Perl 6 - the future is here, just unevenly distributed

IRC log for #perl6, 2017-01-24

Perl 6 | Reference Documentation | Rakudo

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

All times shown according to UTC.

Time Nick Message
00:05 Geth doc: 4a8f8f5af5 | (Samantha McVey)++ | 2 files
00:05 Geth doc: Fix --no-proc-async and allow sassc as one of our sass compilers
00:05 Geth doc: review: https://github.com/perl6/doc/commit/4a8f8f5af5
00:05 samcv [Coke], --no-proc-async should be fixed now
00:06 samcv oh nice, and highlighting on the site is back up \o/
00:16 mr_ron joined #perl6
00:16 BenGoldberg joined #perl6
00:18 mr_ron .tell lizmat for the p6weekly the ecosystem addition is US-ASCII not US::ASCII - thanks
00:18 yoleaux mr_ron: I'll pass your message to lizmat.
00:28 dataf3l joined #perl6
00:32 raschipi joined #perl6
00:33 [Coke] samcv: testing now
00:41 raschipi .tell lizmat The weekly has a link to "reddit comments" that doesn't point to reddit.
00:41 yoleaux raschipi: I'll pass your message to lizmat.
00:41 Geth doc: cfcfe5dce4 | raiph++ | doc/Programs/00-running.pod6
00:41 Geth doc: Link to page w/ examples + Windows quoting rules
00:41 Geth doc:
00:41 Geth doc: cale2++ for reporting problem and mentioning linked doc
00:41 Geth doc: review: https://github.com/perl6/doc/commit/cfcfe5dce4
00:53 pmurias joined #perl6
00:56 lep-delete joined #perl6
00:58 sammers hi #perl6
01:05 [Coke] Ho.
01:16 pyrimidine joined #perl6
01:26 geraud joined #perl6
01:39 skids joined #perl6
01:46 Geth doc: antquinonez++ created pull request #1161: Style updates to faq and about pod.
01:46 Geth doc: review: https://github.com/perl6/doc/pull/1161
01:52 travis-ci joined #perl6
01:52 travis-ci Doc build passed. Samantha McVey 'Fix --no-proc-async and allow sassc as one of our sass compilers'
01:52 travis-ci https://travis-ci.org/perl6/doc/builds/194661035 https://github.com/perl6/doc/compare/e3ca965b8fcb...4a8f8f5af5f2
01:52 travis-ci left #perl6
01:55 AlexDaniel oh wow, these doc pull requests :o
02:05 agentzh joined #perl6
02:09 labster joined #perl6
02:11 samcv so it looks like the changes to highlights were pretty small https://github.com/atom/highlights/commit/77e03c42ccc34cfda402b4b3a61e266b7d638515#diff-79e70f7be06fce1120c13431b70ddfd4R302
02:11 samcv like a few lines
02:12 diakopter joined #perl6
02:17 kalkin- joined #perl6
02:24 pyrimidine joined #perl6
02:40 MasterDuke joined #perl6
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!
03:00 matiaslina joined #perl6
03:04 xtreak joined #perl6
03:05 * [Coke] tries to figure out why we're getting html/.html files, and man is this a twisty maze. :(
03:05 b7j0c joined #perl6
03:06 b7j0c howdy! is there a way to have one logical package span many files?
