»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.
Aggressively_Alp Like, if you're defining some logic about some data you have, it's nice to be able to bring that up into the top level... I've got to be honest, I don't know what macros will look like for p6. Are the6 compiled in the host machine? Mine? What are the rules?!? That's kinda what I'm curious about. Also yes. Ace language. Too marks. Very fond. 00:02
pilne i've been peeking at perl6 for a while, decided to give this a look, really until somewhat recently i only knew enough perl to mostly grok shell scripts, dipped my toes in 5 for a while, liked it, a bit archaic in a few places, but not without reasons. read some of mr. wall's writings, about perl and perl6, and decided to give perl6 a really solid look 00:04
the last thing to even remotely impress me this much was dart, but it just irks me a bit in a few of the same ways python does 00:05
(i'm a hobbyist dabbler of languages)
i have exactly 0loc of anything truly productive on this planet lol (:
but i'm going to give at least some fun things a whirl in perl6 that i've been tossing around in my head. 00:06
pilne and, is MRO related to the order listed by "is" when using multiple inheritence? 00:10
timotimo perl6 has something called the C3 MRO 00:11
timotimo if that helps you any 00:12
pilne it tells me "no" to the positional thing of the is statements (: which fully answers all the tangents my mind had sprung onto as well (: thank you. 00:13
pilne waitasecond... i can just use role everywhere instead of class and lose nothing in terms of... getting shit done? 00:16
without having to fret about footshotgunning myself?
or is there some advantage to is instead of does, that would make refactoring later into proper classes where it makes sense, a better idea? 00:19
timotimo roles are the best tool for code re-use 00:24
ugexe flatten your inheritance tree instead of deepening it 00:25
timotimo yup
that also makes adhering to the liskov substitution principle easier
MasterDuke but right now are there some limitations with using natives and roles?
Aggressively_Alp As a Scala fanboy myself... Traits/roles 4 lyfe. There are always going to be restrictions when your abstractions don't map neatly, do you have any issues in mind? 00:26
MasterDuke *aren't
pilne i love a lot about scala, i just always have issues doing things on the jvm scale, everything feels too big for a small fry like myself lol. 00:28
lookatme morning .o/ 00:30
bioduds Hello friends. Would someone be interested in see what I wrote? 01:05
It is the beginning of a Merkle Tree Module
here is the article
steemit.com/bitcoin/@bitworkers/de...or-bitcoin
I would very much appreciate all suggestions to improve the code as I am a newbie in Perl6
and want to contribute with a quality module to the repository 01:06
samcv bioduds, that sounds pretty neat. i'd be curious to see how it looks, though i'm a bit busy so not sure how much feedback i may or may not be able to give 01:14
bioduds Thank you samcv! Any help is appreciated and I don't keep bothering. Any suggestions I will go find out how to implement without bothering you. 01:22
samcv no problem :) always good to see enthusiastic new perl 6 people :)
bioduds perl6 is a killer language, my goal is to build my blockchain project in it 01:23
long way to go though
:D
pilne timotimo=> is this your libtcod bindings attempt i found on github? (roguelikes are something that tickle my fancy frequently) 01:26
(i say attempt because it says not working).
pilne hmmmmm, maybe i'll just abuse the docs of libtcod and try and write it native perl6... 01:36
lookatme bioduds, I have read your post, learn a lot. 01:44
Aggressively_Alp Biodiesel m 02:05
BenGoldberg m: subset my_callback_t of &cb (int32 --> int32); 02:08
camelia 5===SORRY!5=== Error while compiling <tmp>
Malformed trait
at <tmp>:1
------> 3subset my_callback_t of7⏏5 &cb (int32 --> int32);
BenGoldberg m: subset my_callback_t of &cb (int32 --> int32) where *; 02:09
camelia 5===SORRY!5=== Error while compiling <tmp>
Malformed trait
at <tmp>:1
------> 3subset my_callback_t of7⏏5 &cb (int32 --> int32) where *;
BenGoldberg m: subset my_callback_t of Callable &cb (int32 --> int32) where *;
camelia 5===SORRY!5=== Error while compiling <tmp>
Confused
at <tmp>:1
------> 3set my_callback_t of Callable &cb (int327⏏5 --> int32) where *;
expecting any of:
statement end
statement modifier
statement modif…
BenGoldberg m: subset my_callback_t of Callable &cb (int32);
camelia 5===SORRY!5=== Error while compiling <tmp>
Undeclared routine:
cb used at line 1
ParsonsNose m: class A { has $!a ; method BUILD(:$!a) { say $!a } ; method shoot { say $!a } }a 02:12
camelia 5===SORRY!5=== Error while compiling <tmp>
Strange text after block (missing semicolon or comma?)
