»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.
timotimo haha 00:00
do it like me; stay around when smarter people talk and grab all the smarts for yourself using osmosis
that's the only way i got any smart at all :)
lambd0x timotimo: it's sure a good tip ;) 00:02
Later guys o/
timotimo there's already a much simpler way to write all this
timotimo m: $*IN.words.Bag.list.sort(-*.value).say 00:03
camelia rakudo-moar 0de139: OUTPUT«(mo => 9 a => 8 le => 5 is => 5 ag => 4 an => 4 Ghaoth => 4 gan => 3 na => 3 ghlas’ => 3 slán => 3 Dhún => 3 i => 3 Dobhair. => 3 go => 3 gach => 3 do => 2 dom => 2 bhí => 2 ina => 2 nGall => 2 nár => 2 mé => 2 gleanntáin => 2 ar => 2 beag => 2 fá…»
sena_kun timotimo, I did it since my childhood, to be honest, so jk. That's why I learned haskell, for example, and did other strange things in my life. And you know, it's not always a best way to spend a night(putting haskell aside, it was great). 00:04
AlexDaniel m: ‘🚱’.uniname.say 00:04
camelia rakudo-moar 0de139: OUTPUT«NON-POTABLE WATER SYMBOL␤»
timotimo non-portable? 00:05
AlexDaniel that's what I read the first time too 00:06
gfldex timotimo: www.qsafesps.com.au/images/P/P205.jpg
timotimo ah 00:07
AlexDaniel
.oO( non-portable program. Do not drink )
 00:08
timotimo %)
also, i believe the .list in my m: was superfluous
as .sort will already call .list for you
sena_kun Ah, I've understand the '-' sign meaning. Seems like a golf. Thanks for a trick, timotimo.
timotimo gfldex already showed it a bit earlier 00:09
sena_kun timotimo, he did, but I catched it up a bit later. (: 00:10
timotimo OK :)
AlexDaniel 👶🚽 – looks great here
AlexDaniel but incorrect semantically 00:11
timotimo hm? 00:20
AlexDaniel timotimo: well, here is the screenshot: files.progarm.org/2016-08-07-03233..._scrot.png 00:24
timotimo oh, i didn't see the icons at all 00:25
timotimo i thought it was spaces :) 00:25
AlexDaniel it's a followup to “Do not drink” thingy :)
timotimo ah, i missed that connection 00:26
good one :)
gfldex AlexDaniel: using fonts for picrograms is a bad idea: imgur.com/a/Yk1Jg 00:28
AlexDaniel gfldex: yes
timotimo baby on the back of a toilet
lambd0x timotimo: I think I finally got what you meant earlier, my code: bpaste.net/show/0e9df598ffed 00:59
timotimo do you know yo udon't have to pull the value out, change it, then put it back in? :) 01:00
lambd0x timotimo: actually I didn't... 01:01
timotimo: hahah, how? 01:02
timotimo just %myhash{$thekey}++ 01:03
i still have no idea why you were using :kv :)
lambd0x timotimo: whahhhh...
timotimo: you know what...neither do I 01:04
hahah
timotimo and "my $tmp++" is the same as "my $tmp = 1", and i see no reason why you have $tmp at all
lambd0x timotimo: ah, that was just because I initially was unsure about how to insert in a hash variable, afterward I did forget about it there... 01:06
thanks for tips :)
timotimo ah 01:06
lambd0x timotimo: wow the code got a lot shorter now 01:07
timotimo: about the %myhash{$thekey}++, does it work with arrays too? 01:08
m:
m: my @array <1>; @array[0]++; 01:09
camelia rakudo-moar 77724b: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Missing required term after infix␤at <tmp>:1␤------> 3my @array <1>7⏏5; @array[0]++;␤ expecting any of:␤ prefix␤ term␤»
lambd0x m: my @array = <1>; @array[0]++;
camelia ( no output )
lambd0x m: my @array = <1>; say @array[0]++; 01:10
camelia rakudo-moar 77724b: OUTPUT«1␤»
timotimo don't forget it's returning the value that it was before incrementing 01:11
that's why it's 1 and not 2
lambd0x timotimo: yes :) 01:17
lambd0x timotimo: is this kinda of variable that is subjected to collision? 01:20
b2gills lambd0x: You can remove all of the conditional logic from your code. ( hint %h{'a'}++ will leave %h{'a'} == 1 the first time that 'a' is seen ) 01:30
timotimo collision?
b2g is correct 01:31
the magic behind that is quite impressive, though :)
auto-vivification is black voodoo
lambd0x b2gills: :) 01:36
timotimo: yes collision of values in the same key.. a large set of values for the same key that gets much bigger than the actual number of keys in the hash function. 01:39
b2gills lambd0x: If there is a collision the runtime hides it for you 01:42
lambd0x b2gills: even if the input is huge? 01:43
b2gills It can store more than one key-value in the same bucket. There have been CVEs against Perl5 because it didn't split the buckets up in a few cases 01:45
lambd0x b2gills: That's so nice \o/
lambd0x b2gills: how that for a code? bpaste.net/show/006f17d9c1aa 01:47
b2gills In fact I would expect that if you have more than ten or twenty key-values in a hash that it will almost always have some of them sharing buckets
lambd0x b2gills: true. 01:48
b2gills lambd0x: the unless could have ｢%hash{$word}++｣ in it and it would still work. 01:49
AlexDaniel lambd0x: by the way, can you paste your snippets without “$ cat cal_001.pl6” thingy? 01:50
so that we can ask one of the bots to eval it
lambd0x yes I can :P 01:51
lambd0x here: pastebin.com/qXvcT9VP 01:53
Thanks everyone :) 01:54
AlexDaniel committable: HEAD pastebin.com/raw/qXvcT9VP
committable AlexDaniel: Successfully fetched the code from the provided URL.