03:08 b2gills m: https://gist.github.com/b2gills/b877658f38dae51e3c88fbdc0990f581
03:08 camelia rakudo-moar 700a07: OUTPUT«ok 1 - ␤ok 2 - ␤ok 3 - ␤ok 4 - ␤ok 5 - ␤ok 6 - ␤ok 7 - ␤ok 8 - ␤ok 9 - ␤ok 10 - ␤»
03:09 b2gills b7j0c: The best way would be to have a role per file and compose them into the resulting class
03:09 b7j0c thanks b2gills, I'll look into that
03:11 b7j0c yeah a quick glance at an old Advent post on Roles seems to indicate that is what I need
03:26 pyrimidine joined #perl6
03:27 agentzh joined #perl6
03:36 geekosaur joined #perl6
03:48 noganex joined #perl6
03:49 BenGoldberg .u ᛊᚾᛁᚲᛖᚱᛊᚾᛇᚲ
03:49 yoleaux U+16B1 RUNIC LETTER RAIDO RAD REID R [Lo] (ᚱ)
03:49 yoleaux U+16B2 RUNIC LETTER KAUNA [Lo] (ᚲ)
03:49 yoleaux U+16BE RUNIC LETTER NAUDIZ NYD NAUD N [Lo] (ᚾ)
03:50 BenGoldberg u: ᛊᚾᛁᚲᛖᚱᛊᚾᛇᚲ
03:50 unicodable6 BenGoldberg, U+16CA RUNIC LETTER SOWILO S [Lo] (ᛊ)
03:50 unicodable6 BenGoldberg, U+16BE RUNIC LETTER NAUDIZ NYD NAUD N [Lo] (ᚾ)
03:50 unicodable6 BenGoldberg, https://gist.github.com/4117b1032a9a03f0660399cad252256c
04:13 beatdown joined #perl6
04:32 xtreak joined #perl6
04:49 Cabanoss- joined #perl6
04:57 MasterDuke joined #perl6
05:23 bkd joined #perl6
05:50 troys joined #perl6
06:07 xtreak joined #perl6
06:11 Unavowed_ joined #perl6
06:18 wamba joined #perl6
06:25 dataf3l joined #perl6
06:26 matt_ joined #perl6
06:31 pyrimidi_ joined #perl6
06:31 CIAvash joined #perl6
06:48 cschwenz joined #perl6
06:48 bjz joined #perl6
06:50 RabidGravy joined #perl6
06:51 mr_ron joined #perl6
07:08 xtreak joined #perl6
07:13 domidumont joined #perl6
07:20 domidumont joined #perl6
07:20 bjz joined #perl6
07:49 cpage_ joined #perl6
07:59 n0xff__ left #perl6
08:05 trnh joined #perl6
08:23 zakharyas joined #perl6
08:31 AlexDaniel joined #perl6
08:35 lizmat .
08:35 yoleaux 00:18Z <mr_ron> lizmat: for the p6weekly the ecosystem addition is US-ASCII not US::ASCII - thanks
08:35 yoleaux 00:41Z <raschipi> lizmat: The weekly has a link to "reddit comments" that doesn't point to reddit.
08:35 lizmat fixed!
08:35 * lizmat clickbaits https://p6weekly.wordpress.com/2017/01/23/2017-04-welcome/
08:35 moritz lizmat++
08:37 dakkar joined #perl6
08:41 CIAvash lizmat++
09:03 jonas1 joined #perl6
09:06 ChoHag joined #perl6
09:14 is_ joined #perl6
09:16 salva joined #perl6
09:18 mr-foobar joined #perl6
09:20 zacts joined #perl6
09:27 andrzejku joined #perl6
09:33 espadrine_ joined #perl6
09:38 ChoHag joined #perl6
09:45 jonas1 joined #perl6
09:48 wamba joined #perl6
09:57 mr_ron joined #perl6
09:59 obfusk_ joined #perl6
10:02 jonas1 joined #perl6
10:04 domidumont joined #perl6
10:13 Geth ecosystem: bd2c869a8a | (Ahmad M. Zawawi)++ | META.list
10:13 Geth ecosystem: Remove totem ( Fun search bar written in Perl 6 )
10:13 Geth ecosystem:
10:13 Geth ecosystem: The project is 2-years old and I am focusing my efforts where it matters.
10:13 Geth ecosystem: review: https://github.com/perl6/ecosystem/commit/bd2c869a8a
10:18 wamba joined #perl6
10:30 trnh joined #perl6
10:30 agentzh joined #perl6
10:33 Vynce joined #perl6
10:50 xtreak joined #perl6
10:53 xtreak joined #perl6
10:56 xtreak joined #perl6
11:00 raschipi joined #perl6
11:00 pierrot joined #perl6
11:48 sufrostico joined #perl6
11:52 rindolf joined #perl6
11:53 mr_ron joined #perl6
11:54 lukaramu joined #perl6
11:59 jaguar007 joined #perl6
12:04 andrzejku joined #perl6
12:17 raschipi joined #perl6
12:29 gregf_ joined #perl6
12:32 parv joined #perl6
12:52 ChoHag joined #perl6
13:00 jaguar007 joined #perl6
13:02 pyrimidine joined #perl6
13:02 wamba joined #perl6
13:08 pyrimidine joined #perl6
13:09 sena_kun joined #perl6
13:39 pyrimidine joined #perl6
13:48 kurahaupo__ joined #perl6
13:50 dataf3l joined #perl6
14:06 pyrimidine joined #perl6
14:22 ggoebel joined #perl6
14:23 matiaslina joined #perl6
14:25 brokenchicken We changed the .O stuff semi-recently right? This is no longer the valid way to set up precedence: Eji::Grammar.O(':prec<u=>, :assoc<left>', '%multiplicative'); ?
