»ö« Welcome to Perl 6! | perl6.org/ | evalbot usage: 'p6: say 3;' or rakudo:, or /msg camelia p6: ... | irclog: irc.perl6.org or colabti.org/irclogger/irclogger_logs/perl6 | UTF-8 is our friend!
Set by moritz on 22 December 2015.
00:00 cyphase joined 00:01 itaipu joined
psch m: my @a[1;1]; my @b := @a; say @b.shape 00:02
camelia rakudo-moar 4abc28: OUTPUT«(1 1)␤»
psch m: my @a[1;1]; sub f(@a is raw) { say @a.shape }; f @a
camelia rakudo-moar 4abc28: OUTPUT«(1 1)␤»
psch m: my @a[1;1]; sub f(@a) { say @a.shape }; f @a
camelia rakudo-moar 4abc28: OUTPUT«(1 1)␤»
psch m: my @a[1;1]; sub f(@a where .shape eqv (1,1)) { say @a.shape }; f @a 00:03
camelia rakudo-moar 4abc28: OUTPUT«(1 1)␤»
gfldex if it's Array of Array would it not be all(@a) ~~ Positional?
psch well, if it was Array of Array :(Array @a) would match
but i guess it's probably not nominally Array of Array, just practically
but yes, if all elements are Arrays it should be &all 00:04
seatek I *love* that you can assign default values to a method's signature variables, from the class attributes. So handy. 00:05
00:07 tushar left
seatek weird about this though: Class.new(myVar => 'thisVal').iNeedThatVal -- method iNeedThatVal would not have the value in the class yet though. All the little things to learn. 00:09
00:09 imcsk8 left
psch m: class A { has $.myVar; method iNeedThatVal { $.myVar } }; say A.new(myVar => 'thisVal').iNeedThatVal 00:10
camelia rakudo-moar 4abc28: OUTPUT«thisVal␤»
psch seatek: uh..? :)
00:10 cyphase left
psch oh, probably about the parameter thing too? 00:11
seatek psch: maybe i'm crazy!
let me try again in a minute
psch m: class A { has $.myVar; method iNeedThatVal($myVar = $.myVar) { $myVar } }; say A.new(myVar => 'thisVal').iNeedThatVal
camelia rakudo-moar 4abc28: OUTPUT«thisVal␤»
psch nope, also works vOv
seatek well, it must have just started working then... ;) ;) (oops :) ) 00:14
psch probably TimToady++ again with his time machine :)
seatek yeah it does work. i must have been doing something else wrong at the time
hehe
00:15 cyphase joined 00:18 tushar joined
seatek I had to coerce a unix time into an integer for DateTime.new to accept it without an error. 00:23
00:23 cyphase left
seatek the timestamp value came out of a split 00:23
but there was no explicit typing going on 00:24
psch m: use NativeCall; my $a = cglobal('c', 'time', int32); say $a # booo 00:27
camelia rakudo-moar 4abc28: OUTPUT«Cannot locate native library 'libc.so': /usr/lib64/libc.so: invalid ELF header␤ in block at /home/camelia/rakudo-m-inst-1/share/perl6/sources/24DD121B5B4774C04A7084827BFAD92199756E03 (NativeCall) line 422␤ in block <unit> at <tmp> line 1␤␤»
psch probably some guesswork going wrong there
i mean, i know we default to libc for native subs that we don't give a libname for
seatek m: my $ts = "1475362095"; say DateTime.new($ts) 00:28
camelia rakudo-moar 4abc28: OUTPUT«Invalid DateTime string '1475362095'; use an ISO 8601 timestamp (yyyy-mm-ddThh:mm:ssZ or yyyy-mm-ddThh:mm:ss+01:00) instead␤ in block <unit> at <tmp> line 1␤␤»
psch might be neat if we had something similar for cglobals, if that's possible
00:28 cyphase joined
seatek m: my $ts = "1475362095"; say DateTime.new($ts.Int) 00:28
camelia rakudo-moar 4abc28: OUTPUT«2016-10-01T22:48:15Z␤»
seatek m: my $ts = 1475362095; say DateTime.new($ts)
camelia rakudo-moar 4abc28: OUTPUT«2016-10-01T22:48:15Z␤»
psch m: say DateTime.can('new').candidates>>.signature 00:29
camelia rakudo-moar 4abc28: OUTPUT«No such method 'candidates' for invocant of type 'List'␤ in block <unit> at <tmp> line 1␤␤»
psch m: say DateTime.can('new')>>.signature
camelia rakudo-moar 4abc28: OUTPUT«((DateTime $: | is raw) (Mu $: | is raw))␤»
psch m: say DateTime.can('new')[0].candidates>>.signature
camelia rakudo-moar 4abc28: OUTPUT«((DateTime $: \y, \mo, \d, \h, \mi, \s, :$timezone = 0, :&formatter, *%_) (DateTime $: :$year!, :$month = 1, :$day = 1, :$hour = 0, :$minute = 0, :$second = 0, :$timezone = 0, :&formatter, *%_) (DateTime $: Date:D :$date!, *%_) (DateTime $: Instant:D $i, :…»
psch feh
ETOOMANYCANDS
timotimo it's too candid 00:33
00:34 rgrinberg joined 00:36 cyphase left
grondilu (candid and candidate probably have a common ethymology. Never thought of that) 00:37
00:37 Zoffix joined
psch anyway, yeah, probably could solve that problem with a Int(Cool) coercer on the appropiate params 00:38
Zoffix Happy Hacktoberfest! It's an event where you can get a l33t shirt by sending 4 PRs to any repos on GitHub (see hacktoberfest.digitalocean.com/ ). We have many Perl 6 organization issues ready for Hacktober Hacking! Get fame and profit by helping us out: github.com/issues?utf8=%E2%9C%93&a...ktoberfest
psch wooo
tushar I am trying to remove elements from an array by preserving the indexes. Meaning if I delete first element, rest of the elements hold their original position. I wrote a function for it and you can find it here pastebin.com/Rfc32sh5 . I am just curious, is there any better way of doing this because Perl 6 is very rich in features and shortcuts. 00:39
psch m: my @a = 1,2,3,4; @a:delete(2); say @a.perl
camelia rakudo-moar 4abc28: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Variable '@a:delete<2>' is not declared␤at <tmp>:1␤------> 3my @a = 1,2,3,4; 7⏏5@a:delete(2); say @a.perl␤»
psch erg
m: my @a = 1,2,3,4; @a[2]:delete; say @a.perl
camelia rakudo-moar 4abc28: OUTPUT«[1, 2, Any, 4]␤»
psch tushar: ^^^ that maybe?
