Perl 6 - the future is here, just unevenly distributed

IRC log for #perl6, 2017-01-26

Perl 6 | Reference Documentation | Rakudo

| Channels | #perl6 index | Today | | Search | Google Search | Plain-Text | summary

All times shown according to UTC.

Time Nick Message
00:02 agentzh thanks
00:05 sena_kun last summer I introduced a typo in Exception.pod6 and noticed it only tonight. What a shame.
00:05 timotimo :D
00:05 timotimo i know what you did last summer
00:06 sena_kun lol
00:08 brokenchicken hsh
00:08 brokenchicken *hah
00:09 sena_kun ah, there goes another. I need to check my commits twice.
00:10 timotimo we do have spell checking now at least :)
00:12 sena_kun timotimo, the first case is a missing word, the second case is a missing bracket. God bless spell checker that can point out such cases! inb4: it's just you.
00:15 Cabanossi joined #perl6
00:15 pierrot joined #perl6
00:16 agentzh where can i find my reported issues to  rakudobug@perl.org?
00:16 sena_kun ah, no, the second case *always* was there. \0/
00:16 agentzh never mind, found https://rt.perl.org/Public/Search/Results.html?Query=Queue=%27perl6%27%20AND%20Status=%27new%27
00:17 sena_kun agentzh, I was not quick enough, sorry.
00:18 agentzh sena_kun: no need to apologize at all :) i should have diged harder :)
00:18 agentzh just filed a ticket for the infinite looping with RAKUDO_MODULE_DEBUG=1: https://rt.perl.org/Public/Bug/Display.html?id=130645
00:19 agentzh btw, is there a way to get perl 6 backtrace from inside gdb?
00:19 agentzh while being attached to the moar process?
00:21 brokenchicken agentzh: we also have perl6.fail that's a bit easier to get all-tickets-at-a-glance from
00:22 bjz joined #perl6
00:23 gfldex agentzh: jnthn could help you with your moarvm woes if he didn't need sleep
00:29 MasterDuke agentzh: there is a way, timotimo probably knows
00:30 agentzh brokenchicken: oh, nice!
00:30 agentzh MasterDuke: for parallel compilation?
00:31 MasterDuke agentzh: perl6 level backtrace in gdb
00:31 agentzh MasterDuke: oh great
00:32 BenGoldberg joined #perl6
00:33 MasterDuke something like there should be a tc variable, enter its frame and then type `bt`
00:33 MasterDuke but i haven't done it, so i'm just remembering bits and pieces of what i've seen other people say
00:35 MasterDuke agentzh: try `(gdb) print MVM_dump_backtrace(tc)`
00:39 agentzh MasterDuke: segfault
00:39 agentzh MasterDuke: seems like calling into the userland function is unsafe per se?
00:39 agentzh since gdb can interrupts into the process in any state.
00:40 agentzh *interrupt
00:42 agentzh MasterDuke: tried another p6 script, segfault as well.
00:42 agentzh see https://gist.github.com/agentzh/0708ffe8de7a13fd66379a474676845f
00:54 Ben_Goldberg joined #perl6
00:56 lep-delete joined #perl6
01:00 agentzh seems like rakudo has issues handling timestamps of the dependency modules.
01:00 agentzh it always compiles some of the dep modules even though they have never been edited since the last compilation.
01:02 pyrimidine joined #perl6
01:05 agentzh everytime i edit a single .pm6 file of only 146 lines of code (including empty lines and comment lines), it always takes ~7 sec to compile.
01:05 agentzh this is very annoying.
01:06 agentzh though i must say it's much faster than re-compiling rakudo when changing the Core.setting module.
01:06 agentzh the latter takes more than a 1min :)
01:08 timotimo are you using Inline::Perl5 perhaps?
01:08 agentzh timotimo: nope.
01:08 agentzh pure Perl 6
01:08 timotimo there are modules that export "no precompilation" into their users
01:08 agentzh not even P5 regexes.
01:08 timotimo those don't matter
01:08 timotimo have you tried the module debug output thingie?
01:09 timotimo RAKUDO_MODULE_DEBUG
01:09 timotimo environment variable
01:13 pyrimidine joined #perl6
01:15 agentzh timotimo: yes. and it triggers a bug which makes moar hot looping: https://rt.perl.org/Public/Bug/Display.html?id=130645#ticket-history
01:15 Cabanossi joined #perl6
01:15 agentzh *enter hot looping
01:16 agentzh from the partial module debug output before it enters the infinite loop, it seems that rakudo is indeed trying to recompile other p6 modules which at least indirectly depend on the changed .pm6 file.
01:19 kurahaupo__ joined #perl6
01:29 sena_kun m: my %pairs = also-named => 4; sub f(|c) {}; f |%pairs;
01:29 camelia rakudo-moar 0ad05c: ( no output )
01:47 Geth doc: 423308f731 | (Samantha McVey)++ | doc/Language/classtut.pod6
01:47 Geth doc: Add X<> for ^methods method
01:47 Geth doc: review: https://github.com/perl6/doc/commit/423308f731
01:53 bjz joined #perl6
01:54 Geth doc: 8a3b45ef86 | (Samantha McVey)++ | doc/Language/unicode.pod6
01:54 Geth doc: Add some missing quotation marks to the Unicode page
01:54 Geth doc: review: https://github.com/perl6/doc/commit/8a3b45ef86
01:55 samcv Geth, source
01:55 Geth samcv, Source at https://github.com/perl6/geth To add repo, add an 'application/json' webhook on GitHub pointing it to http://geth.perl6.party/?chan=#perl6 and choose 'Send me everything' for events to send | use `ver URL to commit` to fetch version bump changes
01:58 Gothmog_ joined #perl6
02:02 astj joined #perl6
02:15 astj joined #perl6
02:15 Cabanossi joined #perl6
02:16 kalkin- joined #perl6
02:23 Geth doc: 8167550f35 | (Samantha McVey)++ | htmlify.p6
02:23 Geth doc: Make --no-proc-async default for MacOS
02:23 Geth doc:
02:23 Geth doc: Also warn the user the build will be very slow.
02:23 Geth doc: review: https://github.com/perl6/doc/commit/8167550f35
02:23 samcv .tell [Coke] I have set --no-proc-async as default for MacOS
02:23 yoleaux samcv: I'll pass your message to [Coke].
02:32 samcv i forget guys, what's the way to find the name of the variable?
02:32 timotimo .VAR.name?
02:32 samcv k
02:32 samcv yes!
02:32 * timotimo disappears to bed immediately
02:34 MasterDuke .tell agentzh according to timotimo, 'will probably have to "up" a few times until you find a frame that has "tc" in it', before the `print MVM_dump_backtrace(tc)`
02:34 yoleaux MasterDuke: I'll pass your message to agentzh.
02:46 agentzh joined #perl6
02:47 ilbot3 joined #perl6
02:47 Topic for #perl6 is now »ö« Welcome to Perl 6! | https://perl6.org/ | evalbot usage: 'p6: say 3;' or rakudo:,  or /msg camelia p6: ... | irclog: http://irc.perl6.org or http://colabti.org/irclogger/irclogger_logs/perl6 | UTF-8 is our friend!
02:48 hartenfels joined #perl6
02:53 pyrimidine joined #perl6
03:04 cale2 joined #perl6
03:04 cale2 Does p6 have type aliases? Or would you just use subset for that?
03:05 cale2 Also, how can I make a subset that is a List of Strings
03:05 cale2 m: subset WordList of List where *all items are strings*
03:05 camelia rakudo-moar 0ad05c: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Two terms in a row␤at <tmp>:1␤------> 3subset WordList of List where *7⏏5all items are strings*␤    expecting any of:␤        infix␤        infix stopper␤        postfix␤        statement…»
03:08 cale2 I guess I'm basically wondering if the syntax is [Str] or List(Str)
03:17 hartenfels cale2 I think you can do subset WordList of List where { .all ~~ Str }
03:18 cale2 hartenfels: thanks! I got this so far
03:18 cale2 m: subset WordList of List where { $_.map: { $_ ~~ Str } }
03:18 camelia rakudo-moar 0ad05c: ( no output )
03:18 hartenfels Well, that's similar.
03:19 cale2 I figured there should be an all method though :)
03:19 hartenfels all will create an &-Junction, so that'll deal with the iteration for you.
03:19 azertus joined #perl6
03:21 hartenfels subset WordList of List where { .all ~~ Str }; say ("asdf", "jklö") ~~ WordList; say () ~~ WordList; say ("asdf", 123) ~~ WordList;
03:21 hartenfels m: subset WordList of List where { .all ~~ Str }; say ("asdf", "jklö") ~~ WordList; say () ~~ WordList; say ("asdf", 123) ~~ WordList;
03:21 camelia rakudo-moar 0ad05c: OUTPUT«True␤True␤False␤»
03:23 agentzh joined #perl6
03:26 cale2 hartenfels: in the type system, is there an easier way to express a list of a type?
03:26 cale2 like [Str] or List(Str)
03:26 hartenfels Maybe you can do List[Str]
03:26 hartenfels I'm not sure.