14:25 brokenchicken What's the new way?
14:25 brokenchicken Right now that dies with 'Too many positionals passed; expected 1 argument but got 3'
14:26 * brokenchicken activates the arnsholt -signal
14:28 timotimo did we actually get that merged?
14:29 brokenchicken yeah
14:29 arnsholt Starting to be a while now, I think
14:29 brokenchicken Looks like this is the new way to do it: https://github.com/edumentab/rakudo-and-nqp-internals-course/blob/master/examples/rubyish-3.nqp#L71-L80
14:30 arnsholt Yup
14:30 brokenchicken cool
14:31 * brokenchicken is writing an emoji-based language :)
14:31 brokenchicken ???? 'Hello, World!'
14:31 arnsholt Was merged in early September last year, FWIW =)
14:31 brokenchicken That prints hello world.
14:32 agentzh joined #perl6
14:41 cibs joined #perl6
14:45 Actualeyes joined #perl6
14:53 brokenchicken $ cat test.eji
14:53 brokenchicken ???? 2.4 ✖ 2.5
14:53 brokenchicken $ ./eji test.eji
14:53 brokenchicken 6
14:53 brokenchicken :)
14:54 rindolf joined #perl6
14:57 pyrimidine joined #perl6
14:57 brokenchicken m: "????".uninames.say
14:57 camelia rakudo-moar e99510: OUTPUT«(SQUARED COOL)␤»
14:57 brokenchicken That's pretty... cool
14:58 mr_ron joined #perl6
15:00 brokenchicken Gawd dammit! Someone stole my idea: http://www.emojicode.org/
15:03 brokenchicken Hm, looks a lot less cool actually.
15:03 perigrin you just don't understand it ... nobody understands it ... it'll just be in it's room GAH!
15:04 * perigrin has a teenager who puts the emo into emoji
15:04 * brokenchicken is lucky to still be offspring-free
15:04 itcharlie joined #perl6
15:04 committable6 joined #perl6
15:04 evalable6 joined #perl6
15:04 benchable6 joined #perl6
15:04 bisectable6 joined #perl6
15:04 unicodable6 joined #perl6
15:04 statisfiable6 joined #perl6
15:12 brokenchicken Well... lacking any original ideas, I'm gonna write Perl 7 :)
15:14 pyrimidine joined #perl6
15:17 b2gills brokenchicken: Just pick a random word from the dictionary and try to implement that as a language
15:18 mst just make the random number generator run again if it picks 'felch'
15:19 huf yeah, dont call it that
15:19 huf however, moist would be a great name
15:20 skids joined #perl6
15:20 brokenchicken ... moist?
15:21 cdg joined #perl6
15:21 huf yeah, lots of people seem to hate the word
15:21 huf for no real reason
15:21 huf well, i believe their main reason is "just listen to it: moist *ewwwww*"
15:24 mr_ron m: https://gist.github.com/ronaldxs/901b9b802dcd85c8b588b202c36bb5ac
15:24 camelia rakudo-moar e99510: OUTPUT«True␤True␤err: No such method 'g::t' for invocant of type 'Cursor'␤err: No such method 'g::t' for invocant of type 'Cursor'␤»
15:25 mr_ron anyone know how to do such indirect grammar reference?
15:25 girafe joined #perl6
15:26 b2gills just imagine all of the keywords in a language called moist
15:27 mr_ron m: grammar g { token t { <[2]> } }; $rname = "g::t"; say so "2" ~~ /<::($rname)>/ # simplified version of question
15:27 camelia rakudo-moar e99510: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Variable '$rname' is not declared. Did you mean '&rename'?␤at <tmp>:1␤------> 3grammar g { token t { <[2]> } }; 7⏏5$rname = "g::t"; say so "2" ~~ /<::($rna␤»
15:27 mr_ron m: grammar g { token t { <[2]> } }; my $rname = "g::t"; say so "2" ~~ /<::($rname)>/ # simplified version of question
15:27 camelia rakudo-moar e99510: OUTPUT«No such method 'g::t' for invocant of type 'Cursor'␤  in block <unit> at <tmp> line 1␤␤»
15:28 mr_ron Is that example supposed to work?