tushar hmm.. 00:41
00:41 cyphase joined
Zoffix m: my @a = 1,2,3,4; my @indx = 1, 3; @a[@indx]:delete; say @a.perl 00:42
camelia rakudo-moar 4abc28: OUTPUT«[1, Any, 3]␤»
00:42 Zoffix left
tushar psch and Zoffix thanks. But I do not want Any object in the output. 00:46
psch tushar: what do you want instead?
tushar for your example code, I want 1,2,4; 00:47
psch tushar: oh, so you *don't* want the gap
tushar: that's not "hold their original position"
tushar: if i have @a = 1,2,3; and i remove the 1, then either 2,3 move up or you get a hole at the start 00:48
00:48 buggable left
tushar yes. That's why I wrought a solution. Did you check that? My solution might be lengthy. 00:48
00:48 buggable joined
psch i don't get the solution 00:49
tushar For the same reason, I can not use splice either.
I provided a pastebin link
here it is again pastebin.com/Rfc32sh5
psch yes, i don't understand it :P
not your fault, probably
00:51 cyphase left
psch m: my @a = [[1,2,3],[4,5,6]]; @a[0] = @a[0].splice(1); say @a.perl 00:53
camelia rakudo-moar 4abc28: OUTPUT«[[2, 3], [4, 5, 6]]␤»
psch why does something like that not work?
geekosaur what might have helped is some examples of using it, both inputs and outputs 00:54
psch geekosaur++
psch probably needs to get to bed anyway o/
00:55 cyphase joined
tushar I can provide explanation. My function accepts two array -- one array holds the data which can be an array or array of arrays and another array holds the indexes. I convert the data array into hash key and value. I am using hash keys as indexes and removing them. Then I am looping through remaining hash keys and pushing their values to a new array. Replace old array with new array and return it. 00:57
01:01 canopus left 01:03 cyphase left
tushar geekosaur: pastebin already has an example. 01:04
do you prefer more?
timotimo i wonder if you can just @incoming_array[@my_list_of_key-lists]:kv:delete 01:05
01:05 itaipu left 01:06 itaipu joined 01:07 cyphase joined
tushar m: my @a = [[1..4],[5..8],[9..12],[13..17]]; @a[0,2]:kv:delete; @a.say; 01:08
camelia rakudo-moar 4abc28: OUTPUT«[(Any) [5 6 7 8] (Any) [13 14 15 16 17]]␤»
01:08 canopus joined
timotimo don't forget to say the result of @a[0,2]:... 01:08
tushar timotimo: it keep the gaps. I don't want them.
timotimo: yes. it returns key and corresponding values. 01:09
timotimo then i'd go ahead and grep for defined stuff later
tushar m: my @a = [[1..4],[5..8],[9..12],[13..17]]; @a[0,2]:kv:delete.say; @a.say;
camelia rakudo-moar 4abc28: OUTPUT«(0 [1 2 3 4] 2 [9 10 11 12])␤[(Any) [5 6 7 8] (Any) [13 14 15 16 17]]␤»
01:11 a3r0 left 01:12 a3r0 joined
tushar m: my @a = [[1..4],[5..8],[9..12],[13..17]]; @a[0,2]:kv:delete; for @a -> $ele { if $ele.defined { $ele.say }}; 01:12
camelia rakudo-moar 4abc28: OUTPUT«[5 6 7 8]␤[13 14 15 16 17]␤»
tushar timotimo: ^^^ like this
timotimo m: my @a = [[1..4],[5..8],[9..12],[13..17]]; @a[0,2]:kv:delete; .say for @a.grep(*.so) 01:13
camelia rakudo-moar 4abc28: OUTPUT«[5 6 7 8]␤[13 14 15 16 17]␤»
tushar timotimo++ 01:14
timotimo: can you provide explanation about ".say for @a.grep(*.so)"? 01:15
timotimo .say is syntactic sugar for "$_.say", a for loop at the end of the statement will just put stuff into $_ for you 01:16
and grep(*.so) will only give you things that return True when you call .so on them
and .so is the boolean coercer (opposite of "not")
tushar timotimo++ 01:17
timotimo: thanks for your explanation.
timotimo no problem at all
i'll be going to bed real soon now
tushar so I am the last luckiest one. :D
timotimo there's enough of me to go around usually ;) 01:18
tushar haha
tbrowder is there a way to get a 01:27
an unreleased module into travis somehow via the .travis.yaml file? 01:28
i'm testing a new module that requires another of my modules that is not yet in the ecosystem 01:30
timotimo just put "panda install ." into the travis.yml
it doesn't have to be in the ecosystem to be installed that way 01:31
tbrowder ok, i assume i can do "panda install <local path to repo> then? thanks, timotimo 01:32
timotimo that's right 01:33
01:34 itaipu left 01:35 itaipu joined 01:46 ilbot3 left 01:48 ilbot3 joined
tushar timotimo: as you are still around. I can little more of your help. I am trying to capture the deleted elements into another array. When I used your solution "@a[0,2]:kv:delete", it returns a list with key and values where keys are the indexes and values are the values that removed. I only want values and not indexes. How can i achieve that? 01:50
tbrowder hm, not working...what is CWD on travis?
