»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.
00:03 mr-foobar left 00:05 labster joined 00:17 parisba left, parisba joined, TakinOver joined 00:19 Spot__ left, SmokeMachine____ left 00:22 SmokeMachine____ joined 00:23 Spot__ joined 00:36 rurban1 joined 00:37 rurban left 00:38 rurban joined 00:41 rurban1 left 00:42 rurban1 joined 00:43 rurban left 00:48 tbrowder joined
colomon grondilu: I have to say at first blush, overriding infix:<*> if you have three objects seems very wrong to me. are you trying to improve the accuracy of floating point or something? 00:50
00:52 tbrowder left 00:54 rurban1 left, rurban joined 01:00 rurban left 01:01 rurban joined 01:05 kalkin- left, kalkin-_ joined 01:11 rurban left 01:12 rurban joined 01:16 rurban left 01:17 rurban joined 01:20 molaf left 01:21 rurban left, kurahaupo_ joined, rurban joined 01:23 kurahaupo left 01:33 molaf joined
grondilu colomon: I want to optimize expressions like a*b*a 01:34
or rather a.transpose*b*a, as discussed then 01:35
01:39 kurahaupo_ left
grondilu I don't want it very hard though, but a simple solution would be nice. 01:40
colomon I’m not sure of the wisdom of trying to squeeze matrix multiplication in with normal multiplication. 01:42
But it seems like if you really want to do something like this, the right way to do it is to add another layer of abstraction 01:43
so that infix:<*> (Matrix, Matrix) doesn’t return a Matrix, it returns some sort of proxy object which can generate a Matrix 01:44
and the optimizations are built on the proxy level
01:45 ilbot3 left 01:46 sufrostico left 01:47 ilbot3 joined
grondilu yeah I was considering that, but it's kind of ugly. 01:55
though now that I rethink about it, it makes some sense for my application. 01:56
02:00 rurban left 02:03 kid51 left 02:05 labster left 02:09 colomon left 02:11 colomon joined
timeless Type Order not found in type-graph data 02:18
02:22 Actualeyes joined
BenGoldberg Personally, I would think about porting PDL to perl6. 02:30
02:33 noganex joined 02:36 noganex_ left
grondilu again it's not actually for linear algebra, but for my Clifford module (that is, geometric algebra) 02:40
02:40 pecastro joined
timeless doc: make html is really slow :( 02:42
anyone know how practical it would be for `make html` to be parallelized? 02:43
i have access to >100 cores and would love to have this process run faster (right now, i'm assuming it's using one core) 02:44
02:44 Zero_Dogg left, Zero_Dogg joined 02:45 labster joined
Juerd If it's already parallelizable, try make -j 100 html 02:45
timeless html:
perl6 htmlify.p6
Juerd Oh :( 02:46
02:46 yqt left
timeless is perl6 happy to fork? 02:52
02:54 huggable left, huggable joined
grondilu I vaguely remember someone posted a neat trick to perform a fork lately 02:58
it might have involved v5 though. Can't quite remember. 02:59
timeless grondilu: wait, forking isn't supported in v6 yet? 03:01
(is threading?)
grondilu I think it will never be because it's not portable or something.
but there is a trick using a system call 03:02
timeless language/5to6-perlfunc.html#fork
grondilu m: say "I'm a child" if run fork 03:03
camelia rakudo-moar 87e91c: OUTPUT«5===SORRY!5=== Error while compiling /tmp/D0oQRTQZ0T␤Undeclared routine:␤ fork used at line 1␤␤»
grondilu m: say "I'm a child" if run 'fork'
camelia rakudo-moar 87e91c: OUTPUT«run is disallowed in restricted setting␤ in sub restricted at src/RESTRICTED.setting line 1␤ in sub run at src/RESTRICTED.setting line 14␤ in block <unit> at /tmp/BrJ6DSETH0 line 1␤␤»
timeless m: use NativeCall; sub fork returns int32 is native { * }; say fork; 03:07
camelia rakudo-moar 87e91c: OUTPUT«19709␤0␤»
03:08 skids joined
geekosaur it won't be in core. a POSIX module is projected but nonexistent as yet (likewise a module for Win32) 03:09
timeless m: eval 'use NativeCall; sub fork returns int32 is native { * }; $x=fork; say $x if $x > 0'
camelia rakudo-moar 87e91c: OUTPUT«5===SORRY!5=== Error while compiling /tmp/ecLHOG0NVG␤Undeclared routine:␤ eval used at line 1. Did you mean 'EVAL', 'val'?␤␤»
geekosaur but you can get at either of those with NativeCall of course
grondilu yes I remember now, it was indeed using NativeCall.
timeless m: EVAL 'use NativeCall; sub fork returns int32 is native { * }; $x=fork; say $x if $x > 0' 03:10
camelia rakudo-moar 87e91c: OUTPUT«5===SORRY!5=== Error while compiling /home/camelia/EVAL_0␤Variable '$x' is not declared␤at /home/camelia/EVAL_0:1␤------> 3sub fork returns int32 is native { * }; 7⏏5$x=fork; say $x if $x > 0␤»
grondilu m: use NativeCall; sub fork returns int32 is native { * }; say "I'm the parent" if fork;
camelia rakudo-moar 87e91c: OUTPUT«I'm the parent␤»
geekosaur I would expect rakudo's core to be slightly confused at times though, if it's using anything that doesn't expect fork (rakudo itself might not, but libuv may)
timeless m: EVAL 'use NativeCall; sub fork returns int32 is native { * }; my $x=fork; say $x if $x > 0' 03:11
camelia rakudo-moar 87e91c: OUTPUT«19772␤»
timeless m: EVAL 'use NativeCall; sub forkz returns int32 is native { * }; my $x=forkz; say $x if $x > 0'
camelia rakudo-moar 87e91c: OUTPUT«Cannot locate symbol 'forkz' in native library ''␤ in method setup at /home/camelia/rakudo-m-inst-2/share/perl6/sources/24DD121B5B4774C04A7084827BFAD92199756E03 (NativeCall) line 281␤ in method CALL-ME at /home/camelia/rakudo-m-inst-2/share/perl6/sou…»
grondilu timeless: why do you EVAL it?
timeless because that's what i'd do in perl5 to handle `use` failing 03:12
i want to write code which would work w/ fork if i can fork, and not die if i can't fork
grondilu there has to be a better way to do it.
timeless hopefully :) 03:13
clearly this didn't work :)
grondilu if there is no better way to deal with conditional NativeCall bindings, then it's clearly LTA 03:14
timeless LTA? 03:15
timeless needs to fix an error in the EVAL docs
so, what's the perl equivalent of `try: import X / except ImportError:` ? 03:18
(perl6)
03:19 sherwool left
timeless visits design.perl6.org/S04.html#Exceptions 03:20
grondilu timeless: Less Than Awesome 03:21
that's my favorite euphemism in S99
timeless localhost:3000/language/.html.html.....html.html 03:26
timeless kicks the `make run` script 03:27
03:31 kurahaupo joined 03:40 nowan left
dj_goku_ evening 03:41
03:46 skids left 03:48 nowan joined
timeless grumbles 03:53
m: EVAL 'use NativeCall; sub forkz returns int32 is native { * }; my $x=forkz; say $x if $x > 0' or say "oops"; 03:54
camelia rakudo-moar 87e91c: OUTPUT«Cannot locate symbol 'forkz' in native library ''␤ in method setup at /home/camelia/rakudo-m-inst-2/share/perl6/sources/24DD121B5B4774C04A7084827BFAD92199756E03 (NativeCall) line 281␤ in method CALL-ME at /home/camelia/rakudo-m-inst-2/share/perl6/sou…»
timeless m: EVAL 'use NativeCall; sub forkz returns int32 is native { * }; my $x=forkz; say $x if $x > 0'; CATCH {default {say .WHAT.perl, do given .backtrace[0] { .file, .line, .subname }}} 04:03
camelia rakudo-moar 87e91c: OUTPUT«X::AdHoc(/home/camelia/rakudo-m-inst-2/share/perl6/sources/24DD121B5B4774C04A7084827BFAD92199756E03 (NativeCall) 281 setup)␤»
04:04 maybekoo2 left
timeless grondilu: so, this seems to work, i guess 04:05
m: use NativeCall; sub forkz returns int32 is native { * }; CATCH {default {say .WHAT.perl, do given .backtrace[0] { .file, .line, .subname }}}; my $x=forkz; say $x if $x > 0; 04:06
camelia rakudo-moar 87e91c: OUTPUT«X::AdHoc(/home/camelia/rakudo-m-inst-2/share/perl6/sources/24DD121B5B4774C04A7084827BFAD92199756E03 (NativeCall) 281 setup)␤»
timeless m: use NativeCall; sub forkz returns int32 is native { * }; CATCH {X::AdHoc {say .WHAT.perl, do given .backtrace[0] { .file, .line, .subname }}}; my $x=forkz; say $x if $x > 0; 04:13
camelia rakudo-moar 87e91c: OUTPUT«5===SORRY!5=== Error while compiling /tmp/gh0wdOyVJ7␤Unexpected block in infix position (missing statement control word before the expression?)␤at /tmp/gh0wdOyVJ7:1␤------> 3s int32 is native { * }; CATCH {X::AdHoc7⏏5 {say .WHAT.perl, do giv…»
timeless m: use NativeCall; sub forkz returns int32 is native { * }; CATCH {when X::AdHoc {say .WHAT.perl, do given .backtrace[0] { .file, .line, .subname }}}; my $x=forkz; say $x if $x > 0;
camelia rakudo-moar 87e91c: OUTPUT«X::AdHoc(/home/camelia/rakudo-m-inst-2/share/perl6/sources/24DD121B5B4774C04A7084827BFAD92199756E03 (NativeCall) 281 setup)␤»
04:13 molaf left, BenGoldberg left 04:15 BenGoldberg joined, khw left
timeless m: use NativeCall; sub waitpid returns int32 is native { * }; CATCH {when X::AdHoc {say .WHAT.perl, do given .backtrace[0] { .file, .line, .subname }}}; say "hi"; 04:17
camelia rakudo-moar 87e91c: OUTPUT«hi␤»
04:18 Cabanossi left 04:19 Cabanossi joined 04:25 ssotka left
timeless m: use NativeCall; sub waitpid(int32, Pointer[int32], int32) returns int32 is native { * }; CATCH {when X::AdHoc {say .WHAT.perl, do given .backtrace[0] { .file, .line, .subname }}}; say "hi"; 04:30
camelia rakudo-moar 87e91c: OUTPUT«hi␤»
04:31 kurahaupo left
timeless m: use NativeCall; sub fork returns int32 is native { * }; sub waitpid(int32, Pointer[int32], int32) returns int32 is native { * }; my $x=fork; say "hi" if ($x == 0); if ($x) {my Pointer[int32] $p; my $z = waitpid($x, $p, 1); say $p.deref; 04:37
camelia rakudo-moar 87e91c: OUTPUT«5===SORRY!5=== Error while compiling /tmp/2RfJvo8_5T␤Missing block␤at /tmp/2RfJvo8_5T:1␤------> 3y $z = waitpid($x, $p, 1); say $p.deref;7⏏5<EOL>␤»
timeless m: use NativeCall; sub fork returns int32 is native { * }; sub waitpid(int32, Pointer[int32], int32) returns int32 is native { * }; my $x=fork; say "hi" if ($x == 0); if ($x) {my Pointer[int32] $p; my $z = waitpid($x, $p, 1); say $p.deref; }
camelia rakudo-moar 87e91c: OUTPUT«hi␤Invocant requires an instance of type NativeCall::Types::Pointer[int32], but a type object was passed. Did you forget a .new?␤ in method deref at /home/camelia/rakudo-m-inst-2/share/perl6/sources/51E302443A2C8FF185ABC10CA1E5520EFEE885A1 (NativeCal…»
timeless m: use NativeCall; sub fork returns int32 is native { * }; sub waitpid(int32, int32 as rw, int32) returns int32 is native { * }; my $x=fork; say "hi" if ($x == 0); if ($x) {my $p; my $z = waitpid($x, $p, 1); say $p; } 04:40
camelia rakudo-moar 87e91c: OUTPUT«5===SORRY!5=== Error while compiling /tmp/QSTvJAtbYd␤Malformed parameter␤at /tmp/QSTvJAtbYd:1␤------> 3s native { * }; sub waitpid(int32, int327⏏5 as rw, int32) returns int32 is native {␤ expecting any of:␤ constraint␤ …»
timeless m: use NativeCall; sub fork returns int32 is native { * }; sub waitpid(int32, int32 is rw, int32) returns int32 is native { * }; my $x=fork; say "hi" if ($x == 0); if ($x) {my $p; my $z = waitpid($x, $p, 1); say $p; }
camelia rakudo-moar 87e91c: OUTPUT«hi␤Native call expected argument that references a native integer, but got P6opaque␤ in method CALL-ME at /home/camelia/rakudo-m-inst-2/share/perl6/sources/24DD121B5B4774C04A7084827BFAD92199756E03 (NativeCall) line 303␤ in block <unit> at /tmp/bpt8…»
timeless m: use NativeCall; sub fork returns int32 is native { * }; sub waitpid(int32, int32 is rw, int32) returns int32 is native { * }; my $x=fork; say "hi" if ($x == 0); if ($x) {my $p=int32.new; my $z = waitpid($x, $p, 1); say $p; } 04:41
camelia rakudo-moar 87e91c: OUTPUT«hi␤Method 'BUILDALLPLAN' not found for invocant of class 'Perl6::Metamodel::NativeHOW'␤ in block <unit> at /tmp/RGOeQIFE3e line 1␤␤»
timeless ?!