15:29 [Coke] if it's not tested in roast, no guarantees? I see you quoted the syn in the gist.
15:29 [Coke] code from there is often just speculative.
15:31 mr_ron m: my $rname = "digit"; say so "2" ~~ /<::($rname)>/; # that works
15:31 camelia rakudo-moar 4ef37c: OUTPUT«True␤»
15:32 dalek joined #perl6
15:33 cdg joined #perl6
15:34 mr_ron https://github.com/perl6/roast/blob/master/S05-metasyntax/angle-brackets.t#L171
15:34 synopsebot6 joined #perl6
15:37 brokenchicken That works because digit is a method on Cursor, while "g::t" ain't
15:38 brokenchicken m: grammar g { token t { <[2]> } }; dd ::("g::t")
15:38 camelia rakudo-moar 4ef37c: OUTPUT«Failure.new(exception => X::NoSuchSymbol.new(symbol => "g::t"), backtrace => Backtrace.new)␤»
15:38 brokenchicken m: grammar g { our token t { <[2]> } }; dd ::("g::t")
15:38 camelia rakudo-moar 4ef37c: OUTPUT«Failure.new(exception => X::NoSuchSymbol.new(symbol => "g::t"), backtrace => Backtrace.new)␤»
15:38 brokenchicken doesn't seem to be even visible outside of it
15:38 AlexDaniel .tell dalek Please switch this bot to Geth
15:38 yoleaux AlexDaniel: I'll pass your message to dalek.
15:38 brokenchicken m: grammar g { our sub t { } }; dd ::("g::t")
15:38 camelia rakudo-moar 4ef37c: OUTPUT«Failure.new(exception => X::NoSuchSymbol.new(symbol => "g::t"), backtrace => Backtrace.new)␤»
15:38 brokenchicken m: grammar g { our sub t { } }; dd ::("&g::t")
15:38 camelia rakudo-moar 4ef37c: OUTPUT«Sub t = sub t () { #`(Sub|54816568) ... }␤»
15:38 brokenchicken m: grammar g { our token t { ^} }; dd ::("&g::t")
15:39 camelia rakudo-moar 4ef37c: OUTPUT«Regex t = token t { ^}␤»
15:39 mr_ron m: grammar g { token t { <[2]> } }; say so "2" ~~ /<g::t>/ # simplified version of question
15:39 camelia rakudo-moar 4ef37c: OUTPUT«True␤»
15:39 brokenchicken m: grammar g { our token t { <[2]> } }; my $rname = "g::t"; say so "2" ~~ /<::("&$rname")>/
15:39 camelia rakudo-moar 4ef37c: OUTPUT«No such method '&g::t' for invocant of type 'Cursor'␤  in block <unit> at <tmp> line 1␤␤»
15:40 b2gills m: grammar g { our token t { <[2]> } }; my $rname = "&g::t"; say so "2" ~~ /<{::($rname)}>/
15:40 camelia rakudo-moar 4ef37c: OUTPUT«True␤»
15:40 * mr_ron noticed people found time for Perl 7 ...
15:40 mr_ron b2gills: thanks much
15:41 * [Coke] doesn't think that's really a thing.
15:41 brokenchicken It is! I'm working on it.
15:43 mr_ron I think the technique is useful.  If no objection will look to document and add test(s?).
15:44 brokenchicken I'm surprised that stuff even works
15:44 brokenchicken m: grammar g { token t { <[2]> } }; dd g::t
15:44 camelia rakudo-moar 4ef37c: OUTPUT«Could not find symbol '&t'␤  in block <unit> at <tmp> line 1␤␤Actually thrown at:␤  in block <unit> at <tmp> line 1␤␤»
15:44 brokenchicken ... yet you can access it from inside a grammar
15:44 mr_ron m: grammar g { token t { <[2]> } }; say so "2" ~~ /<g::t>/
15:44 camelia rakudo-moar 4ef37c: OUTPUT«True␤»
15:44 moritz tokens are has-scoped
15:44 brokenchicken moritz: what does that mean?
15:44 moritz so <a::b> inside a regex is special[tm]
15:45 moritz brokenchicken: it means they have the same lookup semantics as methods
15:45 moritz they live in the method table, not the namespace
15:45 brokenchicken moritz: OK, but why can some random regex look up a symbol in some other random grammar?
