»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.
dpk p6: say 'xß' ~~ rx:i/XSS|YS/ 00:01
camelia Nil
dpk p6: say 'xss' ~~ rx:i/XSS|YS/ 00:02
camelia ｢xss｣
dpk p6: say 'xß' ~~ rx:i/Xß|YS/
camelia ｢xß｣
dpk p6: say 'xss' ~~ rx:i/Xß|YS/
camelia Nil
dpk ⬑ this seems like a bug
dpk though beware there is also the more complicated case (from an implementation perspective) of rx:i/[XS|Y]S/ 00:03
none of the regexp engines i've tried has got that one right, and only Ruby (oniguruma) has got the previous examples right too 00:04
dpk (i've tried re2, pcre, perl 5, and egrep) 00:05
lookatme o/ 00:21
tyil \o 00:23
dpk okay, even ICU's regexp engine does not bother to try to match (?i:(XS|Y)S) against xß 'correctly' 00:41
so i think that case can safely be ignored
lookatme I am curious if I collect the everything(include lib, precomp) moar needed, is it possible run it in another pc. 00:58
ryn1x When uploading to CPAN on PAUSE do I need to put Perl6 in the 'Target Directory' field? 01:17
ZzZombo m: say "\n" ~~ /<ws>/ 01:58
camelia ｢
｣
ws => ｢
｣
ZzZombo Eh, why does the regex doc page omit info for `<ww>` despite using the negated form in 4 places towards the end of it? 02:03
What does exactly it do?
buggable New CPAN upload: Terminal-Spinners-0.0.3.tar.gz by RYNIX cpan.metacpan.org/authors/id/R/RY/...0.3.tar.gz
geekosaur ZzZombo, it's in there. look for 'Predefined subrules:' 02:05
geekosaur notes a typo, reading on therefrom... 02:06
ZzZombo Why is it written with the '?'?? It makes no sense, and made Ctrl+F to fail. 02:08
Also, shouldn't the dot be used to suppress capture?
geekosaur that is explained in the description 02:10
*any* symbol character prevents capture; '?' is used because it implies 'this is an assertion that matches a *state*, as opposed to something that actually matches one or more characters' 02:11
lookatme m: say "123" ~~ /<.ww>/; say "123" ~~ /<?ww>/; say "123" ~~ /<ww>/; 02:12
camelia ｢｣
｢｣
｢｣
ww => ｢｣
geekosaur whereas <.ww> would suggest that <ww> matches some character or characters
lookatme What's the difference `.` and `?`
geekosaur in this case they;re the same from a perl 6 standpoint, it's consistency for the reader that matteers
ok, what I said above was mumblenonsense
lemme see if I can figure out a different way to say it
perl 6 actually does not care what symbol you use to say 'this is non-capturing'. (in some specific circumstances using ! has meaning.) 02:13
lookatme I found ` <alpha> # match a letter and capture to $alpha (eventually $<alpha>) 02:14
<.alpha> # match a letter, don't capture
<?alpha> # match null before a letter, don't capture` in design doc
geekosaur from a *human* standpoint, using '.' is intended to say 'this matches some text, and would give it to us but we told it not to'
from a *human* standpoint, using '?' means 'this matches a particular state instead of a particular set of characters'
so like <?before foo> means "match the state of being just before 'foo'", as opposed to "match 'foo'" 02:15
geekosaur <ww> is an assertion that you are between word characters. it doesn't match and wouldn't capture any actual characters; it is a state. as such, it doesn't make much sense normally to use it in default capturing mode, since the capture would aways be empty 02:17
geekosaur *but* in some contexts (like building regexes programmatically) it is useful for all such things to behave consistently, so you don't need to have special cases "this <xxx> captures, but this <xxy> doesn't" 02:18
so all rules behave the same way: they add *something* to the list of captuires, unless they start with a symbol in which case they do not. 02:19
so, if you say <ww> in a regex, if it matches it doesn't match *haracters*, but it still records an empty capture for consistency 02:20
lookatme m: say "fp" ~~ /<alpha><alpha>/; say "fp" ~~ /<.alpha><alpha>/; say "fp" ~~ /<?alpha><alpha>/; # this make sense
camelia ｢fp｣
alpha => ｢f｣
alpha => ｢p｣
｢fp｣
alpha => ｢p｣
｢f｣
alpha => ｢f｣
geekosaur and . and ? don't actually mean anything different. but for human readers . implies it would capture characters but ? implies it's about a certain condition being true instead of about what character(s) it would match 02:21
it's just convention
lookatme Hmm 02:21
ZzZombo But the question what exactly constitutes "within word" still stands. 02:25
geekosaur that's defined a little before that in \w 02:28
"matches a single word character; i.e., a letter (Unicode category L), a digit or an underscore."
geekosaur so <?ww> matches between two things that match \w, and <?wb> matches between two things where one matches \w and the other matches \W 02:29
lookatme m: say "123abc" ~~ /\d+<?ww>\w+/ 02:30
camelia ｢123abc｣
ZzZombo thanks 02:31
lookatme m: say "123 abc" ~~ /<?wb>\w+/
camelia ｢123｣
geekosaur huh, maybe it's smarter than p5's, that would be nice. (one thing that always annoyed me about the p5 version was I had to separately handle the "start/end of line" case( 02:32
ok, filed doc bug :) 02:33
geekosaur (unrelated, I just noticed it while ooking at the doc) 02:33
that said, the doc wshould probably be clearer about what those chars mean if it's not already stated somewhere 02:34
the speculation was clear about it
dmaestro m: use Inline::Perl5 03:09
camelia ( no output )
dmaestro m: use Inline::Perl5 ; my $a = EVAL "[ 1, 2 ]", :lang<Perl5>; say $a.elems 03:10
camelia 2
dmaestro m: use Inline::Perl5 ; my $a = EVAL "[ ]", :lang<Perl5>; say $a.elems 03:11
camelia 0
dmaestro Interesting, I don't get the above result (0 elements for an empty Inline::Perl5::Array) on my Mac running Rakudo Star 2017.10 03:13
dmaestro > my $a = EVAL "[]", :lang<Perl5> ; say $a.elems 03:14
4294967296
Inline::Perl5:ver<0.29>:auth<github:niner> 03:19
geekosaur looks like a fixed bug to me
dmaestro How recently? version? 03:20
Is there a way to ask camelia what version of a module is loaded? 03:24
geekosaur I think there is b ut it's not easy and I'm not sure of details. means messing with CompUnitRepos 03:28
dmaestro ok, thanks. I haven't found a reference to that bug in Inline::Perl5 - could it have been a rakudo bug on Mac? 03:30
geekosaur also possible, yes 03:34
it could even be a perl 5 bug
(or worse, not a bug but somehow a mismatch because perl 5 got replaced/upgraded/it's running the wrong one) 03:35
dmaestro Well, turns out an attempted re-install of Inline::Perl5 failed because of too-new version of Module::Runtime - I had the latest version in my local::lib. 03:53
I removed the local::lib version so it would fall back on the system installed module. Inline::Perl5 reinstalled OK and the bug disappeared. 03:54
geekosaur there you go then. probably mismatched versioning, then; different perl 5 releases have subtly different internals and macros like (iirc) PvAV will not necessarily work right against a different perl library 03:57
dmaestro I removed the local::lib version so it would fall back on the system installed module. Inline::Perl5 reinstalled OK and the bug disappeared. 04:02
Nope, only seemed to disappear - the correct result is given the first time I run it! Subsequent attempts fail. 04:03
geekosaur wow. that sounds like a precomp issue 04:03
.tell nine colabti.org/irclogger/irclogger_log...11-24#l119 possible precomp bug? (cf colabti.org/irclogger/irclogger_log...1-24#l156) 04:04
yoleaux geekosaur: I'll pass your message to nine.