tushar m: my @a = [[1..4],[5..8],[9..12],[13..17]]; @a[0,2]:kv:delete.say;
camelia rakudo-moar 4abc28: OUTPUT«(0 [1 2 3 4] 2 [9 10 11 12])␤»
tushar I want only [[1..4], [9..12]].
tbrowder going to read info on travis again... 01:51
02:03 noganex joined 02:06 noganex_ left
hackedNODE tushar: why'd ya have :kv on it then? :) 02:07
geekosaur because timotimo had it
hackedNODE tsk tsk
geekosaur (we were all still trying to figure out what tushar wanted...) 02:08
02:08 skids joined
hackedNODE tbrowder: have you seen docs.travis-ci.com/user/languages/perl6 ? 02:10
tbrowder: simple version that'll do it's thing. No ecosystem inclusion needed: github.com/zoffixznet/perl6-IRC-Cl...travis.yml
tbrowder that seems to imply my deps are inside my repo being tested, no. 02:21
no?
in that case i'll have to move some stuff around
i'll experiment some more tomorrow--nite all 02:24
hackedNODE tbrowder: are you deps in the ecosystem? Then it's fine. panda will find them 02:25
02:25 itaipu left
gfldex please keep in mind that changes to modules are delayed by up to 40min in the ecosystem 02:26
02:27 itaipu joined 02:30 cyphase left 02:34 cyphase joined 02:40 rgrinberg left 02:55 pierre_ joined 02:59 noganex_ joined, pierre_ left 03:02 noganex left 03:07 TakinOver left 03:10 tushar left 03:13 lostinfog joined 03:14 lostinfog left 03:27 itaipu left
dalek href="https://perl6.org:">perl6.org: 651ab0c | (Jota Sprout)++ | source/documentation/index.html:
Created new section for screencasts

Added YouTube lists and moved current screencasts link.		 03:28
href="https://perl6.org:">perl6.org: 45cd8db | (Zoffix Znet)++ | source/documentation/index.html:
Merge pull request #58 from jotasprout/master

Created new section for screencasts
href="https://perl6.org:">perl6.org: d373b31 | (Zoffix Znet)++ | source/documentation/index.html:
Fix typo
 03:29
03:44 grondilu left 03:58 kent\n left 04:10 kent\n joined 04:13 bjz_ joined, bjz_ left 04:15 bjz left 04:16 bjz_ joined, bjz_ left 04:17 bjz joined 04:18 bjz left 04:20 bjz joined 04:27 rgrinberg joined 04:41 khw left 04:57 Cabanossi left 05:00 Cabanossi joined 05:28 kaare_ joined 05:37 mohae joined 05:38 mempko joined 05:39 araujo left, araujo joined 05:41 mohae_ left 05:42 araujo left, araujo joined 05:44 araujo left 05:45 araujo joined 05:53 BenGoldberg left 06:09 seatek left 06:16 cyphase left 06:18 andrzejku joined 06:21 cyphase joined 06:36 bjz left 06:37 pierre_ joined 06:39 bjz joined 06:49 rgrinberg left 06:55 bjz left 06:56 bjz joined, domidumont joined 06:57 AlexDaniel left 07:00 domidumont left, skids left 07:01 xinming_ joined, domidumont joined 07:04 xinming left 07:05 domidumont left 07:07 pierre_ left, wamba joined 07:30 labster joined 07:41 MARTIMM joined
andrzejku hey, need help with 9 english senteces to fix them 07:45
;P
07:47 bjz left 07:50 bjz joined, setty1 joined 07:52 bjz left 07:53 bjz joined 07:54 kaare_ left 07:55 kaare_ joined, rindolf joined 08:05 pyrimidine left 08:08 pierre_ joined 08:11 nadim_ joined 08:12 pierre_ left
moritz andrzejku: I'm not a native speaker, but I can try 08:20
andrzejku moritz, already fixed in perl chat but thanks :P 08:21
08:30 mempko left 08:42 pierre_ joined 08:50 FROGGS joined 08:52 pyrimidine joined
FROGGS o/ 08:54
08:54 eliasr joined
moritz \o 08:58
raydiak o/
09:01 pierre_ left 09:06 pierre_ joined 09:15 kurahaupo joined 09:22 pierre_ left 09:33 ggoebel left 09:40 pierre_ joined 09:43 pmurias_ joined, pmurias_ left, pmurias joined
pmurias hi, rakudo doesn't support the -i flag? 09:43
yoleaux 1 Oct 2016 17:18Z <grondilu> pmurias: you're right, passing text between stages is not the way to go.
09:46 ggoebel joined
tbrowder .tell hackedNODE: no they're not, but i'm going to try to fake it with some sym linking 09:54
yoleaux tbrowder: What kind of a name is "hackedNODE:"?!
tbrowder .tell hackedNODE no they're not but i'm trying to fake it with some sym linking 09:55
yoleaux tbrowder: I'll pass your message to hackedNODE.