m: use NativeCall; sub fork returns int32 is native { * }; sub waitpid(int32, int32 is rw, int32) returns int32 is native { * }; my $x=fork; say "hi" if ($x == 0); if ($x) {my int32 $p; my $z = waitpid($x, $p, 1); say $p; } 04:42
camelia rakudo-moar 87e91c: OUTPUT«hi␤0␤»
geekosaur NativeCall has some weird corner cases as yet; you hit one of them 04:43
I think it, or a close relative, is already rakudobug-d
timeless is this likely close to correct? www.irccloud.com/pastebin/z1N7ziCH...fy.diff.pl 04:49
===SORRY!=== Error while compiling /home/timeless/hg/perl6/doc/htmlify.p6 04:50
Only one CATCH block is allowed
at /home/timeless/hg/perl6/doc/htmlify.p6:50
ok, so i guess i want try{} 04:51
m: constant $x=3; say $x 04:55
camelia rakudo-moar 87e91c: OUTPUT«3␤»
timeless m: const $WNOHANG = 1; say $WNOHANG;
camelia rakudo-moar 87e91c: OUTPUT«5===SORRY!5=== Error while compiling /tmp/NM71BCMKAe␤Variable '$WNOHANG' is not declared␤at /tmp/NM71BCMKAe:1␤------> 3const 7⏏5$WNOHANG = 1; say $WNOHANG;␤»
timeless ?! 04:56
oh constant!=const
> This appears to be Perl 5 code 04:57
heh
BenGoldberg m: use NativeCall; sub fork returns int32 is native { * }; sub waitpid(int32, Pointer[int32], int32) returns int32 is native {*}; my $x = fork; if( $x ) { my Pointer[int32] $p .= new; waitpid( $x, $p, 1 ); say $p } else { say "hi" }
camelia rakudo-moar 87e91c: OUTPUT«5===SORRY!5===␤Word 'if' interpreted as 'if()' function call; please use whitespace instead of parens␤at /tmp/jq4su7d0QK:1␤------> 3ns int32 is native {*}; my $x = fork; if7⏏5( $x ) { my Pointer[int32] $p .= new; wa␤Unexpected block in in…»
BenGoldberg m: use NativeCall; sub fork returns int32 is native { * }; sub waitpid(int32, Pointer[int32], int32) returns int32 is native {*}; my $x = fork; if ( $x ) { my Pointer[int32] $p .= new; waitpid( $x, $p, 1 ); say $p } else { say "hi" } 04:58
camelia rakudo-moar 87e91c: OUTPUT«hi␤NativeCall::Types::Pointer[int32]<NULL>␤»
04:59 ufobat joined
timeless m: use NativeCall;try { sub fork returns int32 is native { * }; CATCH { when X::AdHoc { sub fork() { return -1; } } } }; say fork; 05:03
camelia rakudo-moar 87e91c: OUTPUT«5===SORRY!5=== Error while compiling /tmp/zFh4f1zq2f␤Undeclared routine:␤ fork used at line 1␤␤»
05:04 mohae left
BenGoldberg m: my $x = 4; say $x.WHERE; 05:05
camelia rakudo-moar 87e91c: OUTPUT«140131859372752␤»
BenGoldberg m: use NativeCall; sub fork returns int32 is native { * }; sub waitpid(int32, (Pointer), int32) returns int32 is native {*}; my $x = fork; if ( $x ) { my $p = -1; waitpid( $x, $p.WHERE, 1 ); say $p } else { say "hi" } 05:06
camelia rakudo-moar 87e91c: OUTPUT«Potential difficulties:␤ In 'waitpid' routine declaration - Not an accepted NativeCall type for parameter [2] : Any␤ --> For Numerical type, use the appropriate int32/int64/num64...␤ at /tmp/w8tuB54aXM:1␤ ------> 3ter), int32) return…»
BenGoldberg m: use NativeCall; sub fork returns int32 is native { * }; sub waitpid(int32, Pointer[int32], int32) returns int32 is native {*}; my $x = fork; if ( $x ) { my $p = 42; waitpid( $x, Pointer[int32].new($p.WHERE), 1 ); say $p } else { say "hi" } 05:07
camelia rakudo-moar 87e91c: OUTPUT«hi␤42␤»
05:07 sno left 05:08 Praise left 05:20 BenGoldberg left 05:38 domidumont joined, domidumont left, ssotka joined, domidumont joined 05:42 cpage_ left 05:43 cpage_ joined 05:44 cpage_ left 05:46 domidumont left 05:56 ssotka left 05:59 domidumont joined 06:00 domidumont left, domidumont joined 06:02 geraud left, sno joined 06:09 _mg_ joined 06:10 grondilu left 06:11 grondilu joined 06:13 _mg__ joined 06:14 _mg_ left, _mg__ is now known as _mg_ 06:26 firstdayonthejob joined 06:36 rindolf joined 06:40 rurban joined
masak good morning, #perl6 06:40
masak almost wrote "good Missing or wrong dependency, #perl6", but decided that was too snarky ;)
06:41 firstdayonthejob left 06:46 rindolf left
moritz I don't think I've got that error in a while 06:50
06:53 cpage_ joined 06:54 huggable left, huggable joined 06:55 wamba joined 07:01 brrt joined 07:07 aries_liuxueyang left 07:09 aries_liuxueyang joined 07:14 xinming left 07:15 darutoko joined, xinming joined 07:29 bjz joined
masak moritz: I'm currently getting it dozens of times per day 07:30
moritz huh 07:31
seems I don't use enough Rakudo these days :-)
masak this is happening when modifying a 12-module project with dependencies between the modules 07:36
moritz seems to warrant a bug report 07:39
is nine++ aware of this?
masak I've talked to FROGGS about it, and FROGGS++ submitted a bug report
I'm not 100% sure his report has the same root cause as mine, but it seems close enough 07:40
07:41 zakharyas joined 07:45 bjz left 07:52 brrt left 08:00 mr-foobar joined, bjz joined 08:01 ocbtec joined 08:03 rurban1 joined 08:04 rurban left 08:05 RabidGravy joined 08:09 rurban joined 08:11 rurban1 left 08:13 abraxxa joined 08:23 abraxxa left, abraxxa joined
RabidGravy hahaha, I've been fretting because this thing for generating SQL was making too many parentheses in comparison to the output SQL::Abstract tests against, all the while not realising that the latter actually parses the output and the expected SQL and compares the AST :-\ 08:34
so if someone could come up with an SQL parser that would be great kthx 08:40
jnthn Well, I did a lolsql one a while back, but... :)
08:48 AlexDaniel joined 08:52 zakharyas left 08:54 zakharyas joined 09:01 AlexDaniel left 09:02 mr-foobar left 09:07 rindolf joined
Ulti can someone who's an admin on reddit maybe deal with the latest post www.reddit.com/r/perl6 09:20
or maybe bump me up so I can deal with it
since I'm on there quite a bit
09:20 bjz left 09:21 user9 joined
RabidGravy not sure if I even have a reddit account 09:23
Ulti not entirely sure why the puppet account wasnt blocked immediately from the first post
looks like it was removed though 09:24
09:26 rurban1 joined, rurban left 09:31 zakharyas left 09:32 zakharyas joined
stmuk_ the reddit video doesn't even play! 09:39
RabidGravy probably because the malware payload is only just enough of a video to have the browser think it is a video 09:41
psch looks like it's a thumbnail, actually 09:45
09:47 bjz joined
stmuk_ I liked the pocgtfo executable pdfs 09:48
09:51 bjz_ joined 09:52 bjz left 10:01 maybekoo2 joined
RabidGravy so Grammar::BNF can't hack the SQL grammar I can find, I think it's make the dumbest possible grammar that works by hand 10:08
10:08 kid51 joined 10:21 rurban joined 10:24 TEttinger left, jack_rabbit left, rurban1 left 10:25 rurban1 joined, mr-foobar joined 10:26 rurban left 10:31 rurban joined 10:33 rurban1 left
RabidGravy or, even slacker, adjust the expected results to match the expected amount of parenthesizification 10:34
10:35 rurban left 10:36 rurban joined 10:37 labster left, kaare_ joined 10:40 rurban left, rurban joined 10:42 mr-foobar left 10:45 rurban left, rurban joined 10:50 perlpilot left, kent\n left 10:52 perlpilot joined 10:54 TimToady joined, huggable left, huggable joined 10:59 donaldh joined 11:00 yqt joined 11:01 Zero_Dogg left 11:02 Zero_Dogg joined, Zero_Dogg left, Zero_Dogg joined, telex left 11:03 telex joined 11:09 kid51 left 11:11 rindolf left 11:12 kid51 joined, _mg_ left 11:20 tbrowder joined 11:21 AlexDaniel joined 11:24 tbrowder left 11:26 huf joined 11:27 rindolf joined 11:36 rurban1 joined 11:38 rurban left 11:46 kid51 left 11:48 _mg_ joined 11:49 donaldh left 11:51 donaldh joined 11:59 awwaiid joined 12:04 g4 joined
DrForr I think I've found a way to break lib/.precomp - Are there known bugs? 12:10
12:10 Praise joined, Praise left, Praise joined
masak DrForr: yes, I'm getting that one several times daily 12:12
DrForr: basically, I change a .pm file in my project, and when I run things again I get "Missing or wrong version of module" and have to nuke lib/.precomp
lizmat DrForr: if you have a golf on how to do that, please report
masak FROGGS++ reported something like that
DrForr Okay, this is different then.
lizmat output with running RAKUDO_MODULE_DEBUG=1 is helpful :-) 12:13
masak oh, I will try that next time -- thanks for the tip :) 12:14
is RAKUDO_MODULE_DEBUG documented somewhere?
moritz masak: yes, docs/running.pod 12:15
... which should be installed as a manpage, but isn't
and I think we have a bug report for that too 12:16
12:16 rurban_ joined
masak :) 12:17
DrForr I'll have to play here because it's essentially using awwaliid's repl code but breaking the loader somehow.
12:18 setty1 joined
DrForr I'm going to have to do this later, but it's a little odd. 12:24
12:27 skids joined 12:28 g4 left
DrForr Basically if I have a grammar with a term, the code works. I remove the term from the grammar, and rerun. Now I get 'Could not find symbol '&here'' in the script that loads the grammar. Nothing to do with the script because I can fix it by deleting the .precomp directory. 12:30
12:31 mr-foobar joined
masak yes, that does sound different 12:31
please try to golf and submit an RT ticket
DrForr Will do this weekend.