AlexDaniel: gist.github.com/d89e8d6b6cf0a4d367...fe8422f13d
b2gills .tell lambd0x I was trying to get you to realize it could be written as ｢my %h; for $*IN.words -> $word { %h{$word}++ }｣ then I was going to tell you about ｢my %h := $*IN.words.BagHash｣ irclog.perlgeek.de/perl6/2016-08-06#i_12978231 02:04
yoleaux b2gills: I'll pass your message to lambd0x.
AlexDaniel b2gills++ # ｢｣ 02:05
timotimo i'd advise for Bag instead of BagHash, as you probably don't have to modify it afterwards 02:09
b2gills That would have been immediately after BagHash 02:11
timotimo fair enough 02:41
awwaiid uses tau constant with joy 02:43
timotimo someone found out on stackoverflow, that the "sum" method ought to be marked "is nodal" 02:44
not sure if it's been reported to RT, though.
killbill expression s/(<:No+:Nd+:Nl>)$0$0/{($0+1) x 3}/ works with 666 but don't work with ⁶ ⁶ ⁶ ( Cannot convert string to number: base-10 number must begin with valid digits or '.' in '⏏⁶) 02:52
yoleaux 11 Jul 2016 06:31Z <TimToady> killbill: It's <:super>, not <:superscript>.
AlexDaniel m: 'my $x = ‘❸❸❸’; $x ~~ s/(<:No+:Nd+:Nl>)$0$0/{($0.Str.unival+1) x 3}/; say $x 03:00
camelia rakudo-moar 77724b: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Unable to parse expression in single quotes; couldn't find final "'" ␤at <tmp>:1␤------> 3l>)$0$0/{($0.Str.unival+1) x 3}/; say $x7⏏5<EOL>␤ expecting any of:␤ single quotes␤ …»
AlexDaniel m: my $x = ‘❸❸❸’; $x ~~ s/(<:No+:Nd+:Nl>)$0$0/{($0.Str.unival+1) x 3}/; say $x
camelia rakudo-moar 77724b: OUTPUT«444␤»
AlexDaniel by the way why not **3 ? 03:03
ah, I see now 03:04
killbill AlexDaniel: it was just synthetic test 03:08
b2gills m: my $x = ‘❸❸❸’; $x ~~ s/(<:N>)$0$0/{($0.Str.unival+1) x 3}/; say $x
camelia rakudo-moar 77724b: OUTPUT«444␤»
timotimo aye, we allow you to get "exotic" numbers as numbers in perl6 source code, but not in Intifying strings with a + or .Numeric or anything like that 03:09
Ben_Goldberg m: <❸❸❸>.perl.say 03:32
camelia rakudo-moar 77724b: OUTPUT«"❸❸❸"␤»
Ben_Goldberg m: < ❸❸❸ >.perl.say
camelia rakudo-moar 77724b: OUTPUT«"❸❸❸"␤»
Ben_Goldberg m: ❸❸❸.perl.say
camelia rakudo-moar 77724b: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Bogus postfix␤at <tmp>:1␤------> 3❸7⏏5❸❸.perl.say␤ expecting any of:␤ infix␤ infix stopper␤ statement end␤ statement modifier␤ statement modi…»
Ben_Goldberg m: (❸❸❸ + 1).perl.say
camelia rakudo-moar 77724b: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Bogus postfix␤at <tmp>:1␤------> 3(❸7⏏5❸❸ + 1).perl.say␤ expecting any of:␤ infix␤ infix stopper␤ statement end␤ statement modifier␤ statemen…»
TimToady you can't use non-digits as if they were digits 03:34
AlexDaniel it's kinda funny 03:43
m: ‘❸’.uniname.say
camelia rakudo-moar 77724b: OUTPUT«DINGBAT NEGATIVE CIRCLED DIGIT THREE␤»
AlexDaniel … calls it a digit!
m: ‘❸’.uniprop.say
camelia rakudo-moar 77724b: OUTPUT«No␤»
AlexDaniel m: ‘⓭’.uniname.say 03:44
camelia rakudo-moar 77724b: OUTPUT«NEGATIVE CIRCLED NUMBER THIRTEEN␤»
Ben_Goldberg m: 'a'.uniprop.say;
camelia rakudo-moar 77724b: OUTPUT«Ll␤»
Ben_Goldberg m: '42'.uniprop.say; 03:44
camelia rakudo-moar 77724b: OUTPUT«Nd␤»
AlexDaniel by the way 03:45
m: ‘4a’.uniprop.say
camelia rakudo-moar 77724b: OUTPUT«Nd␤»
AlexDaniel ↑ shouldn't this warn or something?