at <tmp>:1
------> 3{ say $!a } ; method shoot { say $!a } }7⏏5a
expecting any of:
infix
infix stopper
s…
ParsonsNose Sorry. Keeping hitting enter too early by mistake. Phones suck for writing code. 02:13
BenGoldberg Even laptops suck for writing code -- I often hit enter when I mean to hit shift. 02:15
BenGoldberg m: my $x = (1..3).pick; say $x ~~ 1|3; 02:39
camelia True
zengargoyle curse the laptop where PgUP/PgDn/Home/End take a Fn key on the opposite side of the keyboard.
zengargoyle my tiny hands just can't stretch that far. 02:40
lookatme m: subset IntS where .signature ~~ :(Int); my IntS $a = sub (Int){};
camelia ( no output )
lookatme m: subset IntS where .signature ~~ :(Int); my IntS $a = sub (Int $x){ say $x;}; $a.(5)
camelia 5
lookatme m: subset IntS where .signature ~~ :(Int); my IntS $a = sub (Int $x, Int $y){ say $x;}; 02:42
camelia Type check failed in assignment to $a; expected IntS but got Sub (sub (Int $x, Int $y) ...)
in block <unit> at <tmp> line 1
lookatme m: subset IntS where .signature ~~ :(Int); my IntS &a = sub (Int $x){ say $x;};
camelia Type check failed in assignment to &a; expected Callable[IntS] but got Sub (sub (Int $x) { #`(Sub...)
in block <unit> at <tmp> line 1
BenGoldberg m: subset IntS where .signature ~~ :(Int); my IntS &a := sub (Int $x){ say $x;}; 03:53
camelia Type check failed in binding; expected Callable[IntS] but got Sub (sub (Int $x) { #`(Sub...)
in block <unit> at <tmp> line 1
BenGoldberg m: subset IntS where .signature ~~ :(Int); dd IntS;
camelia IntS
BenGoldberg m: subset IntS where .signature ~~ :(Int); IntS.WHAT.say; 03:54
camelia (IntS)
BenGoldberg m: subset IntS where .signature ~~ :(Int); say sub (Int) {*} ~~ IntS;
camelia True
BenGoldberg m: subset IntS where .signature ~~ :(Int); say sub (Int $x) {*} ~~ IntS; 03:55
camelia True
BenGoldberg m: subset IntS where .signature ~~ :(Int); sub foo(IntS) {}; foo(sub (Int $x) {*})
camelia ( no output )
BenGoldberg m: subset IntS where .signature ~~ :(Int); sub foo(IntS) {42}; say foo(sub (Int $x) {*})
camelia 42
BenGoldberg m: subset IntS where .signature ~~ :(Int); sub foo(IntS) {42}; dd &foo;
camelia Sub foo = sub foo ($ where { ... }) { #`(Sub|72089696) ... }
BenGoldberg m: subset IntS where .signature ~~ :(Int); sub foo(IntS) {42}; dd &foo.signature.params[0].sub_signature; 03:59
camelia Any
BenGoldberg lookatme, The problem with a subset is that it's not sufficient/suitable for use with a NativeCall sub's callback parameter, because NativeCall is presently designed to introspect using .sub_signature. 04:03
BenGoldberg m: constant S = sub foo(Int $i) {*}; sub bar(S $x) {42}; bar(S); 04:11
camelia Type check failed in binding to parameter '$i'; expected Int but got Sub (sub foo (Int $i) { #`...)