12:33 Sgeo_ left 12:34 maybekoo2 left 12:37 kurahaupo joined, brrt joined 12:42 skids left
[Coke] (can't email a bug to RT) - please let me know if this is still an issue. 12:43
12:44 kurahaupo left 12:48 Praise left 12:50 Praise joined, Praise left, Praise joined 12:51 abraxxa left
dalek c: 9b72be0 | (Zoffix Znet)++ | app.pl:
Prevent infinite redirect on 404 files

Related to #576
12:51
13:02 cdg joined 13:11 mr-foobar left 13:15 brrt left
dalek c: fef2e8e | (Zoffix Znet)++ | app.pl:
Fix another infinite redirect bug
13:16
13:16 avalenn joined 13:22 sufrostico joined
stmuk_ votes "in" to Wombles 13:26
13:35 rurban joined, rurban1 left
dogbert17 o/ #perl6, does anyone want to take a quick look at a doc gist for Baggy.elems here? gist.github.com/dogbert17/d6bc8023...6ca7a27c4a 13:47
13:51 salv0 joined
AlexDaniel what is mixhash? 13:52
what is mix!? 13:53
psch m: my $x = new MixHash: 1, 2, 3; say $x;
camelia rakudo-moar 9cfcb9: OUTPUT«MixHash.new(3, 1, 2)␤»
psch m: my $x = new MixHash: 1, 2, 3; say $x.perl
camelia rakudo-moar 9cfcb9: OUTPUT«(3=>1,1=>1,2=>1).MixHash␤»
dogbert17 AlexDaniel: haven't you read the docs :)
psch its kinda like a SetHash but with different-than-one for the values i think? 13:54
well, "not only 1"*
AlexDaniel :o
dogbert17 the doc page for Baggy is a bit pathetic atm 13:55
AlexDaniel ok, so *Hash is mutable while just * is immutable 13:56
dogbert17 yes
AlexDaniel why not call it BagMut or something, uh… nevermind
dogbert17 maybe BagWomble 13:57
AlexDaniel then we hav Bag and Mix, and the difference is just that Bag has int weights and Mix has non-int weights
have*
psch m: (new BagHash: a => .5).perl.say 13:58
camelia rakudo-moar 9cfcb9: OUTPUT«().BagHash␤»
psch m: (new MixHash: a => .5).perl.say
camelia rakudo-moar 9cfcb9: OUTPUT«().MixHash␤»
psch m: (new MixHash: 1 => .5).perl.say
camelia rakudo-moar 9cfcb9: OUTPUT«(1 => 0.5=>1).MixHash␤»
AlexDaniel what
psch m: my $x = new MixHash: 5; $x<5> = .5; say $x.perl
camelia rakudo-moar 9cfcb9: OUTPUT«(5=>1,IntStr.new(5, "5")=>0.5).MixHash␤»
AlexDaniel m: say ("b" => 9, "b" => 2).Mix.perl;
camelia rakudo-moar 9cfcb9: OUTPUT«("b"=>11).Mix␤»
psch m: my $x = new MixHash: 5; $x{5} = .5; say $x.perl # actually... 13:59
camelia rakudo-moar 9cfcb9: OUTPUT«(5=>0.5).MixHash␤»
AlexDaniel m: say ("b" => 9.5, "b" => 2).Mix.perl;
camelia rakudo-moar 9cfcb9: OUTPUT«("b"=>11.5).Mix␤»
AlexDaniel m: say ("b" => 9.5, "b" => 2).Bag.perl;
camelia rakudo-moar 9cfcb9: OUTPUT«Type check failed in assignment; expected Int but got Rat (9.5)␤ in block <unit> at /tmp/FVTFGKm1gM line 1␤␤»
psch .new threw me for a loop there
m: (new MixHash: %(1 => .5)).perl.say
camelia rakudo-moar 9cfcb9: OUTPUT«("1" => 0.5=>1).MixHash␤»
psch well, not like that either apparently vOv
AlexDaniel m: my $n = ("b" => 9, "b" => 2).MixHash; say $n.grab 14:00
camelia rakudo-moar 9cfcb9: OUTPUT«.grab is not supported on a MixHash␤ in block <unit> at /tmp/M4HRsHFGcq line 1␤␤Actually thrown at:␤ in block <unit> at /tmp/M4HRsHFGcq line 1␤␤»
14:01 skids joined 14:05 girafe joined, ptolemarch joined
dalek c: 2bd9117 | (Jan-Olof Hendig)++ | doc/Type/Baggy.pod:
Added docs for Baggy.elems
14:05
14:15 rurban left
AlexDaniel dogbert17: the graph here is not very promising :) github.com/perl6/doc/issues/561 14:20
14:23 Praise left, Praise joined, Praise left, Praise joined
dalek c: 6f028bd | (Aleks-Daniel Jakimenko-Aleksejev)++ | lib/Pod/Htmlify.pm6:
Shorter commit SHA on doc footer

Fixes #578
14:27
[Coke] what browser are we trying to support by providing png support in addition to svg support on the docs site, OOC? 14:33
just wondering if we can remove one of the moving parts.
dalek c: ee767e6 | (Aleks-Daniel Jakimenko-Aleksejev)++ | lib/Pod/Htmlify.pm6:
Shorter date on doc footer
14:34
dogbert17 AlexDaniel: indeed, numbers actually got worse after fixing a bunch of 404's
[Coke] docs - we are inconsistent in adding ␤ in the comment showing what the output of something was.
e.g. :
say 42.WHICH eq 42.WHICH; # True
vs.
say 42; # 42␤
I would recommend just removing ␤ - it's not adding value, IMO 14:35
dogbert17 [Coke]: i'm afraid I'm one of the culprits :(
[Coke] if we do feel it's adding value, write up something and note it in the contributions guidelines.
14:35 _mg_ left 14:36 ZoffixW joined
[Coke] is it desirable on docs that we have a max column width for the docs? 14:37
14:37 rurban joined
ZoffixW [Coke], pngs can be dropped IMO. They were likely added 'cause IE8 wasn't supporting it, but it's been EOLed by Microsoft since January. Everything else supports it: caniuse.com/#feat=svg 14:37
dogbert17 [Coke]: I agree, it doesn't add any value IMHO
AlexDaniel keep PNGs please 14:38
[Coke] AlexDaniel: ok, but why?
ZoffixW [Coke], [max column] yes, the lines would be a tad too long on my 1440p 27" monitor :P 14:39
But to that point, I noticed on mobile the margins are way too fat.
AlexDaniel [Coke]: because my text-based browser does not display SVGs 14:40
ZoffixW AlexDaniel, does it display PNGs?
AlexDaniel ZoffixW: yes
ZoffixW AlexDaniel, what browser is it?
AlexDaniel ZoffixW: w3m
ZoffixW: you need a terminal with support for images though
ZoffixW :/ 14:41
DrForr Sounds like a cruel way to cheat with aalib+doom :)
AlexDaniel but even if you don't care about text-based browsers there's still IE 8
DrForr Oh, that'd be interesting. regular aa text on the back layer and a cropped image of the original Doom app somewhere embedded. 14:42
AlexDaniel don't remove features that are already there
[Coke] AlexDaniel: no, there is not still IE8. :P 14:43
ZoffixW No, there isn't IE8. We support it neither in our support policy nor should we, because it's no longer secure.
I have nothing against keeping PNGs for text browsers. I mean, we do have an actual user right here :)
b2gills
.oO( was IE8 ever secure )
14:44
[Coke] if you can't use your text based browser directly for pngs, I'm not sure that counts as a reason, either. if there was a standalone svg file, would that suffice?
AlexDaniel where does that need to remove pngs comes from again? 14:45
what are you trying to solve by removing one of the potentially useful features?
b2gills Does anyone know if it is simple to use only some of the candidates for multi tokens in a grammar? 14:46
[Coke] Because we have limited resources - if we can remove a thing that requires support, it's one less thing for us to deal with
AlexDaniel how much resources does png generation take?
[Coke] I am ok with leaving it in, if we have one contributor who needs it for local development. That's fine.
AlexDaniel: dead code is best code
b2gills removed dead code is best code 14:47
[Coke] It slows down local development.
b2gills: yes, thanks you for clearly up my sloppy sentiment there.
14:48 wamba left
AlexDaniel [Coke]: OK, if you make these SVGs external then it's probably good enough 14:48
right now it is inlined, which hurts a lot actually 14:49
e.g. I can't even zoom it properly in firefox
[Coke] In general, I like inline (except for Any); having separate .svg external links alongside the .pngs would be a nice to have, aye 14:50
14:50 ZoffixW left
b2gills
.oO( r̶e̶m̶o̶v̶e̶d̶ exterminated dead code is the best code )
14:51
14:51 mohae joined
DrForr waits for the inevitable dalek checkin :) 14:51
AlexDaniel [Coke]: ok, so both shortened commit hash and shortened date are now live
14:51 rurban1 joined
AlexDaniel [Coke]: any other issues with the footer? 14:51
14:52 FROGGS joined, rurban left 14:54 huggable left
[Coke] AlexDaniel: Artistic License 2.0 should be a link to same. 14:56
copyright should be spelled: 14:57
AlexDaniel [Coke]: well, that's the file to edit :) github.com/perl6/doc/blob/147a284c...ooter.html
[Coke] ©
gregf_ hello 14:58
[Coke] ... Yes, I can make the edits myself, but then why ask?
gregf_: hello
gregf_ role Foo { method should_be_able_to_access_ivars(){ say $!bar; }}; class FooImpl does Foo { has $.bar is rw; } <== should this not work?
[Coke] Thanks for fixing the ones I opened tickets for.
AlexDaniel [Coke]: thought that there would be something slightly more complicated
[Coke] m: role Foo { method should_be_able_to_access_ivars(){ say $!bar; }}; class FooImpl does Foo { has $.bar is rw; }
camelia rakudo-moar 9cfcb9: OUTPUT«5===SORRY!5=== Error while compiling /tmp/i9VUPXVzJO␤Attribute $!bar not declared in role Foo␤at /tmp/i9VUPXVzJO:1␤------> 3be_able_to_access_ivars(){ say $!bar; }}7⏏5; class FooImpl does Foo { has $.bar is ␤ expecting any of:␤ …»
psch m: role Foo { method should_be_able_to_access_ivars(){ say $!bar; }}; class FooImpl does Foo { has $.bar is rw; }
camelia rakudo-moar 9cfcb9: OUTPUT«5===SORRY!5=== Error while compiling /tmp/Pgx3D7sJoV␤Attribute $!bar not declared in role Foo␤at /tmp/Pgx3D7sJoV:1␤------> 3be_able_to_access_ivars(){ say $!bar; }}7⏏5; class FooImpl does Foo { has $.bar is ␤ expecting any of:␤ …»
psch argh
[Coke] heh
psch m: role Foo { method should_be_able_to_access_ivars(){ say $.bar; }}; class FooImpl does Foo { has $.bar is rw; } 14:59
camelia ( no output )
psch didn't mean to copy the \n, but apparently did..? vOv
[Coke] the role has to compile by itself first, so that seems correct
psch gregf_: the ! twigil means the Attribute. the . twigil means the (potentially auto-generated) accessor
gregf_ m: role Foo { method should_be_able_to_access_ivars(){ say $.bar; }}; class FooImpl does Foo { has $.bar is rw; }; Foo.new(:bar(100)).should_be_able_to_access_ivars()
camelia rakudo-moar 9cfcb9: OUTPUT«Method 'bar' not found for invocant of class 'Foo'␤ in method should_be_able_to_access_ivars at /tmp/N9TQv_Q1d7 line 1␤ in block <unit> at /tmp/N9TQv_Q1d7 line 1␤␤»
gregf_ psch: ah - i see 15:00
wow, such subtle differences :)
psch yeah, it compiles but still doesn't find the method
m: role R { method bar { $.b } }; class C does R { has $.b is rw }; C.new(:2b).bar.say 15:01
camelia rakudo-moar 9cfcb9: OUTPUT«2␤»
gregf_ correct, but it should i guess ?
psch well, i can't see through your long names :P
gregf_ heh
*just so that im clear in making my point* :) 15:02
psch gregf_: oh, you instantiated Foo, not FooImpl
gregf_ ah - good catch
oh so :b(100) and :100b are the same 15:03
o_O
psch gregf_: well, it's neat for e.g. s///
m: $_ = "foo"; s:2nd/o/u/; .say
camelia rakudo-moar 9cfcb9: OUTPUT«fou␤»
gregf_ my, that took me a while to figure out :) 15:04
psch m: Date.today.later(:2days)
camelia ( no output )
psch m: Date.today.later(:2days).say
camelia rakudo-moar 9cfcb9: OUTPUT«2016-06-12␤»
psch or that
gregf_ m: say CORE::.keys.grep(/nd/) 15:05
camelia rakudo-moar 9cfcb9: OUTPUT«(&append &prepend &infix:<notandthen> &rindex &indir &rand &indices IterationEnd &trait_mod:<handles> &infix:<and> PositionalBindFailover &METAOP_TEST_ASSIGN:<andthen> &roundrobin &index &infix:<andthen> &undefine &find-reducer-for-op &srand &round &end Se…»
psch m: say later Date.today : :2days : # heh
camelia rakudo-moar 9cfcb9: OUTPUT«2016-06-12␤»
gregf_ psch++ 15:06
[Coke]++
15:08 zakharyas left
gregf_ m: $_ = "foofoo"; s:5th/o/u/; .say 15:09
camelia rakudo-moar 9cfcb9: OUTPUT«foofoo␤»
psch m: $_ = "foofoo"; s:3th/o/u/; .say
camelia rakudo-moar 9cfcb9: OUTPUT«foofuo␤»
psch m: $_ = "foofoo"; s:2x/o/u/; .say
camelia rakudo-moar 9cfcb9: OUTPUT«fuufoo␤»
15:10 zakharyas joined
psch gregf_: see doc.perl6.org/routine/subst 15:10
(although apparently :$st, :$nd, and :$rd are undocumented)
15:10 bbkr joined, Zero_Dogg left
gregf_ oh, thats just sugar for subst? 15:10
psch well, actually s/// is subst-rw, and S/// is subst iirc 15:11
15:11 Zero_Dogg joined
gregf_ rw as in it modifies the string? 15:11
bbkr hi. any idea why successful test suite is treated by Travis-CI as failed? travis-ci.org/bbkr/GeoIPerl6/builds/99810697
gregf_ anyways psch++ 15:12
psch bbkr: i see "Failed tests: 2-3" and "Done. Your build exited with 1.", which seems to indicate failure..?