TimToady it's a feature, if ord is a feature 03:46
AlexDaniel m: ‘a4’.uniname.say
camelia rakudo-moar 77724b: OUTPUT«LATIN SMALL LETTER A␤»
AlexDaniel OK then 03:47
Ben_Goldberg m: .chr.say for first { *.chr.uniprop eq 'No' }, ^Inf;
camelia rakudo-moar 77724b: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Malformed double closure; WhateverCode is already a closure without curlies, so either remove the curlies or use valid parameter syntax instead of *␤at <tmp>:1␤------> 3.say for first { *.chr.uniprop eq…»
TimToady m: ‘a4’.uninames.say
camelia rakudo-moar 77724b: OUTPUT«(LATIN SMALL LETTER A DIGIT FOUR)␤»
Ben_Goldberg m: .chr.say for first { $_.chr.uniprop eq 'No' }, ^Inf;
camelia rakudo-moar 77724b: OUTPUT«²␤»
Ben_Goldberg m: .chr.print for grep { $_.chr.uniprop eq 'No' }, ^Inf; 03:48
AlexDaniel and we still have no uniprops
camelia rakudo-moar 77724b: OUTPUT«(timeout)²³¹¼½¾৴৵৶৷৸৹୲୳୴୵୶୷௰௱௲౸౹౺౻౼౽౾൰൱൲൳൴൵༪༫༬༭༮༯༰༱༲༳፩፪፫፬፭፮፯፰፱፲፳፴፵፶፷፸፹፺፻፼៰៱៲៳៴៵៶៷៸៹᧚⁰⁴⁵⁶⁷⁸⁹₀₁…»
Ben_Goldberg No means "number, other"
AlexDaniel haha
Ben_Goldberg m: '৴'.uniname.say; 03:49
camelia rakudo-moar 77724b: OUTPUT«BENGALI CURRENCY NUMERATOR ONE␤»
Ben_Goldberg m: '᧚'.uniname.say;
camelia rakudo-moar 77724b: OUTPUT«NEW TAI LUE THAM DIGIT ONE␤»
Ben_Goldberg m: .chr.print for grep { $_.chr.uniname ~~ /TURNED/ }, gather { take $++ while time < ((INIT time) + 1) }; 03:55
camelia rakudo-moar 77724b: OUTPUT«ƍƜǝɅɐɒɥɯɰɹɺɻʇʌʍʎʞʮʯʴʵ􏿽xCA􏿽xBB􏿽xCC􏿽x92􏿽xE0􏿽xA3􏿽xA3􏿽xE1􏿽x83􏿽xB9ᴂᴈᴉᴔᴚᴟᵄᵆᵌᵎᵚᵷᶛᶣᶭᶺ℩Ⅎ⅁⅂⅄⅋ⅎ↊↋⌙⛉⛊❛❝⦢⯂ⱯⱰⱹⱻ⸲⸵⸸ꝾꝿꞀꞁꞍꞰꞱꟺ꧍ꭁꭂꭃꭄꭑ»
Ben_Goldberg m: .chr.print for grep { $_.chr.uniname ~~ m/FRAKTUR/ }, gather { my $c = 100000; take $c++ while time < (INIT {2 + time}) }; 04:08
camelia rakudo-moar 77724b: OUTPUT«𝔄𝔅𝔇𝔈𝔉𝔊𝔍𝔎𝔏𝔐𝔑𝔒𝔓𝔔𝔖𝔗𝔘𝔙𝔚𝔛𝔜𝔞𝔟𝔠𝔡𝔢𝔣𝔤𝔥𝔦𝔧𝔨𝔩𝔪𝔫𝔬𝔭𝔮𝔯𝔰𝔱𝔲𝔳𝔴𝔵𝔶𝔷𝕬𝕭𝕮𝕯𝕰𝕱𝕲𝕳𝕴𝕵𝕶𝕷𝕸𝕹𝕺𝕻𝕼…»
AlexDaniel “Goo” is actually not a bad name for a programming language ( github.com/golang/go/issues/9 ) 05:19
ah, already taken? Just use “Gooo” then… 05:20
doublec AlexDaniel: goo was a lisp/dylan hybrid 06:08
andrzejku hello 06:09
andrzejku hi 07:11
smls Hi 07:48
dalek osystem: 7d2ff14 | (Shoichi Kaji)++ | META.list:
Add MetaCPAN::Favorite to ecosystem

See github.com/skaji/perl6-MetaCPAN-Favorite		 07:49
RabidGravy boom 08:46
skaji hi 09:44
m: my $*D; my $i = Supply.interval(1); $i.tap: { say $*D }; $i.wait;
camelia rakudo-moar 77724b: OUTPUT«Unhandled exception in code scheduled on thread 4␤Dynamic variable $*D not found␤ in block at <tmp> line 1␤␤»
skaji is this expected? 09:45
vimal2012 Is there a function to convert first letter of every word of a string to upper case? For example my $string="The quick brown fox" should be converted to "The Quick Brown Fox" 10:09
llfourn m: vimal2012 try .wordcase 10:12
camelia rakudo-moar 77724b: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Undeclared routine:␤ vimal2012 used at line 1␤␤»
llfourn m: "the quick brown fox".wordcase.say
camelia rakudo-moar 77724b: OUTPUT«The Quick Brown Fox␤»
llfourn vimal2012: ^
sena_kun . 10:25
llfourn skaji: That's a bug. see RT #127033. 10:33
synopsebot6 Link: rt.perl.org/rt3//Public/Bug/Displa...?id=127033
skaji llfourn: thanks! 10:35
llfourn you're welcome :) 10:36
timotimo .seen skarsnik 12:09
yoleaux I saw Skarsnik 6 Jul 2016 13:58Z in #perl6: <Skarsnik> RabidGravy, do you have an example of using it async? I'd like to have a Discord module that use the websocket api x)
timotimo :(
RabidGravy boom 12:10
masak how do I mark an exception as handled in a CATCH? 12:17
timotimo you successfully smartmatch against it, for example
masak what's the idiomatic way to say "if we get to this point, mark it as handled" 12:19
?