15:46 moritz brokenchicken: because it's public
15:46 brokenchicken OK
15:56 mr_ron m:grammar g { our token t { <[2]> } }; say so "2" ~~/<&g::t>/; say so "2" ~~ /<g::t>/ #second one not work - problem
15:56 mr_ron m: grammar g { our token t { <[2]> } }; say so "2" ~~/<&g::t>/; say so "2" ~~ /<g::t>/ #second one not work - problem
15:56 camelia rakudo-moar 4ef37c: OUTPUT«True␤No such method 't' for invocant of type 'g'␤  in block <unit> at <tmp> line 1␤␤»
15:57 brokenchicken https://github.com/perl7/perl7 :)
15:57 kyan joined #perl6
15:58 brokenchicken With the naming problem finally solved, all I need is a kickass logo, and I'm done!
16:01 b2gills m: grammar g { token t { <[2]> } }; my $r-name = 'g::t'; my ($r-class,$r-method) = ($r-name ~~ /(.*)'::'(.*)/)[]».Str; say grammar :: is ::($r-class) {token TOP {<::($r-name)>}}.parse('2') # this >< close
16:01 camelia rakudo-moar 4ef37c: OUTPUT«===SORRY!===␤This type cannot unbox to a native string: P6opaque, QAST::Var␤»
16:02 b2gills brokenchicken: seems a bit ... anemic
16:02 jast minimalism!
16:04 brokenchicken b2gills: Perl 6 is too damn complex. So the obvious solution for Perl 7 is to make it as simple as possible.
16:04 brokenchicken it can add, subtract, multiple, divide, and print strings. What else could you possibly want?
16:05 perigrin brokenchicken: http://t06.deviantart.net/dwjaHECiFqRTw0e3vI6PNu-DT-8=/fit-in/700x350/filters:fixed_height(100,100):origin()/pre07/1d86/th/pre/f/2012/232/e/1/teffla_motivational_poster_by_tacogirl456-d5bur4p.jpg
16:05 perigrin logo for you.
16:05 brokenchicken eeewwww... TIP: do not leave your sandwiches over the weekend in company's fridge... There's a chance they'll smell like strawberries when you come back for them :(
16:06 jeek but I like strawberries
16:06 perigrin you know subtraction is just addition with negative numbers, and division is just multiplication with fractions ... so you could simplify the language by 40% there.
16:07 brokenchicken perigrin: pffft.. That picture fails to learn from the suckage of Perl 6's logo. The most awesomest logo is "P7" in a bold solid font, inside a rectangle. That's the type of logos that managers like and clearly the language will be in top 5 langs of 2017 based on just logo alone.
16:08 ggoebel joined #perl6
16:08 brokenchicken Hm. Or maybe it's not strawberries and my sandwich just got bad?
16:09 brokenchicken Is 4 days too long a time to leave a sandwich in the fridge for :/
16:09 ugexe throw some new bread on it
16:09 brokenchicken Aint got any
16:12 luiz_lha joined #perl6
16:13 mr_ron m: grammar g { token t { <[2]> } }; my $r-name = "t"; grammar g2 is g { token TOP { <t> } }; say g2.parse("2")
16:13 camelia rakudo-moar 4ef37c: OUTPUT«「2」␤ t => 「2」␤»
16:13 mr_ron m: grammar g { our token t { <[2]> } }; my $r-name = "t"; grammar g2 is g { token TOP { <t> } }; say g2.parse("2") # add our
16:13 camelia rakudo-moar 4ef37c: OUTPUT«No such method 't' for invocant of type 'g2'␤  in regex TOP at <tmp> line 1␤  in block <unit> at <tmp> line 1␤␤»
16:17 brokenchicken mr_ron: what does `our` do for methods/tokens?
16:17 brokenchicken m: class Foo { our method bar { say "hi" } }; class Bar is Foo { method stuff { self.bar } }.stuff
16:17 camelia rakudo-moar 4ef37c: OUTPUT«No such method 'bar' for invocant of type 'Bar'␤  in method stuff at <tmp> line 1␤  in block <unit> at <tmp> line 1␤␤»
16:18 brokenchicken Doesn't work with normal methods either
16:18 mr_ron I was looking at the suggestions from b2gills ...
16:19 jaguar007 joined #perl6
16:30 wamba joined #perl6
16:31 mr_ron m: grammar g { token t { <[2]> } }; $rname = "g::t"; say so "2" ~~ /<::($rname)>/ # our scoping not answer, other ideas? rfc? rt?