dmaestro m: use Inline::Perl5; my $a = EVAL "[]", :lang<Perl5>; $b = EVAL "[]", :lang<Perl5>; say [ $a.elems, $b.elems ] 04:05
camelia 5===SORRY!5=== Error while compiling <tmp>
Variable '$b' is not declared
at <tmp>:1
------> 3Perl5; my $a = EVAL "[]", :lang<Perl5>; 7⏏5$b = EVAL "[]", :lang<Perl5>; say [ $a.e
dmaestro m: use Inline::Perl5; my $a = EVAL "[]", :lang<Perl5>; my $b = EVAL "[]", :lang<Perl5>; say [ $a.elems, $b.elems ]
camelia [0 0]
dmaestro doug$ perl6 -e 'use Inline::Perl5; my $a = EVAL "[]", :lang<Perl5>; my $b = EVAL "[]", :lang<Perl5>; say [ $a.elems, $b.elems ]' --> [0 4294967296] 04:07
geekosaur hm, that's not even precomp, it's just weird. 04:13
well, left message for nine, maybe they can decipher it
dmaestro Posted details in github.com/niner/Inline-Perl5/issues/106 04:22
ZzZombo m: say 'asd' ~~ /(.*)/ 04:35
camelia ｢asd｣
0 => ｢asd｣
ZzZombo m: say 'asd' ~~ /(.)*/
camelia ｢asd｣
0 => ｢a｣
0 => ｢s｣
0 => ｢d｣
ZzZombo Why does this keep on matching indefinitely: 06:29
token string { \h* [<header> || <key-value>]? <comment>? \h* [\n || $] }
on the last line of my file, which is empty.
m: say '' ~~ /$/ 06:31
camelia ｢｣
ZzZombo m: my token a { \h* [\n || $] };say '' ~~ &a 06:42
camelia ｢｣
ZzZombo what
Geth doc: 523f616499 | (Alex Chen)++ (committed using GitHub Web editor) | doc/Language/regexes.pod6
Fix typo, geekosaur ++

close #1683		 06:51
synopsebot Link: doc.perl6.org/language/regexes
ZzZombo m: my token a { (\h* [\n || $])* };say '' ~~ &a 07:01
yeah, this works to show my issue.
camelia (timeout)
ZzZombo ^
ZzZombo m: my token a { (\h* [\n || $])+ };say '' ~~ &a 07:10
camelia (timeout) 07:11
geekosaur the expression can match nothing, and can do so repeatedly 07:11
geekosaur and will do so at the end of input because the $ will always match there and match empty string for \h* 07:12
lookatme m: say "" ~~ /(<?>)/ 07:13
camelia ｢｣
0 => ｢｣
lookatme m: say "" ~~ /(<?>)*/ # will timeout
camelia (timeout)
geekosaur (potentially it's worse than that because iirc $ matches before newline whereas $$ means actual end of input, but the alternation with \n catches that screw case)
lookatme m: say "" ~~ /(<?>)**1..3/ # will timeout 07:14
camelia ｢｣
0 => ｢｣
0 => ｢｣
0 => ｢｣
lookatme Interesting :)
geekosaur basically anything that can match zero characters and succeed, that is then repeatable with no stop case (that is, nothing following that can match), will always loop forever matching end of input
and yes, that too, if instead of * or + you give it a range, it'll match the maximum number of times 07:15
ZzZombo so, what can I do to avoid that? Keep in mind that for my purposes, empty or consisting of only whitespace documents are valid. 07:16
geekosaur I would generally make sure the action tokens always match something, and write the grammar with whitespace tokens in positions that don't get to repeat forever. (this means for example letting the TOP token end with a single, nonrepeated, optional whitespace match --- and, again, the whitespace rule should always match *something*) 07:19
it's rather difficult to make grammar engines or parser generators handle this for you; the halting problem will always find some way to bite you 07:20
buggable New CPAN upload: Terminal-Spinners-0.0.4.tar.gz by RYNIX cpan.metacpan.org/authors/id/R/RY/...0.4.tar.gz 07:43
Geth ecosystem: e28a057086 | (Jarkko Haapalainen)++ (committed using GitHub Web editor) | META.list
App::Platform moved to CPAN
 08:30
Geth mu: 97bba7fbcd | (Moritz Lenz)++ | misc/perl6advent-2017/schedule
p6advent: claim day 8
 09:46
Geth doc: gfldex++ created pull request #1686:
be precise when to early is to early for constant		 10:36
ZzZombo m: my $b=Str;my $a is default(Int)=$b;say $a.new; 11:16
camelia
ZzZombo m: my $b=Nil;my $a is default(Int)=$b;say $a.new;
camelia Any.new
ZzZombo what is this? 11:17
I expect the second to be `Int.new` in effect. 11:18
nvm
m: my %b;my $a is default(Int)=%b<asd>;say $a.new; 11:19
camelia Any.new
ZzZombo hm
this is fucked up
how then can I pass a TYPE as a parameter, and have a default value for it at the same time? Obviously I can't do `my $class=$arg // $default` for this. 11:21
El_Che ZzZombo: you should look into mult 11:35
multi
jnthn m: sub foo($x = 'default value') { dd $x }; foo 42; foo Int; foo; 11:37
camelia Int $x = 42
Int $x = Int
Str $x = "default value"
gfldex m: my $b := Nil; my $a is default(Int) = $b; say $a.new; 12:07
camelia 0
gfldex m: say Int.new;
camelia 0
gfldex ZzZombo: ^^^ 12:08
ZzZombo I do not want a multi. I get the parameter from a hash.