timotimo hm, i don't know of an idiomatic way yet 12:20
masak m: class X::Custom is Exception { method message { "OH NOES" } }; { die X::Custom.new; CATCH { say "in catch" } }
camelia rakudo-moar 77724b: OUTPUT«in catch␤OH NOES␤ in block <unit> at <tmp> line 1␤␤»
masak m: class X::Custom is Exception { method message { "OH NOES" } }; { die X::Custom.new; CATCH { say "in catch"; $_ ~~ Any } }
camelia rakudo-moar 77724b: OUTPUT«in catch␤OH NOES␤ in block <unit> at <tmp> line 1␤␤»
masak "successfully smartmatch" doesn't seem to be enough
timotimo hm. how does default { ... } do it, then?
masak m: class X::Custom is Exception { method message { "OH NOES" } }; { die X::Custom.new; CATCH { say "in catch"; default {} } }
camelia rakudo-moar 77724b: OUTPUT«in catch␤»
masak gonna go with `default {}` for now
timotimo oh 12:22
m: class X::Custom is Exception { method message { "OH NOES" } }; { die X::Custom.new; CATCH { say "in catch"; succeed } }
camelia rakudo-moar 77724b: OUTPUT«in catch␤succeed without when clause␤ in block at <tmp> line 1␤ in block <unit> at <tmp> line 1␤␤»
timotimo m: class X::Custom is Exception { method message { "OH NOES" } }; { die X::Custom.new; CATCH { say "in catch"; proceed } }
camelia rakudo-moar 77724b: OUTPUT«in catch␤proceed without when clause␤ in block at <tmp> line 1␤ in block <unit> at <tmp> line 1␤␤» 12:23
timotimo hmm.
masak yeah; those only make sense inside a `when` clause, I guess
timotimo the default { } does a succeed at the end, i expect that just skips over the rethrow we code-gen for the CATCH block
masak aye 12:24
timotimo neat, pypy got a flag for giving you a traceback for when the GC closes a file for you 12:46
so you can properly put "with" statements or close method calls where they need to go
i think we could easily build something similar
dalek ateverable: 10eee4e | (Daniel Green)++ | / (2 files):
Perl 6 rewrite of benchable
 13:40
brrt hmmm, how? do you want to record the code locations at every file open? 13:50
dalek c/revert-814-signature-test-option: 753a50d | Altai-man++ | / (18 files):
Revert "New pod config option: 'signature'"
 13:59
travis-ci Doc build passed. Altai-man 'Revert "New pod config option: 'signature'"' 14:19
travis-ci.org/perl6/doc/builds/150421053 github.com/perl6/doc/commit/753a50d41da0
dalek c: 753a50d | Altai-man++ | / (18 files):
Revert "New pod config option: 'signature'"
 14:21
c: 69198a3 | Altai-man++ | / (18 files):
Merge pull request #815 from perl6/revert-814-signature-test-option

Revert "New pod config option: 'signature'"
dalek c: ce5b9d9 | Altai-man++ | util/extract-examples.p6:
The examples extractor was updated to complete signatures without config option
 14:27
dalek c: 55b76fc | Altai-man++ | doc/Type/ (16 files):
We don't need `{}` after signature to make it correct anymore
 14:36
travis-ci Doc build failed. Altai-man 'We don't need `{}` after signature to make it correct anymore' 14:45
travis-ci.org/perl6/doc/builds/150425166 github.com/perl6/doc/compare/ce5b9...b76fc76cff
sena_kun m: floor(1.1) := 42; CATCH { default { put .^name, ': ', .Str } }; 15:26
camelia rakudo-moar 77724b: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Cannot use bind operator with this left-hand side␤at <tmp>:1␤------> 3floor(1.1) := 427⏏5; CATCH { default { put .^name, ': ', .S␤»
dalek c: 653b91c | Altai-man++ | doc/Language/functions.pod6:
Index "is assoc", closes github.com/perl6/doc/issues/817
 15:54
dalek c: 62e617f | Altai-man++ | doc/Type/ (9 files):
Now Backtrace, Bag, BagHash, Baggy, Blob, Block, Bind and two exceptions are compilable
 15:58
travis-ci Doc build passed. Altai-man 'Index "is assoc", closes github.com/perl6/doc/issues/817' 16:16
travis-ci.org/perl6/doc/builds/150435939 github.com/perl6/doc/compare/55b76...3b91ce3611
travis-ci Doc build passed. Altai-man 'Now Backtrace, Bag, BagHash, Baggy, Blob, Block, Bind and two exceptions are compilable' 16:19