03:27 MasterDuke m: my Str @a
03:27 camelia rakudo-moar 0ad05c: ( no output )
03:27 MasterDuke m: my Str @a; @a[0] = 4
03:27 camelia rakudo-moar 0ad05c: OUTPUT«Type check failed in assignment to @a; expected Str but got Int (4)␤  in block <unit> at <tmp> line 1␤␤»
03:28 cale2 m: my [Str] @a; @a[0] = 4
03:28 camelia rakudo-moar 0ad05c: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Malformed my␤at <tmp>:1␤------> 3my7⏏5 [Str] @a; @a[0] = 4␤»
03:28 cale2 m: my List(Str) @a; @a[0] = 4
03:28 camelia rakudo-moar 0ad05c: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Coercion List(Str) is insufficiently type-like to qualify a variable␤at <tmp>:1␤------> 3my List(Str) @a7⏏5; @a[0] = 4␤    expecting any of:␤        constraint␤»
03:28 hartenfels That works when declaring a variable at least.
03:28 hartenfels But it's not the same to have an Array[Str] as it is to have an array that happens to have a bunch of strings in it.
03:29 cale2 Pretty annoying when you want to add a guard that this is a list of certain items
03:30 hartenfels m: my List[Str] $doesnt-work-either;
03:30 camelia rakudo-moar 0ad05c: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤An exception occurred while parameterizing List␤at <tmp>:1␤Exception details:␤  5===SORRY!5=== Error while compiling <tmp>␤  List cannot be parameterized␤  at <tmp>:1␤  ------> 3my List[Str]…»
03:31 cale2 I guess it's not terrible to write `where { .all ~~Str }` all over
03:31 hartenfels You'd probably make a subset type.
03:31 cale2 but you'd have to do that for every list
03:31 cale2 I'm translating a haskell program from a while ago. That's why I'm asking
03:32 hartenfels You'd also have to specify [Str] for every list :P
03:33 cale2 hartenfels: maybe you could write a macro for type objects
03:33 cale2 if it's wrapped in square brackets, check to make sure it's a list of this type
03:34 cale2 I was gonna write Str anyway though lol
03:34 hartenfels I dunno, for me just specifying `subset Strs where { .all ~~ Str}` and then using `sub (Strs $bla)` seems fine.
03:34 hartenfels And Strs is one less character than [Str]!
03:36 hartenfels And it doesn't require you to write any macros of course.
03:36 cale2 That's true. Good call
03:37 cale2 Isn't there a github repo for comparing perl6 to other languages?
03:37 hartenfels There's RosettaCode.
03:37 cale2 No, I'll find it
03:45 noganex joined #perl6
03:48 cale2 nevermind, i can't find it lol
03:49 travis-ci joined #perl6
03:49 travis-ci Doc build errored. Samantha McVey 'Make --no-proc-async default for MacOS
03:49 travis-ci https://travis-ci.org/perl6/doc/builds/195409042 https://github.com/perl6/doc/compare/8a3b45ef86a0...8167550f359b
03:49 travis-ci left #perl6
03:55 pyrimidine joined #perl6
03:57 cale2 Is there a functional difference between Strs and Chars in p6?
03:58 cale2 Would you typically just use the Str type even for individual graphemes?
03:58 agentzh joined #perl6
03:58 hartenfels cale2 There's a Char type?
03:59 cale2 hartenfels: I don't see it in the docs
03:59 hartenfels Then the functional difference between a Str and a Char is that the latter doesn't exist :P
04:00 cale2 I'll make a subtype
04:01 cale2 m: subset Char of Str where { $_.chars ~~ 1 }; say Char "cale";
04:01 camelia rakudo-moar 0ad05c: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Two terms in a row␤at <tmp>:1␤------> 3of Str where { $_.chars ~~ 1 }; say Char7⏏5 "cale";␤    expecting any of:␤        infix␤        infix stopper␤        postfix␤        statement end…»
04:01 cale2 m: subset Char of Str where { $_.chars ~~ 1 }; my Char $c = "cale";
04:01 camelia rakudo-moar 0ad05c: OUTPUT«Type check failed in assignment to $c; expected Char but got Str ("cale")␤  in block <unit> at <tmp> line 1␤␤»
04:02 hartenfels You should probably be using == if you're doing numeric comparison.
04:12 cale2 hartenfels: feel free to contribute, or just look at what i'm doing https://github.com/WildYorkies/perl6-mess
04:15 hartenfels cale2: I could probably whip up something in Perl 5, but I can't check what the Haskell one does since ghc takes about 3 years to compile in Gentoo. Is it supposed to let the user guess the letters or just play hangman with itself?
04:16 cale2 hartenfels, it's been a while since I touched the haskell code. maybe I'm leaving off the user input part by accident haha
04:18 hartenfels It doesn't look like it play at all actually.
04:18 cale2 hartenfels: I was mainly looking for single file programs from functional languages
04:19 cale2 I'll see if I can find the rest of that code lol
04:24 ilbelkyr joined #perl6
04:29 Cabanossi joined #perl6
05:01 astj joined #perl6
05:05 kalkin- joined #perl6
05:06 geekosaur joined #perl6
05:08 pyrimidine joined #perl6
05:09 Tonik joined #perl6
05:10 solarbunny joined #perl6
05:46 pyrimidine joined #perl6
06:12 CIAvash joined #perl6
06:17 TimToady joined #perl6
06:25 parv joined #perl6
06:30 AlexDaniel joined #perl6
06:46 brillenfux joined #perl6
06:47 pyrimidine joined #perl6
06:48 RabidGravy joined #perl6
06:51 hartenfels1 joined #perl6
06:59 Cabanossi joined #perl6
07:04 imcsk8 joined #perl6
07:08 domidumont joined #perl6
07:11 labster joined #perl6
07:15 domidumont joined #perl6
07:25 z8 joined #perl6
07:29 jjido joined #perl6
07:34 abraxxa joined #perl6
07:37 ufobat joined #perl6
07:39 darutoko joined #perl6
07:40 abraxxa joined #perl6
07:59 grondilu joined #perl6
08:00 grondilu m: my %h = hash <a b c> Z=> ^3; for %h X %h -> $a, $b { say "$a:$b" }
08:00 camelia rakudo-moar 0ad05c: OUTPUT«a    0 a     0:a     0 c     2␤a     0 b     1:c     2 a     0␤c     2 c     2:c     2 b     1␤b     1 a     0:b     1 c     2␤Too few positionals passed; expected 2 arguments but got 1␤  in block <unit> at <tmp> line 1␤␤»
08:00 grondilu ^not sure what I'm doing wrong
08:00 grondilu m: my %h = hash <a b c> Z=> ^3; for %h X %h -> ($a, $b) { say "$a:$b" }
08:00 camelia rakudo-moar 0ad05c: OUTPUT«Too few positionals passed; expected 2 arguments but got 0 in sub-signature␤  in block <unit> at <tmp> line 1␤␤»
08:00 melezhik_ joined #perl6
08:00 melezhik_ HI Perl6 gurus!
08:00 grondilu hello
08:01 grondilu m: my %h = <a b c> Z=> ^3; for %h X %h -> ($a, $b) { say "$a:$b" }
08:01 camelia rakudo-moar 0ad05c: OUTPUT«Too few positionals passed; expected 2 arguments but got 0 in sub-signature␤  in block <unit> at <tmp> line 1␤␤»
08:01 grondilu m: my %h = <a b c> Z=> ^3; for %h X %h -> $a, $b { say "$a:$b" }
08:01 camelia rakudo-moar 0ad05c: OUTPUT«a    0 a     0:a     0 c     2␤a     0 b     1:c     2 a     0␤c     2 c     2:c     2 b     1␤b     1 a     0:b     1 c     2␤Too few positionals passed; expected 2 arguments but got 1␤  in block <unit> at <tmp> line 1␤␤»
08:01 grondilu m: my %h = <a b c> Z=> ^3; for flat %h X %h -> $a, $b { say "$a:$b" }
08:01 camelia rakudo-moar 0ad05c: OUTPUT«a    0:a     0␤a     0:c     2␤a     0:b     1␤c     2:a     0␤c     2:c     2␤c     2:b     1␤b     1:a     0␤b     1:c     2␤b     1:b     1␤»
08:01 melezhik_ how to tuncate string to contain no more N characters ?
08:01 melezhik_ s/tuncate/truncate/
08:02 grondilu substr
08:02 grondilu m: say "fooooo".substr(0, 3)
08:02 camelia rakudo-moar 0ad05c: OUTPUT«foo␤»
08:03 melezhik_ grondilu: thanks
08:04 melezhik_ m: say "fooooo".substr(0, 20)
08:04 camelia rakudo-moar 0ad05c: OUTPUT«fooooo␤»
08:04 melezhik_ ok, works for me! thanks
08:06 parv is it possible to specify an exact version of a module/package (not just the minimum as is the case in Perl 5)?
08:07 moritz yes. I just don't know how well it works at the moment
08:08 parv the working part, was that comment meant in context of rakudo?
08:09 moritz yes
08:09 parv ok. thanks, moritz.
08:11 krakan_ joined #perl6
08:14 xinming_ joined #perl6
08:14 Cabanossi joined #perl6
08:23 curt_ joined #perl6
08:26 wamba joined #perl6
08:28 zakharyas joined #perl6
08:28 rindolf joined #perl6
08:29 ChoHag joined #perl6
08:32 faraco joined #perl6
08:32 espadrine_ joined #perl6
08:37 mr-foobar joined #perl6
08:39 parv is there any other comprehensive & current doc about version specification other than "https://design.perl6.org/S11.html#Versioning"? That URL says to consult docs.perl6.org for current state; didn't find anything suitable.