16:31 camelia rakudo-moar 4ef37c: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Variable '$rname' is not declared. Did you mean '&rename'?␤at <tmp>:1␤------> 3grammar g { token t { <[2]> } }; 7⏏5$rname = "g::t"; say so "2" ~~ /<::($rna␤»
16:31 mr_ron m: grammar g { token t { <[2]> } }; my $rname = "g::t"; say so "2" ~~ /<::($rname)>/ # our scoping not answer, other ideas? rfc? rt?
16:31 camelia rakudo-moar 4ef37c: OUTPUT«No such method 'g::t' for invocant of type 'Cursor'␤  in block <unit> at <tmp> line 1␤␤»
16:32 moritz it doesn't like fully-qualified names with the <::()> syntax
16:35 mr_ron I guess that's a "no".  As a last mention the use case I had was here: https://github.com/ronaldxs/Perl6-US-ASCII/blob/master/t/01-basic.t
16:36 moritz m: grammar g { token t { <[2]> } }; my $rx = g.^find_method('t'); say so "2" ~~ /$rx/
16:36 mr_ron It would be nice to repeat for each rule - is $latin-chars.comb(/<US-ASCII::alpha>/).join, %charset-str< alpha > # again for upper, lower, ...
16:36 camelia rakudo-moar 4ef37c: OUTPUT«Type check failed in binding to <anon>; expected g but got Cursor (Cursor.new)␤  in regex t at <tmp> line 1␤  in block <unit> at <tmp> line 1␤␤»
16:37 mr_ron s/to repeat/not to repeat/
16:37 moritz m: grammar g { token t { <[2]> } }; my $rx = g.^find_method('t'); say so "2" ~~ /<$rx>/
16:37 camelia rakudo-moar 4ef37c: OUTPUT«Type check failed in binding to <anon>; expected g but got Cursor (Cursor.new)␤  in regex t at <tmp> line 1␤  in block <unit> at <tmp> line 1␤␤»
16:44 brokenchicken m: grammar g { token t(Any:) { <[2]> } }; my $rx = g.^find_method('t'); say so "2" ~~ /<$rx>/
16:44 camelia rakudo-moar 4ef37c: OUTPUT«True␤»
16:47 mr_ron m: grammar g { token t(Any:) { <[2]> } }; my $rx = g.^find_method("t"); say so "2" ~~ /<g::t>/ # looks better
16:47 camelia rakudo-moar 4ef37c: OUTPUT«True␤»
16:51 lichtkind joined #perl6
16:54 mr_ron m: grammar g { token t(Cursor:) { <[2]> } }; my $rx = g.^find_method("t"); say so "2" ~~ /<g::t>/; say so "2" ~~ /<$rx>/
16:54 camelia rakudo-moar 4ef37c: OUTPUT«True␤True␤»
17:02 mr_ron So at this point it comes down to which is uglier ... adding (Cursor:) signatures to all the US-ASCII rules or leaving in the repetition for tests.
17:03 agentzh joined #perl6
17:08 moritz or using EVAL in the tests
17:08 moritz m: grammar g { token t(Any:) { <[2]> } }; my $rx = g.^find_method("t"); say so "2" ~~ $rx
17:08 camelia rakudo-moar 4ef37c: OUTPUT«True␤»
17:08 moritz ... or using that :-)
17:08 moritz mr_ron: ^^
17:13 brokenchicken And now I have a kickass Perl 7 logo: https://github.com/perl7
17:14 brokenchicken I guess I better write a book too. I heard that was a requirement for success.
17:15 moritz it just doesn't pay well
17:15 moritz instead you should publish courses on Pluralsight or whatever it's called
17:16 timotimo i just set up syncthings to synchronize my ecosystem (i.e. checkouts of a bunch of modules), moarvm, nqp, and rakudo folders across my laptop and desktop
17:19 brokenchicken moritz: so what was with people ++ you the other day with comment "book"? You finished your book?
17:20 timotimo not finished
17:20 moritz brokenchicken: I wish :-)
17:20 brokenchicken So what were ++ about?
17:20 moritz brokenchicken: propbably just because of an update, or general appreciation of the effort
17:20 st_elmo joined #perl6
17:20 brokenchicken Ah
17:21 brokenchicken Well, when you're done, you can write one for Perl 7 :)
17:21 brokenchicken It's the language of the Future!