gfldex, thanks. 12:09
m: my %b;my $c:=%b<asd>;my $a is default(Int)=$c;say $a.new;
camelia Any.new
ZzZombo ? 12:10
gfldex ZzZombo: `is default` reacts only to Nil, not just any undefined value. And Hash will autovivify to Any. 12:12
m: my %b of Nil; my $c:=%b<asd>; my $a is default(Int)=$c; say $a.new; 12:13
camelia 0
gfldex ZzZombo: ^^^ 12:13
ZzZombo nice
jnthn m: my %b of Nil; %b<x> = 42 12:14
camelia Type check failed in assignment to %b; expected Nil but got Int (42)
in block <unit> at <tmp> line 1
jnthn Only so nice ;)
ZzZombo oh
m: my %b;my $c:=%b<asd> // Nil;my $a is default(Int)=$c;say $a.new;
camelia 0
ZzZombo this
jnthn m: my %b is default(Nil); my $c:=%b<asd>; my $a is default(Int)=$c; say $a.new;
camelia 0
jnthn Also this
jnthn lunch & 12:20
timotimo i'd use :exists to see if something's in there rather than relying on a specific value being returned by accessing 12:26
Altreus Why doesn't the if version require a block? docs.perl6.org/language/py-nutshel...rehensions 13:25
teatime I think in perl5 at least that form of if was described as a "statement modifier" 13:28
Zoffix Altreus: because those are statement modifiers and none of them require a block. I think the writer of that section simply didn't realize `for` doesn't need it either 13:33
raschipi m: .say for 1..4
camelia 1
2
3
4
Zoffix What happened with ugexe? I got a flood of notifications about closed PRs from him this morning 13:34
m: sub foo { $^a * 2 }; say (&foo for ^3) 13:36
camelia (sub foo ($a) { #`(Sub|61255088) ... } sub foo ($a) { #`(Sub|61255088) ... } sub foo ($a) { #`(Sub|61255088) ... })
Zoffix m: say (-> $ {42} for ^3)
camelia (42 42 42)
Zoffix I would've expected the block form to behave same as the sub form above. Basically the same as a block at the end of a routine (it'd get returned as a block, unless it's a bareblock 13:37
Geth doc: 403d0fea02 | (Zoffix Znet)++ (committed using GitHub Web editor) | doc/Language/py-nutshell.pod6
Tweak list comprehension examples

  - These are blocks, not anonymous functions
  - Show a block-less `for` example irclog.perlgeek.de/perl6/2017-11-24#i_15489899		 13:39
synopsebot Link: doc.perl6.org/language/py-nutshell
Zoffix ) 13:39
MasterDuke Zoffix: re ugexe, i would guess it's related to irclog.perlgeek.de/perl6/2017-11-23#i_15486662 and the subsequent conversation
Altreus Zoffix: we tested it, and it complains about sink context
well, without the arrow
m: ( $_ + 1 for 1,2,3 ) 13:40
camelia WARNINGS for <tmp>:
Useless use of "+" in expression "$_ + 1" in sink context (line 1)
Altreus like this
whereas
m: ( $_ + 1 if $_ > 1 for 1,2,3 )
camelia WARNINGS for <tmp>:
Useless use of "+" in expression "$_ + 1" in sink context (line 1)
Altreus oh that's new
on our rakudo that didn't complain
Zoffix m: say ( $_ + 1 for 1,2,3 ) 13:41
camelia (2 3 4)
Altreus oh it's because I didn't store it in an array isn't it
m: my @saddfga = ( $_ + 1 for 1,2,3 )
camelia ( no output )
raschipi Yeah, it detected you're just throwing the value away and warned you of that.
m: ( quietly $_ + 1 for 1,2,3 ) 13:42
camelia WARNINGS for <tmp>:
Useless use of "+" in expression "$_ + 1" in sink context (line 1)
raschipi m: ( quietly ($_ + 1) for 1,2,3 )
camelia WARNINGS for <tmp>:
Useless use of "+" in expression "$_ + 1" in sink context (line 1)
Altreus so it's a feature of the REPL, I guess, that it complains about sink context and then echoes the result anyway?
raschipi REPL is weird 13:43
Zoffix It won't work. `quietly` is for silencing CX::Warn and those ^ I think come from optimizer complaining
Altreus: doesn't do it for me. What code are you using?
raschipi It didn't complain in the REPl for me either 13:44
Zoffix It's possible you're using a very old Rakudo, Altreus
raschipi Oh, sorry. It did
Altreus yes I think it's just a version thing :)
raschipi > ( $_ + 1 for 1,2,3 )␤ WARNINGS for <unknown file>:␤ Useless use of "+" in expression "$_ + 1" in sink context (line 1) 13:45
Zoffix That shows for me too. `( $_ + 1 if $_ > 1 for 1,2,3 )` doesn't show it tho. Weird. 13:46
Altreus aha, must be the REPL and not the eval bot 13:47
raschipi camelia doesn't use the REPL
Zoffix MasterDuke: thought so. Thanks. 13:48
Zoffix &
Geth doc: 80d2f04df9 | (Zoffix Znet)++ (committed using GitHub Web editor) | doc/Language/py-nutshell.pod6
Make references to routines actual links to their docs
 13:49
synopsebot Link: doc.perl6.org/language/py-nutshell
Altreus does perl6 have any built in partial application for functions? 14:03
moritz Altreus: yes, see method "assuming"
and for operators there's a syntax 14:04
m: say $plus-two = * + 2; say $plus-two(5)
camelia 5===SORRY!5=== Error while compiling <tmp>
Variable '$plus-two' is not declared
at <tmp>:1
------> 3say 7⏏5$plus-two = * + 2; say $plus-two(5)
moritz m: my $plus-two = * + 2; say $plus-two(5)
camelia 7
Altreus yo 14:05
imma look this up
can't catch this language out can I?