08:40 dakkar joined #perl6
08:41 astj joined #perl6
08:43 wololoooo joined #perl6
08:44 wololoooo Hello guys, I have a problem with file IO
08:44 jonas1 joined #perl6
08:45 wololoooo perl6 -e 'if "ratings.list".IO.f {say "HI!!";}
08:45 wololoooo my $fg = open ("./ratings.list" ,:r , enc => "latin1");
08:45 wololoooo say $fg.get;'
08:46 wololoooo == OUTPUT ==
08:46 wololoooo HI!!
08:46 wololoooo Failed to open file /home/pi/project/perl6/lear/ratings.list r  True enc        latin1: no such file or directory
08:46 wololoooo in any  at /home/pi/project/perl6/rakudobrew/moar-nom/install/share/perl6/runtime/CORE.setting.moarvm line 1
08:46 wololoooo in block <unit> at -e line 2
08:46 wololoooo Actually thrown at:
08:46 wololoooo in any  at gen/moar/Metamodel.nqp line 3072
08:46 wololoooo in block <unit> at -e line 3
08:46 moritz wololoooo: please use a pastebin
08:46 moritz wololoooo: the problem is the space between open and (
08:46 wololoooo ok, sorry
08:46 moritz wololoooo: leave out the parens, or the space
08:46 geekosaur joined #perl6
08:47 wololoooo ok
08:48 moritz wololoooo: as is, you're calling open() with a single argument, which is a list of a string and two Pair objects
08:49 wololoooo isn't that practically 3 arguments?
08:49 moritz nope
08:50 moritz in Perl 6, lists are first-class objects
08:50 wololoooo then what would multiple arguments look like in perl6?
08:52 g4 joined #perl6
08:52 g4 joined #perl6
08:55 parv S11 missed the chance to use "globalyness" when "globalness" was used instead. 8-[
09:04 cibs joined #perl6
09:04 xiaomiao joined #perl6
09:05 moritz wololoooo: open(a, b, c) or open a, b c
09:12 curt__ joined #perl6
09:17 trnh joined #perl6
09:17 trnh joined #perl6
09:30 ufobat joined #perl6
10:16 wizard_ joined #perl6
10:20 Vynce joined #perl6
10:21 dalek joined #perl6
10:23 AlexDaniel joined #perl6
10:31 darutoko- joined #perl6
10:32 user9 joined #perl6
10:32 xfix joined #perl6
11:00 pyrimidine joined #perl6
11:06 pyrimidine joined #perl6
11:12 bjz_ joined #perl6
11:16 brokenchicken call for a new TPF grants committee member: http://news.perlfoundation.org/2017/01/grants-committee-needs-a-new-m.html
11:16 AlexDaniel joined #perl6
11:17 raschipi joined #perl6
11:19 lukaramu joined #perl6
11:20 wamba joined #perl6
11:24 pyrimidine joined #perl6
11:29 Cabanossi joined #perl6
11:38 * brokenchicken reads backlog
11:39 ugjka joined #perl6
11:39 brokenchicken You guys are overdoing it with type checks.
11:39 brokenchicken Which is why you feel it's "annoying".
11:42 moritz that's what I tell people regularly
11:42 moritz "Perl 6 is not Haskell"
11:43 El_Che moritz: of "Perl 6 is a slower Haskell" :)
11:43 El_Che or
11:45 DrForr Type checking has caught a few things for me, almost enough to justify keeping the return types around.
11:45 jnthn To be fair, it takes time to develop a feel for when to do it and when not to.
11:45 jnthn Just part of growing up as a Perl 6 programmer.
11:45 timotimo sure, but it'd be nice if you listened to the people trying to help you out ;)
11:45 timotimo Just Programmer Things
11:45 jnthn Listening to people who know stuff isn't the in thing these days. :P
11:45 El_Che :)
11:47 arnsholt Neat! http://gravitational.com/blog/instant-ssh-github/
11:49 timotimo that's cute
11:49 FROGGS joined #perl6
11:49 brokenchicken Neat indeed. Now all I need is a remote team to let ssh into my laptop...
11:55 parv brokenchicken: are you looking for people to log on do your 'puter and do stuff? ;->
11:55 arnsholt brokenchicken: Like #perl6? =)
11:55 brokenchicken hehehe :)
11:55 timotimo outsource all your work for cheap?
12:11 parv joined #perl6
12:12 parv joined #perl6
12:14 Cabanossi joined #perl6
12:20 wololoooo_ joined #perl6
12:35 abraxxa1 joined #perl6
12:36 gfldex where could that come from? https://gist.github.com/gfldex/59e957330691744e2c5d90493a567264
12:37 abraxxa joined #perl6
12:39 timotimo does --ll-exception give you anything?
12:39 gfldex an empty file with `use Git::Config` is causing it
12:39 yqt joined #perl6
12:40 gfldex updated with stack trace https://gist.github.com/gfldex/59e957330691744e2c5d90493a567264
12:40 gfldex i will check on a clean vm
12:41 abraxxa joined #perl6
12:44 rindolf joined #perl6
12:51 abraxxa joined #perl6
12:51 gfldex same in a clean VM
12:56 pmurias joined #perl6
13:02 andrzejku joined #perl6
13:11 brokenchicken Hah. Funny: https://stackoverflow.com/admin.php
13:12 rba_ joined #perl6
13:16 arnsholt Hee hee
13:16 arnsholt I should add that rule to my Varnish =D
13:17 brokenchicken m: say v6 before v6.c
13:17 camelia rakudo-moar d06d7c: OUTPUT«False␤»
13:17 brokenchicken gfldex: ^ fwiw, the "perl" meta field is supposed to specify the minimum required perl version and yours won't even be met
13:17 brokenchicken in Git::Cojnfig
13:18 brokenchicken wonder if that Nil is due to something missing in the META file
13:18 gfldex checking right now
13:20 brokenchicken Hm, I don't get any nils with empty file with use Git::Copnfiug  in it on This is Rakudo version 2016.12-376-g69a66b7 built on MoarVM version 2016.12-113-gd1da1ba
13:21 gfldex This is Rakudo version 2017.01-56-g0ad05ce built on MoarVM version 2017.01-13-g4aab506
13:27 gfldex `-I ../perl6-git-config/lib/` hides the problem
13:31 brokenchicken Hm, I see you can't do QAST::Var.new(:name(~$<ident>), :scope<lexical>, |(:decl<var> if $*DECL),) in NQP? Like slipping in some named args based on condition
13:31 jnthn No
13:31 brokenchicken Oh well :)
13:31 jnthn You'd have to put it into a hash and the flatten that in
13:31 brokenchicken Ah
13:32 jnthn NQP does sigil analysis to decide if we're flattening in positionals or nameds
13:32 jnthn So it's a static decisions
13:32 jnthn *decision
13:33 jnthn Unlike in Perl 6 where it's a dynamic decsion (which is why it's an order of magnitude slower :P)
13:33 brokenchicken :D
13:34 * gfldex .oO( Perl 6 is slow to allow humans to keep up. )
13:44 brokenchicken Ain't getting Nils on This is Rakudo version 2017.01-58-gd06d7c1 built on MoarVM version 2017.01-13-g4aab506 either
13:44 brokenchicken in a git clone, with perl6 -Ilib foo.p6, where foo.p6 has just use Git::Config
13:46 Actualeyes joined #perl6
13:53 kurahaupo__ joined #perl6
13:54 trnh joined #perl6
13:54 trnh joined #perl6
13:57 brokenchicken w00t! Just added varibles support to Perl 7. No sigils. No declarators. Varibles are auto-declared on first use. Pretty revolutionary idea; I hope it won't have problems ????
13:58 moritz \o/ revolutionary languge design!
13:58 brokenchicken :)
13:59 Cabanossi joined #perl6
14:03 khw joined #perl6
14:04 salv0 joined #perl6
14:13 raschipi brokenchicken: There's a strong parallel between variables and hash lines. You should make use of that, just give the programmer a single hash and have him use the lines as variables.
14:17 brokenchicken hm, some langs use "func" to declare functions, but usually have braces or "end" to signal end of its body. I'm surprised no one used "tion" to end such declarations. The "tion" matches the length of "func" AND the one and true ident width of 4 spaces. Makes things look pretty.
14:17 brokenchicken or shorten it to 3 chars: "fun" and "ion" :P
14:18 raschipi Identation is made with tabs.
14:18 raschipi And a tab is 8 spaces.
14:19 brokenchicken uhuh. I wanna see you write Perl 6 code using that.
14:20 raschipi It's the only aceptable way to do it, to be quite honest.
14:21 timotimo ===SORRY NOT SORRY!=== It seems like you're indenting your code with spaces. what the fuck is wrong with you!
14:22 brokenchicken raschipi: Perl 6 is not C. If you're using 8-spaced tabs, I hope you also plan on setting your line lengths to 200 chars
14:22 raschipi Yep, the compiler should enbforce it, this would be very helpful for programmers.
14:26 ocbtec joined #perl6
14:29 pmurias brokenchicken: you want to release your toy language as Perl 7? :/
14:29 raschipi brokenchicken: currently my terminal has 264 columns.
14:29 brokenchicken Toy?!
14:30 pmurias * pet language
14:31 raschipi Perl5 and Perl 6 are TimToady's pet languages.