[Coke] bbkr: it's not successful? you failed 2 subtests?
psch as for what failed: floating pointyness, i'd guess from a distance 15:13
bbkr damn, I see two different results on two different browsers. something is messed with session probably :) 15:14
bbkr.org/travisvci.png 15:16
FROGGS o/
psch o/ FROGGS 15:17
15:18 bbkr left
timeless did BenGoldberg find a version of use Nativall ... sub waitpid that worked? 15:18
m: use NativeCall; sub fork returns int32 is native { * }; sub waitpid(int32, Pointer[int32], int32) returns int32 is native {*}; my $x = fork; if ( $x ) { my $p = 42; waitpid( $x, Pointer[int32].new($p.WHERE), 1 ); say $p } else { say "hi" } 15:19
camelia rakudo-moar 9cfcb9: OUTPUT«hi␤42␤»
timeless this doesn't look like what we're expecting 15:20
m: use NativeCall; sub fork returns int32 is native { * }; sub waitpid(int32, int32 is rw, int32) returns int32 is native { * }; my $x=fork; say "hi" if ($x == 0); if ($x) {my int32 $p; my $z = waitpid($x, $p, 1); say $p; }
camelia rakudo-moar 9cfcb9: OUTPUT«hi␤0␤» 15:21
timeless hopes that "is" the correct answer, but isn't really sure how to figure it out
actually, i think what i really want to know is how to properly use `try` to conditionally define something :/ 15:22
psch m: my $x = do { try "foo" + 2 } // 10; say $x; $x = do { try "2" + 2 } // 10; say $x 15:24
camelia rakudo-moar 9cfcb9: OUTPUT«10␤4␤»
psch timeless: ^^^ like that maybe?
15:24 mst is now known as {[{[{[{[, {[{[{[{[ is now known as mst 15:25 _mg_ joined
[Coke] docs.perl6.org/language/traps has several perl 5 specific traps that should be in 5-6 instead. 15:25
timeless try { sub waitpid(int32, int32 is rw, int32) returns int32 is native { * }; CATCH { when X::AdHoc { sub waitpid() { return -1; }}} $waitpid = waitpid;
psch: something like this ^ (which doesn't work at all)
b2gills (10:27:28 AM) b2gills: m: use NativeCall; sub fork returns int32 is native { * }; sub waitpid(int32, int32 is rw, int32) returns int32 is native {*}; if fork() -> $x { my int32 $p = 42; waitpid( $x, $p, 0 ); say $p } else { say "hi"; exit 0 } 15:26
(10:27:28 AM) b2gills: m: use NativeCall; sub fork returns int32 is native { * }; sub waitpid(int32, int32 is rw, int32) returns int32 is native {*}; if fork() -> $x { my int32 $p = 42; waitpid( $x, $p, 0 ); say $p } else { say "hi"; exit 1 }
m: use NativeCall; sub fork returns int32 is native { * }; sub waitpid(int32, int32 is rw, int32) returns int32 is native {*}; if fork() -> $x { my int32 $p = 42; waitpid( $x, $p, 0 ); say $p } else { say "hi"; exit 0 } 15:27
m: use NativeCall; sub fork returns int32 is native { * }; sub waitpid(int32, int32 is rw, int32) returns int32 is native {*}; if fork() -> $x { my int32 $p = 42; waitpid( $x, $p, 0 ); say $p } else { say "hi"; exit 1 }
camelia rakudo-moar 9cfcb9: OUTPUT«hi␤0␤»
rakudo-moar 9cfcb9: OUTPUT«hi␤256␤»
timeless oh, right, exit, thanks b2gills
psch timeless: yeah, that can't work because the sub declaration is compile time, and try only helps at runtime
timeless ok, so, that side's settled,
psch m: use NativeCall; my &waitpid = do { try sub (int32, int32 is rw, int32 --> int32) is native { * } } // sub { return -1 }; say waitpid 15:28
camelia rakudo-moar 9cfcb9: OUTPUT«-1␤»
15:28 rurban1 left
psch i guess NC subs don't like the in-Signature return type..? 15:28
m: use NativeCall; my &waitpid = do { try sub (int32, int32 is rw, int32) returns int32 is native { * } } // sub { return -1 }; say waitpid
camelia rakudo-moar 9cfcb9: OUTPUT«-1␤»
psch well, that's how i'd do it if you want a Callable in the current scope 15:29
i *think* you can export that too, but i'm not sure
timeless export? 15:30
psch nah, can't
m: module Foo { my &foo is export = sub { 1 } }; import Foo; foo.say
camelia rakudo-moar 9cfcb9: OUTPUT«5===SORRY!5=== Error while compiling /tmp/UzNxnW87OW␤Can't apply trait 'is export' on a my scoped variable. Only our scoped variables are supported.␤at /tmp/UzNxnW87OW:1␤------> 3module Foo { my &foo is export7⏏5 = sub { 1 } }; import Foo; f…»
timotimo well, yeah, it has to be "our" i suppose?
psch oh, maybe read those helpful error messages... :)
m: module Foo { our &foo is export = sub { 1 } }; import Foo; foo.say
camelia rakudo-moar 9cfcb9: OUTPUT«1␤»
timeless timotimo: `Only our` would be better as `Only 'our' ..`
the word `our` alone looks like "English"
whereas you are trying to tell the reader it's perl6 15:31
mst use whatever the pod6 equivalent is to C<our> ?
[Coke] our-scoped
timotimo :)
timeless [Coke]: i guess? any of these would be an improvement 15:32
15:32 ufobat left
timotimo i did put it into quotation marks 15:32
oh!
now i see what you mean
[Coke] 'our' is probably better, as suggested by timeless. (since we have 'is export' just prior)
timeless [Coke]: sold, can you make it so? :) 15:33
psch probably same for the "my scoped" bit before
b2gills don't forget to change the 'my' too
timeless probably on a `'my' scoped`too
psch heh
timeless rotfl
b2gills jinx
timeless thanks, my stomach-core muscles needed that exercise/laugh :-)
[Coke] timeless: won't be able to get to it soon, no; please open a ticket. 15:34
[LTA] error..
timeless needs to check to see if LTA is in the glossary (i know someone mentioned it was in S99) 15:35
(it should be, since i tried to make the glossary cover S99...)
[Coke] docs.perl6.org/language/glossary#LTA
timeless TIL when w10 is in the "please select a window to fill this space" mode, ctrl-w will close the currently selected window 15:37
(that's a pretty neat piece of sugar, too bad, it closed this window...)
15:39 setty1 left
timeless psch: so, i don't /think/ i need export here 15:40
and perhaps i don't even need the catch, is the try{} enough?
i just need to either get a callable or define my own replacement (which i could probably give a different name temporarily or stick into an EVAL) 15:41
m: use NativeCall; my &waitpid = do { try sub (int32, int32 is rw, int32) returns int32 is native { * } } // sub { return -2 }; say waitpid 15:42
camelia rakudo-moar 9cfcb9: OUTPUT«-2␤»
15:42 ssotka joined
timeless um, ok, what magic is `//`? 15:42
also, is there a perl6doc? 15:43
localhost:3000/routine///.html.html....html.html
gee, that's a very helpful url
psch m: say True // 1, False // 2, Nil // 3 15:44
camelia rakudo-moar 9cfcb9: OUTPUT«TrueFalse3␤»
timeless doc.perl6.org/routine/%2F%2F
psch timeless: infix:<//>($a, $b) is pretty much '$a.defined ?? $a !! $b'
15:46 khw joined
psch and yes, that broken link is ticketed, i'm pretty sure 15:46
ISTR that there's some apache directive to fix that, but i don't know if doc.perl6.org even runs on apache vOv
timeless files a ticket 15:50
[Coke]: sorry, GitHub is a lot easier to use than RT
psch timeless: github.com/perl6/doc/issues/155
[Coke] timeless: the app runner isn't intended to be a fully functional website, btw. 15:51
timeless [Coke]: sure
[Coke] so those are just nice to have. (if it happens on the prod site, sure) 15:52
timeless but you can't honestly expect a drive by contributor to set up a fully functional website just to test fixes
[Coke] .. no. but I don't expect you to hammer on the not-for-production thing when that's not what you're trying to fix. 15:53
Like I said, feel free to report 'em, but those are super low priority
just setting expectation levels.
timeless nods
from my perspective, if i report something that's low priority, i don't expect a fix 15:54
but if i provide a fix, i kinda expect someone to seriously consider taking it
and as a drive-by contributor, i expect not to be forced to rewrite it 100 times
[Coke] Sure
timeless :-)
[Coke] Has that happened to you here?
timeless github.com/perl6/doc/pull/576 was pretty close 15:55
timeless still can't reopen
15:55 maybekoo2 joined
timeless and yes, i /think/ zoffixznet fixed it, although I couldn't figure out what his changeset did 15:56
(this is where if i had written that code and showed it to someone, they'd complain that it was undocumented-unreadable-perl)
timeless has lots of such code
anyway, i need to update and see how it works 15:57
15:58 zakharyas left, ZoffixW joined
ZoffixW timeless, there was a mess with redirects and I tossed them and made the app serve static files instead of redirecting to them for the Static handler to catch them. 15:59
timeless slowly tries to parse that
ZoffixW timeless, as for why I closed that PR is because I didn't know you were trying to fix the .html.html.html stuff. I thought you were just trying to avoid the extra HTTP request generated by the redirect. 16:00
timeless yeah, i don't blame you, i should have explained
but, over-eagerly-closing isn't helpful
RabidGravy as a matter of interest how much effort would be involved in knocking up a docker/vagrant/whatever the fashionable thing for containerisation is today to mo mimic the docs server set-up?
timeless you could ask first and give it a bit before closing instead of closing at the same time
psch timeless: i hope closing dupes is fine with you :)
timeless psch: that i appreciate 16:01
I also appreciated ZoffixW 's link to his fix
even though i still don't understand it
ZoffixW timeless, oh, my bad. I didn't realized closed PRs don't have a reopen button :S
timeless ZoffixW: fwiw, i still can't parse/digest your explanation here
ZoffixW: i'm not sure
it's possible only members or something can
RabidGravy so are all the types in the core setting documented yet? 16:02
timeless i've certainly had the button in other contexts
psch ZoffixW: i think the PR-opener can only reopen if they closed themselves
timeless RabidGravy: NaN isn't iirc
ZoffixW Ah
timeless i have a thing for it
also, Order is in a phantom zone (i have a patch for that too)
psch timeless++ # docs
timeless ZoffixW: do you want one pr per thing, or a chain where you have to selectively something?
my tree is getting messy, and i'm only working on ~10 thins 16:03
timeless can easily end up working on ~100
timeless is currently an hg refugee
(my hg hg repo has too many WIPs not making progress)
ZoffixW timeless, I'm not the Ultimate Ruler of PRs, but in my own projects I expect a PR to include just one thing. It can be multiple commits but they should be related to each other in the goal they're trying to achieve 16:04
timeless anyway, while those q's are on the table, i need to figure out how to get waitpid to work
ZoffixW: so, the problem i have w/ that model, is that if i'm making 100 improvements
i don't want the overhead of 100 pr's
ZoffixW Sure.
timeless, I guess do what you think is best :) 16:05
timeless (there isn't a good model that handles these cases, so, i'm not blaming you/anyone else for not having a good answer)
psch well, in the docs space i can see "document @a-bunch-of-types" as "one thing", fwiw
timeless psch: yeah, the problem is that it easily becomes "i like 3/4 or 15/21"
psch but "document @a-bunch-of-types, improve 'make run' handling, $something-else" isn't one thing
timeless i hope i didn't include make-run in the document pr's 16:06
(it's possible, but i do /try/ to split on category)
psch timeless: i don't know, it was just an example at the top of my head :)
16:06 donaldh left
psch in any case, nobody's gonna tear into anyone for making a mistake in a PR 16:06
16:06 domidumont left
ZoffixW timeless, as for the fix: first the Static file handler tries to find the file, if it doesn't it goes route matching. We used to have three routes that changed the URL and sent an HTTP redirect, restarting the process. So if Static didn't find, say foo.html, It'd reach the route, append another .html and restart the process, throwing the whole thing in a loop. With my change, when Static fails to find the file, I modify the URL and tell Static to ser 16:06
ve it. And if that file isn't there you get a 404 instead of that redirect loop.
psch i know, cause i did lots of mistake when i still had to PR :P
timeless ZoffixW: ok, thanks for the explanation 16:07
psch (well, also afterwards, but those were also largely free of being torn into :) )
s/largely/completely/ # i suppose
at worst it was "isn't that wrong? can you fix it?" iirc 16:08
16:08 rurban_ left
perlpilot psch: we're saving it up for that one time where the mistake is so big that we need to remind you of all of the little mistakes you've made over the years ;) 16:09
yoleaux 9 Jun 2016 21:12Z <AlexDaniel> perlpilot: right now it does not handle gists and the way it handles long-running programs is non-ideal. But you can send your stuff to me and I'll run it in the terminal
timeless here's my current repo www.irccloud.com/pastebin/Z1NztIr1/ 16:10
mst: was i supposed to remind you to look at something? :)
psch perlpilot: please no :/ 16:11
timeless if not, then i should probably start using yoleaux for reminding, because my memory sucks :)
mst oh, yeah, the design.perl6.org
timeless it's a good thing your memory is better than mine 16:12
mst anybody know which repo that's built from?