moritz well, you can ask for homocionicy or whatever it's called :) 14:06
timotimo homoiconicity i think 14:06
Altreus well I'd never heard of it to ask 14:08
Altreus now you've ruined it ;) 14:08
raschipi That is what Python does, using whitespace to represent syntax. 14:09
lucs Am I supposed to keep synchronized the version number of my module both in the .pm6 file (unit module Foo:ver<1.1>) and in the META6.json file ("version" : "1.0",)? (whoops! they're different!) 14:10
lucs s/to keep/to manually keep/ 14:11
pmurias raschipi: nope, homoiconicity means the program text has the same structure as the program AST 14:15
yoleaux 23 Nov 2017 00:26Z <samcv> pmurias: that is a little bit complex. simply see this table unicode.org/reports/tr10/#Comparis...able_Table the full definition is complex, but non-ignorable just means nothing is ignored except the baseline levels of the UCD spec
SmokeMachine m: react whenever Supply.zip(Supply.interval(1), Supply.from-list(^10)) {.say} 14:17
camelia (timeout)(0 0) 14:18
SmokeMachine m: react { whenever Supply.zip(Supply.interval(1), Supply.from-list(^10)) {.say}; whenever Promise.in: 11 {say "timed out"; done} 14:19
camelia 5===SORRY!5=== Error while compiling <tmp>
Missing block
at <tmp>:1
------> 3r Promise.in: 11 {say "timed out"; done}7⏏5<EOL>
SmokeMachine m: react { whenever Supply.zip(Supply.interval(1), Supply.from-list(^10)) {.say}; whenever Promise.in: 11 {say "timed out"; done}} 14:20
camelia (0 0)
(1 1)
(2 2)
(3 3)
(4 4)
(5 5)
(6 6)
(7 7)
(8 8)
(9 9)
timed out
SmokeMachine m: react { whenever Supply.zip(Supply.interval(1), Supply.from-list(^10)) {.say}; whenever Promise.in: 15 {say "timed out"; done}} 14:21
camelia (0 0)
(1 1)
(2 2)
(3 3)
(4 4)
(5 5)
(6 6)
(7 7)
(8 8)
(9 9)
timed out
SmokeMachine m: react { whenever Supply.zip(Supply.interval(1), Supply.from-list(^2)) {.say}; whenever Promise.in: 15 {say "timed out"; done}} 14:22
camelia (0 0)
(1 1)
timed out
jnthn If you're trying to get it to not bother wiht the timed out thing once the zip is over, then add ;LAST done after the .say 14:23
*with
raschipi pmurias: Like in Lisp? 14:24
pmurias raschipi: yes 14:30
SmokeMachine jnthn: I think it's different from the docs... it says that the resulted supply will be done when any of the supplies be done... 14:31
m: react { whenever Supply.zip(Supply.interval(.1), Supply.from-list(^10)) {.say}}
camelia (timeout)(0 0)
jnthn Hm, interesting 14:35
SmokeMachine I think its will only be done when every supply is done: github.com/rakudo/rakudo/blob/mast...y.pm#L1224
jnthn Yup, that's what the impl reads like to me also 14:36
SmokeMachine s/its/it/
jnthn Wonder what roast things
*thinks
SmokeMachine docs.perl6.org/type/Supply#method_zip
jnthn Yeah, agree the impl isn't in line with the docs 14:41
Also, what the docs want is surely the more useful thing
SmokeMachine it does not test with live/on-demand supplies... github.com/perl6/roast/blob/d1baf2...pply/zip.t 14:42
I agree...
jnthn Doesn't test the current behavior that we don't like though, so we're free to fix it :) 14:43
SmokeMachine :) 14:44
jnthn Goodness, the folks doing construction work outside my office have a machine that's creating so much vibration, my keyboard is shaking... 14:45
SmokeMachine :-o
I think zip-latest has the same problem... 14:46
timotimo jnthn: yes, for the other thing, see "zip-latest" i guess? 14:47
SmokeMachine adding a LEAST {done} on each whenever would fix that? 14:49
jnthn LAST, but yes
Think so, anyway
I don't have time to try it right now 14:50
SmokeMachine yes, LAST, sorry! 14:50
SmokeMachine Ill do that when I come home... 14:50
jnthn Cool :)
SmokeMachine++ 14:51
SmokeMachine :)
raschipi LAST but not LEAST 14:56
jnthn ;)
SmokeMachine :) 14:58
comborico1611 Question on printing unsigned integer into bits. value = 65000; putchar( value & displaymask ? '1' : '0' ) ; Is value still 65000, or somehow miraculously binary? 15:48
comborico1611 Raschipi, Question on printing unsigned integer into bits. value = 65000; putchar( value & displaymask ? '1' : '0' ) ; Is value still 65000, or somehow miraculously binary? 15:52
raschipi Is that Perl6? 15:53
comborico1611 Heh, no.
raschipi So I have no idea 15:53
comborico1611 Oh, alright. Thanks, though. 15:53
raschipi Everything is binary, modern computers know nothing else. 15:54
comborico1611 Hmm. 15:54
raschipi In the past there were decimal computers and even before that there was analog computers, But it has been decades that's not used anymore. 15:55
So I can answer with confidence it is indeed binary. 15:56
comborico1611 But the purpose of the program is to print an integer into binary. 15:57
I suppose it is possible that giving an integer to bitwise& automatically converts integer into binary. 15:58
raschipi integer is binary 15:59
gfldex m: my Int $i = 49152; say $i.base(2); 16:00
camelia 1100000000000000
raschipi comborico1611: If displaymask has only one bit set, doing an AND on it with a number will tell if the same bit is set on the number
comborico1611 That's a nice clean binary. 16:01
comborico1611 I don't know why this is so confusing to me. 16:02
raschipi here is something similar in P6: 16:04
m: my $num = 116; say so $num +& 2** $_ for 1,2,3...30
camelia False
True
False
True
True
True
False
False
False
False
False
False
False
False
False
False
False
False
False
False
False
False
False
False
False
False
False
False
False
False
raschipi Or with 0 and 1:
m: my $num = 116; say + so $num +& 2** $_ for 1,2,3...30 16:05
camelia 0
1
0
1
1
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
raschipi Do you get the algorithm? 16:08
ggg hi? 16:13
tyil hi
comborico1611 I'm sorry, no. Thanks for trying. 16:15
raschipi let's go over it, one bit at a time
comborico1611 I do understand how to bitwise and operator works. But how a particular code works, I don't understand. 16:16
raschipi You didn't get the one you posted or the ones in Perl6 too? 16:17
comborico1611 1 & 0 = 0 1 & 1= 1. And so on.