14:32 brokenchicken pmurias: Nah, bruh. It's a legit lang. I listened to complains about Perl 6 and solved them.... Like the logo stuff: https://github.com/perl7/perl7
14:33 brokenchicken Actually, the logo was the majorest issue. From what I understand this is what managers pay attention too. And with that solved, Perl 7 will flourish. The rest of the features are just details, though I think the single-character print function is sure to turn some of the heads, in today's fast-paced, no-time-for-nonsense world.
14:33 profan man, is it a sad indicator when it's actually hard to tell if obvious satire is supposed to be satire or not :D
14:33 mst brokenchicken++ # finally, somebody who understands
14:34 mst pmurias: I would love him to; given the general stupidity levels of conversations involving 'perl7' ending the conversation by linking this would actually be an improvement
14:35 brokenchicken :)
14:35 SmokeMachine brokenchicken: why not use something like ???? for comments?
14:36 pmurias mst: I'm just worried someone will try to install that by mistake
14:38 mst pmurias: if you're worried about confusing naming, start a campaign to rename perl6. if you're not going to start such a campaign, it's clearly not that important to you and you're worrying about nothing :P
14:39 SmokeMachine brokenchicken: I mean for perl7
14:41 brokenchicken Nah, the comments will stay as '#' because I don't know how to configure my editor to use something else when I press CTRL+D :)
14:41 huf the logo is particularly spiffy
14:41 huf possibly the best proglang logo i've ever seen
14:42 brokenchicken See? People notice!
14:42 sammers joined #perl6
14:42 huf the only way to improve it would be if the image format were .txt
14:42 pmurias mst: I view the Perl 6 naming as already done damage
14:45 raiph .oO ( fun ... `nuf )
14:46 lee_ joined #perl6
14:46 huf that's good but sub/bus is even better because "bus"
14:46 brokenchicken I went with 'fuc' and 'ton'
14:47 raiph :)
14:47 mst huf: having a compiler that can say 'bus error at line 13' would be AWESOME
14:47 huf mst: i was hoping to fit "fuck you i'm a bus" in there somewhere
14:47 sena_kun joined #perl6
14:48 sena_kun m: my UInt $u = 1; say $u.WHAT;
14:48 camelia rakudo-moar d06d7c: OUTPUT«(Int)␤»
14:48 cdg joined #perl6
14:48 sena_kun it should be int, not uint?
14:48 brokenchicken sena_kun: UInt is just a core subset of Int
14:48 raiph you what?
14:48 sena_kun brokenchicken, hmmm, okay. I'll fix docs then, thanks.
14:49 brokenchicken mc: my UInt $u = 1; say $u.WHAT;
14:49 committable6 brokenchicken, ¦«2015.12»: (Int)
14:50 brokenchicken sena_kun: they also list wrong definition. It's https://github.com/rakudo/rakudo/blob/nom/src/core/Int.pm#L6
14:50 cdg joined #perl6
14:51 brokenchicken We also have UInt64, but IIRC it needs a bit of loving first
14:51 * sena_kun noted.
14:51 brokenchicken liek that .defined check
14:52 Gasher joined #perl6
14:55 brokenchicken m: say Int ~~ UInt64
14:55 camelia rakudo-moar d06d7c: OUTPUT«Invocant requires an instance of type Int, but a type object was passed.  Did you forget a .new?␤  in block <unit> at <tmp> line 1␤␤»
14:55 brokenchicken mhm
14:55 brokenchicken m: say Int ~~ UInt
14:55 camelia rakudo-moar d06d7c: OUTPUT«True␤»
14:56 brokenchicken And the other one is to check if it got a off-by-one issue on its max value. I think the constant it uses is 1-too-many, but then the check in it is `<` so it's fine. Or something or other
14:57 brokenchicken Don't even know if it's meant to be public tho :)
14:57 brokenchicken Or whether it's even roast
14:57 brokenchicken edc
14:57 sena_kun afaik uint is in roast.
14:59 Cabanossi joined #perl6
14:59 sena_kun https://github.com/perl6/roast/blob/b4da789b94c74f6fa3d9e7612ab99c2b36150056/S32-num/int.t#L147
15:01 skids joined #perl6
15:02 brokenchicken yeah
15:08 AlexDaniel brokenchicken: :P https://gist.github.com/AlexDaniel/3e7ad0adac5d646bd786aeab5332b570
15:09 brokenchicken heh
15:09 FROGGS joined #perl6
15:09 mst AlexDaniel++ # yaaaaaaaaaaaasss
15:12 * brokenchicken steals it
15:12 AlexDaniel feel free :P
15:13 raiph .oO ( Perl ל aka Perl Lamed -- https://en.wikipedia.org/wiki/Lamedh)
15:13 AlexDaniel brokenchicken: though maybe you want to slap this : on your own version, there are some tiny issues inside the P hole due to tracing
15:14 brokenchicken Yes, I will have the logo designer add it. I paid $3000 for this logo design, so he should have high quality files to edit.
15:15 AlexDaniel should've paid $6000 for svg
15:15 bwisti joined #perl6
15:16 brokenchicken Pfft... The real high quality logos don't use SVGs. They use .ai
15:16 brokenchicken Everybody knows it ;)
15:16 AlexDaniel forgive my incompetence
15:17 brokenchicken It's OK. That's why I have the guy making $3000 logos for me.
15:18 AlexDaniel samcv: so where's your grant proposal? :o
15:19 cdg_ joined #perl6
15:24 [Coke] AlexDaniel: AIUI, waiting for the board to decide if she can apply against the core perl 6 fund instead of the hague fund. (or the general grant committee fund, which brokenchicken submitted against)
15:24 yoleaux 02:23Z <samcv> [Coke]: I have set --no-proc-async as default for MacOS
15:24 AlexDaniel oh
15:26 [Coke] but the original submission against hague won't work, as apparently that fund is depleted. (which I am surprised to hear, and would love to see if there's a list of outstanding in-progress-but-unpaid grants.)
15:27 AlexDaniel oh… ouch…
15:29 sena_kun is our Language docs section compileable?
15:29 mst [Coke]: ah, yes, "trying to get financial information out of TPF" ... good luck
15:30 mspo our logo guy produced .eps
15:31 brokenchicken AlexDaniel: my logo guy said your version sucked, but for another $1,000 he'd make me an improved logo that better captures the spirit of the language and emboldens the brand identity or something or other. I said good deal, and behold! https://github.com/perl7/perl7#perl-7-the-up-to-date-perl
15:32 [Coke] mst: It's super challenging, and I'm on the GC.
15:33 brokenchicken mspo: clearly you're not getting the bestest logos!
15:37 brokenchicken sena_kun: compilable to what?
15:40 sena_kun brokenchicken, I meant ability to automatically compile examples and check their validness(this point is still LTA). it is possible for Type/ section, but I didn't work on Language/, hence the question.
15:41 brokenchicken Ah, no idea.
15:41 sena_kun it's time to try it to check!
15:41 alnewkirk joined #perl6
15:53 tokomer joined #perl6
15:56 tinita joined #perl6
15:56 sena_kun m: my %pairs = also-named => 4; sub f(|c) {}; say f |%pairs;
15:56 camelia rakudo-moar d06d7c: OUTPUT«Nil␤»
15:57 sena_kun should it be Nil here?
15:57 brokenchicken Yup
15:57 sena_kun doc page is lying!
15:57 brokenchicken What's it saying/
15:57 sena_kun (:$also-named)
15:58 brokenchicken :S
15:58 brokenchicken m: my %pairs = also-named => 4; sub f(|c) {c}; say f |%pairs;
15:58 camelia rakudo-moar d06d7c: OUTPUT«\(:also-named(4))␤»
15:58 sena_kun Signature page.
15:58 domidumont joined #perl6
15:58 brokenchicken That if anything.
15:58 brokenchicken c: 2015.07 my %pairs = also-named => 4; sub f(|c) {c}; say f |%pairs;
15:58 committable6 brokenchicken, ¦«2015.07»: \(:also-named(4))
15:58 brokenchicken been that way for ages
16:00 brokenchicken Looks like it's just indicating the meaning instead of showing output
16:01 sena_kun hmm... Let it be like that for a bit until my future pr merge/rejection.
16:10 lichtkind joined #perl6
16:14 user9 joined #perl6
16:19 araujo joined #perl6
16:26 cale2 joined #perl6
16:27 grw joined #perl6
16:27 cale2 I'm having an issue with Str.comb
16:28 cale2 It seems to spit out a Seq. I thought it would be a List.