in sub foo at <tmp> line 1
in sub bar at <tmp> line 1
in block <unit> at <tmp> line 1
BenGoldberg m: subset S of Callable; 04:16
camelia ( no output )
finanalyst where has Task::Star gone. It's not in the modules directory and I am just upgrading. 06:40
ufobat good morning 06:45
Geth Inline-Python: a02d3569c7 | (Stefan Seifert)++ | 3 files
Add missing meta data
07:06
lookatme afternoon \o. 07:09
parv nighty0. 07:09
Geth Inline-Python: 19e8cceb5e | (Stefan Seifert)++ | 2 files
Remove leading ./ from names in MANIFEST
07:17
Geth Inline-Perl5: a4f4275d9c | (Stefan Seifert)++ | 2 files
Add a MANIFEST file
07:25
lizmat clickbaits p6weekly.wordpress.com/2017/05/22/...-21-yap6b/ 07:27
moritz takes the bait, clicks 07:28
actually I read it earlier in the subway :-)
lizmat++
parv using RatStr instead of just plain Rat is smelly (near the bottom of www.learningperl6.com/2017/05/21/d...ogramming/ ) 08:03
... to check a number of arbitrary precision between a range. 08:04
parv don't know why "Does this have any meaning? (www.reddit.com/r/perl/comments/6c9...meaning/)" wa even merit a mention on the blog post. 08:17
s/wa even merit/even merited/ 08:18
lookatme parv++ good post
parv well i did not write/contribute to anything. i am just opining here about other URLs in lizmat's clickbait 08:21
lookatme I haven't read that post yet. :) 08:24
I make a ping-pong sample in Perl 6 08:26
s/make/made/
Geth doc: 1774c09413 | (Jan-Olof Hendig)++ | doc/Type/Map.pod6
Fix incorrect signatures for .keys, .values, .kv and pairs. Brian.d.Foy++

Closes #1323
lookatme The code: gist.github.com/araraloren/929e60b...3287a02174 and the test: ideone.com/i7yUQQ 08:31
ping-pong is a erlang sample in its document : erlang.org/doc/getting_started/conc_prog.html 08:32
lizmat parv: I mentioned it because of Zoffix's reply 08:37
parv ah. thanks for the clarification, lizmat 08:38
Geth Pod-To-HTML: c5847eab0a | (Moritz Lenz)++ | .travis.yml
Use zef in travis
08:49
Geth doc: 0b32e6764b | (Jan-Olof Hendig)++ | doc/Type/Map.pod6
Forgot to change Pairs
09:00
lookatme bye . 09:35
Zoffix "<parv> using RatStr instead of just plain Rat is smelly" well, 9.99999999999999999999999 isn't a Rat. It's denominator is too big for it 09:36
finanalyst, just install modules you need instead of random stuff. Task::Star is gone 09:39
timotimo .tell pilne i already completely forgot that i even did anything with tcod ... 09:47
yoleaux timotimo: I'll pass your message to pilne.
zengargoyle lizmat: Damian Conway will be given a freely-accessible -- s/given/giving/ 10:32
bioduds good morning 12:24
How do I write a Buf to a file in binary mode?
I'm trying this with no success 12:25
my $fh = open "blocks/bl000001.dat", :bin;
my $data = "test data";
my Blob $sha256 = sha256 $data.encode: 'ascii';
$fh.print( $sha256 );
$fh.close;
jnthn Use .write 12:27
bioduds I tried .write, let me try again I must have used it wrong 12:28
Failed to write bytes to filehandle: bad file descriptor 12:29
my $fh = open "blocks/bl000001.dat", :bin;
my $data = "test data";
my Blob $sha256 = sha256 $data;
$fh.write( $sha256 );
$fh.close;
do I need to encode?
jnthn Ah, I think :w is missing from the open statemnet too, to open it in write mode 12:30
s/statement/function call/
bioduds it should be :wbin ? 12:31
jnthn no
bioduds only :w
jnthn :w, :bin
Or just :w
bioduds ok
jnthn If you're using .write it already means writing binary data
So you don't really need :bin
bioduds voila
worked
thanks jnthn!! 12:32
excellent
jnthn Welcome :)
bioduds precisely what I needed
this community is awesome
bye
jnthn o/ 12:33
[Coke] jnthn++ bioduds++ 12:43
tbrowder hi #perl6, i'm fiddling with a rudimentary pod2man using pod2text as a model. my idea is to have a man src file in pod6 to generate the man page and have zef install it on *nix systems. a default page could be generated from any README, although it might not be pretty. thoughts? (nipping out for a while) 12:54
Geth doc: 2746a1bb89 | (Zoffix Znet)++ (committed using GitHub Web editor) | doc/Language/unicode_entry.pod6
Remove out-of-place reference

Renders `"say routine"` below the code block... "say routine" what?