The one i posted. 16:18
raschipi Can you provide context for it? 16:19
What it does depends on the value displaymask has.
comborico1611 User enters an integer. The program simply outputs it's binary equivalent. It uses a for Loop to go bit by bit, and it is that for Loop that is strange to me. 16:20
raschipi So the part that is strange you didn't post?
It probably multiples displaymask by 2 or bit shifts it one position over (which is the same thing for integers) 16:21
comborico1611 I did: value = 65000; putchar( value & displaymask ? '1' : '0' ) ; Is value still 65000, or somehow miraculously binary?
The worst part of this is I'm typing from my phone. 16:22
raschipi putchar probably outputs to the screen and has no side effects
probably
comborico1611 Correct. 16:23
raschipi So, what's the matter? 16:23
comborico1611 But bit mask is 10000000 00000000 . That & with 65000 (binary) yields all zeros. Yet his program prints 65000 in binary. 16:24
So there is something I'm missing. 16:25
I wish i could post photos on here.
El_Che I am happy you can't 16:26
:)
comborico1611 Haha. Good point!
El_Che many people have slack/mattermost/... allergies 16:27
raschipi m: say + so 0b0000000000100000 +& 65000
camelia 1
raschipi 10000000 00000000 & with 65000 is 0 because the corresponding bit in the number is 0. Do you know AND can be applied to numbers bigger than 1 bit, right? 16:28
comborico1611 I wasn't sure if it could. 16:29
It treats them as 1?
raschipi You do it "bit by bit", like when summing you go digit by digit summing them. 16:30
comborico1611 Correct.
raschipi So, if a number has just one of it's own bits set, the result will have a 1 only if the other number has a 1 in the corresponding position 16:31
comborico1611 Correct. Hey, thank you for trying, but i need to take a break. 16:32
raschipi Right, another time then
comborico1611 If i figure this thing out, I'll try to remember to tell you what my hang-up was.
Morfent when i run make rakudo-test when compiling rakudo star from source on freebsd 12, t/04-nativecall/02-simple-args.t fails with "Dubious, test returned 1 (wstat 256, 0x100)" 16:36
timotimo can you ./perl6 t/04-nativecall/02-simple-args.t and see what it outputs?
Morfent hastebin.com/rupepocuxo.go 16:37
i'm wondering if the issue's related to the tests that are skipped on mac os x, since its kernel uses parts of freebsd's 16:38
timotimo this should be too far from the kernel to matter :\
Morfent ah
is this something i'd need to worry about though? 16:39
timotimo i could imagine perhaps "unsigned short" not meaning "16bit integer" on your system 16:40
that would certainly mess this test up
maybe our C code for these tests should fprint diagnostics when they don't get the right value 16:41
so maybe putting an fprintf(stderr, "# value received: %hd\n", x); before the "return 0" in 02-simple-args.c tells us something interesting 16:42
Morfent i'll give that a shot 16:45
timotimo though perhaps %hx is nicer to look at than %hd 16:46
Morfent # value received: -2 16:48
timotimo hm, isn't that correct though?
like, if you use %hx it'll use unsigned logic and display as hexadecimal and that should correspond to 0xFFFE 16:49
well, 0xfffe in the case of %hx, 0xFFFE with %hX
Morfent i was in the middle of running make when you mentioned to use %hx
timotimo you shouldn't need to run make again i think 16:50
timotimo though i'm not sure how exactly compile_test_lib is implemented 16:50
Morfent i wasn't sure, so i was running it again just to be safe 16:51
Morfent # value received: fffe 16:52
timotimo OK, so the C code we have there is not what we expect
i mean, it doesn't compile the same way on your system as on others
i.e. we're relying on undefined behaviour?
raschipi Morfent: which compiler do you have? 16:53
Morfent clang 4.0.1
i have clang 5.0.0 installed, i'll check again with that 16:54
raschipi Is Perl6 even tested on Clang?
timotimo it is 16:55
oh
moarvm is, but that only runs nqp's test suite
and those don't contain nativecall stuff, or at least not much of it 16:56
raschipi Morfent: Can you try GCC? 16:56
Morfent sure
timotimo you're on freebsd, right? any more details than that? 17:01
freebsd 12
Morfent: just to verify, the fprint you have goes between both return statements, right? 17:02
Morfent it goes before both 17:03
timotimo oh
well, that changes things a whole lot
timotimo now it could be it's correctly running the "return 11" but we're somehow not taking the return value correctly 17:04
rather than giving us 0xfffe but comparing against 0xfffe giving "false"
can you make that change and re-run?
Morfent sure 17:07
wait what change do you mean? 17:08
timotimo put the fprintf between "return 11" and "return 0" 17:10
such that it only gets run when the code is about to return 0
Morfent same deal, # value received: fffe 17:12
timotimo now that is interesting!
um, could you change the wording inside the message so we can be sure that it actually compiled the change you made?
timotimo i don't want to be surprised by compile_test_lib being "clever" about caching or something :) 17:13
scimon Can I apologise in advance for anything I say in my talk at LPW tomorrow that's wrong. :D 17:16
Morfent # value received, different message: fffe 17:16
timotimo okay that's weird!
Geth ecosystem: a34a96fb5c | (David Warring)++ (committed using GitHub Web editor) | META.list
Font::AFM to CPAN
 17:18
timotimo can you operate gdb and/or lldb? 17:19
Morfent never used them before
timotimo how complicated is it to get a freebsd 12 VM up and running containing a working perl6? 17:21
timotimo doesn't look trivial to get at a freebsd 12 17:22
buggable New CPAN upload: Font-AFM-1.24.2.tar.gz by WARRINGD cpan.metacpan.org/authors/id/W/WA/...4.2.tar.gz 17:23
raschipi This download links in the PAUSE new upload messages are kinda useless. I think linking to the module home page would be better. 17:25
Morfent the os itself is relatively simple to get installed, but there's a bit of work involved to get it to run properly in virtualbox
timotimo damn (:
:(
Morfent i'm not sure if virtualbox additions will run correctly or not since 12 is on the dev branch atm 17:26
i can install virtualbox and get a vm set up to see if it'll work
timotimo i'm not sure if i'll have much time to look into this today 17:27
timotimo would you be okay with a guided gdb session over irc? 17:28
Morfent sure 17:30
timotimo cool. we have a ./perl6-gdb-m script that'll land you right inside a gdb session 17:31
it'll run the program to completion if you don't ctrl-c, but we can just re-run it after setting up break points and such
so you'd run ./perl6-gdb-m t/04-nativecall/...t
then set a break point with "break TakeUInt16" 17:32
when you "run" it'll reach that point and wait for us to do whatever
dmaestro I'm trying to do a deep comparison (eqv), but I want cmp semantics between Cool values. Any way to lexically modify the eqv op behavior to do that?
timotimo oh, but a moarvm without --debug=3 and with anything but --optimize=0 will make debugging a pain
Morfent Could not open t/04-nativecall/...t. Failed to stat file: no such file or directory 17:33
timotimo well, the ... would be ... what, 02-simple-args i think?