09:56 bjz left 09:57 bjz joined 09:59 bjz left 10:00 domidumont joined, bjz joined 10:02 bjz left 10:03 bjz joined 10:04 bjz left 10:06 bjz joined 10:07 bjz_ joined 10:08 bjz left 10:10 zakharyas joined 10:11 labster left
tbrowder not that sym linking doesn't work in git, had to cp -r from the other repo lib dir and temp add it to the repo i'm trying to test with an out-of-the ecosystem module; maybe a travis yml before_install cmd can do the trick? will check later if this works 10:12
10:21 bjz_ left 10:22 bjz joined, bjz left 10:23 bjz joined, bjz left 10:24 bjz joined 10:27 bjz left 10:30 bjz joined, bjz left, dayangkun joined 10:36 bjz joined 10:41 bjz left 10:44 pyrimidi_ joined 10:46 bjz joined 10:47 pyrimidine left 10:49 bjz left 10:50 bjz_ joined 10:51 itaipu joined
tbrowder hm, travis build is stuck, cannot cancel, back to drawing board... 10:52
has anyone used travis cache successfully. 10:53
10:54 bjz_ left, dayangkun left 10:55 bjz joined
tbrowder has anyone used cache successfully with a travis build? did it speed up the build? 10:56
11:04 domidumont left 11:12 Actualeyes left, darutoko joined 11:13 Actualeyes joined 11:17 zakharyas left, zakharyas joined 11:21 zakharyas left 11:40 kaare_ left, TEttinger left 11:42 grondilu joined
grondilu hello #perl6 11:42
FROGGS hi grondilu
tbrowder no one on #perl6-toolchain now, can anyone here talk about travis and p6 tool chain? 11:43
hi grondilu
for instance, using travis on github what is $HOME to travis for a github repo? 11:47
if so, then any scripting, etc., used in the yml file must be relative to that, but maybe github locations need to be by html link? 11:50
11:52 kaare_ joined 11:57 perlawhirl joined
dalek c: c04988c | titsuki++ | doc/Language/typesystem.pod6:
Add indexes for both :ver and :auth
 12:06
c: da5ad5b | titsuki++ | doc/Language/typesystem.pod6:
Merge pull request #932 from titsuki/add-ver-n-auth-index

Add indexes for both :ver and :auth
12:08 kaare_ left 12:29 itaipu left 12:30 itaipu joined
perlawhirl .tell tushar like this? gist.github.com/0racle/abbc0d2f7c7...7aacc7a5f8 12:31
yoleaux perlawhirl: I'll pass your message to tushar.
12:39 MARTIMM left 12:49 itaipu left 12:55 spebern joined 13:00 canopus left
spebern does anyone have any idea on the issue why panda's bootstrap.pl exits withot installing panda? 13:02
13:03 zakharyas joined
spebern I'm on arch, default login shell is zsh and I am using perlbrew for perl5 13:03
I have tried rakudobrew and rakudo-star-2016.07 13:07
13:08 canopus joined 13:09 pmurias left 13:16 pmurias joined 13:17 pierre_ left
SmokeMachine____ FROGGS: I didn't understand the problem you was trying to solve, but some one sad that a solver could help. So, do you think that this (github.com/FCO/ProblemSolver) could help? 13:20
13:21 kaare_ joined
timotimo spebern: does it give any output that looks like errors? did you add the path it mentions at the end to your PATH? 13:24
13:25 kaare__ joined
grondilu in src/Perl6/Grammar.nqp line 2750 I see the rule param_sep beginning with an empty string C<''>. Is that a hack or something? 13:25
13:26 kaare_ left, spebern left
grondilu what I mean is does an empty string have any effect in a regex? 13:26
timotimo "rule" is for - among other things - sigspace 13:27
so i expect it's there to allow sigspace at the beginning
grondilu as with :s ? 13:29
timotimo yes
andrzejku hi :) 13:35
stmuk_ timotimo: I think spebern is suffering the dreaded *BSD crash on linux 13:40
timotimo damn it :( 13:41
13:49 lucasb_ joined
stmuk_ it may affect linux more on newer processors due to an Intel processor bug but I've seen something v similar on lower spec processors 13:49
timotimo that's the microcode problem with lock elision? 13:50
stmuk_ yes 13:51
timotimo foams a little at the mouth
stmuk_ but a N2840 probably doesnt have HLE 13:52
13:59 cibs left 14:01 cibs joined, telex left 14:02 telex joined 14:04 rosso joined
dalek c: be918cf | titsuki++ | doc/Language/typesystem.pod6:
Fix a typo
 14:04
c: 3a88e95 | titsuki++ | doc/Language/typesystem.pod6:
Merge pull request #933 from titsuki/fix-grammar-typo

Fix a typo
rosso Hi there
timotimo greetings rosso
14:05 rosso is now known as Guest82778 14:06 AlexDaniel joined
Guest82778 which perl6 or perl 5 ? 14:06
moritz Guest82778: for what?
14:07 domidumont joined
Guest82778 which perl5 or perl 6 for you 14:08
hey 14:09
can you hear me ? 14:10
andrzejku hello my friends :)
perlawhirl we can. The answer for me is... whichever is more suitable for the task. the answer for you might be different.