13:01
Geth doc: 4fabb6c763 | (Zoffix Znet)++ (committed using GitHub Web editor) | doc/Language/unicode_entry.pod6
Make it more obvious «$baa» splits on whitespace in var
13:03
Zoffix bioduds: it looks like you have waaaay more code than you need for that operation. It can be written much shorter as: 'blocks/bl000001.dat'.IO.spurt: sha256 $data 13:06
it will do the opening and the binning for you. 13:07
and the closing :)
lizmat zengargoyle++ # fixed 13:32
Geth doc: jecki++ created pull request #1324:
Invalid Python code in class-inheritance-example
13:38
doc: 58124f9f7c | (Eckhart Arnold)++ (committed by Zoffix Znet) | doc/Language/py-nutshell.pod6
Invalid Python code in class-inheritance-example (#1324)

The lines class Animal: def jump: print "I am jumping"
... (7 more lines)
13:39
tbrowder ref pod2man: the META6.json could be used to provide the canonical man src file names and output man names. among other things, if a man src were not provided by the author, a default man page could have as a minimum the module source link. 13:42
araraloren tbrowder, pod2man ? Is the pod mean Perl5 Pod ? 14:11
tbrowder no, perl6 pod
it doesn't exist yet 14:12
araraloren Oh
Sounds great, the idea. 14:13
Mithaldu i think github just recognized a file as perl 6 because it starts with "use 5.006;" 14:30
lizmat yuck :-) 14:33
eater :'))) 14:34
moritz use.*6 ? 14:41
Mithaldu that's my suspicion 14:43
araraloren lazy github :) 14:45
Mithaldu hmmm, i know it's this breaking: github.com/github/linguist#troubleshooting 14:49
but i can't easily find where in the code the detection is being done
Mithaldu mmmmaybe this github.com/github/linguist/blob/ma...#L347-L353 14:50
might have to narrow down why it's messing up with my file
tbrowder araraloren: thanks! 14:51
araraloren tbrowder, Em, will you make a module for that idea ?
tbrowder yes
araraloren tbrowder++ 14:52
tbrowder on my TODO, fiddling with working code now...
Mithaldu in case anyone wants to bug into this a little: github.com/github/linguist/issues/3637 15:44
raiph m: say %( 'a' => 42 ){ 'a' ; } 16:28
camelia 42
raiph m: say %( 'a' => 42 ){ 'a' ; ; } 16:29
camelia Type Int does not support associative indexing.
in block <unit> at <tmp> line 1

Actually thrown at:
in block <unit> at <tmp> line 1
raiph o/ Does anyone know of a matching RT? ^^^ 16:30
raschipi Because of the LTA message or the unexpected behavior? 16:35
jnthn Is it even a bug? 17:06
I'd expect { 'a' ; ; } to compile as if it were { 'a' ; Nil } or so 17:08
It's a semilist, so it's things separated by ; with a trailing one allowed
I guess we could choose to refuse an empty thing 17:09
TimToady "Empty subscript not allowed (did you mean * for a wildcard?)" or some such... 17:19
cygx o/ 17:20
yoleaux 22 May 2017 14:19Z <Zoffix> cygx: would you recommend .seek/.tell to throw if called on a handle that isn't in binary mode? I recall you saying they are technically binary operations when I had that case where .tell was giving wrong result after some text read operation
TimToady mind, the carpage should only be about syntactic emptiness, not semantic, since we want to allow null lists as a degenerate case
cygx m: say do { my $i = 42; LEAVE $i = 0; $i }; say do { my $i = 42; LEAVE $i = 0; +$i } 17:22
camelia 0
42
cygx ^ bug?