Morfent ok, tried again with ./perl-gdb-m t/04-nativecall/02-simpleargs.t 17:38
timotimo Morfent: did you get my last message? you'd have t/04-nativecall/02-simple-args.t in the argument list
Morfent i don't think i did
timotimo needs one more hyphen i believe?
dmaestro Wishing I could hook into the (presumed) recursive behavior of eqv, but this doesn't work: 17:39
say { multi sub infix:<eqv>(Cool $l, Cool $r) { $l cmp $r === Same }; [42.0] eqv ['42'] }()
timotimo dmaestro: aye, the implementations of eqv all look only in *their* lexical scope for the other operators, which is the core setting
dmaestro m: say { multi sub infix:<eqv>(Cool $l, Cool $r) { $l cmp $r === Same }; [42.0] eqv ['42'] }()
camelia False
timotimo lexically speaking your own implementations are unreachable 17:40
dmaestro Yeah, I figured that was the reason ...
evalable6 False
dmaestro Any other features of Perl6 that conveniently reach deep into nested structures I might use? 17:41
timotimo hypers kind of do that 17:42
m: say [[1, 1], 2] »==« [[1, 2], 2]
camelia [[True False] True]
timotimo m: say [[1, 1], 2] »==« [1, [2, 2]]
camelia Lists on either side of non-dwimmy hyperop of infix:<==> are not of the same length while recursing
left: 2 elements, right: 1 elements
in block <unit> at <tmp> line 1
dmaestro hmm...
timotimo if you catch that exception in case things are odd-shaped 17:43
dmaestro That might actually work for what I'm doing ...
timotimo and you'll then have to break the structure down
timotimo because [False] is a list with one element, and therefor it'll booleanify to True 17:43
Morfent i'm not really sure what you mean by needing one more hyphen timotimo 17:44
timotimo the file name
you wrote "simpleargs" i thought it should be "simple-args"
Morfent oh, that was a typo 17:45
timotimo OK
Morfent Function "TakeUInt16" not defined. Make breakpoint pending on future shared library load? (y or [n])
timotimo oh, the i must be lower case 17:46
dmaestro m: say [42.0] »cmp« ['42'] 17:47
camelia [Same]
Morfent that was it 17:48
Thread 1 hit Breakpoint 1, 0x000000080dc9ca54 in TakeUint16 () from /usr/home/morfent/Downloads/rakudo-star-2017.10/rakudo/lib02-simple-args.so
timotimo OK, good 17:48
now "print x" gives you what? 17:49
Morfent No symbol "x" in current context.
timotimo oh?
hm, compile_test_lib is perhaps not compiling any debug info into the files
AlexDaniel squashable6: next 17:50
squashable6 AlexDaniel, ⚠🍕 Next SQUASHathon in 6 days and ≈16 hours (2017-12-02 UTC-12⌁UTC+14). See github.com/rakudo/rakudo/wiki/Mont...Squash-Day
AlexDaniel If someone has any suggestions for the next squashathon, please let me know
moritz maybe Perl 6 websites? 17:52
perl6.org, examples.perl6.org, modules.perl6.org etc.
dunno if that's a good idea 17:53
timotimo Morfent: try editing t/04-nativecall/CompileTestLib.pm to have -O0 -g3 after the $cfg<cflags> and $cfg<ldflags> in lines 16 and 16 17:54
16 and 17
Morfent done 17:57
dmaestro timotimo: I think this will do 90% of what I want. Thanks!
m: say so all([42] »cmp« ['42']) === Same
camelia True
timotimo cool
Morfent The spawned command 'clang -shared -fPIC -O3 -DNDEBUG -Wl,-rpath,"//usr/home/morfent/Downloads/rakudo-star-2017.10/install/lib" -lm -lpthread -lkvm -o -O0 -g3 lib02-simple-args.so 02-simple-args.o' exited unsuccessfully (exit code: 1) 18:00
timotimo bleh, does it hate multiple -O flags? 18:01
Morfent yeah, -O3 and -O0
er
timotimo okay, put .subst("-O3", "") after the closing " in those lines
that's terrible, but it'll get us what we want for now 18:02
Morfent still fails
timotimo oh 18:02
Morfent oh, those flags are showing up after -o
timotimo yes, that's the problem here
OK, then put it *before* the cflags and ldflags :D
timotimo i mean ideally you'd rebuild moarvm with --optimize=0 --debug=3 in the flags for its Configure.pl 18:04
Morfent ...now the test passes
timotimo :o
Morfent oh wait i forgot to remove the second set of flags on line 16 18:05
Morfent but now i can print x 18:05
$1 = 65534
timotimo m: say 0xfffe 18:06
camelia 65534
timotimo that's correct
which makes sense because the test passes
timotimo without disassembling the code and looking directly at register contents we won't make headway here it seems? 18:07
Morfent i was running this after fixing my typo
the test still fails
timotimo oh? the test fails but x prints 65534?
Morfent yeah
timotimo what does "print 0xfffe" give you?
Morfent $2 = 65534 18:08
timotimo OK, now you can step through the code with "n" and see what it does. "list" will output a piece of source code 18:09
Morfent it segfaults at some point 18:11
hastebin.com/yoceqikulu.cs 18:12
timotimo the trap would have been interesting. it's supposed to output information before crashing though i think? 18:14
SmokeMachine im having a problem compiling rakudo... 18:15
www.irccloud.com/pastebin/jC1gQLFO/
Morfent this is what list shows just after the trap hastebin.com/kaxacerewa.cpp 18:16
SmokeMachine does any one knows why?
timotimo Morfent: it doesn't have line number or source code info for where it actually is (which is moarvm's interpreter loop) so it just outputs what it last had
SmokeMachine: are you working from a clean state? i.e. "make clean" and everything? 18:17
SmokeMachine trying to make clean...
timotimo Morfent: how were you compiling rakudo? did you have a Configure.pl with --gen-moar in the commandline? 18:18
Morfent perl Configure.pl --backends=moar --gen-moar
timotimo OK, you can give moar options there, too, let me look 18:19
Morfent perl version's v5.24.3
AlexDaniel moritz: interesting idea. There are only 27 open issues (from all mentioned repos) so maybe that does not provide enough room for the squashathon
timotimo --moar-version=--optimize=0 --moar-version=--debug=3
SmokeMachine timotimo: make cleaned... im trying to make again...
same error... 18:20
AlexDaniel I was also thinking about a bot squashathon, but at the moment you can't really run whateverable locally… :S
timotimo SmokeMachine: that's very strange because the step right before it should have used Perl6/Grammar.moarvm just the same way as this one? 18:23
SmokeMachine www.irccloud.com/pastebin/gxZwnh85/timotimo 18:24
timotimo SmokeMachine: but it already gave you a step with "stage parse" and other timings, right?