mst both 14:13
moritz most likely both, yes
Guest82778 you are bitch 14:15
perlawhirl lovely
Guest82778 I Question not answer
fucking bobs 14:16
bye
14:16 Guest82778 left, zakharyas left 14:17 khw joined
mst wtf 14:17
14:18 pierre_ joined 14:21 bioduds joined 14:23 pierre_ left
pmurias do we have a replacement for -i in rakudo or should I work on implementing it? 14:24
moritz pmurias: I don#t think we have anything set
14:24 domidumont left
moritz *yet 14:24
14:25 domidumont joined 14:31 bjz left 14:32 bjz joined 14:43 domidumont left 14:52 spebern joined 15:00 rgrinberg joined
lucasb_ just a small nit: see this line here: github.com/rakudo/rakudo/blob/nom/...on.pm#L954 15:01
my $m = "Redeclaration of $.what '$.symbol$.postfix'";
maybe the intention was to write '$.symbol'$.postfix ? 15:05
see these errors: 15:06
m: -> $x { $^x }
camelia rakudo-moar 4abc28: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Redeclaration of symbol '$^x as a placeholder parameter'␤at <tmp>:1␤------> 3-> $x { $^x7⏏5 }␤»
lucasb_ m: sub f(--> Str) returns Int {}
camelia rakudo-moar 4abc28: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Redeclaration of return type for 'f (previous return type was Str)'␤at <tmp>:1␤»
AlexDaniel pmurias, moritz: #124259
synopsebot6 Link: rt.perl.org/rt3//Public/Bug/Displa...?id=124259
lucasb_ I would prefer that the single quotes was only surrounding the symbol, not the explanation 15:07
15:07 zacts joined 15:09 rindolf left
moritz lucasb_: open a pull request 15:11
lucasb_ about this -i flag, is the variable $*INPLACE or $*INPLACE-EDIT specced? 15:12
15:12 MARTIMM joined
moritz $ git grep INPLACE 15:12
S28-special-names.pod: $^I $INPLACE_EDIT $*INPLACE_EDIT ...or some such
lucasb_ moritz: oh, right. so this feature was already considered... just NYI. thanks 15:14
15:15 rindolf joined
Juerd uses -i with perl5 a lot 15:19
$^I too
MARTIMM hi, I was wondering if the module info using 'use' would be visible in the module 'use'd 15:21
gfldex MARTIMM: it can be if you provide your own EXPORT see docs.perl6.org/language/modules#in...sub_EXPORT 15:27
MARTIMM gfldex: thanks, I will look into it 15:28
15:40 BenGoldberg joined 15:41 cyphase left 15:44 domidumont joined 15:45 cyphase joined 15:46 domidumont left 15:47 domidumont joined 15:57 spebern left
FROGGS SmokeMachine____: I guess this could be a performance problem... I need to do this in Javascript when object are drawn to the calendar, and when stuff gets moved around 16:01
16:12 Ven_ joined 16:16 kaare__ is now known as kaare_ 16:23 aindilis joined 16:28 spebern joined
pmurias FROGGS: in your calendar problem the start time and and time of the spans are set and just need to assign each timespan a row? 16:32
16:35 labster joined 16:36 skids joined
FROGGS pmurias: aye 16:36
that's why I think I can just loop over the rows several times and I'm done
16:40 perlawhirl left 16:42 BenGoldberg left
pmurias FROGGS: would sorting the rows on starting time and then inserting them in the row with the biggest ending time they fit it work (and adding a new one if they don't)? 16:44
* they don't fit anywhere 16:45
FROGGS yes, that sounds like what I am up to 16:46
16:47 MilkmanDan left 16:48 Lucas_One joined, wamba left, MilkmanDan joined 16:51 spebern left 16:52 khw left 16:56 Ven_ left 16:58 labster left 17:08 khw joined 17:11 wamba joined
hackedNODE lucasb_: just checking, will you be submitting a PR for that exception message? 17:12
yoleaux 09:55Z <tbrowder> hackedNODE: no they're not but i'm trying to fake it with some sym linking
lucasb_ hackedNODE: can you do it? :) 17:13
hackedNODE lucasb_: OK
tbrowder hackedNODE: disregard the tell msg, I'm working on modified .travis.yml now to do what I need
lucasb_ when client code instantiates an X::Redeclaration exception, they have to pass the parameter with an leading space 'postfix => " some problem"' 17:14
it would be interesting to pass without space and handle this inside the message method of the X::Redeclaration class
hackedNODE sure
lucasb_ idk, something like: $.postfix.defined ?? " $.postfix" !! '' 17:15
hackedNODE ~ (" $.postfix" if $.postfix})
s/}// 17:16
lucasb_ hackedNODE: yes, works too :)
17:22 isBEKaml joined 17:31 isBEKaml left
timotimo hackedNODE: i just posted a summary of the profile of loading the gigantic repository and my conclusion to the perl6-compiler list 17:37
17:38 nadim_ left
hackedNODE cool 17:41
timotimo in short: canonpath is destroying our performance 17:42
slurping and concating is most probably only a percent or three of total run time 17:43
mst which people had already noticed as a bizarre pessimisation IIRC
timotimo the what now?
hackedNODE slurping and concating also has implication for memory use
mst the canonpath()ing lots in various IO handling
timotimo yes, true
i already gave it a nice little optimization path, but that's mostly for very simple paths 17:44
hackedNODE I could have a zillion files in my ~/ and if I include it in lib bad things will happen
So IMO, we shouldn't be slurping random files at all, but I don't know anything about CUR to offer any solutions
timotimo did you see how many entries we have to canonpath? 17:45
that's a little bit bizarre
there's only 4633 files under my mu 17:46
but it's entered 25500 times
hm, well, i guess since we recurse in dir i'd have to factor that in 17:48
hackedNODE m: say 25500 / 4633 17:49
camelia rakudo-moar 28c23a: OUTPUT«5.503993␤»
timotimo 4 is about the average depth directory-wise 17:50
17:51 wamba left
timotimo huh 17:52
did someone revert my canonpath optimizations?
hackedNODE s: "x".IO, 'canonpath' 17:53
SourceBaby hackedNODE, Something's wrong: ␤ERR: Type check failed in binding to &code; expected Callable but got Nil (Nil)␤ in sub do-sourcery at /home/zoffix/services/lib/CoreHackers-Sourcery/lib/CoreHackers/Sourcery.pm6 (CoreHackers::Sourcery) line 42␤ in sub sourcery at /home/zoffix/services/lib/CoreHackers-Sourcery/lib/CoreHackers/Sourcery.pm6 (CoreHackers::Sourcery) line 33␤ in block <unit> at -e line 6␤␤
hackedNODE s: $*SPEC, 'canonpath'
SourceBaby hackedNODE, Sauce is at github.com/rakudo/rakudo/blob/28c2...Unix.pm#L3
hackedNODE timotimo: they seem to be still there.