travis-ci.org/perl6/doc/builds/150436431 github.com/perl6/doc/compare/653b9...e617f4c2d7
dalek c: 56acfc8 | (Wenzel P. P. Peppmeyer)++ | doc/Language/typesystem.pod6:
close C<>
 17:36
c: 2cc567c | (Wenzel P. P. Peppmeyer)++ | doc/Language/typesystem.pod6:
doc role arguments
c: 612cdb5 | (Wenzel P. P. Peppmeyer)++ | doc/ (2 files):
index type capture
dalek ateverable: 3c2feb5 | (Daniel Green)++ | Benchable.p6:
Fixed a typo and such
 17:47
grondilu was thinking of a way to test if an integer is a power of an other. Turns out log($a) %% log($b) works. Nice. 18:01
m: say log(9) %% log(3)
camelia rakudo-moar 77724b: OUTPUT«True␤»
grondilu m: say log(9) %% log(2)
camelia rakudo-moar 77724b: OUTPUT«False␤»
dalek c: 52cb8a5 | (Wenzel P. P. Peppmeyer)++ | doc/Language/typesystem.pod6:
close C<>
 18:10
c: 804011a | (Wenzel P. P. Peppmeyer)++ | doc/Language/typesystem.pod6:
remove left behind heading
c: 3e40128 | (Wenzel P. P. Peppmeyer)++ | doc/Language/typesystem.pod6:
add missing heading
c: 3624102 | (Wenzel P. P. Peppmeyer)++ | doc/Language/typesystem.pod6:
fix example
andrzejku Perl is friend, Perl is love 19:05
x)
labster m: say "\x[0e53]".succ eq "\x[0e53]" 19:50
camelia rakudo-moar 77724b: OUTPUT«True␤»
labster m: say "\x[0e53] ", uniname("\x[0e53]"); 19:51
camelia rakudo-moar 77724b: OUTPUT«๓ THAI DIGIT THREE␤»
masak labster: ...so three follows on three? 19:52
labster m: say "\x[0e54] ", uniname("\x[0e54]"); # Usually four follows three 19:53
camelia rakudo-moar 77724b: OUTPUT«๔ THAI DIGIT FOUR␤»
labster Also hi masak! 19:54
masak ahoj 19:55
moritz aloha! 19:59
labster Is moritz coming or going? 20:01
moritz or is he here to stay? who knows?
timotimo gekommen um zu bleiben. 20:02
moritz three weeks of vacation ahead \o/ 20:05
including some travel to Denmark
labster sounds great 20:07
awwaiid niec!
er, nice!
labster Nice isn't in Denmark, though. 20:09
awwaiid give or take
DrForr But Billund is :)
labster files thai digits rakudobug 20:10
moritz DrForr: indeed, we'll visit there as well 20:14
DrForr moritz: I don't know if they're still available, but you might want to follow the sculpture path into town; the first restaurant you find sells (or at least did sell) a LEGO rarity - The model of the LEGO museum is only available in Billund. 20:17
pmurias DrForr: is the model of the LEGO museum made of LEGOs? 20:18
DrForr Of course. I have 3 :) 20:19
grondilu was writing |@somearray and was wandering if we could create a sigil for "flat" arrays.
grondilu also wondered if it is possible to create sigils at runtime. 20:20
s/create/define/
timotimo with slangs, everything is possible
moritz though not at runtime
syntactic features can only be added at compile time
timotimo one man's run time is another's compile time 20:21
grondilu ok runtime was not the correct expression. How would you call a modification that is made by the program and not by the core?
"user code" or something? 20:22
sena_kun grondilu, you mean reflection?
moritz use-space code, yes
*user
grondilu "user space" indeed 20:23
brrt i've been thinking of macros for a bit again 20:28
guess that means i should try out 007
timotimo probably :) 20:29
killbill what I should change in the regex in /𝐀𝐁𝐂/ === /ABC/ to get True? ( former is Mathematical Alphanumeric Symbols block, latter is Latin one) 20:30
pmurias brrt: re finding unclosed filehandles, wouldn't recording the locations where they are opened be a resonable implementation? 20:36
brrt pmurias: reasonable, yes 20:37
timotiom's suggestion was that pypy gave full tracebacks
which means you have to maintain and keep alive full stack traces every time you open a file
just for debugging
seems... expensive 20:38
timotimo yeah, for pypy the flag is optional
we can have this as a module that wraps the constructor of IO::File :P 20:39
we don't need support in the core or backend or anything
user code can do that
brrt hmm 20:40
that
makes sense
actually, do we have destructors?