SmokeMachine yes 18:25
www.irccloud.com/pastebin/FNWCTq0Z/
timotimo that doesn't make sense to me :( 18:26
Morfent: with those flags to Configure.pl you'll have to rebuild, but it'll let you actually use gdb properly 18:27
Morfent alright
SmokeMachine :( 18:28
raschipi Can you imagine a bot squashaton, 300 bots connected at the same time to the channel for debugging? 18:35
timotimo we'll just set up an irc server docker container :P
Morfent timotimo: after running perl Configure.pl --backends=moar --gen-moar --moar-option=--debug=3 --moar-option=--optimize=0, make still compiles with -O3 18:49
timotimo did it recompile moar at all?
i.e. did you see many short lines with "3rdparty" and such? 18:50
Morfent yep 18:51
Morfent hastebin.com/rorafedeni.m 18:52
timotimo hm, yes, indeed :\ 18:55
well, you can cd into the nqp/moarvm folder or where it is exactly
check out the Makefile for a list of options that were originally passed to Configure.pl in there
tweak that and re-run Configure.pl inside moarvm, then "make install"
after that, no real need to recompile anything else
SmokeMachine timotimo: I cloned again, and that worked... 18:57
timotimo i'm afraid i have to leave for a bit now 18:58
AlexDaniel raschipi: in debug mode *ables only join #whateverable tho :) 19:02
raschipi AlexDaniel: Why would people run them on debug mode? 19:12
AlexDaniel IIRC it's harder to run them otherwise :D 19:12
raschipi Right, then problem solved 19:14
jstuder Hey all, for the docs I rewrote a section on "Truncating slices" in the subscripts page to address some inaccurate info and submitted the PR a few days back. I was hoping that someone more senior could give it a once over to make sure everything is accurate. 19:23
Or should I just commit it and any inaccuracies can be addressed after the fact?
Morfent timotimo: after doing that, gdb gives something a bit different hastebin.com/fecosawula.cs 19:27
AlexDaniel jstuder: OK, first of all thanks for the awesome PR 19:30
jstuder: PRs that have lots of effort put to them often require lots of effort to review them, that's why it took so long :) 19:31
jstuder: so it is totally correct to come here and to kick some asses to get the job done
jstuder: here's one thing: “produced one-after-another up until the point that the subscript no longer yields a defined value from the collection” 19:32
m: my @letters = <a b c d e f>; @letters[4]:delete; say @letters[3..*]
camelia (d (Any) f)
AlexDaniel jstuder: so [4] was actually not defined, but it didn't stop
jstuder AlexDaniel: no problem. I'm not trying to kick any asses, just wasn't sure anyone was aware. 19:33
AlexDaniel: i see.
AlexDaniel my point is that you should feel free to kick some asses :)
jstuder ok i appreciate it.
AlexDaniel m: my @letters = <a b c d e f>; @letters[4]:delete; say @letters[lazy 3..*]' 19:34
camelia 5===SORRY!5=== Error while compiling <tmp>
Two terms in a row
at <tmp>:1
------> 3tters[4]:delete; say @letters[lazy 3..*]7⏏5'
expecting any of:
infix
infix stopper
postfix
statement end
…
AlexDaniel m: my @letters = <a b c d e f>; @letters[4]:delete; say @letters[lazy 3..*]
camelia (d)
AlexDaniel hmmm… but that's a bug, right?
c: 2015.12 my @letters = <a b c d e f>; @letters[4]:delete; say @letters[lazy 3..*] 19:35
committable6 AlexDaniel, ¦2015.12: «(d)»
AlexDaniel or is it?
jstuder Hmm I guess there could be more to this than I thought. 19:42
I appreciate you looking into it.
AlexDaniel jstuder: ok, here's something to consider 19:49
c: 2015.12 my @letters = <a b c d e f>; @letters[4]:delete; say @letters[3..*]; 19:50
committable6 AlexDaniel, ¦2015.12: «(d)»
AlexDaniel c: HEAD my @letters = <a b c d e f>; @letters[4]:delete; say @letters[3..*];
committable6 AlexDaniel, ¦HEAD(3166400): «(d (Any) f)»
AlexDaniel let's see what was the justification for the change
bisect: my @letters = <a b c d e f>; @letters[4]:delete; say @letters[3..*];
bisectable6 AlexDaniel, Bisecting by output (old=2015.12 new=3166400) because on both starting points the exit code is 0
AlexDaniel, bisect log: gist.github.com/2bca83b3f25f8043ca...d09264a640
AlexDaniel, (2017-10-08) github.com/rakudo/rakudo/commit/45...51c42a52ff
AlexDaniel c: 456358e3c3^,456358e3c3 my @letters = <a b c d e f>; @letters[4]:delete; say @letters[3..*]; 19:51
committable6 AlexDaniel, ¦456358e3c3^: «(d)» ¦456358e: «(d (Any) f)»
AlexDaniel :S
AlexDaniel jstuder: alright, I created this ticket: github.com/rakudo/rakudo/issues/1268 19:56
jstuder I guess we'll find out what the intended behavior is and I'll revise the PR accordingly. 19:59
AlexDaniel: that bisectable bot is pretty cool.