ZoffixW And the // and the rest of the ilk need a more serious fix. The apache directive would only solve those links that actually have files generated, but begin with a dot. The // doesn't even have a file. I
psch perlpilot: but i guess that proves my point, anyway. it still remains friendly, even with eager but only barely competent devs like me :)
ZoffixW I'm still leaning for making that into slash-slash.html and basically changing all iffy punctuation into words
perlpilot mst: it doesn't say on the page somewhere?
mst timeless: remembering the thing isn't my problem. remembering to remember the thing while in front of a computer is my problem.
timeless mst: default = git://github.com/perl6/specs ?
mst aha 16:13
ta
ZoffixW Yes, that's the repo and there's a script in Mu that runs the update: github.com/perl6/mu/blob/master/ut...rl6.org.sh
Keep in mind: perl.css is used on multiple websites
ZoffixW learned that the hard way :) 16:14
timeless heh
ZoffixW: did you add a comment to perl.css to help the next poor soul?
ZoffixW: i favor slash-slash and similar
it makes google searches much more practical 16:15
... unfortunately, a tiny spider just crawled into my keyboard
dalek ecs: 55ef20a | (Zoffix Znet)++ | html/perl.css:
Add a note about used on multiple sites.
timeless is outside
ZoffixW There is now :) 16:16
timeless bonus points for considering listing the actual sites...
i know i'm greedy
.*
timotimo really? .perl.org?
not just .perl6.org?
timeless but, i'm the kind of person likely to change such a file 16:17
timeless is likely to do just about anything
oh well, i think the tiny spider will probably live/die in this keyboard
dalek ecs: f9061b3 | (Zoffix Znet)++ | html/perl.css:
Clarify note
ZoffixW Nah, it was just a typo :)
timeless arg, why do i keep missing phone calls? 16:18
ZoffixW Because it's 2016 and no one uses "phone calls" anymore? :)
Heh, the Womble thing is now at 8 contributors :) 16:21
16:21 donaldh joined
ZoffixW *participants 16:21
psch but what *color* should the womble be?
RabidGravy Underground, overground, wombling free
ZoffixW psch, that's the real question! 16:22
RabidGravy well they're mostly a sort of mousey brown colour, but there's one ginger one IIRC
stmuk_ there are at least two North American Wombles - one of which is Canadian according to www.tidybag.uk/information/whos-who...e-wombles/ 16:24
RabidGravy I think we should branch out to tellytubbies however, it's incontrovertible which colours they are 16:25
16:28 ZoffixW left 16:29 sno left 16:30 domidumont joined 16:31 raiph left 16:34 beatdown is now known as mosh, mosh left, mosh joined
mst oh, excellent, it's mostly pod except the pod6 bits 16:35
arnsholt Assuming you mean the specs/synopses/whateverthey'recalledthesedays, yeah I believe that was a concious decision precisely to make them renderable 16:37
Xliff Here's a question. If I am trying to pull a version constant from a c-lib, how would I do that with NativeCall? 16:40
Of course, full disclosure -- I don't see the symbol name I'm expecting via readelf
timeless i'm not sure those are generally there
Xliff I was afraid of that.
timotimo how is it defined?
Xliff Via #define
timotimo yeah, those don't show up by themselves 16:41
timeless definitely not available
Xliff :S
timotimo i.e. they disappear before the compiler gets to the file
timeless it would have to be `int x=...`
timotimo if you don't have the header that goes with it, you're basically SOL
Xliff Yeah. The header is available, tho.
timotimo and of course you can use a header that has a different version definition than the one your .so file was compiled with, and you wouldn't notice.
Xliff Right.
Well that sucks. 16:42
timotimo if your library doesn't have a "get the version we've linked against" function ... too bad
Xliff Ooh! Will check. Thanks.
dalek c: f0d7c81 | (Aleks-Daniel Jakimenko-Aleksejev)++ | template/footer.html:
Add © symbol to the copyright notice

Fixes #581
16:43
16:43 _mg_ left
[Coke] Regarding the double <a><a>'s in the htmlify'd pod for docs - I golfed down the generated Pod:: object to a snippet that shows the double linking. Hopefully this can help someone trim up the pod we're generating to avoid the duplicative link, or fix a bug in Pod::To::HTML (if that's where the issue lies): gist.github.com/coke/bae3843be5593...dba3afbb5e 16:45
Xliff Shucks! Doesn't.
[Coke] wonders if the word copyright should be capitalized. 16:46
(when used in a copyright message)
Xliff OK. How do you properly capitalize that? 16:49
Google says just "Copyright" and not "CopyRight"
timotimo yeah
Xliff Of course, I could be riding too many camels.
timotimo maybe with small capitals :P 16:50
dalek c: e8811b5 | (Aleks-Daniel Jakimenko-Aleksejev)++ | template/footer.html:
License notice should link to the actual license text

Fixes #580
16:51
ugexe m: class Foo { has %!bar; method baz { %!bar //= :a(1), :b(2); }; }; say Foo.new.baz # tricked me
camelia rakudo-moar 9cfcb9: OUTPUT«({} b => 2)␤»
16:52 rurban joined
AlexDaniel [Coke]: perhaps add a link to your gist here github.com/perl6/doc/issues/561 16:52
ugexe m: class Foo { has %!bar; method baz { %!bar = :a(1), :b(2); }; }; say Foo.new.baz
camelia rakudo-moar 9cfcb9: OUTPUT«{a => 1, b => 2}␤»
psch m: class Foo { has %!bar; method baz { %!bar //= %(:a(1), :b(2)); }; }; say Foo.new.baz 16:53
camelia rakudo-moar 9cfcb9: OUTPUT«{}␤»
psch m: say % // "foo"
camelia rakudo-moar 9cfcb9: OUTPUT«{}␤»
16:53 rurban_ joined
ugexe right, but its a trick in that you could look at it and think it should work, especially given the second example 16:54
psch yeah, looks pretty much like a trap to me 16:56
16:56 Sound joined
[Coke] AlexDaniel: my golf may have been too extreme. :| 16:57
AlexDaniel omg this womble discussion…
I want to nuke the whole issue to prevent further bikeshedding, but people are making some good points… 16:58
/o\
16:58 mr-foobar joined, kurahaupo joined
[Coke] opened github.com/perl6/doc/issues/584, subset of 561 16:59
Xliff .tell FROGGS Ran into an issue with pointers vs references in one of the tests for XML::LibXML. Let me know when you have some time to chat.
yoleaux Xliff: I'll pass your message to FROGGS.
[Coke] (to give us something smaller and actionable) 17:00
AlexDaniel [Coke]: ah, good!
17:00 Sound left
AlexDaniel [Coke]++ 17:00
FROGGS Xliff: is there a small piece of code that triggers it? 17:02
yoleaux 16:59Z <Xliff> FROGGS: Ran into an issue with pointers vs references in one of the tests for XML::LibXML. Let me know when you have some time to chat.
17:02 maddingue joined 17:03 TreyHarr1 joined 17:05 firstdayonthejob joined 17:06 TreyHarris left
Xliff FROGGS: Yeah. I'll gist it up later tonight. 17:07
17:07 rurban left
timeless [Coke]: thank for the doubled thing 17:08
there were a couple of things the w3 link checker screamed about
timeless should dump its output somewhere
doc.perl6.org link checker report www.irccloud.com/pastebin/kKy7oBRG/ 17:09
the issue to replace tokens [.?/] with words would fix a bunch 17:10
AlexDaniel timeless: yeah. Yeah. Yeah.
17:10 ZoffixW joined
timeless also, at some point we (hey, i'm using we) need to standardize on `.html` or not 17:10
the fact that the htmlify script generates both forms is annoying 17:11
timeless hasn't spent the time figuring that out
ZoffixW timeless, is that a motion to volunteer fixing it? :) I think it was pointed out as a problem even before last Christmas but the issue was drowned in bikeshedding I think and no one is yet to fix it.
timeless ZoffixW: it's half a motion to
ZoffixW AlexDaniel, I'm laughing at the Womble issue :) But mostly because Womble is a funny word :D
timeless i'm pretty sure i could do it
ZoffixW timeless++
timeless it's a pretty easy map in htmlify 17:12
also [#]
ZoffixW Sweet. Do it.
timeless i'm not sure about [$]
AlexDaniel do what? Change these symbols into words?
timeless yeah
n.b. not today -- i need to go north to get my gf and drive her here for the weekend
ZoffixW AlexDaniel, yes, so, say, docs.perl6.org/routine/// would be docs.perl6.org/routine/slash-slash
timeless this weeken's a holiday
ZoffixW And actually work as a URL
timeless so i'm gone until monday (some wouldn't return until tuesday)
AlexDaniel in #155 I recommended to use www.muppetlabs.com/~breadbox/interc...nsila.html 17:13
timeless and i'm traveling for an interview thursday
AlexDaniel that would be routine/mesh-mesh
timeless so i /might/ be able to look M/T, but otherwise possibly until +2w
ZoffixW /slat-slat 17:14
AlexDaniel ah, oops
timeless AlexDaniel: `half-mesh` bothers me
AlexDaniel yea, slat-slat
timeless a *lot*
[Coke] updated github.com/perl6/doc/issues/584 with a slightly better snippet of Perl6 code demonstrating the problem. might be a P2H bug.
timeless double-oh-seven also bothers me
ZoffixW My issue with that is I know a '/' is a slash, so I can type the URL by hand, but now I'd have to learn that whole table.
AlexDaniel timeless: bothers you in what sense? It is funny
timeless in fact, anything w/ `-`s in it will make doubled words painful
ZoffixW OTOH that's really more bikeshedding and we need to fix it at least with *something*
timeless AlexDaniel: double-oh-seven-double-oh-seven
how does one parse that? 17:15
percent-percent is parsable
also, `right angle` isn't doing anyone any favors
since space has another use
AlexDaniel use rightangle instead
timeless personally, going back to another discussion, i'd favor en-us terms over odd ones 17:16
AlexDaniel I'm not going to insist that we must use intercal terminology 17:17
but it is much funnier than wombles
timeless heh
ZoffixW mst, what was the stupid thing you wanted to ask me? It now bothers me that I don't know. RE twitter.com/shadowcat_mst/status/7...3941015556
stmuk_++ # for killing wombles :) 17:18
mst now it's bothering me that I don't remember
trout.me.uk/mstcat2.jpg
AlexDaniel well no, I have a commit comming that will actually kill wombles… though it will make everyone happy I hope
RabidGravy it's Friday and after six, if you're doing remembering properly then you're not doing it right
timeless m: my %tokenmap = ( % => percent, # => hash, . => dot, ? => question, ); print %tokenmap[.]; 17:19
camelia rakudo-moar 9cfcb9: OUTPUT«5===SORRY!5=== Error while compiling /tmp/T49XvXStDJ␤Unable to parse expression in parenthesized expression; couldn't find final ')' ␤at /tmp/T49XvXStDJ:1␤------> 3t, ? => question, ); print %tokenmap[.];7⏏5<EOL>␤ expecting any of:␤ …»
timotimo no, you need to quote those
RabidGravy KILL WOMBLES? BAD BAD HEARTLESS MAN
timeless timotimo: that was xargs not doing me any favors
ZoffixW m: my %tokenmap = <% percent # hash . dot ? question>; print %tokenmap{.}; 17:20
camelia rakudo-moar 9cfcb9: OUTPUT«5===SORRY!5=== Error while compiling /tmp/d1smp7IXAw␤Malformed postfix call␤at /tmp/d1smp7IXAw:1␤------> 3 . dot ? question>; print %tokenmap{.7⏏5};␤»
ZoffixW m: my %tokenmap = <% percent # hash . dot ? question>; print %tokenmap{'.'};
camelia rakudo-moar 9cfcb9: OUTPUT«dot»
timeless m: my %tokenmap = ( '%' => 'percent', '#' => 'hash', '.' => 'dot', '?' => 'question', ); print %tokenmap['.'];
camelia rakudo-moar 9cfcb9: OUTPUT«Cannot coerce NaN to an Int␤ in block <unit> at /tmp/h2ZQYS4VAA line 1␤␤Actually thrown at:␤ in block <unit> at /tmp/h2ZQYS4VAA line 1␤␤»
timeless there, that's the error i got :)
ZoffixW timeless, that only works in JavaScript or PHP :)
psch m: my %tokenmap = ( '%' => 'percent', '#' => 'hash', '.' => 'dot', '?' => 'question', ); print %tokenmap<.>; 17:21
camelia rakudo-moar 9cfcb9: OUTPUT«dot»
ZoffixW timeless, hashes are looked up with %hash<foo> or %hash{'foo'}
timotimo ZoffixW: isn't # actually comments inside < >?
psch m: say <# foo>
camelia rakudo-moar 9cfcb9: OUTPUT«(# foo)␤»
ZoffixW I hope not
timeless ZoffixW: yeah, i'm a js guy at heart
timotimo oh, ok
psch m: say <<# foo>>
camelia rakudo-moar 9cfcb9: OUTPUT«5===SORRY!5=== Error while compiling /tmp/wTXSi7ITxR␤Unable to parse expression in shell-quote words; couldn't find final '>>' ␤at /tmp/wTXSi7ITxR:1␤------> 3say <<# foo>>7⏏5<EOL>␤ expecting any of:␤ argument list␤ sh…»
timeless could camelia replace `/tmp/..../` with `script`? 17:22
it eats too many chars from the output :(
AlexDaniel timeless: good point
timotimo i wanted that, too, at some point
AlexDaniel timeless: create an issue here github.com/perl6/evalbot/issues
ZoffixW timeless, well, she's here if you want to take a crack at it :) github.com/perl6/evalbot/
AlexDaniel and I'll take care of it 17:23
psch replace it with "{$user}\'s script" imo
well, minus the \
timeless psch: too many chars
psch timeless: but useful information
17:23 mr-foobar left
ZoffixW Well, the ===SORRY!=== stuff can go too. 17:23
AlexDaniel go where?