timotimo don't think so 17:23
m: say (do { my $i = 42; LEAVE $i = 0; $i }) = 10
camelia 10
timotimo you're getting passed back the container
cygx so blocks are implicitly rw? 17:24
timotimo that's how i'd interpret that 17:26
jnthn Yes, blocks don't decontainerize, like subs 17:30
cygx jnthn: or rather, unlike subs (I assume) 17:36
confusing sentence is confusing ;)
jnthn haha 17:47
Subs do decontainerize unless marked "is rw" 17:48
Blocks do not
Geth doc: jecki++ created pull request #1327:
tabs converted to spaces in lines 550-552
17:52
doc: 3faba8f7cb | (Eckhart Arnold)++ (committed using GitHub Web editor) | doc/Language/py-nutshell.pod6
tabs converted to spaces in lines 550-552
doc: 2cff5dd033 | lizmat++ (committed using GitHub Web editor) | doc/Language/py-nutshell.pod6
Merge pull request #1327 from jecki/patch-1

tabs converted to spaces in lines 550-552
andrzejk_ hi people 18:04
lizmat andrzejk_ o/ 18:05
andrzejk_ what o? 18:06
lizmat it's waving :-) 18:07
saying hi :-) 18:08
andrzejk_ oh ok
I need C++ expert
lizmat the #perl6 channel doesn't appear to be the place to find those 18:09
andrzejk_ I know but I will try 18:10
perlpilot I guess #perl6 has inherited the same thing from #perl where people think the channel is full of really smart people who know lots of things .... ;-) 18:12
pilne i do my best to stay far far away from c++ 18:14
yoleaux 09:47Z <timotimo> pilne: i already completely forgot that i even did anything with tcod ...
pilne timotimo=> fair enough, i used the python version ages ago.
timotimo i'm sitting on many unfinished things :| 18:16
pilne i have those too, happens when you have more languages installed on your computer than... well anything else added after a clean ubuntu mate install 18:16
cbk_ Is there a bug with ZEF, not installing modules to the right place? I keep getting an error saying it cant find a module which was just installed....? 18:19
lizmat did you update to the latest zef ? 18:23
cbk_ I do believe so. Had to install it 18:24
Using rakudobrew
lizmat hmmm... I seem to recall seeing problem reports with that, so maybe backlogging the past 4 days or so will give an answer 18:25
cbk_ Backlogging...? Well I'm more then happy to help with that if I can, but really don't know/understand what your asking. 18:28
ugexe sounds like rakudobrew is mixed up and invoking zef with a different version of perl6 than what you are trying to use 18:31
cbk_ I'll go back and nuke my install of perl6 and try again from scratch. 18:32
ugexe might try `rakudobrew rehash` as well 18:33
cbk_ I may have a few different installs of Perl6 on the same system.
timotimo i thought rehash only does stuff for binaries
cbk_ Ok I can that now...
ugexe timotimo: it does, but i think there is a way for the perl6 shim to get mixed up for a command 18:34
cbk_ Updating shims......
timotimo ah 18:35
ugexe do `zef install zef --force` and note where it says it installs the bin/ script. does this match where you think it should?
cbk_ Ok...
azawawi . 18:36
..
hi :)
cbk_ The reshash thing does not appear to have worked.
ZEF installed to ~/.rakudobrew/moar-nom/install/share/perl6/site/bin 18:37
azawawi is thinking whether to write a gnuplot or plplot library... so many choices... so little time :)
cbk_ Looks good to me
ugexe `which perl6`? 18:38
cbk_ 2017.40.3.293-ga7c23aa
ugexe no, run that actual command 18:39
cbk_ Perl6
timotimo it should give you a path 18:39
cbk_ My perl6 lives at ~/.rakudobrew/bin/perl6 18:40
timotimo right, that's the forwarder 18:41
cbk_ That is the output from which perl6 18:42
azawawi ponders with the obscure "it should give you a path"... gnuplot vs plplot :)
raiph m: say %( 'a' => 42 ){ 'a' ; 42 } 18:44
camelia Type Int does not support associative indexing.
in block <unit> at <tmp> line 1

Actually thrown at:
in block <unit> at <tmp> line 1
raiph ^^^ It doesn't look to me like the issue is emptiness
timotimo is this in reference to something? 18:45
raiph timotimo: irclog.perlgeek.de/perl6/2017-05-23#i_14625002
timotimo m: sub postcircumfix:<{ }>($subj, |c) { say c.perl }; %( a => 42 ){ 'a'; 42 } 18:46
jnthn m: say (%( a => 42)).WHAT 18:47
Lazy bots... 18:48
timotimo i'm not sure how i broke it
jnthn But yeah, I'm not sure why it's surprising that trying to do 42<42> wouldn't work :)
timotimo locally i immediately get "int does not support associative indexing"
yeah
jnthn Which is what that code boils down to before it fails
{x ; y} means the same as {x}{y} 18:49
timotimo i wonder why it doesn't use my own postcircumfix:<{ }> in that code, though
jnthn Because that form dispatches to postcircumfix:<{; }> iirc 18:49
jnthn Otherwise the dispatch got very confusing :) 18:50
timotimo ah 18:51
jnthn raiph: What were you expecting it to do, btw?