timotimo yes, we do have submethod DESTROY
brrt aha
timotimo we have that in Failure already to warn when Failure objects get destroyed before they get handled
which usually means a .sink is missing or something 20:41
brrt but i suppose IO::File already has a DESTROY? can we augment that and still call the original implementation?
i recall not having destructors some eons ago 20:42
timotimo we can probably wrap it
yeah, we used to not have them
and now they're sot guaranteed to run, even
moritz ok, weird idea 20:46
have a separate memory pool for objects of classes with DESTROY methods
then, before interpreter exit, iterate over them all, and call of their DESTROY methods (without doing a full GC run)
brrt hmmm 20:49
don't we already have a full-clean exit? 20:50
mode
also, why a separate pool? that means you have to know at memory allocation time
meaning you have a problem if a DESTROY role is ever mixed in 20:51
i'm not 100% sure that is true, though
gfldex m: class B {...}; class A { trusts B; has $!foo }; class B { submethod BUILD { my $an_A = A.new(); $an_A!A::foo = 42 } }; dd B.new; 20:52
camelia rakudo-moar 77724b: OUTPUT«No such private method '!foo' for invocant of type 'A'␤ in submethod BUILD at <tmp> line 1␤ in block <unit> at <tmp> line 1␤␤»
timotimo i don't think trusts allows you to get at private attributes, only methods? 20:56
gfldex running S12-attributes/trusts.t fails, so that may NYI 20:57
timotimo mhh
gfldex m: class B {...}; class A { trusts B; has $!foo = 42; method !foo { return-rw $!foo }; method perl { "A.new(foo => $!foo)" } }; class B { has A $!a .= new; method change { dd $!a; my $an_A = $!a; $an_A!A::foo } }; dd B.new.change; 20:59
camelia rakudo-moar 77724b: OUTPUT«A $!a = A.new(foo => 42)␤42␤»
timotimo try method !foo is rw { $!foo } instead 21:00
i think return-rw without "is rw" is trouble
(and we should probably do something about that)
El_Che I am looking into porting a perl5 module I use often (5 minutes here, 5 minutes there), and all the perl6 enhancements make it impossible toeven try a 1:1 port 21:02
that bad and great at the same time :)
timotimo aaw
DrForr What module? 21:03
El_Che perl-ldap if you keep it quiet. I have new method and some attributes :) 21:04
DrForr Oh, right, should've guessed.
El_Che eg., one of the modules accepts potitional parameters for new, but that pretty non-perl6 style 21:05
El_Che DrForr: yeah, I wrote the program I needed using the perl5 modules :) 21:05
timotimo i don't have a problem with positional arguments to .new
i've biult a few things with that kind of interface (optionally, usually)
DrForr Incidentally Perl6::Tidy roundtrips all basic operators and declarations (epsilon whitespace) 21:06
El_Che timotimo: I am very fond of the 'has' way of doing thing of Moose and p6
dalek c: f96ce35 | (Wenzel P. P. Peppmeyer)++ | doc/Language/typesystem.pod6:
doc trusts
timotimo fair enough 21:07
El_Che an other example, there is a method that accepts one parameter as the subject (dn) and the second as a values (str) or values (arrayref). Could be done with a multimethod, but varargs seems also a good alternative 21:09
DrForr You can override creation with a custom new() method if you need to, but ... Is the first 'values' a string with a comma or otherwise delimited list, vs. an aray of single items? 21:13
El_Che the first is "$entry->add ( 'sn' => 'Barr' )", the second " $entry->add ( 'street' => [ '1 some road','nowhere' ] );" 21:18
the fat comma is justa comma ofcourse, so I was thinking of "method add (Str $attribute, *@values) {...}"
DrForr Oh, I wouldn't separate them in that case. 21:19
El_Che indeed
dalek c: 8e4234a | (Wenzel P. P. Peppmeyer)++ | util/extract-examples.p6:
shield better against redeclarations
 21:30
c: 8cac1b2 | (Wenzel P. P. Peppmeyer)++ | doc/Language/typesystem.pod6:
make examples compile
timotimo $entry{'street'} = [ '1 some road', 'nowhere' ]; # how about this? :) 21:31
El_Che timotimo: that's why I had at first. I am looking how it's used in the rest of the codebase 21:33
El_Che I am correcting this (docs.perl6.org/language/5to6-nutsh...-constant) to align it with this (docs.perl6.org/language/terms#inde...Terms%29). I am removing the sigil of the perl6 constant in the perl5to6 doc. Question: The perl5to6 doc uses all capitals like in perl5 while the Term doc capitalizes only the first letter. What's common practive for 21:47