ZoffixW This isn't Canada, eh
Go away :)
AlexDaniel no
I have to know that it was a compile-time thing 17:24
psch yeah, i'd prefer only changing meaningless into meaningful information
timeless AlexDaniel: github.com/perl6/evalbot/issues/8
psch not ditching meaningful information
timeless ZoffixW: i'm in canada, i wouldn't be sorry to see that sorry go
psch and "too many chars" isn't really a valid argument for me there - you probably have text replacement functionality in your irc client
timotimo psch: camelia cuts off text before your irc client gets it 17:25
timeless psch: it's eating from the tail of the error
ZoffixW Replace it with a shorter meaningful version. The q{Replace ===SORRY!=== Error while compiling /tmp/wTXSi7ITxR␤} can be replaced with a single char really.
timeless there's a limit in irc
ZoffixW timeless, where in Canada?
timeless TO
AlexDaniel no, SORRY is not going anywhere
17:25 TakinOver left
psch timeless: well, if camelia is *all* you're using to perl6 for then i'm sorry vOv 17:25
ZoffixW timeless, Brampton. Are you going to Damian's Perl 6 talks June 27/28?
timeless ZoffixW: link?
ZoffixW 1 sec 17:26
timeless, first www.meetup.com/Toronto-Perl-Monger...231418224/ and second www.meetup.com/Toronto-Perl-Monger...231418278/
timeless is there a way to get .vcal / add to google calendar links from there? 17:27
ZoffixW IDFK
timeless could definitely go (assuming no other scheduling things turn up)
timeless could walk there
ZoffixW timeless, there isn't, but if you're not giving any data to the server, it'll ping you and failing to respond to that ping will disconnect you. RE <timeless> there's a limit in irc 17:28
ping you as in IRC command PING
17:28 donaldh left
timeless ZoffixW: wrong kind of limit 17:28
there's a message length limit 17:29
ZoffixW Oh, I think I've commingled your nickname with your question. I thought you were asking about time limit lol
timeless (although it technically varies by ircd)
psch i do recognize that camelia could get closer to freenodes MSG limit, yeah
ZoffixW There's a 512-char per message limmit give or take
timeless anyway, it's pretty clear that on that thing i complained about camelia hit it
17:29 ZoffixW left
timeless shortening /tmp (twice) buys characters for more useful stuff 17:29
AlexDaniel yes, that's a good point 17:30
that does not apply for removing SORRY, that's a bad point
psch irclog.perlgeek.de/perl6/2016-06-10#i_12643274 seems to be the freenode msg length limit
so camelia could expand to that, instead of tossing useful information 17:31
timeless oh
i can't generally do mondays
timotimo we could replace ===SORRY!=== with ==SRY==
dalek c: 64bc5f3 | (Aleks-Daniel Jakimenko-Aleksejev)++ | doc/Language/testing.pod:
Less wombles, more butterflies

Fixes #577
timeless 17:32
AlexDaniel sorry to all fans of wombles
timeless ok, i really have to go 17:34
maybe see some real people that tuesday @moco community space
AlexDaniel: that explanation is *much* better 17:35
17:35 ZoffixW joined
AlexDaniel timeless: see you! 17:35
timeless it actually *helps* people understand about camelia !
killing two birds w/ one stone == huge win
ZoffixW psch, that's HexChat's message limit :)
AlexDaniel yea, it's cool. Now everyone shut up about this thing
psch ZoffixW: well, it's still definitely lower than freenodes limit, in any case :)
well, or equal to, i suppose
17:36 TelnetZoffix joined
TelnetZoffix dassdasdasdsadasdasddasdsadsadsadsadsadsadasddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddd 17:36
ddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddd
ZoffixW That's the max
17:37 TelnetZoffix left
ZoffixW Which is probably the same as the message psch linked to making my spam effectively useless :P 17:37
psch nah, it looks longer from here i think? 17:38
well, there's 448 'd's
and 'my $max_output_len = 290;' is camelia code 17:39
ZoffixW Aye this part ddddddddddddddddddddddddddddddddd" doesn't fit when I try to send the same thing from HexChat. The nick is part of the max length tho
17:40 kent\n joined, kent\n left, kent\n joined
ZoffixW Hey, AlexDaniel ... 17:40
...
AlexDaniel what
mst the channel name is definitely part of the max length
ZoffixW ... Womble *trollface*
psch m: say chars 'Zoffix@static-67-226-172-41.ptr.terago.net msg #perl6'
camelia rakudo-moar 9cfcb9: OUTPUT«53␤»
psch m: say chars 'Zoffix@static-67-226-172-41.ptr.terago.net msg #perl6' + 448
camelia rakudo-moar 9cfcb9: OUTPUT«Cannot convert string to number: base-10 number must begin with valid digits or '.' in '3⏏5Zoffix\@static-67-226-172-41.ptr.terago.net msg #perl6' (indicated by ⏏)␤ in block <unit> at /tmp/bQz4v1Ii_2 line 1␤␤Actually thrown at:␤ in block <…»
psch m: say chars('Zoffix@static-67-226-172-41.ptr.terago.net msg #perl6') + 448
camelia rakudo-moar 9cfcb9: OUTPUT«501␤»
psch well, i'm not learning RFC 1459 for that now :P 17:41
17:41 domidumont left
psch in any case, it probably also depends on the other networks camelia is on, if any 17:41
ZoffixW Yeah, the servers shove around things like :Angel!wings@irc.org PRIVMSG #perl6 :MESSAGE 17:43
And it's RFC2812 now :)
AlexDaniel ==SRY== eh… by using the same logic people can say that dalek should replace my name with something shorter…
ZoffixW Finally, my hours spent reading all of them for IRC::Client were not in vain! :P 17:44
timotimo yeah, AD, shorten your name, dood
AlexDaniel A.-D. J.-A. eh
something like that is written on my bank card…
gnull hello everyone 17:45
timotimo greetings gnull
17:46 telex left
gnull m: sub f([$a, [$b, $c]]) {say "$a, $b, $c"}; f [1, [2, 3]] 17:46
camelia rakudo-moar 9cfcb9: OUTPUT«1, 2, 3␤»
gnull Looks like this [] constructor can also decompose elements 17:47
Is it possible to do this with arbitary classes?
timotimo yes, with a sub-signature and named parameters you can grab attributes out of instances 17:48
psch m: class Foo { has $.bar }; sub f(Foo $a (:$bar)) { say $bar }; f Foo.new(:2bar)
camelia rakudo-moar 9cfcb9: OUTPUT«2␤»
17:48 telex joined
gnull cool 17:48
timotimo m: sub test($p (:$key, :$value)) { say "$key $value" }; test "foo" => "bar"
camelia rakudo-moar 9cfcb9: OUTPUT«foo bar␤»
ZoffixW 0.o 17:49
perlpilot because I find that required space slightly annoying, I've taken to using a colon instead ...
m: sub test($p:(:$key, :$value)) { say "$key $value" }; test "foo" => "bar"
camelia rakudo-moar 9cfcb9: OUTPUT«foo bar␤»
RabidGravy yeah, I didn't realise the generalised version of this existed until the other day when I was thinking about pairs
perlpilot Same for Foo $a:(:$bar)
psch m: class Foo { has $.bar }; sub f(Foo $a(:$bar)) { say $bar }; f Foo.new(:2bar)
camelia rakudo-moar 9cfcb9: OUTPUT«5===SORRY!5=== Error while compiling /tmp/6gsB4OVP1n␤Shape declaration with () is reserved;␤ please use whitespace if you meant a subsignature for unpacking,␤ or use the :() form if you meant to add signature info to the function's type␤at /t…»
RabidGravy lorra lorra pairs out there 17:50
psch perlpilot: the error message says they mean different things..?
perlpilot: ohh, that's probably only for &-params
perlpilot without the space it means something different
psch somewhat misleading
gnull m: sub out(Proc $p (:$out)) {say $out}; output(shell 'echo hello', :out)
camelia rakudo-moar 9cfcb9: OUTPUT«5===SORRY!5=== Error while compiling /tmp/0fYLSLon8T␤Undeclared routine:␤ output used at line 1␤␤»
timotimo camelia doesn't let you use "shell"
gnull I want to extract :out from Proc this way
timotimo only works if out is declared as an attribute 17:51
psch m: sub f(&a:($,$)) { say "yup" }; f -> $, $ { } 17:52
camelia rakudo-moar 9cfcb9: OUTPUT«yup␤»
psch m: sub f(&a ($,$)) { say "yup" }; f -> $, $ { }
camelia rakudo-moar 9cfcb9: OUTPUT«Too few positionals passed; expected 2 arguments but got 0 in sub-signature of parameter &a␤ in sub f at /tmp/HNo34C6n6w line 1␤ in block <unit> at /tmp/HNo34C6n6w line 1␤␤»
psch perlpilot: i think :() working there for non-&-params is kind of weird
perlpilot why? 17:53
17:53 mr-foobar joined, sno joined
psch perlpilot: well, the forms aren't equivalent, and the error message suggests that :() is only for "signature info [of] the function's type" 17:53
i'm not willing to argue i understand that completely correctly, and it might just be fine as a fallback, but then at least the error for the reserved case is LTA vOv 17:54
ZoffixW m: class Wombles {method womble {self}}.womble.womble.womble.womble.womble.womble.womble.womble.womble.womble.womble.womble.womble.womble.say
camelia rakudo-moar 9cfcb9: OUTPUT«(Wombles)␤»
ZoffixW Well... I'm officially bored outta my mind :)
gnull Thank you 17:55
perlpilot psch: your example was making me think that not only was I right to favor $aggregate:($a,$b), but arguably "more correct" for the case where it's a &thingy
17:56 ZoffixW left
dalek c: 8ebabd4 | (Steve Mynott)++ | doc/Language/testing.pod:
Reverting since the consensus in #577 was clearly in favour of no change. In case of further argument this should be decided by either Pumpking or BDFL
17:57
AlexDaniel reverting? What the f?
psch m: sub f(&a (:$name)) { say $name }; f &say
camelia rakudo-moar 9cfcb9: OUTPUT«(Any)␤»
17:57 yqt left
psch perlpilot: but :() means "the &-param before has this signature" 17:58
perlpilot: and ' ()' means "destructure this param into what's inside the parens"
[Coke] stmuk++
psch ...i'm not quite sure why my last example gave Any, but i'd guess it's Attributes of the param, and not methods..? 17:59
RabidGravy I've totally lost the will to live but I now have "the sql generating thing that is loosely based on SQL::Abstract" down to 24 genuine test failures rather than a squillion due to not having a proper SQL parser
psch m: my &say-foo = &say but role { has $.a = "foo" }; sub f(&a (:$a)) { say $a }; f &say-foo 18:00
camelia rakudo-moar 9cfcb9: OUTPUT«foo␤»
psch m: my &say-foo = &say but role { has $.a = "foo" }; sub f(&a:(:$a)) { say $a }; f &say-foo
camelia rakudo-moar 9cfcb9: OUTPUT«5===SORRY!5=== Error while compiling /tmp/aYGfV0dqx9␤Variable '$a' is not declared. Did you mean '&a'?␤at /tmp/aYGfV0dqx9:1␤------> 3as $.a = "foo" }; sub f(&a:(:$a)) { say 7⏏5$a }; f &say-foo␤»
18:02 _mg_ joined
AlexDaniel just showed both examples to someone. Guess which example received “what the?” reaction and which one was “oh, ok, now it's clear” 18:02
timotimo yeah, that's not destructuring
perlpilot psch: oh, I see. I was misreading. I still find that required space annoying :)
18:03 cdg left
psch guesses we don't roast ($a:($b)) 18:04
so, well, it'll probably keep working in v6.c and v6.d..? :) 18:05
18:06 wamba joined, robinsmidsrod joined
mst AlexDaniel: and? 18:06
AlexDaniel: if you put the same effort into rewriting it while keeping the wombles, and then see if people still find it unclear, that would have a point 18:07
AlexDaniel: as it is, this is now twice that there's been an active discussion about this that you've tried to circumvent
AlexDaniel: you might even be right, but your behaviour has been antisocial and anticollaborative and so the merits have been obscured 18:08
[Coke] wonders what "needs fixing" indicates in the perl6/doc/issues queue. Surely everything in there needs fixing 18:10
ditto "affects content" vs. "needs writing"
mst [Coke]: "has been triaged and determined to be a doc error rather than a bug in rakudo" ?