timotimo oh lord 17:54
my local rakudo is 4 weeks old
hackedNODE :}
timotimo i'll need to do the profile over ...
lucasb_ when a sub has return type constraints, like sub f(--> Foo) or sub f() returns Foo 17:56
what values/types can bypass this constraint? only Nil and Failure?
hackedNODE Yeah
timotimo those two or anything derived from them
lucasb_ is that intended behaviour, right?
hackedNODE Yes
timotimo yup 17:57
lucasb_ ok, thanks
17:57 setty1 left 17:59 optikalmouse joined
optikalmouse raiph, I saw your comments on my article, the perl6 json one. thx for the corrections :D 17:59
18:02 cyphase left 18:05 rgrinberg left
timotimo hackedNODE: followed it up with some new insights 18:11
hopefully that helps someone find the problem
a LHF would be for someone to port github.com/rakudo/rakudo/commit/6e...5e502ecbdd and the three commits before it to the Windows Spec 18:12
18:21 MasterDukeLaptop joined 18:26 wamba joined 18:30 dataangel left
nine I'm moving my root file system to a new disk while playing Diablo II. Linux is just awesome. 18:36
FROGGS heh 18:37
18:39 itaipu joined 18:41 sjoshi joined, sjoshi left 18:46 MARTIMM left 18:47 girafe joined 18:50 Actualeyes left 18:52 cyphase_eviltwin joined 18:59 Woodi_ left 19:00 nyarmith joined
moritz nine: recently I was burning a CD (old-fasioned, I know) while running an "apt-get dist-upgrade". I know that on Windows, I had to care to not do anything potential harmful while doing that 19:01
because something might interrupt the process, and the CD would be lost 19:02
(ftr I'm burning a CD because my $work workstation can't boot from USB sticks, even though the bios thinks it can)
bioduds I'm thinking ELIZA-like program would work very nicely using Grammars, what do you guys think? 19:03
19:05 Sqirrel joined
awwaiid bioduds: like grammar to extract out keywords? 19:10
bioduds yes 19:11
and process "quantum" style
so to speak
awwaiid I think ELIZA chops into words as it's only parsing
bioduds correct 19:12
perhaps there Markov Chains associated with Grammar Styles might be applicable 19:13
awwaiid that is possibly overkill for grammar then, though maybe you could encode some stemming? hmm
moritz yes, you'd use a p6 grammar to tokenize, not to parse
19:14 MasterDukeLaptop left
bioduds yes 19:14
arnsholt ELIZA only searches for key terms, IIRC
bioduds but then, associate with classes and roles to sort of create personas
arnsholt Regex-based, I think 19:15
awwaiid ah. you're likely going well beyond ELIZA at that point
bioduds have different "ELIZA"s so to speak
yes
also metaprogram its core in order to have ELIZA "learn" and "forget" things. Or rather yet, dinamically build it's core withing certain rules 19:17
awwaiid I guess looking at the __DATA__ in cpansearch.perl.org/src/NEILB/Chatb...t/Eliza.pm I start to imagine what you mean -- use grammars to encode these rules?
bioduds I'm guessing doing that in c would be a nightmare and in Perl6 would be much closer to something achievable 19:18
19:18 optikalmouse left
bioduds awwaiid : exaclty 19:19
19:19 darutoko left 19:20 rindolf left
bioduds once these rules are encoded into grammars, it becomes much easier to play around with them 19:21
have, say, slightly different sets, introduce perhaps a genetic algorithm to come out with better improved ones 19:22
the idea is exciting, right?
tailgate I'm a little terrified you're going to come up with some therapist that mind-controls its patients 19:24
psch weee machine overlords \o/
bioduds lol 19:25
19:25 rindolf joined
nyarmith lol, If it's mutual it's ok right? 19:27
both parties consent to the mind control
19:29 itaipu left
bioduds whereas both minds are in actual change 19:30
influenced by one another
DrForr Hey, have we put up your install script somewhere? (just curious, thinking about it) 19:33
bioduds yes 19:34
DrForr Cool!
bioduds it stands in the git repo waiting to be used in install.perl6.org
raw.githubusercontent.com/bioduds/...install.sh
this can be directly called already 19:35
but now someone with access to the perl6.org servers and https must carry out
gfldex lolibloggedalittle: gfldex.wordpress.com/2016/10/02/th...e-finding/
bioduds :)
DrForr Cool.