El_Che constants in perl6. I'll like the doc to be consistent 21:47
dalek c: a79f40d | (Wenzel P. P. Peppmeyer)++ | doc/Language/typesystem.pod6:
remove TODO
 21:50
lizmat El_Che: not sure: internally both ALLCAPS as well as CamelCase is used (e.g. IterationEnd vs REQUIRE_DYNAMIC) 21:51
dalek c: 3c165ca | (Claudio Ramirez)++ | doc/Language/5to6-nutshell.pod6:
Remove sigil from perl6 constant in accordance with the Term doc
c: 8ed43c4 | nxadm++ | doc/Language/5to6-nutshell.pod6:
Merge pull request #820 from nxadm/master

Remove sigil from perl6 constant in accordance with the Term doc
El_Che lizmat: I'll replace the example of docs.perl6.org/language/terms#Constants as it a confusing one. It defines a Pi constant while there is one in core 21:55
lizmat El_Che++
gfldex sena_kun: counting the examples that still fail: time find examples/ -type f -exec nohup perl6 '{}' '2>/dev/null 1>/dev/null' ';' -or -print 2>/dev/null | wc -l 22:02
dalek c: 0397c1f | (Claudio Ramirez)++ | doc/Language/terms.pod6:
Choose beter constant examples (pi is alreay a builtin term)
c: 2e26deb | nxadm++ | doc/Language/terms.pod6:
Merge pull request #821 from nxadm/master

Choose beter constant examples (pi is alreay a builtin term)
BenGoldberg m: (1+sqrt 5)/2.say 22:06
camelia rakudo-moar 77724b: OUTPUT«WARNINGS for <tmp>:␤Useless use of "/" in expression "(1+sqrt 5)/2.say" in sink context (line 1)␤2␤»
BenGoldberg m: ((1+sqrt 5)/2).say
camelia rakudo-moar 77724b: OUTPUT«1.61803398874989␤»
BenGoldberg m: ((1+sqrt 5)/2).Rat.say 22:07
camelia rakudo-moar 77724b: OUTPUT«1.618034␤»
BenGoldberg m: ((1+sqrt 5)/2).Rat.perl.say
camelia rakudo-moar 77724b: OUTPUT«<1597/987>␤»
BenGoldberg m: my $phi = ((1+sqrt 5)/2); ($phi - $phi.Rat).say 22:07
camelia rakudo-moar 77724b: OUTPUT«-4.59071787028975e-07␤»
BenGoldberg m: my $phi = ((1+sqrt 5)/2); ($phi - 8944394323791464/5527939700884757).say 22:11
camelia rakudo-moar c8c27e: OUTPUT«0␤»
BenGoldberg wonders how we presently convert Num to Rat. 22:12
Surely it can't be that hard, to make .Rat.Num round trip precisely. 22:16
lizmat BenGoldberg: feels to me that would require mixing in the original Rat into the Num 22:20
sena_kun gfldex, I'm fixing it in an alphabet fashion, but thanks. Much of them were fixed by signature detection, by the way.
BenGoldberg m: sub ratify(Num $n is copy) { my $s = sign($n) or return 0.Rat; $n .= abs; my $denom = 2**(53-log( $n, 2 ).Int); $s * ($n*$denom).Int / $denom }; say ratify(pi) - pi; 22:30
camelia rakudo-moar c8c27e: OUTPUT«0␤»
BenGoldberg m: sub ratify(Num $n is copy) { my $s = sign($n) or return 0.Rat; $n .= abs; my $denom = 2**(53-log( $n, 2 ).Int); $s * ($n*$denom).Int / $denom }; say ratify(pi).perl;
camelia rakudo-moar c8c27e: OUTPUT«3.141592653589793115997963468544185161590576171875␤»
BenGoldberg m: sub ratify(Num $n is copy) { my $s = sign($n) or return 0.Rat; $n .= abs; my $denom = 2**(53-log( $n, 2 ).Int); $s * ($n*$denom).Int / $denom }; say ratify(pi).nude;
camelia rakudo-moar c8c27e: OUTPUT«(884279719003555 281474976710656)␤»
dalek c: fca225c | (Wenzel P. P. Peppmeyer)++ | doc/Language/containers.pod6:
make examples compile
 22:35
BenGoldberg m: sub ratify(Num $n is copy) { my $s = sign($n) or return 0.Rat; $n .= abs; my $denom = 2**(53-log( $n, 2 ).Int); $s * ($n*$denom).Int / $denom }; say ratify(pi) - pi.Rat;; 22:36
camelia rakudo-moar c8c27e: OUTPUT«-0.000000266764189185␤»
BenGoldberg m: pi.perl.say; 22:37
camelia rakudo-moar c8c27e: OUTPUT«3.14159265358979e0␤»
dalek c: c4e3f84 | (Wenzel P. P. Peppmeyer)++ | doc/Language/contributors.pod6:
make examples compile
 22:48
skids What do we use to pass a NULL to a nativecall "int32 $foo is rw"? "int32" does not work. 23:00
"Native call expected argument that references a native integer, but got P6int" 23:01
geekosaur Pointer[int32]? 23:05
sena_kun skids, I haven't used nativecall for a while, but maybe "Pointer[Int32]"? Since NULL is just a pointer. It works in my code, at least. See github.com/Altai-man/perl6-Compres...Raw.pm6#L7
timotimo can't have Int32, just int32 23:06
sena_kun timotimo, ah, yes, it's a typo.
skids I did already try Pointer[int32] but that gives same error message, just s/P6int/CPointer/ 23:07
sena_kun And "uint32" to be precise.