timotimo you are right
timotimo but yeah, when you have { 'a'; } it goes through the { } candidate, but if you have { 'a' ; ; } it goes through the {; } candidate 18:52
jnthn I think that's probably right 18:53
raiph jnthn: Whatever it was, I was confused. Your explanation was perfect. The error message makes sense. My apologies for the noise. :) 18:54
timotimo and it gives you a Nil in the second slot, just like you said earlier
raiph jnthn: A little knowledge is a dangerous thing. I knew enough, and pay enough attention to detail, to check a `{...;...}` subscript while researching the footnote in my stackoverflow.com/a/44102980/1077672 answer then erroneously jumped to the island of conclusions about the error message... /o\ 19:00
timotimo hmm, Buf doesn't do any nums 19:34
vytautas No candidates found matching identity: Task::Star 20:25
timotimo yeah, Task::Star was deprecated
but not yet taken off of the how-to-get-rakudo page
vytautas and what is aprecated? 20:26
vytautas made a new word
timotimo also not sure if rakudobrew should still be recommended on rakudo.org at all
it's recommended to grab whatever modules you're interested in from modules.perl6.org 20:27
vytautas ok 20:27
timotimo alternatively modules.zef.pm/
just "zef install NameFromThatList"
El_Che zef install LibraryMake -> Aborting due to test failure: LibraryMake:ver('1.0.0'):auth('github:retupmoca') (use --force to override) 20:35
cloing the repo from github and running the test manuall is OK though
[Coke] continues the tradition of asking p5 questions here, and wonders if there's a useful way to report a test failure in a perlbrew of the 5.26 RC 20:36
El_Che ok, nuking the SHA resources dirs fixed it (Oh, I the dislike of dirs like C3893DD6E92E8A64304F0D65AF5173B7BC3531CA) 20:38
timotimo i changed the rakudo download page to no longer mention Task::Star 20:39
El_Che why?
timotimo because Task::Star is no longer in the ecosystem
so you can't install it any more
El_Che ah crap, I need to update the doc of my pkgs 20:40
thx for the pointer
timotimo :)
so many modules still refer to panda in their readme 20:42
and even more in their .travis.yml
El_Che I just met one referring to ufo
timotimo fantastic :) 20:43
El_Che I revived some old perl6 code that probably doesn't run anymore, but still has a use case :)
timotimo how was the experience? how long before christmas was that code? 20:48
El_Che actually, I am not there yet 20:49
El_Che stuck in failed zef installs 20:49
zef --verbose install Config::Parser::yaml Digest::SHA1::Native SSH::LibSSH 20:50
fails: JSON::Unmarshal
timotimo installs fine for me 20:53
El_Che 0.08? 20:54
timotimo yeah
El_Che zef --verbose install Config::Parser::yaml Digest::SHA1::Native SSH::LibSSH 20:55
oops
No such method 'bla' for invocant of type 'Any' in block <unit> at t/parameterised.t line 16
timotimo t/parameterised.t .. ok 20:57
*shrug* :(
El_Che it even fails when $ perl6 -Ilib t/parameterised.t 20:58
timotimo: what rakudo ver are you running? 21:01
timotimo This is Rakudo version 2017.05-55-gdd6d6cc built on MoarVM version 2017.05-24-gd86a35b
thundergnat .tell azawawi Thanks for the Terminal::Libcaca bindings. It's been fun to play with. Some code in the wild: rosettacode.org/wiki/Draw_a_rotating_cube#Perl_6 21:34
yoleaux thundergnat: I'll pass your message to azawawi.
pilne i might regret running the "prime conspiracy" perl6 code from rosetta code >.< 23:20
Geth doc: edd7431734 | (Zoffix Znet)++ | doc/Type/Pair.pod6
Fix up Pair.invert docs

  - It (since a few months ago) always returns a Seq; old behaviour
   was not specced.
  - Simply examples
Spec: github.com/perl6/roast/commit/f4828aaf64 Fixes RT#131349: rt.perl.org/Ticket/Display.html?id=131349
23:45
synopsebot6 Link: rt.perl.org/rt3/Public/Bug/Display...?id=131349