AlexDaniel: nobody has said "this cannot be improved", only "we would like, if possible, to keep the wombles in the process" 18:12
I'm not sure why you're working so hard to ignore everybody else's opinion, but it doesn't appear to be a winning strategy right now
AlexDaniel mst: instead of discussing the thing in a team of 9 participants (where one half is saying that the example is hard to understand and another half is saying that “womble” is a funny word) I decided to change the example to something that is cool, very clear, not too weird, and still very perl6-ish. Closing the issue, as it has been done, was pretty unfair to one half (remember that this problem was originally pointed out by a person who 18:13
is new to perl 6). Instead I just fixed the example. That's it. You may say that it is unsocial, but it gets the job done
RabidGravy if only the same amount of effort went into making the missing documentation
AlexDaniel you may insist on keeping the funny word in the documentation, whatever. That's my last word on that topic, I'm not participating in this bullshit anymore 18:14
18:15 rurban_ left, tomboy64 left, rurban_ joined
mst AlexDaniel: if you'd presented the alternative as clearer-in-general and asked for comments, there would have been no 'bullshit' 18:15
18:15 telex left, telex joined
mst telling half the people involved in a conversation 'fuck you' is not generally a winning strategy 18:16
(and I do sympathise on the basis of really wanting to do that on a fairly regular basis, but it still doesn't tend to work out how I'd hope ;) 18:17
RabidGravy it's fine, we can stop writing the documentation as it's clearly all in hand
18:18 tomboy64 joined
[Coke] I'd like to point out that snark also hardly ever has the desired effect. 18:20
perlpilot Where's the hugbot when you need it? :) 18:21
AlexDaniel perlpilot: there's bisectbot instead
18:21 ZoffixW joined 18:22 huggable joined
moritz hugs perlpilot 18:22
thereyago
ZoffixW huggable, hug perlpilot
huggable hugs perlpilot
ZoffixW :)
perlpilot really, I'm not the one who needs it. 18:23
But it looks like all those involved in the womble discussion could use some hugs
RabidGravy eugh, if I thought it was that kind of channel
ZoffixW I hope we can all walk away from this amicably. There's one thing more stupid than bikeshedding and that's holding grudges over bikeshed issues.
huggable, hug me
huggable hugs ZoffixW
perlpilot ZoffixW: so you're calling everyone stupid now?!?! ;> 18:24
RabidGravy we are all stupid
mst we are all stupid
18:24 yqt joined
ZoffixW Fun fact: googling for "how to fight bikeshedding" gives you results for "how to fight climate change" :) 18:24
RabidGravy actually that's probably right 18:25
[Coke] btw, AlexDaniel++ for quickly resolving all the doc tickets I opened today
18:25 labster joined
AlexDaniel these were LHFs 18:25
RabidGravy most of the "climate conferences" do seem to be mass colour swatch discussion fests
dogbert17 here's my attempt to document Baggy.keys, does it look ok? gist.github.com/dogbert17/acbbd419...cf4402ac81 18:26
[Coke] still, thanks. They were high enough I didn't have time to do 'em. :)
DrForr lib/.precomp bug golfed down reasonably far, and the autoreply did work. 18:27
RabidGravy dogbert17, looks great but <supressed sarcastic comment> ;-)
dogbert17 RabidGravy: is it the spam? 18:28
AlexDaniel RabidGravy: yea… I was about to say that :D
RabidGravy I'd just go with it then be prepared to defend the monty python allusion
perlpilot dogbert17: I dunno. "without taking their individual weights into account" seems like it might warrant some more explanation because of the second example. 18:29
dogbert17: I mean, naively, if I didn't know anything, the first question I'd ask would be "what happened to a => 0?"
but maybe that's just in the main BagHash docs? 18:30
ZoffixW m: my $n = ("a" => -1, "b" => 2).BagHash; say $n.keys # seems to freeze
Oh, never mind. camelia is just ded 18:31
18:31 andreoss joined
perlpilot ZoffixW: here, I get a LTA error. 18:31
dogbert17 perlpilot: the second example was actually intentional, zero weight keys are removed but perhaps I should mentions that
s/mentions/mention/
ZoffixW ditto. I just assumed it was a freeze, 'cause camelia didn't respond when I tried it :)
dogbert17 perlpilot: i'll mention something about it 18:32
perlpilot dogbert17++
camelia rakudo-moar 9cfcb9: OUTPUT«Use of Nil in string context in block <unit> at /tmp/v40TeDI8Vs line 1␤Use of Nil in string context in block <unit> at /tmp/v40TeDI8Vs line 1␤Found negative values for in ␤ in block <unit> at /tmp/v40TeDI8Vs line 1␤␤Actually thrown at:␤ in…»
18:34 grondilu left 18:35 domidumont joined 18:42 Actualeyes left
andreoss m: our $x = 1; { my $y := $x; temp $x = $y; $y++; say $x }; say $x; 18:42
camelia rakudo-moar 9cfcb9: OUTPUT«2␤1␤»
andreoss m: our $x = 1; { my $y := $x; $y++; say $x }; say $x;
camelia rakudo-moar 9cfcb9: OUTPUT«2␤2␤»
18:43 ZoffixW left, donaldh joined 18:44 apathor joined
andreoss m: our $x = 1; { INIT my $y := $x; temp $x = $y; $y++; say $x }; say $x; 18:44
camelia rakudo-moar 9cfcb9: OUTPUT«Cannot resolve caller postfix:<++>(); none of these signatures match:␤ (Mu:D $a is rw)␤ (Mu:U $a is rw)␤ (Int:D $a is rw)␤ (int $a is rw)␤ (Bool:U $a is rw)␤ (Bool:D $a is rw)␤ (Num:D $a is rw)␤ (Num:U $a is rw)␤ …»
andreoss why temp is executed before my? 18:45
18:46 domidumont left 18:47 itaipu joined
andreoss m: our $x = 1; { my $y := $x; let $x = $y; $y++; say $x }; say $x; 18:48
camelia rakudo-moar 9cfcb9: OUTPUT«2␤2␤»
andreoss why temp and let don't have the same behavior here? 18:49
perlpilot m: our $x = 1; try { my $y := $x; let $x = $y; $y++; say $x; fail }; say $x 18:55
camelia rakudo-moar 9cfcb9: OUTPUT«2␤1␤»
andreoss i see 18:57
let keeps the value
m: our $x = 1; { my $y := $x; { temp $x = $y; $y++; say $x } }; say $x; 18:58
camelia rakudo-moar 9cfcb9: OUTPUT«2␤1␤»
18:59 regreg left
andreoss m: our $x = 1; my $y := $x; { temp $x = $y; $y++; say $x }; say $x; 19:00
camelia rakudo-moar 9cfcb9: OUTPUT«2␤1␤»
19:01 sherwool joined 19:02 _mg__ joined
gnull I'm reading source of JSON::Tiny. And can't understand what does ':sym<something>' appended to token name do. For example here: github.com/moritz/json/blob/master...mar.pm#L12 19:03
Or is it just arbitary part of token name?
19:04 _mg_ left, _mg__ is now known as _mg_
timotimo no, it's actually more than that 19:05
when you have a <sym> in your token, it'll match the literal that you have in the :sym<...> part
so you can have multi token value:sym<True> { <sym> }; multi token value:sym<False> { <sym> }
and you can't accidentally copy-paste that wrong
dj_goku_ 00~design.perl6.org/S05.html01~ maybe?
timotimo huh, how did that happen? 19:06
dj_goku_ design.perl6.org/S05.html
timotimo: good question.
copy + paste in irssi
timotimo i know vt100 has a "give me escape codes at the beginning and end of data copied in" setting that an application can turn on and off via an escape sequence 19:07
well, not vt100, but whatever some terminals implement
dj_goku_ timotimo: yeah it just started happening. 19:08
gnull dj_goku_: thanks for the link 19:09
dj_goku_ gnull: I don't eve know if it is valid. :D
gnull timotimo: thanks you too :)
dj_goku_ is a master googler
gnull It is better than nothing :) 19:10
^^^ Someone told this about open-source
dj_goku_ wow that grammar is very short. 19:12
gnull I found a more beginner friendly description of these ':sym's. They are called protos. en.wikibooks.org/wiki/Perl_6_Progr...g/Grammars
19:13 spider-mario joined, cdg joined 19:18 darutoko left
perlpilot gnull: Well ... sym isn't the proto, it's more of a modifier for the proto. In the example at that link, the proto is "protocol" and the sym is an expression of that proto (one of http, https, etc) 19:23
gnull Got it 19:24
perlpilot gnull: see doc.perl6.org/syntax/proto for more about proto 19:25
gnull m: my proto token expr {*}; my token expr:sym<plus> { \d+ '+' \d }; '1+1' ~~ &expr
camelia ( no output )
19:25 donaldh left
moritz gnull: use a 'say' before the last expression 19:25
m: my proto token expr {*}; my token expr:sym<plus> { \d+ '+' \d }; say '1+1' ~~ /<expr>/ 19:26
camelia rakudo-moar 9cfcb9: OUTPUT«Nil␤»
gnull m: my proto token expr {*}; my token expr:sym<plus> { \d+ '+' \d }; say '1+1' ~~ &expr
camelia rakudo-moar 9cfcb9: OUTPUT«Nil␤»
19:26 skids left
perlpilot doc.perl6.org/language/regexes could probably use a section on protos too 19:27
gnull I expect it to match. What is wrong? 19:28
perlpilot or maybe doc.perl6.org/language/grammars
gnull m: my token expr { \d+ '+' \d }; say '1+1' ~~ &expr 19:30
camelia rakudo-moar 9cfcb9: OUTPUT«「1+1」␤»
perlpilot I dunno ... maybe the proto machinery has a bug. 19:33
m: grammar G { proto token expr {*}; token expr:sym<plus> { \d+ '+' \d+ } }; say G.parse('1+1', :rule<expr>); 19:34
camelia rakudo-moar 9cfcb9: OUTPUT«「1+1」␤»
perlpilot well, that works at least :)
gnull Very strange
timotimo yeah, i think proto only makes sense inside a grammar 19:40
or maybe it's just implemented that way at the moment
19:42 donaldh joined
gnull ok 19:42
perlpilot Hmm... maybe proto requires a package 19:44
timotimo well, the unit of he running script is also a package, to be fair 19:45
AlexDaniel oh look: $tmp_res =~ s|/tmp/\w{10}|/tmp/tmpfile|g; 19:46
something is not working here :)
perlpilot timotimo: but he used "my proto token" which is lexical and not bound to a package.
timotimo mhm 19:50
*shrugs*
diakopter dares to wonder about the womble discussion 19:51
19:53 sherwool left
diakopter gives up backlogging to find information on it 19:53
nemo hugs huggable 19:54
hmph
it needs more functions
AlexDaniel nemo: like?
nemo AlexDaniel: didn't hug me back 19:55
also maybe random hugs or based on commits?