19:37 domidumont left, pmurias left 19:41 nyarmith left 19:45 Amendil joined 19:50 itaipu joined 19:52 benjikun joined
benjikun I'm running perl6 w/ bailador & http::server::simple; how do I make sure that the web server continues running even if it encounters errors? 19:53
tbrowder hackedNODE: and timotimo: I have a solution to the uninstalled dep problem for Perl 6 modules on gitbub/travis; see gist gist.github.com/tbrowder/a8e416a34...606d6ceb7b
19:54 dataangel joined
benjikun preferably without external scripts 19:54
nine Amazing what difference an upgrade of one's SSD can make. From "I cannot use my computer while our CMS' tests are running due to 20 second hangs and stopping mouse cursor" with the Crucial M4 128GB to "I don't notice anything besides the fan spooling up" with the Intel SSD 750 400GB PCIe 19:55
timotimo tbrowder: you don't have to cd into the folder 19:56
before you "panda install .", you just have to "panda install thefolder/"
tbrowder okay, i thought i tried that and it didn't work...i'll mod and try again... 19:57
gfldex benjikun: looking at it (need to read quite a bit code)
19:58 nadim_ joined 19:59 gregf_ left 20:01 Amendil left
timotimo maybe you didn't make it look like a path enough 20:01
try ./foobarblahblubb
gfldex benjikun: Bailador got a CATCH that should get everything (in /lib/Bailador/App.pm:method dispatch). As I see it, it should not stop on errors in your app, as long as they are no compile time errors. If it causes you problems, you need to file a github-issue on github.com/ufobat/Bailador/issues 20:03
20:04 Lucas_One left
benjikun Use of uninitialized value of type Any in numeric context 20:05
in block at /root/.rakudobrew/moar-nom/install/share/perl6/site/sources/100A1B43D02574570A9FA5
343A0679EF35FBD229 (HTTP::Easy) line 101
that is the error that is causing the application to quit
20:05 cyphase_eviltwin is now known as cyphase 20:06 cyphase left
timotimo that's not an error, that's just a warning 20:06
20:06 cyphase joined
benjikun it quits whenever that happens though 20:07
I've found another instance of stuff happening that causes it to quit
timotimo right, it could get confused by the value being used being actually 0 instead of something 20:08
benjikun thepb.in/p/y8h6GmPAprPHO
How would I keep the application running even when this happens, or even avoid it completely? 20:09
20:09 robertle left
tbrowder timotimo: you're correct, at least one slasg is required; travis buld error msg: 20:10
www.irccloud.com/pastebin/BPtnd6Vx/
timotimo that's helpful and nice :) 20:11
BBIAB
gfldex benjikun: that's a bug in HTTP::Easy github.com/supernovus/perl6-http-e...y.pm6#L101 20:12
benjikun what do I do
gfldex benjikun: file an issue at github.com/supernovus/perl6-http-easy/issues 20:14
benjikun: you could also fork HTTP::Easy and change „my $msg-body-pos;“ to „my $msg-body-pos = 0;“ (HTTP/Easy.pm6:58) and see if that helps 20:16
20:21 Ven_ joined
lucasb_ bioduds: I took a look at your install script. I guess you are brazillian, right? Cool, me too :) 20:25
bioduds opa, sim sim
BH e vc?
lucasb_ interior de SP 20:26
bioduds já começou a brincar de Perl6?
gfldex so you are the 2 guys from brazil who read my blog? :->
lucasb_ yes, playing with P6 on and off for a few months
gfldex: I opened in a tab here but didn't read yet :) 20:27
bioduds gfldex : by exclusion, yes lol
gfldex 2 out of 206,440,850. I'm getting there. :D 20:28
hackedNODE harmil_wk: unfortunately, we can't fix it as there are multiple tests in 6.c that expect it to throw a NoMulti exception rather than a better Assignment::RO RE: irclog.perlgeek.de/perl6/2016-09-29#i_13310847 20:29
bioduds :D very nice blog man! you sent me another link I guess earlier 20:30
gfldex i try to blog once a week
bioduds lucasb_ : have you actually tried the script? works ok for you?
i really wish I could keep up with a blog but I'm so lazy 20:31
lucasb_ bioduds: no, I didn't try. It installs rakudo star, right? I usually use the monthly releases or build from the repo.
bioduds yes, it does, correct :) 20:33
20:38 Ven_ left, Ven_ joined 20:42 FROGGS left
gfldex m: say X::IO::DoesNotExist.new(path=>'foo', trying=>'bar') ~~ X::IO 20:43
camelia rakudo-moar 3a6cd8: OUTPUT«True␤»
20:44 kaare_ left
gfldex m: say X::AdHoc.new(payload=>'foo') ~~ X::IO 20:44
camelia rakudo-moar 3a6cd8: OUTPUT«False␤»
gfldex m: X::IO.^name.say
camelia rakudo-moar 3a6cd8: OUTPUT«X::IO␤»
gfldex m: X::IO.HOW.say 20:45
camelia rakudo-moar 3a6cd8: OUTPUT«Perl6::Metamodel::ParametricRoleGroupHOW.new␤»
hackedNODE m: X::IO.^mro.say; X::AdHoc.^mro.say
camelia rakudo-moar 3a6cd8: OUTPUT«No such method 'mro' for invocant of type 'Perl6::Metamodel::ParametricRoleGroupHOW'␤ in block <unit> at <tmp> line 1␤␤»
hackedNODE m: X::IO.^pun.^mro.say; X::AdHoc.^pun.^mro.say
camelia rakudo-moar 3a6cd8: OUTPUT«((IO) (Exception) (Any) (Mu))␤No such method 'pun' for invocant of type 'Perl6::Metamodel::ClassHOW'␤ in block <unit> at <tmp> line 1␤␤»
hackedNODE m: X::IO.^pun.^mro.say; X::AdHoc.^mro.say
camelia rakudo-moar 3a6cd8: OUTPUT«((IO) (Exception) (Any) (Mu))␤((AdHoc) (Exception) (Any) (Mu))␤»
gfldex m: class X::IO::Foo is Exception { method message {} }; say X::IO::Foo ~~ X::IO; 20:46
camelia rakudo-moar 3a6cd8: OUTPUT«False␤»
gfldex m: class X::IO::Foo is Exception { method message {} }; say X::IO::Foo.new ~~ X::IO;
camelia rakudo-moar 3a6cd8: OUTPUT«False␤»
hackedNODE m: class X::IO::Foo does X::IO { method message {} }; say X::IO::Foo.new ~~ X::IO; 20:47
camelia rakudo-moar 3a6cd8: OUTPUT«True␤»
gfldex ENODOC
20:51 skids left
dalek c: fa9885a | gfldex++ | doc/Language/exceptions.pod6:
explain why smartmatching against exception categories work
 20:51
20:54 eliasr left
tbrowder timotimo: for my two modules at least, there is a difference in travis behavior between these two lines: 20:56
www.irccloud.com/pastebin/3oMnYJeV/
the first works and the second doesn't; the error msg on the second is during the test of the primary module being tested and the log says the just-installed dependency can't be found, so i'm assuming the resulting environment is different somehow. any ideas? i can try to specify PERL6LIB i guess after i find out where the two different locations are (or 21:00
find i made a false report!!)