AlexDaniel: is there some way to use it without displaying all the results on the channel? I'd love to give it a try but wouldn't want to drive everyone crazy. 20:01
moritz jstuder: you can try to /msg the bot 20:05
jstuder moritz: thanks. I'm definitely going to give that a try. 20:06
jstuder bisectable says it's too private in there :( 20:13
oh well, there is always git bisect
AlexDaniel jstuder: you can go to #whateverable and do as many queries as you like :) 20:21
jstuder AlexDaniel: thanks 20:22
Geth mu: 85fa828441 | (Jonathan Worthington)++ (committed using GitHub Web editor) | misc/perl6advent-2017/schedule
Add myself to the Advent shcedule
 21:49
jnthn Such typing! 21:50
moritz jnthn: for the advent post you'll have the option to correct your typing before it goes live :-) 21:53
Geth doc: f4b6bd9707 | (Tom Browder)++ (committed using GitHub Web editor) | doc/Language/glossary.pod6
add item LGTM
 22:10
synopsebot Link: doc.perl6.org/language/glossary
jnthn moritz: Phew! :) 22:11
timotimo Morfent: hey 22:25
Morfent hey
timotimo Morfent: can you send me the .so file that gets compiled for the test that was failing? 22:26
i have a compiler expert sitting next to me :P
alternatively, go back to the break point in TakeUint16 and ask gdb to "disassemble" 22:27
Morfent this is what disassemble outputs hastebin.com/wekabonemu.pl 22:30
timotimo can you also give the output of "info registers" 22:31
Morfent hastebin.com/yecewegimi.go 22:32
Thrush Has anybody here tried the challenges on code-golf.io ? 22:37
Perl 6 often dominates. 22:38
AlexDaniel well… yes 22:39
s0me0ne-unkn0wn But still cannot beat golf-oriented languages like Jelly 22:40
s0me0ne-unkn0wn Perl 6 with its whitespace enforcement is not the best golf language imho :) 22:41
AlexDaniel golf-oriented languages make no sense to me really. I remember there was one that mapped typical golfing problems to each ascii character, so every solution was 1 byte…
Thrush There's the prime-number puzzle, and the top score is 22 chars. I can get it down to 23 chars, but I can't figure out how to squeeze out an extra char.
Perl 6 is nice for golf puzzles with prime numbers, due to the .is-prime method. 22:42
AlexDaniel Thrush: use unicode character instead of a normal numeric
m: say 10; say ⑩
camelia 10
10
s0me0ne-unkn0wn golf-oriented languages make no sense beyond code gold, to be precise :)
AlexDaniel Thrush: no, for the number 22:43
s0me0ne-unkn0wn AlexDaniel: They usually count bytes, not characters in golf competitions
Geth mu: deepnext++ created pull request #29:
Update schedule
AlexDaniel s0me0ne-unkn0wn: not that particular website
s0me0ne-unkn0wn: well, “bytes” makes no sense to me either… 22:44
Thrush AlexDaniel: There's a unicode character for 10? And for 100?
AlexDaniel like, in what encoding, right?
timotimo Morfent: hi, this is timotimos so-called "expert" speaking
AlexDaniel Thrush: that you'll have to figure out :)
Thrush: also relevant: github.com/perl6/mu/blob/master/mi...hedule#L25
timotimo the trouble is that the c function expects the upper bits of the 16-bit argument to be clean, but as you can see from "info registers", they're clearly not
s0me0ne-unkn0wn Probably some language somewhere in India have a dedicated character for "100" and if it's on unicode table you can use it with Perl 6 :) 22:45
timotimo i have had a look in the ABI, but it unfortunately remains quiet on the issue.
timotimo the safe thing is probably for the caller to clear those bits if some callees expect that 22:45
and with that, back to timotimo 22:46
AlexDaniel I'll just leave this here… :P gist.github.com/AlexDaniel/52b203c...6ec335336c 22:47
Thrush This is very interesting. I never knew there was a unicode character for 10 (or 100) before.
s0me0ne-unkn0wn Wow, 1000000000000 even 22:49
AlexDaniel s0me0ne-unkn0wn: so -0.5 is not that interesting, huh? :P 22:51
timotimo Morfent: can you do another recompile of moarvm? this time with --moar-option=--has-libffi in the mix? 22:52
er, i mean, you're using moarvm's configure.pl directly, so use --has-libffi directly 22:53
Thrush m: say 100; say ௱ 22:54
camelia 100
100
Thrush Wow... I got it!
I found the symbol for "௱" (TAMIL NUMBER ONE HUNDRED) at www.fileformat.info/info/unicode/ca...o/list.htm . 22:55
s0me0ne-unkn0wn AlexDaniel: Well, right now I live in Croatia and I have the clock with glagolic characters in place of normal digits on my wall so digits lower than 13 represented by single character don't surprise me a lot, but 1000000000000... 22:55
Thrush Glagolic? Do you mean "glagolitic"? 22:57
Morfent timotimo: probing whether your compiler thinks that it is gcc Can't compile simple gcc probe, so something is badly wrong at build/probe.pm line 147. 23:01
Geth mu: 570da66916 | deepnext++ (committed using GitHub Web editor) | misc/perl6advent-2017/schedule
Update schedule
mu: 077e632f57 | (Juan Julián Merelo Guervós)++ (committed using GitHub Web editor) | misc/perl6advent-2017/schedule
Merge pull request #29 from deepnext/patch-1

Update schedule Thanks!
timotimo Morfent: hg.dyncall.org/pub/dyncall/dyncall/..._x64.c#l74 - check this out 23:04
see how it passes short args by pretending they are longlong
Morfent yeah
ryn1x Does CPAN store statistics anywhere? Like number of downloads? 23:05
s0me0ne-unkn0wn Thrush: Exactly, excuse me for my English, I speak mostly Slavic languages :)
timotimo that's where the upper bytes of the regsiter get filled by sign extension
the question becomes: do we have functions that would expect sign extension for the upper bytes? 23:07
Thrush Quite all right, s0me0ne-unkn0wn. I just googled the term, and google was correcting me slightly.
s0me0ne-unkn0wn: Ĉu vi parolas Esperanton? (Do you speak Esperanto?) 23:08
El_Che utf8 pr0n hour at #perl6, /me thinks :) 23:09
s0me0ne-unkn0wn Thrush: Unfortunatelly no, my first language is Russian, I also speak Croatian, Bosnian an Serian (which are mosly the same language), very limited Polish and Bulgarian and also Perl (as native) and Perl 6 (as foreign) :) 23:11
AlexDaniel El_Che: softcore, real unicode porn will be in the advent post :P
El_Che AlexDaniel: haha 23:12
lizmat . 23:15
El_Che lizmat: that the most boring utf8 of the evening :) 23:18
lizmat sorry, it was only intended as ascii :-)
El_Che :) 23:20
(there was a utf8 conversation going on before you joined) 23:21
Thrush Thanks for your help, guys. I have to leave now. 23:23
Herby_ o/ 23:57
mudman \ o
raschipi \o

Powered by Cro and the Raku® Programming Language.

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

Thank you!