i got it it was when i passed a class that's not composed yet! 21:34
i'm still getting wrong results and segfaults, though 21:35
PerlJam timotimo: pastebin the whole code? 21:36
timotimo haha, in the GC of all places :)
21:36 colomon left 21:37 colomon joined 21:38 bruges left 21:40 bruges joined
timotimo paste.ee/p/JdjGY there you go, PerlJam 21:40
FROGGS timotimo: it must now the structure of Color in order to unpack a piece of memory to its layout 21:42
21:43 labster joined
census hi labster! 21:43
timotimo how do i tell it the structure of Color? (originally called TCOD_color_t)
labster hi census
I saw the backlog. That was an awful lot of ++es. Thanks guys. 21:44
PerlJam timotimo: just like in README.markdown in the zavolaj distribution.
census labster++
PerlJam labster++ have some more. :)
21:44 xilo left
labster lol census++ (PerlJam++)++ 21:45
FROGGS timotimo: line 33-35 tells it its structure
timotimo oooh, a readme. that would have been too obvious for me :D 21:46
PerlJam timotimo: why are all of your subs declared with "our"?
timotimo i'm not good at this modules thing. 21:47
i thought it would be okay to expose both a class-based interface and the methods from the library "directly"
should i have used "is export" instead?
21:48 bluescreen10 left, bluescreen10 joined
PerlJam timotimo: I probably would have, but that doesn't mean it's the right thing for you :) 21:48
timotimo right. so what's wrong about the way i created the class? 21:49
ah, in the paste i had the traits only once, in the stub
FROGGS nqp: say ord('?')
p6eval nqp: OUTPUT«Confused at line 2, near "say ord('?"␤current instr.: 'panic' pc 14698 (src/stage2/gen/NQPHLL.pir:5230) (src/stage2/gen/NQPHLL.pm:278)␤»
timotimo but it doesn't make it work if i repeat the "is repr" trait in the implementation
FROGGS nqp: say(ord('?'))
p6eval nqp: OUTPUT«Could not find sub &ord␤current instr.: '' pc 46 ((file unknown):154144071) (/tmp/0wTRfuiU4q:1)␤»
FROGGS nqp: say(nqp::ord('?'))
p6eval nqp: OUTPUT«63␤»
21:51 PacoAir left 21:54 bluescreen10 left 21:55 Chillance left 21:56 Chillance joined 22:03 LlamaRider left
labster r: {my $*CWD; chdir '/home'}; say $*CWD; say cwd; 22:12
p6eval rakudo ffe441: OUTPUT«/home/p6eval␤/home␤»
labster okay, how am I supposed to change the working directory in only a local scope?
Tene have an exit handler change back. 22:14
that's dynamic scope, though, not lexical.
lizmat wrt to earlier underscore discussion: S02L2526 states: "Underscore is always considered alphabetic"
I wonder whether that shouldn't say "alphanumeric" instead 22:15
rn: say 100a100a100
p6eval rakudo ffe441: OUTPUT«===SORRY!===␤Confused␤at /tmp/kkTaKqNRyT:1␤------> say 100⏏a100a100␤ expecting any of:␤ whitespace␤»
..niecza v24-35-g5c06e28: OUTPUT«===SORRY!===␤␤Whitespace is required between alphanumeric tokens at /tmp/b_gwEsURSZ line 1:␤------> say 100⏏a100a100␤␤Two terms in a row at /tmp/b_gwEsURSZ line 1:␤------> say 100⏏a100a100␤␤Parse fail…
22:17 snearch left
FROGGS nr: say '_' ~~ /<alpha>/ 22:20
p6eval rakudo ffe441, niecza v24-35-g5c06e28: OUTPUT«「_」␤ alpha => 「_」␤␤»
FROGGS nr: say '_' ~~ /<alnum>/ 22:21
p6eval rakudo ffe441, niecza v24-35-g5c06e28: OUTPUT«「_」␤ alnum => 「_」␤␤»
FROGGS nr: say '_' ~~ /<digit>/
p6eval niecza v24-35-g5c06e28: OUTPUT«Nil␤»
..rakudo ffe441: OUTPUT«#<failed match>␤»
22:22 skids joined, Chillance left
lizmat rn: my ($a,$c); ($a, $, $c) = 1..3 22:23
p6eval rakudo ffe441: OUTPUT«===SORRY!===␤Unsupported use of $, variable; in Perl 6 please use .join() method␤at /tmp/yfoQnzrLgP:1␤------> my ($a,$c); ($a, $,⏏ $c) = 1..3␤»
..niecza v24-35-g5c06e28: OUTPUT«===SORRY!===␤␤Unsupported use of $, variable; in Perl 6 please use .join() method at /tmp/hQlgta0UNw line 1:␤------> my ($a,$c); ($a, $,⏏ $c) = 1..3␤␤Parse failed␤␤»
lizmat according to S02L2574, this should give the message "Anonymous variable requires declarator" 22:24
I assume S02 is wrong here?
labster Tene: Exit handlers are all great until you start threading. I've had issues with cwd changing underneath me in mod_perl.