17:21 moritz speaking of which, selling copies before the book is finished is really motivating
17:22 moritz 28 copies sold so far
17:22 brokenchicken :o
17:23 moritz and now a publisher has asked whether I'm interested in them adopting the book
17:23 domidumont joined #perl6
17:23 perlpilot joined #perl6
17:23 * brokenchicken doesn't see a pre-order link on https://perl6book.com/
17:23 moritz I guess I should update the website, eh? :-)
17:23 brokenchicken :)
17:23 moritz https://leanpub.com/perl6
17:25 moritz m: grammar g { token t { <[2]> } }; my $rx = g.^find_method("t"); say so "2" ~~ $rx
17:25 camelia rakudo-moar 4ef37c: OUTPUT«Type check failed in binding to <anon>; expected g but got Cursor (Cursor.new)␤  in regex t at <tmp> line 1␤  in block <unit> at <tmp> line 1␤␤»
17:25 espadrine_ joined #perl6
17:28 * brokenchicken buys 1 copy of moritz's motivation
17:28 sftp joined #perl6
17:30 agentzh joined #perl6
17:33 moritz brokenchicken++ thanks!
17:34 timotimo cool
17:36 kyan joined #perl6
17:49 itcharlie left #perl6
17:53 jdoege joined #perl6
17:54 moritz https://perl6book.com/ updated
17:56 Actualeyes joined #perl6
18:07 mspo can you add left: 0 to the content-wrapper if you're going to use it in splash-container?
18:07 mspo FF has horiz scroll
18:10 brokenchicken heh
18:10 agentzh joined #perl6
18:14 lukiramu joined #perl6
18:20 mspo this is actually a common problem on firefox
18:20 domidumont joined #perl6
18:22 brokenchicken That's just margins on body. Any self-respecting web dev sets those to zero :)
18:23 agentzh joined #perl6
18:27 mr-foobar joined #perl6
18:30 brokenchicken TIL perl6 takes `--profile-compile` option :o
18:31 timotimo yup
18:31 timotimo that's a thing we've been hoping to be able to use with the core setting
18:32 timotimo but it consistently generates profiles that are out of reach for the html profiler
18:32 timotimo we can use the sql target, though
18:32 brokenchicken neat
18:33 timotimo i haven't played around with the sql target enough
18:33 timotimo but it's a great addition to the whole thing
18:38 domidumont joined #perl6
18:41 wamba joined #perl6
18:44 brokenchicken How come my `method sign` actions don't get triggered? Here are the tokens: https://github.com/perl7/perl7/blob/master/src/perl7.nqp#L17-L23 and here are the actions: https://github.com/perl7/perl7/blob/master/src/perl7.nqp#L57-L67
18:44 brokenchicken And here's the output when compiling ???? 2.4 × −2.5  https://gist.github.com/zoffixznet/9af07d053b25208d019005254d2034d9
18:44 brokenchicken The sign matches, but its actions don't get executed? dafuq?
18:47 jnthn brokenchicken: 'cus you forgot the :sym in your grammaar
18:48 timotimo might only be calling the proto tokens action method
18:48 jnthn *grammar
18:48 jnthn .oO( grammARGH! )
18:48 brokenchicken oh damn :) I knew it had to be something dumb :)
18:48 El_Che perl7? oh oh
18:48 brokenchicken jnthn++
18:48 timotimo ah, nideed
18:50 brokenchicken El_Che: yup. I solved all the version and logo debates with a swift stroke of brilliance! :)
18:50 FROGGS joined #perl6
18:53 mr_ron How do I run a single roast test file with Rakudo perl6.  I know there is fudging involved. Using a rakudo repo and make m-sometests is a hacky way.  Anything better?
18:53 El_Che brokenchicken: you must have been the 1st that thought of that :)
18:54 brokenchicken mr_ron: well, the fudger lives in rakudo's repo
18:54 brokenchicken mr_ron: I don't see what's hacky about it. Check out the roast to t/spec and you're good to roll
18:54 brokenchicken mr_ron: make t/spec/some-test-file or t/fudgeandrun t/spec/some-test-file
18:55 brokenchicken t/fudgeandrun /some/arbitrary/dir might work too
18:55 brokenchicken t/fudgeandrun /some/arbitrary/dir/test-file.t might work too
18:55 cdg_ joined #perl6
18:58 mr_ron It seems odd to me that running roast tests depends on rakudo.  There is a $*VM variable and it seems hacky that you can't actually run any tests without the cooperation of the implimentation.