AlexDaniel huggable: source
huggable AlexDaniel, See github.com/zoffixznet/huggable
AlexDaniel nemo: ↑
masak diakopter: irclog.perlgeek.de/perl6/2015-03-18#i_10299276 seems to be some sort of originating moment
but I could be wrong 19:56
mst diakopter: it can be summed up as "people think the womble example is confusing, other people agree but want to keep the wombles, AlexDaniel proposed a version that fixed the confusing but lost the wombles, then we all made idiots of ourselves" 19:57
AlexDaniel that sums it up, yes
RabidGravy I'd put it down to not enough beer
[Coke] ah, beer. 19:58
AlexDaniel though it feels like we're going to act like even bigger idiots for the next few hours…
anyway, why is this substitution not working…
mst AlexDaniel: I'm assuming that either (a) somebody will come up with a deconfused womble version (b) in a few days we'll shrug, note the absence of a, take a moment to feel sad about the absence of a, and adopt your version 19:59
AlexDaniel I'm actually proposing a third option 20:00
RabidGravy I'd rather someone wrote a half decent sql parser
AlexDaniel let's just stop the bikeshedding for 3 days, then think again…
20:01 andreoss left
mst well this was kinda the 'in a few days' part 20:01
you will notice I'm attempting to carefully avoid restarting the bikeshedding in the way I'm describing things :)
20:02 jameslen_ joined
mst ... meanwhile a .uk-centric channel is discussing stealth womblification of other documentation to compensate 20:02
AlexDaniel :|
20:03 cpage_ left
RabidGravy stealth wombles! 20:03
20:04 maybekoo2 left 20:06 rurban joined
b2gills There is also a discussion of stealth Fraggle-Rock-ification of the docs by people who were kids in the 80's ( currently only in my own head ) 20:06
20:06 rurban left
mst YES! 20:07
b2gills: beware the MONKEY-WRENCH
RabidGravy oh if we're going there I want clangers, trumpton, flowerpt mean the works
20:08 donaldh left
RabidGravy (for those of us who were born in the sixties and all) 20:08
20:09 maybekoo2 joined 20:10 rurban_ left
timotimo we should preface the docs website with a selection "in what era did you grow up?" 20:14
or "which of these kids' shows do you remember from your own childhood?"
[Coke] and what's your DVD region?
AlexDaniel and radio buttons to toggle between, yeah 20:15
timotimo :D
stmuk_ the w***** actually had 1970s and 1990s versions 20:17
20:18 rurban joined 20:20 rurban left
RabidGravy yeah the 90s one was a bit messed up 20:22
like the recent magic roundabout 20:23
mst stmuk_: next you're going to try and tell me there's more than one highlander film
masak mst: that's impossible! 20:27
there can only be One!
mst www.trout.me.uk/facepaw.jpg 20:28
20:28 cpage_ joined
geekosaur ob "there should have been only one!" 20:28
huf the number of highlander movies depends entirely on your will 20:29
it's similar to star wars, really
RabidGravy there is actually a TV series, which is kind of weird really 20:30
stmuk_ just as well there was only one Matrix film too
masak yeah, was gonna say 20:31
well. unless you count that Animatrix thing
[Coke] the 2nd matrix movie is ok if you combine it with the ps2 video game. :) 20:32
masak [Coke]: I finished that game. I beg to differ. 20:34
[Coke] aww, I have fond memories of it. perhaps only becaues I played it so long ago 20:35
masak refreshes his memory with a YouTube video 20:36
generally the rule is "if you try to make a game from your movie, you've already lost"
[Coke] ah, but as I recall, the game actually covered bits that were not present in the movie. 20:38
as opposed to the normal "play the scenes from the movie" tack.
masak that much is true
whether you consider it an advantage (for the movies, mostly) is another matter 20:39
mst I actually quite liked matrix 2 and 3, but I feel like they were effectively disjoint from the first one
masak mst: agreed. 20:41
to me, 1 already forms a coherent whole. 1+2+3 sort of do too, but a different one.
RabidGravy mst, but you've got the long black leather coat, so you have to say that ;-)
masak some of the things 2+3 added to 1 kind of disappointed me. 20:42
20:42 TEttinger joined
masak the polygon count in this game reminds me of the "Money for Nothing" music video 20:42
mst RabidGravy: augh that was my least favourite thing about those films 20:43
RabidGravy: "did you want to be like the guy in the matrix?" NO I BOUGHT THIS BEFOREHAND I AM A FUCKING GOTH WHAT THE FUCKING FUCK
RabidGravy: I hate you
<3
RabidGravy :-* 20:44
AlexDaniel uh, I know that I was screaming words like “bullshit” today… but that ↑ is a little bit too much :)
RabidGravy the goth thing may besomewhat of pyrrhic victory however ;-) 20:45
stmuk_ why do you dress like it's halloween? :P
RabidGravy because it's always halloween in his heart
AlexDaniel OK, so the tempfile was actually supposed to be renamed to /tmp/tmpfile
should I leave it as that or can we come up with something shorter? 20:46
p6: say(
camelia rakudo-moar 9cfcb9: OUTPUT«5===SORRY!5=== Error while compiling /tmp/tmpfile␤Unable to parse expression in argument list; couldn't find final ')' ␤at /tmp/tmpfile:1␤------> 3say(7⏏5<EOL>␤ expecting any of:␤ argument list␤»
AlexDaniel see ↑? It works here
m: say(
camelia rakudo-moar 9cfcb9: OUTPUT«5===SORRY!5=== Error while compiling /tmp/AlBHtzb9vf␤Unable to parse expression in argument list; couldn't find final ')' ␤at /tmp/AlBHtzb9vf:1␤------> 3say(7⏏5<EOL>␤ expecting any of:␤ argument list␤»
AlexDaniel but given that the initial complaint was that it is too long, can I rename it to <tmp> ? 20:47
p6: say ‘/tmp/abcdefghij’ 20:50
camelia rakudo-moar 9cfcb9: OUTPUT«/tmp/tmpfile␤»
AlexDaniel xD
20:51 rindolf left
dalek c: 0d7aab4 | (Jan-Olof Hendig)++ | doc/Type/Baggy.pod:
Added docs for Baggy.keys. #perl6++ for reviewing
20:51
20:54 rurban joined 20:55 _mg_ left 21:01 Xliff left 21:07 vendethiel joined
dalek c: 02f12e2 | (Jan-Olof Hendig)++ | doc/Language/objects.pod:
Fixed broken link
21:09
albot: 282650f | (Aleks-Daniel Jakimenko-Aleksejev)++ | evalbot.pl:
Always shorten the filename, not just sometimes
21:12
albot: fa2e44f | (Aleks-Daniel Jakimenko-Aleksejev)++ | / (2 files):
Make it less likely to stumble upon file name replacement

Of course that's not the right way to fix it, but it is good enough for now. Ideally EvalbotExecuter should return the filename that was used, but current fix is probably good enough (even though it feels weird).
albot: 4ef2b55 | (Aleks-Daniel Jakimenko-Aleksejev)++ | evalbot.pl:
Make the filename shorter to fit more useful info

Original complaint was that the filename is too long and the error message is less useful because the filename appears several times and takes up all space. After 6ff428d the situation got better, but not much.
This commit changes the filename to <tmp>. That's very short but is probably still clear enough.
Closes #8
21:16 camelia joined
AlexDaniel m: say( 21:16
ah, it needs the +v
21:17 ChanServ sets mode: +v camelia
camelia rakudo-moar 5b4470: OUTPUT«5===SORRY!5=== Error while compiling /tmp/ti6cqH5Gse␤Unable to parse expression in argument list; couldn't find final ')' ␤at /tmp/ti6cqH5Gse:1␤------> 3say(7⏏5<EOL>␤ expecting any of:␤ argument list␤» 21:17
21:17 Xliff joined
AlexDaniel ok, it refuses to pull 21:18
right, it claims that the version of evalbot is e600617, which means that somebody touched the config file (freenode.org.conf) without making a proper commit 21:23
my guess is that right now it says “can't pull, you have unstaged changes here!”
diakopter But what acruallt Is a wOmble
AlexDaniel can somebody with access to camelia help with that? 21:24
diakopter camelia: what is my womble
I had enough beer for all of oyu
it makes me funnier 21:25
21:25 ptolemarch left
masak diakopter: "Wombles" is a British TV series from the 70s 21:30
diakopter: they look like people dressed up as silly birds 21:31
stmuk_ youtu.be/VIxkqoNi8I4?t=30 21:34
masak diakopter: stop-motion animation
AlexDaniel
.oO( I should turn it on whenever I see bikeshedding )
21:35
stmuk_ #womblegate
21:35 itaipu left 21:38 wamba left 21:41 regreg joined 21:42 kaare_ left, jdv79 joined
dalek c: 8929182 | (Jan-Olof Hendig)++ | doc/Language/objects.pod:
Fixed typo and missing directive
21:42
21:47 ZoffixMobile joined
ZoffixMobile Jesus. every time I look at the log, all you guys are talking about are wombles :) 21:47
AlexDaniel, did you mean it did not pull after issuing the restart command? 21:48
AlexDaniel ZoffixMobile: well, there's a pull command
ZoffixMobile: and it fails
ZoffixMobile: “version” reports e600617, which means that there are local modifications
ZoffixMobile oh :( 21:49
AlexDaniel because e600617 is before github.com/perl6/evalbot/commit/af...fcda824423
but as we all know camelia is in perl6-dev just fine
so, my conclusion is that somebody broke it :)
21:54 ZoffixMobile left
Hotkeys why does the windows repl hate unicode 22:07
i.imgur.com/cXOsiP6.png
works fine outside repl i.imgur.com/8J7wvNc.png 22:08
22:09 grondilu joined
literal so, Callable.assuming() is really slow 22:10
Hotkeys I don't really know where to look for the repl being unhappy with unicode
literal example: gist.github.com/hinrik/a8ce1c656d4...87004b0f67
geekosaur Hotkeys, backward compat pretty much. default code page is not unicode 22:11
chcp 65001
Hotkeys but the old repl handled it fine
geekosaur note that this will confuse some other programs (notably git)
oh, powershell vs. cmd? I suspect PS forces unicode even when the code page isn't 22:12
but the code page is the only way other programs have to figure out if they are supposed to be using unicode or not, so they become confused
Hotkeys is there any way to make it use unicode? 22:13
if I don't care about whatever backwards compatibility it's going for 22:14
geekosaur seems like Microsoft is pushing for non-unicode code pages to go away, so some future windows version should get this somewhat right
chcp 65001
it's a command
stmuk_ . o O ( making good use of the things that we find. Things that the everyday folks leave behind. )
Hotkeys chcp 65001 just doesn't let me type unicode at all 22:15
geekosaur now *that* sounds wrong
unless you are actually not typing unicode at all, or unless powershell is up to something odd (which I suppose it'd have to be if it's using unicode without changing the code page) 22:16
Hotkeys I paste ² or type ² and it just doesn't exist
idk
geekosaur I don't know windows sufficiently to help at that point, sorry
Hotkeys All I know is that unicode worked before the fancy new repl
masak so, about those wombles 22:17
Hotkeys I haven't been around much
so maybe it was something post-new-repl that made it borked
22:19 Sgeo_ joined
geekosaur [10 22:16] <Hotkeys> I paste ² or type ² and it just doesn't exist 22:20
that is something going wrong at windows level, not perl
Hotkeys it seems like the perl level 22:22
because I can type ² outside the repl still
22:22 huggable left, huggable joined 22:24 bjz_ left 22:25 aries_liuxueyang left 22:27 aries_liuxueyang joined 22:28 skids joined
Hotkeys geekosaur: i.imgur.com/MP0mofy.png 22:29
seems like the repl is doing something funky
oh 22:40
apparently it was something funky with rakudostar
I just built from source and everything is fine
22:40 rurban1 joined
Hotkeys ah 22:43
it's a linenoise thing
22:44 rurban left 22:45 bjz joined 22:47 regreg left
timotimo yes, linenoise doesn't understand unicode, sadly 22:48
Hotkeys is there an alternative that does? 22:49
timotimo Readline 22:50
if you don't have LineNoise installed, but you do have Readline, it'll take that
Hotkeys mmkay I'll try readline 22:51
can panda remove modules?
it doesn't seem to be able to
22:51 rurban joined, hoelzro|phone joined
Hotkeys oh apparently zef can 22:52
hoelzro|phone rakudo will prefer readline even if linenoise is installed
22:52 bjz left, FROGGS left
Hotkeys I guess I'll be grabbing readline 22:52
hoelzro|phone I think you can also specify RAKUDO_LINE_EDITOR, don't remember if I implemented that 22:53
timotimo ah, ok 22:55
22:55 rurban1 left
stmuk_ Hotkeys: rakudo can, zef can, panda can't and I wish I or someone else fixed panda 22:56
Hotkeys mmk 22:57
23:00 bjz joined, spider-mario left 23:01 rurban left 23:08 AlexDaniel left 23:09 ocbtec left 23:15 hankache joined
hankache hola #perl6 23:16
23:18 bjz left 23:25 firstdayonthejob left 23:28 Zero_Dogg left, Zero_Dogg joined
masak ahoj, hankache 23:37
23:37 girafe left
hankache hiya masak 23:38
23:44 jameslen_ left 23:45 jameslenz joined, jameslenz left, sortiz joined 23:46 jameslenz joined
sortiz \o #perl6 23:46
hankache hi sortiz
23:48 mohae_ joined
sortiz \me seem like RT#128156 is reproducing like wombles! :-) 23:48
synopsebot6 Link: rt.perl.org/rt3//Public/Bug/Displa...?id=128156
23:49 RabidGravy left 23:50 jameslen_ joined, jameslen_ left 23:51 mohae left, jameslen_ joined 23:54 jameslenz left 23:58 cdg left 23:59 cdg joined