FROGGS rn: $,
p6eval niecza v24-35-g5c06e28: OUTPUT«===SORRY!===␤␤Non-declarative sigil is missing its name at /tmp/zi_RsGdhDk line 1:␤------> <BOL>⏏$,␤␤Confused at /tmp/zi_RsGdhDk line 1:␤------> <BOL>⏏$,␤␤Parse failed␤␤»
..rakudo ffe441: OUTPUT«===SORRY!===␤Non-declarative sigil is missing its name␤at /tmp/GS7nMqWi1x:1␤------> $⏏,␤ expecting any of:␤ statement list␤ prefix or term␤ prefix or meta-prefix␤»
22:24 tgt left
FROGGS nr: $, $ 22:25
p6eval rakudo ffe441: OUTPUT«===SORRY!===␤Unsupported use of $, variable; in Perl 6 please use .join() method␤at /tmp/I2sg_yQMvR:1␤------> $,⏏ $␤»
..niecza v24-35-g5c06e28: OUTPUT«===SORRY!===␤␤Unsupported use of $, variable; in Perl 6 please use .join() method at /tmp/vb7T4WbT4q line 1:␤------> $,⏏ $␤␤Parse failed␤␤»
FROGGS rn: my ($a,$c); ($a, $ , $c) = 1..3 22:26
p6eval niecza v24-35-g5c06e28: OUTPUT«===SORRY!===␤␤Non-declarative sigil is missing its name at /tmp/xwX8yqtySA line 1:␤------> my ($a,$c); ($a, ⏏$ , $c) = 1..3␤␤Confused at /tmp/xwX8yqtySA line 1:␤------> my ($a,$c); ($a, ⏏$ , $c) = 1..3[0…
..rakudo ffe441: OUTPUT«===SORRY!===␤Non-declarative sigil is missing its name␤at /tmp/u7Rt00H9hd:1␤------> my ($a,$c); ($a, $⏏ , $c) = 1..3␤ expecting any of:␤ postfix␤ infix stopper␤ infix or meta-infix␤ prefix or term␤…
labster lizmat: underscore is alphabetic because subs can start with _ 22:27
lizmat indeed…
and identifiers can's start with numbers
ingy hi lizmat 22:29
lizmat Hi ingy
ingy :)
lizmat calls it a night 22:31
labster The problem with my expectation of unicode classes clashes with the fact that the language uses the grammar more than my code does. You can use the property directly if you have ICU installed (which I don't :( ), or just tell people to match <+alpha-[_]> if they mean to -- which is how I changed the spec.
lizmat night #perl6
labster good night lizmat
ingy lizmat: kisses
masak and hugs 22:34
lue <+alpha-[_]> means match alpha but not underscore, right? 22:35
22:35 woolfy left
FROGGS lue: right 22:35
lue r: say "foo_bar" ~~ /<+alpha-[_]>+/ 22:38
p6eval rakudo ffe441: OUTPUT«「foo」␤␤»
lue r: say "foo_bar" ~~ /[<alpha> !~~ _]+/; # this should be a working alternative
p6eval rakudo ffe441: OUTPUT«===SORRY!===␤Unrecognized regex metacharacter ! (must be quoted to match literally)␤at /tmp/aaMq6R7Wd5:1␤------> say "foo_bar" ~~ /[<alpha> ⏏!~~ _]+/; # this should be a working alt␤Unrecognized regex metacharacter [ (must be qu…
lue n: say "foo_bar" ~~ /[<alpha> !~~ _]+/; # niecza's error is better 22:39
p6eval niecza v24-35-g5c06e28: OUTPUT«===SORRY!===␤␤Action method quantifier:sym<~~> not yet implemented at /tmp/kRFXMdc6lF line 1:␤------> say "foo_bar" ~~ /[<alpha> !~~ _⏏]+/; # niecza's error is better␤␤Unhandled exception: Check failed␤␤ at /home/p6eva…
Tene labster: if you're threading, don't change CWD. Period. CWD is a per-process attribute. 22:41
dalek : 26ff8d5 | (Tobias Leich)++ | lib/Perl6/P5 (2 files):
make substitutions work
: 11b22cf | (Tobias Leich)++ | / (3 files):
support \cA character sequences
22:42 xilo joined
FROGGS tadzik: aww: de.webfail.com/eca15287507 22:47
census this is embarassing ... hope it is ok to ask ... is there a microsoft office/ word room somewhere? 22:49
i tried /msg alis list for searches ...
lue didn't know there was a <sp> rule in the regex
FROGGS census: you might google for office/word problems, you should find your answers there 22:53
census yep ... trying :)
timotimo also there's a built-in help tool that you can ask questions in human language, i believe 22:54
census hahaha yes i need human language :) 22:55
timotimo in the past i've also found answers to things on stack overflow (some excel formula weirdness)
so, i still have no idea what is wrong with my nativecall usage :( 22:56
census stackoverflow.com/questions/tagged/ms-word 22:57
do peopole actually check that?
timotimo i can imagine tha.
census i'm so gonna get my arse downvoted 22:58
i'm just feelign it
hahaha
felher good night, good people :) 22:59
timotimo good night :)
masak night, felher 23:00
'night, #perl6
timotimo 'night masak
colomon \o 23:06
23:08 am0c left 23:09 fgomez left 23:21 fgomez joined 23:28 Timbus left 23:29 autumn left 23:30 sivoais left 23:31 sivoais joined 23:34 Timbus joined, autumn joined