16:28 brokenchicken cale2: what sort of issue
16:28 cale2 https://docs.perl6.org/routine/comb#(Str)_routine_comb
16:29 SCHAAP137 joined #perl6
16:29 cale2 The problem for me is building up subsets. I assumed that everything would be lists, but it turns out that comb (the only way to get a list of graphemes from a Str) returns a Seq
16:30 cale2 subset Char     of Str  where { $_.chars == 1    };
16:30 cale2 subset Word     of Seq  where { $_.all   ~~ Char };
16:30 cale2 subset WordList of List where { $_.all   ~~ Word };
16:30 st_elmo joined #perl6
16:30 brokenchicken Wow, you're sure stepping up the insanity on the type checks :)
16:31 cale2 Having to deal with Lists, Seqs, and (possibly) Arrays sort of muddies the waters. This is where type signatures come in handy
16:31 moritz cale2: so use .comb.List
16:31 brokenchicken cale2: but what's the actual problem you're facing? So far you said you assumed one thing but (on HEAD) it's neither documented nor does return the thing you assumed it does
16:31 moritz cale2: or rely on Iterable or Positional
16:31 cale2 Because I could do a search for "Seq -> List" or "Array -> List"
16:31 moritz cale2: but really, don't treat strings as lists
16:32 moritz cale2: subset Word of Str where /^ \w* $/;
16:32 cale2 Treating strings as a list of chars is quite a natural way to think about textual problems, though
16:34 moritz but it's not natural way in Perl 6
16:34 moritz play to the language's strengths, not its weaknesses
16:35 brokenchicken Seq don't do Positional, only Iterable is shared between Seq and Lists. But note that Maps do it as well and I recall a convo on how about making QuantHash do it too
16:35 moritz m: say "Perl 6 is not $_" for <Haskell Python C C++ Java Perl5>;
16:35 camelia rakudo-moar d06d7c: OUTPUT«Perl 6 is not Haskell␤Perl 6 is not Python␤Perl 6 is not C␤Perl 6 is not C++␤Perl 6 is not Java␤Perl 6 is not Perl5␤»
16:35 * brokenchicken nods
16:35 cale2 moritz: I guess I don't know how to think that way then. For example, I'm very accustomed to mapping a function over a string and having it apply to each character
16:36 moritz cale2: and is that how you think about text in real life, or in your English classes?
16:36 AlexDaniel joined #perl6
16:36 brokenchicken cale2: I saw your repo where you try to recreate programs in other languages to be close to other languages, but I think that's a good recipe for frustration and for missing out on learning all the good idioms of Perl 6
16:37 cale2 I thought one main selling point of perl6 is that everything and nothing is idiomatic and you can basically write it however you want
16:37 brokenchicken Like this typecheck stuff you're doing right now. Along with being frustrating, you'll find the way you're doing right now is also terrible slow.
16:37 cale2 I'm finding that the haskell translations are working well though lol
16:37 brokenchicken Never heard of such a selling point.
16:38 moritz cale2: that's not the Perl 6 way
16:38 Gasher joined #perl6
16:38 brokenchicken The TIMTOADY thing in my view is you can write things many different ways and use the more suitable way depending on situation, same as you use a natural language.
16:38 moritz there's more than one way to do it, but if an experienced programmer tells you that one way works clearly better in the language, you should still listen
16:40 cale2 brokenchicken: https://youtu.be/Nq2HkAYbG5o?t=26m40s
16:41 cale2 I guess I took that presentation too literally lol
16:42 bwisti joined #perl6
16:42 moritz Damian's talk aim to dazzle the audience, more than they try to be accurate, or teach idiomatic programming
16:42 moritz which is fine; if you wanted to learn Perl 6, you could always buy the book :-)
16:44 brokenchicken :) yeah. Same talk shows the quantum computing done in Perl 6... despite there yet not being hardware to do the stuff :)
16:49 cale2 moritz: what book?
16:50 dataf3l joined #perl6
16:50 brokenchicken cale2: https://leanpub.com/perl6
16:50 moritz cale2: https://leanpub.com/perl6 (disclaimer: my own, ebook), https://deeptext.media/perl6-at-a-glance/ (print)
16:50 pmurias joined #perl6
16:50 brokenchicken cale2: though when Damian gave the talk you linked in Toronto, he promised a book too "in a year".... That was almost a year ago :P
16:51 brokenchicken cale2: also https://www.learningperl6.com/ though by reading the preview copies of that one, I'd say it's more geared towards beginner programmers.
16:53 brokenchicken huggable: books :is: "Perl 6 At A Glance" https://deeptext.media/perl6-at-a-glance/ (print only for now); "Perl 6 By Example": https://leanpub.com/perl6 (get preview digital copies now!) ; "Learning Perl 6": https://www.learningperl6.com/ (good for programming beginners)
16:53 huggable brokenchicken, Added books as "Perl 6 At A Glance" https://deeptext.media/perl6-at-a-glance/ (print only for now); "Perl 6 By Example": https://leanpub.com/perl6 (get preview digital copies now!) ; "Learning Perl 6": https://www.learningperl6.com/ (good for programming beginners)
16:53 brokenchicken There's no longer a way to get preview copies of LP6 is there?
16:54 moritz convince bdfoy to give somebody access to the repo :-)
16:57 dataf3l joined #perl6
16:57 astj joined #perl6
17:00 dataf3l joined #perl6
17:01 TimToady .tell grondilu the subsig ($a,$b) is always treating pairs as named args, not positionals; perhaps we could distinguish a [$a,$b] subsig to force positional and let the () form turn pairs to nameds
17:01 yoleaux TimToady: I'll pass your message to grondilu.
17:02 TimToady .oO(if I had access to bdfoy's repo I'd fix the outdents on closing brackets...)
17:03 brokenchicken I can give you the next best thing: open an Issue on bdfoy's repo quoting the above :)
17:04 [Coke] huh. work project using rakudo-star:latest docker image had to go back to rakudo-star:2016.07 (maybe didn't have to go back that far, but that one worked) due to panda not being able to download the ecosystem module.
17:06 brokenchicken Well, damn... http://lifepluslinux.blogspot.de/2017/01/look-before-you-paste-from-website-to.html
17:06 brokenchicken The saddest part is I'll probably still continue copy-pasting commands to terminal >_<
17:07 brokenchicken [Coke]: FWIW there's a ticket of another person having issues with latest star: https://rt.perl.org/Public/Bug/Display.html?id=130339#ticket-history
17:07 brokenchicken Oh. It was lack of prove. Maybe it should be closed now that zef is the Gold Standard
17:08 brokenchicken As is " Rakudo* 2014:08 panda doesn't see bundled modules ": https://rt.perl.org/Public/Bug/Display.html?id=122839#ticket-history
17:19 brokenchicken Done!
17:20 brokenchicken "His Majesty Larry of Wall demands you abandon this weird brace style OR ELSE!"
17:20 brokenchicken :) http://i.imgur.com/XVeLAav.png
17:21 pmurias did learning Perl (5) use that sort of weird brace style?
17:22 brokenchicken Well, he didn't write that one. At least not the first editions.
17:24 astj joined #perl6
17:24 * brokenchicken is failing to find any screencaps of the book on google
17:25 pmurias I found the examples on the internet and it used normal formating
17:26 pmurias I'm not sure where I have my old Learning Perl printed copy
17:28 brokenchicken Well, in the presumably pirated PDF of 5th edition I found online all examples seem to use normal braces EXCEPT for one example with `foreach` that does use that weird bracing
17:29 Cabanossi joined #perl6
17:29 Geth doc: Altai-man++ created pull request #1165: Make comment style the same in Type/ directory
17:29 Geth doc: review: https://github.com/perl6/doc/pull/1165
17:29 brokenchicken Actually just one example of foreach.. The rest all use normal formatting... My guess would be bdfoy used that style for ages and for that book but editor missed this one example
17:29 sena_kun Discussion around this pr is like, super-welcome.
17:30 sena_kun .tell AlexDaniel https://github.com/perl6/doc/pull/1165
17:30 yoleaux sena_kun: I'll pass your message to AlexDaniel.
17:31 brokenchicken Hahaha
17:31 brokenchicken bdfoy replied to my issue
17:32 brokenchicken http://i.imgur.com/56taHJt.png
17:32 cale2 somone make a thug life meme out of that interaction
17:32 brokenchicken I'll assume he's still butthurt about my grant comments and not general dick to people opening issues :P
17:35 brokenchicken m: say (a => 1, 'b', 'c').pairup.perl;
17:35 camelia rakudo-moar d06d7c: OUTPUT«(:a(1), :b("c")).Seq␤»
17:36 brokenchicken TIL we have pairup. Neat
17:39 brokenchicken sena_kun: well, I reviewed up to "doc/Type/Attribute.pod6" and didn't spot anything. Don't have time to review more as I gotta work :)
17:45 astj joined #perl6
17:47 SmokeMachine why is the "master" branch named nom?
17:48 brokenchicken it's more fun that way :)
17:48 brokenchicken SmokeMachine: AFAIK it was first the "new object model" branch but it then became what master should be and well, got left that way
17:49 brokenchicken And if we switch now force of habit + some tools that know it to be master would be confused
17:49 nemsys joined #perl6
17:54 agentzh joined #perl6
17:54 sena_kun brokenchicken, reviews are good, "agree/disagree" comments are better. ;) Thanks for attention anyway.
18:01 samcv joined #perl6
18:02 bstamour joined #perl6
18:11 brokenchicken m: sub foo { fail "foo" }; my $x = foo or dd $x\
18:11 camelia rakudo-moar d06d7c: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Cannot use variable $x in declaration to initialize itself␤at <tmp>:1␤------> 3 foo { fail "foo" }; my $x = foo or dd $7⏏5x\␤    expecting any of:␤        argument list␤        term␤»
18:11 brokenchicken m: sub foo { fail "foo" }; my $x = foo or dd $x
18:11 camelia rakudo-moar d06d7c: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Cannot use variable $x in declaration to initialize itself␤at <tmp>:1␤------> 3 foo { fail "foo" }; my $x = foo or dd $7⏏5x␤    expecting any of:␤        argument list␤        term␤»
18:12 brokenchicken Is this an accurate error? The sub eats up the `or` and it's not lower precedence to be equal to my $x = foo() or dd $x?