sena_kun skids, do you use it as a constant or what? Can you give a piece of the actual code, please? 23:08
skids tries to find a C stdlib with an int *
timotimo it's no longer CPointer 23:09
i think instead of Pointer.new(0) you should just pass the type object
skids right I use "Pointer[int32]", rakudo error message is "but got CPointer"
sena_kun skids, perl6 -v? 23:10
skids This is Rakudo version 2016.07.1-74-g4be4609 built on MoarVM version 2016.07-6-gfd7fe64
timotimo what's the "wanted" part before "but got CPointer"? 23:11
skids "Native call expected argument that references a native integer, but got CPointer'
timotimo oh
in that case it may be enough to use "int32 $foo is rw" in your method declaration
sena_kun skids, ah, okay. You can try to do what timotimo said then. He was too quick for me. (: 23:12
timotimo and when you have that, you just pass a "my int $foo" directly 23:13
i'm going to bed, i'm really tired
have a good one!
skids that's what I have:
sub xcb_connect (str $display, int32 $screen is rw --> xcb_connection_t)
sena_kun timotimo, good night!
skids, in a repl it seems good. Without return type. Something like `sub xcb_connect(Str $dpn, int32 $screen is rw) is native("xcb", v1) { * };`. Was your problem solved? 23:19
skids no, the problem happens durin the call. 23:21
sena_kun Ah, it was expected. 23:22
Can you provide a minimal example gist to reproduce?
skids m: sub xcb_connect(Str $dpn, int32 $screen is rw) is native("xcb", v1) { * }; xcb_connect(Str, Pointer[int32); 23:23
camelia rakudo-moar c8c27e: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Can't use unknown trait 'is native' in a sub declaration.␤at <tmp>:1␤ expecting any of:␤ rw raw hidden-from-backtrace hidden-from-USAGE␤ pure default DEPRECATED inlinable nodal␤ …»
skids m: use NativeCall; sub xcb_connect(Str $dpn, int32 $screen is rw) is native("xcb", v1) { * }; xcb_connect(Str, Pointer[int32); 23:24
camelia rakudo-moar c8c27e: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Unable to parse expression in type parameter; couldn't find final ']' ␤at <tmp>:1␤------> 031) { * }; xcb_connect(Str, Pointer[int327⏏5);␤»
skids m: use NativeCall; sub xcb_connect(Str $dpn, int32 $screen is rw) is native("xcb", v1) { * }; xcb_connect(Str, Pointer[int32]);
camelia rakudo-moar c8c27e: OUTPUT«Native call expected argument that references a native integer, but got CPointer␤ in method CALL-ME at /home/camelia/rakudo-m-inst-1/share/perl6/sources/24DD121B5B4774C04A7084827BFAD92199756E03 (NativeCall) line 328␤ in block <unit> at <tmp> line 1…»
skids there we go
skids builds fresh rakodu JIC.
sena_kun m: use NativeCall; sub xcb_connect(Str $dpn, int32 $screen is rw) is native("xcb", v1) { * }; my int32 $n = 0; xcb_connect("name", $n); 23:27
camelia ( no output )
sena_kun m: use NativeCall; sub xcb_connect(Str $dpn, int32 $screen is rw) is native("xcb", v1) { * }; my int32 $n = 0; say xcb_connect("name", $n); 23:28
camelia rakudo-moar c8c27e: OUTPUT«(Mu)␤»
skids star: use NativeCall; sub xcb_connect(Str $dpn, int32 $screen is rw --> Pointer) is native("xcb") { * }; xcb_connect("0:0",uint32);
camelia star-m 2016.04: OUTPUT«Native call expected argument that references a native integer, but got P6int␤ in method CALL-ME at sources/24DD121B5B4774C04A7084827BFAD92199756E03 (NativeCall) line 301␤ in block <unit> at <tmp> line 1␤␤»
skids star: use NativeCall; sub xcb_connect(Str $dpn, int32 $screen is rw --> Pointer) is native("xcb") { * }; xcb_connect("0:0",Pointer[int32]);
camelia star-m 2016.04: OUTPUT«Native call expected argument that references a native integer, but got CPointer␤ in method CALL-ME at sources/24DD121B5B4774C04A7084827BFAD92199756E03 (NativeCall) line 301␤ in block <unit> at <tmp> line 1␤␤»
sena_kun In the examples here - xcb.freedesktop.org/tutorial/ they used null in both arguments, but C rules of types are loose.
geekosaur I think this si a known shortcoming of NativeCall 23:29
skids is it RTd?
sena_kun Am I the only one who didn't get it? You can't pass a type instead of a value, can you? Or I misunderstood something. 23:31
skids Passing the type is how you get a C NULL.
sena_kun skids, ah. 23:32
geekosaur I don't see it in RT. I'm sure I've seen this discussed before
sena_kun It seems I need to refresh my own knowledge of how current nativecall works before helping someone. (:
TimToady .tell killbill Try something like this: say '𝐀𝐁𝐂'.NFKD.Str === 'ABC' 23:39
yoleaux TimToady: I'll pass your message to killbill.

Powered by Cro and the Raku® Programming Language.

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

Thank you!