21:00 rindolf left 21:04 skids joined
dalek line-Perl5: b69fda4 | niner++ | / (4 files):
Avoid slurpies for less overhead when calling P5 methods

By passing on the Capture instead of packing and unpacking slurpies, we can win another 5 % in csv-ip5xs. Previously an Empty returned by a P5 function without return values was filtered by the interpolation of the slurpies. Now we have to explicitly return a Nil instead (which is more correct anyway), turn it into a NULL and filter that in C code.
Version 0.17		 21:05
21:07 TEttinger joined 21:09 bjz left 21:13 andrzejku left 21:20 Ven_ left 21:23 mcmillhj joined 21:27 BenGoldberg joined 21:35 nowan left 21:37 nowan joined 21:38 labster joined 21:40 lucasb_ left 21:48 labster left, shinobicl joined
shinobicl hi all... how can i override the [] operator? 21:49
psch m: class A { method AT-POS($) { "this and others" } }; say A.new[1] 21:51
camelia rakudo-moar 3a6cd8: OUTPUT«this and others␤»
21:51 nyarmith joined
psch shinobicl: docs.perl6.org/language/subscripts...bscripting for the full list 21:51
shinobicl thanks! :)
gfldex lizmat: per6-docs-september-2016.txt gist.github.com/d20fb824a09abf7bdd...2c8ea0e3f3
21:53 dataangel left
gfldex m: class A { has $.foo = <a b c> }; multi sub postcircumfix:<[ ]>(A:D \SELF, \index){ SELF.foo[index] }; A.new[1].say 21:54
camelia rakudo-moar 3a6cd8: OUTPUT«b␤»
21:55 labster joined
shinobicl but what if i want to overload it for having to indices : $a[$row, $col]? That would be AT-KEY instead of AT-POS? Is there a way to use AT-POS with a pair of numbers? 21:55
gfldex shinobicl: ^^^ you can define your own [] if you like
shinobicl having two* indices
21:56 mcmillhj left
psch m: class A { method AT-POS($) { "this and others" } }; say A.new[1,2,3] 21:56
camelia rakudo-moar 3a6cd8: OUTPUT«(this and others this and others this and others)␤»
gfldex shinobicl: there is a postcircumfix:<[ ]> with Iterable as index type 21:57
psch anyway, g'night o/ 21:58
21:58 dataangel joined, mcmillhj joined
gfldex shinobicl: but it likely doesn't do what you want it to do. It calles .map on the index. You may have to define your own operator. 21:59
21:59 sQuEE joined, labster left
shinobicl m: class C { method AT-POS(Int $r, Int $c) { "$r and $c" } }; my $Obj = C.new; say $Obj.[1,3] 22:00
camelia rakudo-moar 3a6cd8: OUTPUT«Too few positionals passed; expected 3 arguments but got 2␤ in method AT-POS at <tmp> line 1␤ in block <unit> at <tmp> line 1␤␤» 22:01
shinobicl oops i added an extra point, but it returns the same.
gfldex shinobicl: that's the candidate it's calling: github.com/rakudo/rakudo/blob/nom/...ce.pm#L204 22:03
22:03 mcmillhj left 22:04 nyarmith left
shinobicl basically i want to implement an excel-like table with coordinates as ['A' ,3]... i have the inner structure done but i want to call them like that, even as [AB45] later, but the problem is with the former. 22:06
gfldex m: class A { has $.foo = <a b c> }; multi sub postcircumfix:<[ ]>(A:D \SELF, \index){ dd index }; A.new['A', 1]
camelia rakudo-moar 3a6cd8: OUTPUT«("A", 1)␤»
gfldex as long as you define your own class, you are good
with buildin type as SELF, you may face problems 22:07
22:08 dataangel left
shinobicl I've been far awy from perl6 too long i think, never seen this \SELF before 22:08
i guess is time to read the documentation again
gfldex shinobicl: docs.perl6.org/language/variables#..._variables 22:09
22:11 seatek joined, rgrinberg joined 22:13 mcmillhj joined 22:17 dataangel joined 22:18 mcmillhj left, rgrinberg left 22:20 rgrinberg joined 22:23 BenGoldberg left 22:26 dataangel left 22:31 mcmillhj joined 22:35 itaipu left 22:36 mcmillhj left 22:37 itaipu joined 22:43 mcmillhj joined 22:48 mcmillhj left 22:50 dataangel joined
tbrowder timotino: after adding some extra scripts for debugging the build, i have decided to add some more scripts to more tightly control the build. one problem is the chicken and egg situation where we need panda (or zef) installed before they can be used for installation. i want to use zef so that is part of the direction i'm heading without using panda (after 22:58
i get the rest working).
timotimo in the future, we'll be able to use raccoon 22:59
otherwise you can also just -Iothermodule/lib 23:00
if it doesn't have a Build.pm or something
23:00 BenGoldberg joined 23:11 itaipu left 23:13 girafe left 23:17 mcmillhj joined
tbrowder google doesn't help much, link, please? 23:18
23:18 BenGoldberg left
tbrowder hm, i forgot to search github... 23:19
240+ hits... 23:20
23:21 mcmillhj left
timotimo racoon is a tiny installer that's supposed to come with rakudo or something like that? 23:22
i'm not sure myself
i haven't tried it out yet
i think it's being developed from the install-core-modules.pm6 or something?
timotimo heads towards bed 23:25
benjikun night night timotimo, sleep well 23:26
timotimo night :) 23:28
23:30 skids left 23:43 skids joined 23:51 mcmillhj joined 23:53 labster joined 23:56 BenGoldberg joined, mcmillhj left, Actualeyes joined

Powered by Cro and the Raku® Programming Language.

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

Thank you!