18:13 brokenchicken Ah
18:13 brokenchicken m: sub foo { fail "foo" }; my $y = 42; my $x = foo or dd $y
18:13 camelia rakudo-moar d06d7c: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Undeclared routine:␤    or used at line 1␤␤»
18:13 brokenchicken It's seen as a routine and so yeah, gobbled up by foo
18:14 lukiramu joined #perl6
18:14 brokenchicken Hah
18:15 brokenchicken bdfoy now also deleted my pointing out the example is broken.
18:16 brokenchicken Well, I have no compunctions about not recommending his book to people.
18:20 nine brokenchicken: FWIW I have seen this bracing style in the wild. And I find it confusing as hell. But it's all just a matter of getting used to it.
18:23 mst I've only ever seen it on bdf's blog
18:23 espadrine_ joined #perl6
18:23 TimToady all it says to me is "I wish I were python"
18:23 brokenchicken Eh, fuck him. I see him and his pet nanis trash P6 on twitter. I tried to be nice about him and be open minded about his book but if he wants to play penis games on GitHub, I couldn't give two shits.
18:24 brokenchicken Which reminds, I need to load Perl 6 by Example on my ereader to peruse on the bus \o/
18:24 mst brokenchicken: oh, yeah, if you publically disagree with him once he'll be a dick to you forevermore
18:24 mst I've lost count of the number of people he refuses patches from etc. because of some minor disagreement being turned into a vendetta
18:25 El_Che are we talking about the same guy I gave a small amount of money to write a book?
18:27 brokenchicken El_Che: I dunno, there are at least three guys writing a book you may have given a small amount of money to write a book :)
18:27 El_Che hehe
18:28 mst El_Che: we're talking about bdfoy
18:28 El_Che yeah, I know, I was being silly. More of a rhethorical thing
18:29 brokenchicken huggable: books
18:29 huggable brokenchicken, "Perl 6 At A Glance"  https://deeptext.media/perl6-at-a-glance/ (print only for now); "Perl 6 By Example": https://leanpub.com/perl6 (can order preview digital copies)
18:29 brokenchicken ^ the other two guys. The first one is already finished, but no ebook yet and the second one is still in production being produced by our very own morit++
18:29 brokenchicken *moritz++
18:29 robertle joined #perl6
18:30 brokenchicken And the other one is by Anton Shitov
18:33 brokenchicken Ooops. *Andrew Shitov
18:33 astj joined #perl6
18:34 brokenchicken http://act.yapc.eu/ye2015/user/1773
18:35 AlexDaniel joined #perl6
18:52 st_elmo joined #perl6
18:53 jjido joined #perl6
18:58 revere joined #perl6
19:01 pyrimidine joined #perl6
19:13 labster joined #perl6
19:14 Cabanossi joined #perl6
19:19 pyrimidine joined #perl6
19:26 sufrostico joined #perl6
19:27 astj joined #perl6
19:29 jjido joined #perl6
19:36 pyrimidine joined #perl6
19:39 samcv that brace style O.o
19:40 lichtkind_ joined #perl6
19:43 samcv left #perl6
19:43 samcv joined #perl6
19:43 samcv well I found it in this list here https://en.wikipedia.org/wiki/Indent_style#Placement_of_braces
19:43 Geth ecosystem: 274143b6e4 | (Wenzel P. P. Peppmeyer)++ | META.list
19:43 Geth ecosystem: add META6-bin
19:43 Geth ecosystem: review: https://github.com/perl6/ecosystem/commit/274143b6e4
19:43 samcv seems to be Ratliff style: https://en.wikipedia.org/wiki/Indent_style#Ratliff_style
19:44 bstamour left #perl6
19:45 kyan joined #perl6
19:47 brokenchicken https://github.com/gfldex/perl6-meta6-bin/
19:49 brokenchicken gfldex: `perl` is the "minimum perl 6 version this module works on" and you're setting it to v6 by default, which will never be released.
19:49 brokenchicken You meant v6.c (or v6.*)
19:49 girafe joined #perl6
19:51 * brokenchicken sees META6 making the same mistake in SYNOPSIS :/
19:52 brokenchicken m: say Version.new('6') before v6.c
19:52 camelia rakudo-moar d06d7c: OUTPUT«False␤»
19:52 brokenchicken m: say Version.new('6.*') before v6.c
19:52 camelia rakudo-moar d06d7c: OUTPUT«True␤»
19:52 brokenchicken m: say Version.new('6.c') before v6.c
19:52 camelia rakudo-moar d06d7c: OUTPUT«False␤»
19:52 brokenchicken m: say Version.new('6.c') eqv v6.c
19:52 camelia rakudo-moar d06d7c: OUTPUT«True␤»
19:54 gfldex unsurprisingly given that roast doesn't test for v6.c to be younger then v6
19:55 brokenchicken hm?
19:55 gfldex m: say so v6 ~~ v6.0
19:55 camelia rakudo-moar d06d7c: OUTPUT«True␤»
19:55 gfldex m: say so v6.0 cmp v6.c
19:55 camelia rakudo-moar d06d7c: OUTPUT«True␤»
19:55 gfldex m: say v6.0 cmp v6.c
19:55 camelia rakudo-moar d06d7c: OUTPUT«More␤»
19:56 TimToady for normal versions, letters are considered variants of "alpha", basically, hence before .0; however, v6c steals that notation for a different purpose
19:56 brokenchicken Yeah
19:56 brokenchicken m: use v6
19:56 camelia rakudo-moar d06d7c: ( no output )
19:56 pyrimidine joined #perl6
19:56 TimToady .oO(Someday we'll come out with Perl 6.0!)
19:56 brokenchicken ^ that one's a bit confusing tho
19:57 brokenchicken I mean as far as figuring out what to stick in `perl` META field
19:57 Gasher joined #perl6
19:57 brokenchicken Maybe special case it to mean any version? It's a pretty common mistake.
19:57 brokenchicken (not that anything cares about `perl` meta field ATM AFAIK)
19:59 Cabanossi joined #perl6
20:00 gfldex no need to special case it. In a few month we will have v6.d . :->
20:02 Geth doc: 3706a3d8fc | (Wenzel P. P. Peppmeyer)++ | doc/Language/pragmas.pod6
20:02 Geth doc: v6 -> v6.c
20:02 Geth doc: review: https://github.com/perl6/doc/commit/3706a3d8fc
20:02 AlexDaniel .
20:02 brokenchicken Hm, Diwali is October 19th this year... Plenty of time :P
20:02 yoleaux 17:30Z <sena_kun> AlexDaniel: https://github.com/perl6/doc/pull/1165
20:03 brokenchicken And it even aligns with regular release date (will be on Oct 21)
20:04 mspo how do I type a utf8 combining character, like skin tone modifier?
20:04 AlexDaniel mspo: https://docs.perl6.org/language/unicode_entry ?
20:04 AlexDaniel or you can use “\c[…]”
20:04 brokenchicken m: "\c[woman health worker: medium-dark skin tone]"
20:04 camelia rakudo-moar d06d7c: OUTPUT«WARNINGS for <tmp>:␤Useless use of constant string "????????‍⚕️" in sink context (line 1)␤»
20:04 AlexDaniel depending on where you actually want to type it
20:04 brokenchicken m: "\c[woman health worker: medium-dark skin tone]".say
20:04 camelia rakudo-moar d06d7c: OUTPUT«????????‍⚕️␤»
20:05 brokenchicken m: "\c[woman health worker: medium-dark skin tone]".ords.say
20:05 camelia rakudo-moar d06d7c: OUTPUT«(128105 127998 8205 9877 65039)␤»
20:05 mspo m: "\c[pile of poop: medium-dark skin tone]"
20:05 camelia rakudo-moar d06d7c: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Unrecognized character name [pile of poop: medium-dark skin tone]␤at <tmp>:1␤------> 3"\c[pile of poop: medium-dark skin tone7⏏5]"␤»
20:06 bjz joined #perl6
20:06 brokenchicken it's "pile of poo"
20:06 astj joined #perl6
20:06 brokenchicken And I don't think skinton modifiers work just on anything
20:07 brokenchicken m: "\c[snowman: medium-dark skin tone]".say
20:07 camelia rakudo-moar d06d7c: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Unrecognized character name [snowman: medium-dark skin tone]␤at <tmp>:1␤------> 3"\c[snowman: medium-dark skin tone7⏏5]".say␤»
20:07 brokenchicken hehe
20:07 AlexDaniel no that's not how it works
20:08 AlexDaniel here's the list: http://unicode.org/Public/emoji/4.0/emoji-zwj-sequences.txt
20:08 mspo oh well
20:08 brokenchicken And purty chart: http://unicode.org/emoji/charts/emoji-zwj-sequences.html
20:08 Geth doc: a7a81c3d2a | (Wenzel P. P. Peppmeyer)++ | 13 files
20:08 Geth doc: v6 -> v6.c
20:08 Geth doc: review: https://github.com/perl6/doc/commit/a7a81c3d2a
20:08 brokenchicken m: (128169 127998 8205 9877 65039)».chr.say
20:08 camelia rakudo-moar d06d7c: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Two terms in a row␤at <tmp>:1␤------> 3(1281697⏏5 127998 8205 9877 65039)».chr.say␤    expecting any of:␤        infix␤        infix stopper␤        statement end␤        statement modif…»
20:08 AlexDaniel you can slap a tone modifier on a poo character, sure, but not with \c[]
20:09 brokenchicken m: <128169 127998 8205 9877 65039>».chr.say
20:09 camelia rakudo-moar d06d7c: OUTPUT«(???? ???? ‍ ⚕ ️)␤»
20:09 AlexDaniel .oO( and a gender sign, if that matters… )
20:09 brokenchicken How?
20:09 mspo I more more font
20:09 mspo need more fon
20:09 mspo t
20:10 AlexDaniel brokenchicken: poo + zwj + gender sign, that's it
20:10 AlexDaniel brokenchicken: no font will render it, but…
20:10 AlexDaniel you can make one that does
20:11 brokenchicken m: "\c[snowman]\c[ZERO WIDTH JOINER]\c[WOMAN]".say
20:11 camelia rakudo-moar d06d7c: OUTPUT«☃‍????␤»
20:12 st_elmo joined #perl6
20:12 brokenchicken Don't I just print two separate chars with a ZWJ in the middle and... that's it?
20:12 brokenchicken There's no slapping involved?
20:13 AlexDaniel you don't have to slap yourself, yes
20:15 brokenchicken :)
20:15 AlexDaniel font developers may tell you otherwise though
20:17 AlexDaniel but that's actually an interesting question… how often people use that stuff at all?
20:17 AlexDaniel and if their font does not support it, are they going to do it anyway?
20:19 brokenchicken I'd think this is more used by apps that use their own pictures for all of these
20:20 AlexDaniel sena_kun: there are conflicts
20:20 brokenchicken And the user just picks the emoji from emoji selector and is completely unaware of wtf Unicode even is
20:20 sena_kun AlexDaniel, I know, I noted it in the description.
20:20 astj joined #perl6
20:22 sena_kun AlexDaniel, they can be resolved when I finish my struggling with cabal hell. :/
20:23 AlexDaniel sena_kun: I want to just merge the damn thing to be honest
20:23 AlexDaniel sena_kun: the issue was up for 6 months or so, nobody is interested in discussing it. Let's merge and *then* discuss :P
20:23 sena_kun AlexDaniel, I think it is because nobody want such changes. :o
20:24 AlexDaniel nope
20:24 brokenchicken Hm, getting "Cannot find method 'keyword' on object of type Perl7::Grammar" any idea how come? https://github.com/perl7/perl7/blob/master/src/perl7.nqp#L46
20:25 brokenchicken This example has it and it compiles fine: https://github.com/edumentab/rakudo-and-nqp-internals-course/blob/master/examples/rubyish-3.nqp#L37
20:25 brokenchicken Oh... and it defines token keyword. Nevermind, I thought it came from HLL::Grammar
20:27 dj_goku joined #perl6
20:29 Cabanossi joined #perl6
20:31 zakharyas joined #perl6
20:34 [Coke] (diwali) we are NOT release 6.d on the day of diwali
20:34 [Coke] *releasing
20:34 brokenchicken Why not?
20:36 brokenchicken Hm. For some reason function support in Perl 7 don't work right. Fails to parse: https://github.com/perl7/perl7/blob/master/src/perl7.nqp
20:36 * brokenchicken tables it for another day
20:39 [Coke] 1) brokenchicken - well, while diwali is far enough for us to start planning, I don't want us to get locked into releasing on the day of. And we had a lot of pressure on the last day of release that we ended up cutting a few corners - like I said, probably have enough time to do it right this time, but would rather schedule to the work, not the holiday.
20:40 [Coke] ... I started that off as a 1) list and then lost track of that as I kept typing. sorry. )
20:40 [Coke] :)
20:40 geekosaur "a few corners"
20:40 geekosaur it really needed to bake for several more months
20:41 brokenchicken [Coke]: oh right that yeah :)
20:41 brokenchicken Diwali 2018 looks attractive ;)
20:41 [Coke] I don't mean compiler functionality, btw, but language-level things like to deal with cutting a copy of roast, that sort of thing.
20:42 brokenchicken .oO( don't name releases after holidays )
20:42 brokenchicken Name it just The D. The younger generation will love it.
20:42 geekosaur do that but ship it 9 months later. fashionably late...
20:42 brokenchicken heh
20:45 pukku joined #perl6
20:45 pukku Hi! There are two Base64 modules on modules.perl6.org -- MIME::Base64 and Base64. Does anyone know why I would prefer one over the other?
20:46 [Coke] we don't have a quality rating yet, but I would check test coverage, recent-ness of commits...
20:47 dontbe^2 joined #perl6
20:47 Geth doc/master: 4 commits pushed by coke++
20:47 Geth doc/master: c611ffdfe2 | remove trailing whitespace…
20:47 Geth doc/master: 206f9d9496 | use nbsp…
20:47 Geth doc/master: 13f2f59010 | learn new words…
20:47 Geth doc/master: 223e38759c | Remove 'blead' and try to tag a release instead…
20:47 Geth doc/master: review: https://github.com/perl6/doc/compare/a7a81c3d2a...223e38759c
20:47 mspo did perl6 grow its own cpan yet?
20:47 [Coke] (docs) trying something out with 223e38759c - maybe we can use something like when dealing with compiler/language versions.
20:47 mspo zefAN?
20:47 [Coke] mspo: no, we're going to use cpan's cpan
20:48 mspo [Coke]: why?
20:48 [Coke] ... why would we build our own if we didn't have to?
20:48 [Coke] building wheels is HARD
20:48 mspo yeah
20:48 pukku They both pass tests, they both have similar commit recentness. One has 3 stars, and one has 1 star, but I don't know how important stars are...
20:49 mspo does CPAN need to grow a supported-perls or something?
20:49 [Coke] mspo: the perl6 stuff will be invisible to the p5 toolchain
20:49 [Coke] and vice versa.
20:49 mspo interesting
20:50 pukku So I decided to ask and see if anyone has preferences...
20:51 perlpilot pukku: MIME::Base64
20:51 pukku Cool! Any reason, or just flip of the coin?
20:51 perlpilot for no real reason other than it's in perl6  :-)
20:51 pukku Ok
20:52 perlpilot I don't think I've even used ugexe's version
20:52 mspo Base64 looks more "modern"?
20:52 kurahaupo__ joined #perl6
20:53 mspo but MIME:: looks nice too
20:56 pukku Looking deeper, I just noticed the "oneline" option to the MIME::Base64 functions, which makes the choice simpler, as I want back one line.
20:56 pukku Thanks!
20:57 pukku On a different topic, I real somewhere that I should start using zef instead of panda, because that is going to be the new installer?
20:58 gfldex pukku: yes
20:58 [Coke] yes, zef is preferred instead of panda these days
20:59 sena_kun pukku, yes, it seems so.
20:59 pukku Is there something I should do with Homebrew on a Mac to get zef installed? I installed rakudo-star via homebrew,
20:59 pukku (s/,/./)
21:02 RabidGravy panda install zef
21:02 RabidGravy wfm
21:03 pukku Thanks!
21:05 astj joined #perl6
21:07 pmurias joined #perl6
21:14 Cabanossi joined #perl6
21:15 pukku And on another topic, if I want to find out the Content-Type from something fetched with LWP::Simple, is there a way to do that?
21:15 pukku Or is there something more like LWP::UserAgent for Perl 6?
21:16 brokenchicken buggable, eco http::useragent
21:16 buggable brokenchicken, HTTP::UserAgent 'Web user agent': https://github.com/sergot/http-useragent
21:17 pukku Thanks! I also just found Net::HTTP, so I will look at the two.
21:17 cdg joined #perl6
21:18 brokenchicken dont see a way to get content type from httpua (at least by reading the code)
21:20 brokenchicken oh wait .content-type on response
21:23 cdg joined #perl6
21:24 labster joined #perl6
21:25 brokenchicken RabidGravy: are you open to a decent docs PR for HTTP::UserAgent....?
21:25 RabidGravy of course :)
21:25 brokenchicken seeing as current "for documentation links" seem to just point to random locations in source :/
21:26 brokenchicken Oh, maybe there ARE pod docs and the links are just messed up
21:26 brokenchicken yup, there's pod
21:26 RabidGravy The docs did fail to keep up with the code
21:28 RabidGravy so shall I install IBM MQ and have a crack at making a client binding
21:32 brokenchicken vOv why not :)
21:35 pukku Are there any weirdnesses with `qq:to/FOO/`?
21:36 pukku In particular, I'm trying to interpolate some variables into some HTML text, but perl6 is complaining about trying to compile stuff that should be inside the heredoc...
21:36 moritz pukku: if you try to interpolate $foo</p>, the <...> is parsed as a postcircumfix
21:36 moritz (same in double-quoted strings, fwiw)
21:36 geekosaur {} interpolation?
21:37 geekosaur although the other is probably more likely for html, yeh
21:37 gfldex pukku: https://docs.perl6.org/language/faq#%22Type_Str_does_not_support_associative_indexing.%22
21:37 moritz as always, examples and error messages help
21:37 pyrimidine joined #perl6
21:37 pukku Right! OK, there is CSS, and I just copied this out, and forgot to escape the CSS brackets. My bad...
21:38 moritz you can always use Q:s:to/FOO/
21:38 moritz to only interpolate scalars
21:39 brokenchicken m: my $x; say Q:s/$x<y>/
21:39 camelia rakudo-moar d06d7c: OUTPUT«Use of uninitialized value $scalar of type Any in string context.␤Methods .^name, .perl, .gist, or .say can be used to stringify it to something meaningful.␤  in block <unit> at <tmp> line 1␤␤»
21:39 astj joined #perl6
21:41 pukku Many thanks for all the help!
21:41 brokenchicken seems it don't work
21:42 AlexDaniel ?
21:42 brokenchicken I mean :s doesn't prevent <...> from being interpreted as key on scalars
21:42 AlexDaniel m: my $x = 42; say Q:s/$x<y>/
21:42 camelia rakudo-moar d06d7c: OUTPUT«Type Int does not support associative indexing.␤  in block <unit> at <tmp> line 1␤␤Actually thrown at:␤  in block <unit> at <tmp> line 1␤␤»
21:42 AlexDaniel oh
21:43 AlexDaniel m: my $x = 42; say Q:c/{$x}<y>/
21:43 camelia rakudo-moar d06d7c: OUTPUT«42<y>␤»
21:43 AlexDaniel :S
21:48 brokenchicken Hah, this one's funny:
21:48 brokenchicken m: my $x = 42; say Q:s/$x\<y>/
21:48 camelia rakudo-moar d06d7c: OUTPUT«42\<y>␤»
21:49 brokenchicken You can't escape the `<` properly either 'cause of Q :P
21:49 gfldex there is a PR for that
21:50 brokenchicken PR to what?
21:50 brokenchicken Also, am I blind or the adverbs aren't listed in docs :/
21:52 AlexDaniel there is an issue for that
21:54 pyrimidine joined #perl6
21:55 brokenchicken my last eval above isn't a bug since there's no q/qq in adverbs.
21:55 trnh joined #perl6
21:55 brokenchicken or :b
21:55 brokenchicken m: my $x = 42; say Q:bs/$x\<y>/
21:55 camelia rakudo-moar d06d7c: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Unrecognized adverb: :bs␤at <tmp>:1␤------> 3my $x = 42; say Q:bs7⏏5/$x\<y>/␤»
21:55 trnh joined #perl6
21:55 brokenchicken aww
21:55 brokenchicken m: my $x = 42; say Q:b:s/$x\<y>/
21:55 camelia rakudo-moar d06d7c: OUTPUT«42\<y>␤»
21:56 brokenchicken Oh, now I recall a ticket for something like that
21:57 brokenchicken 15 matches on perl6.fail for "interpolation"
21:58 brokenchicken So any idea for a good way to interpolate $ variables in a heredoc with HTML and CSS in it?
21:58 perlawhirl joined #perl6
21:58 SmokeMachine brokenchicken: thanks
21:59 AlexDaniel oh! Oh
21:59 avuserow joined #perl6
21:59 lucs brokenchicken: If you find a good solution, it'll be real useful for me to adapt it to heredocs containing PostScript and LaTeX :-)
22:00 lucs What I'd like in particular is to have a character other than '{' to introduce interpolated evaluation.
22:00 brokenchicken for HTML alone, Q:c works fine with {$whatver} interpolation, but if you add CSS to the mix you're back in backslashing-escape hell
22:00 gfldex brokenchicken: this may help https://github.com/gfldex/perl6-xhtml-writer/blob/master/bin/generate-function-definition.p6#L42
22:01 brokenchicken gfldex: I don't follow. There's a bunch of escapes in that heredoc
22:02 brokenchicken and a confusing use of constants :P
22:02 brokenchicken hmm
22:02 brokenchicken m: my $x = 42; say Q:f/&item($x)<42>/
22:02 camelia rakudo-moar d06d7c: OUTPUT«Type Int does not support associative indexing.␤  in block <unit> at <tmp> line 1␤␤Actually thrown at:␤  in block <unit> at <tmp> line 1␤␤»
22:03 brokenchicken Damn all the ways come with this chaining feature :/
22:03 gfldex m: my $x = 42; constant $foo = '<42>'; say Q:f/&item($x)$foo/
22:03 camelia rakudo-moar d06d7c: OUTPUT«42$foo␤»
22:04 lucs q:v⦃⦄:to/EoT/ ⋯ \begin{Foo}{⦃$bar⦄} ⋯   # If  :v⦃⦄  for example meant to evaluate what is inside those brackets.
22:04 brokenchicken I see. That's far from ideal tho
22:04 gfldex agreed, the nature of all workarounds
22:06 * brokenchicken gives up... for now
22:07 * brokenchicken laughs at this email subject line
22:07 brokenchicken "***SPAM***  Trump attacked by reporters and reveals he took this pill-to become President"
22:07 brokenchicken They make pills for everything these days!
22:09 travis-ci joined #perl6
22:09 travis-ci Doc build errored. Wenzel P. P. Peppmeyer 'v6 -> v6.c'
22:09 travis-ci https://travis-ci.org/perl6/doc/builds/195653620 https://github.com/perl6/doc/compare/8167550f359b...3706a3d8fc71
22:09 travis-ci left #perl6
22:10 agentzh joined #perl6
22:10 brokenchicken "The job exceeded the maximum time limit for jobs, and has been terminated."
22:10 brokenchicken Prolly can be ameliorated by not building rakudo from scratch
22:10 lucs Allowing heredocs to specify their interpolation-introducing characters would be really nice.
22:11 lucs q:k«»:to/EoT ⋯ <div class='foo'>The hash value is $bar«key», right.</div> ⋯ (or something)
22:11 pyrimidine joined #perl6
22:14 Cabanossi joined #perl6
22:15 dj_goku joined #perl6
22:15 dj_goku joined #perl6
22:15 gfldex m: my $c = 42; say Q:c⸨$a<p>{$c}</p>⸩
22:15 camelia rakudo-moar d06d7c: OUTPUT«$a<p>42</p>␤»
22:15 gfldex m: my $c = 42; say Q:c{$a<p>{$c}</p>}
22:15 camelia rakudo-moar d06d7c: OUTPUT«$a<p>{$c}</p>␤»
22:15 brokenchicken w00t
22:15 gfldex strangle the 2nd version doesnt work
22:15 gfldex grammars are hard
22:16 brokenchicken The second version is buggy... I keep meening to look at that ticket
22:16 brokenchicken m: my $c = 42; say Q:c⸨$a<p>{$c}<p>⸩
22:16 camelia rakudo-moar d06d7c: OUTPUT«$a<p>42<p>␤»
22:16 brokenchicken Oh
22:16 brokenchicken Right.
22:16 brokenchicken I thought this was The Solution. But I see it won't work with CSS code :)
22:17 brokenchicken BUT, you can stick a heredoc inside the heredoc, for just the CSS code ('cause it'd be just a separate block) and use different rules on it
22:17 travis-ci joined #perl6
22:17 travis-ci Doc build errored. Wenzel P. P. Peppmeyer 'v6 -> v6.c'
22:17 travis-ci https://travis-ci.org/perl6/doc/builds/195655510 https://github.com/perl6/doc/compare/3706a3d8fc71...a7a81c3d2af4
22:17 travis-ci left #perl6
22:17 AlexDaniel hmmmm
22:18 astj joined #perl6
22:18 AlexDaniel samcv: I figured that greping over the whole range is faster if you 「use nqp」… for @words -> str $word { $sieve .= grep({nqp::isge_i(nqp::index(nqp::getuniname($_), $word, 0), 0)}) };
22:19 AlexDaniel aaabout 1.7s
22:19 AlexDaniel versus 2.2s with no nqp
22:20 labster joined #perl6
22:23 kaare__ joined #perl6
22:25 pmurias brokenchicken: be careful with the pill, it might turn you orange
22:29 rmusial joined #perl6
22:30 brokenchicken hehehe
22:34 kaare__ joined #perl6
22:42 rmusial joined #perl6
22:42 pecastro joined #perl6
22:44 kurahaupo__ joined #perl6
22:46 TEttinger joined #perl6
22:47 samcv AlexDaniel, which whole range?
22:47 samcv oh
22:47 samcv I see
22:54 perlawhirl joined #perl6
22:55 pecastro joined #perl6
23:06 pyrimidine joined #perl6
23:08 cibs joined #perl6
23:08 wamba joined #perl6
23:12 astj joined #perl6
23:16 bwisti joined #perl6
23:20 cdg joined #perl6
23:22 z448 joined #perl6
23:24 pecastro joined #perl6
23:24 hartenfels joined #perl6
23:26 cpage_ joined #perl6
23:28 lep-delete joined #perl6
23:32 cpage_ joined #perl6
23:34 travis-ci joined #perl6
23:34 travis-ci Doc build errored. Will "Coke" Coleda 'Remove 'blead' and try to tag a release instead'
23:34 travis-ci https://travis-ci.org/perl6/doc/builds/195667233 https://github.com/perl6/doc/compare/a7a81c3d2af4...223e38759c5a
23:34 travis-ci left #perl6
23:39 pecastro joined #perl6
23:45 ggoebel joined #perl6
23:49 pyrimidine joined #perl6
23:50 pyrimidi_ joined #perl6
23:51 astj joined #perl6

| Channels | #perl6 index | Today | | Search | Google Search | Plain-Text | summary

Perl 6 | Reference Documentation | Rakudo