18:59 brokenchicken You can.
18:59 brokenchicken The fudges just apply to rakudo. And just rakudo has a list of spectest files it runs.
19:00 brokenchicken Not sure what $*VM's got to do with anything.
19:01 brokenchicken You can run the tests using implementation XYZ to your hearts content, without needing an ounce of rakudo.
19:01 brokenchicken And there's even a fudger in roast repo
19:02 brokenchicken hah.. and a SUPER fudger too :)
19:02 mr_ron working on answering you - 2 minutes please - yes saw fudge program in roast repo
19:03 Actualeyes joined #perl6
19:05 hartenfels joined #perl6
19:07 brokenchicken and t/fudgeandrun is a tiny wrapper for roast's fudger. See what it calls and you don't need rakudo repo
19:08 brokenchicken but 99.9999999999999% of time roast's tests are run for purposes of verifying rakudo they're run from a checkout in t/spec/ of rakludo's repo :)
19:08 brokenchicken 2 minutes passed. You lied
19:08 TimToady he didn't say he'd answer after 2 minutes
19:08 mr_ron sorry - still working
19:09 brokenchicken Oh, right. Fair enough.
19:09 brokenchicken What happens after 2 minutes then?
19:09 TimToady everything, now
19:09 TimToady modulo simultanaeity considerations under relativity
19:11 TimToady but, basically, everything in your downstream lightcone
19:14 cdg joined #perl6
19:19 moritz \o/ relativity
19:21 mr_ron I was trying to run ./fudge and then "prove -eperl6" on the .rakudo and that still doesn't seem to work.  But fudge and perl6 on the .rakudo file seems to work.  Maybe need more familiarity with testing.
19:25 brokenchicken .oO( what .rakudo file? )
19:26 brokenchicken It's usually .rakudo.moar or .rakudo.jvm
19:26 brokenchicken mr_ron: and what does doesn't work mean?
19:28 b2gills whelp 103 days of uptime seems like enough, restarting to let updates finish. be back in a few
19:30 brokenchicken If you're asking for help, saying doesn't work is like coming to the doctor and saying feel ill.
19:30 mr_ron example S05-metasyntax/angle-brackets.t
19:30 mr_ron ./fudge rakudo S05-metasyntax/angle-brackets.t; prove -e perl6 S05-metasyntax/angle-brackets.rakudo
19:31 [Coke] we used to run roast on niecza and pugs and rakudo.parrot and rakudo.moar and rakudo.jvm, so it does work across the board; each of those happened to follow the "checkout roast into t/spec to run it" model, but no reason why you couldn't have a pre-installed compiler you were testing against.
19:32 [Coke] in rakudo, you can "make t/spec/sec/file.t", also, or "prove -v -e t/fudgeandrun t/spec/sec/file.t" or "prove -v -e t/fudgeandrun sec/file.t"
19:32 b2gills joined #perl6
19:32 [Coke] no reason why t/fudgeandrun couldn't be moved into roast and generalized.
19:33 * [Coke] hopes that the perl7 repo doesn't causes any negative publicity. :)
19:33 mr_ron brokenchicken is mostly right - it runs more often than I thought for rakudo.  Not so many files break on rakudo.  I may have just happened to hit them.
19:34 brokenchicken [Coke]: all publicity is good publicity ;)
19:34 [Coke] not every file in roast works on rakudo. that's what t/spectest.data is for
19:35 [Coke] brokenchicken: in that case, best wishes to you. :)
19:36 brokenchicken mr_ron: huh. Weird. Seems ./fudge is adding exit(1) at the end :S
19:36 brokenchicken Which of course makes harness think there've been failures
19:37 * brokenchicken looks closer at code
19:37 [Coke] brokenchicken: you can change the -exit code with fudge.
19:37 [Coke] --keep-exit-code
19:38 domidumont joined #perl6
19:38 brokenchicken aha
19:38 kent\n joined #perl6
19:39 brokenchicken mr_ron: so it's ./fudge --keep-exit-code rakudo S05-metasyntax/angle-brackets.t && prove6 -eperl6 S05-metasyntax/angle-brackets.rakudo
19:40 lichtkind_ joined #perl6
19:42 brokenchicken err s/prove6/prove/
19:45 brokenchicken .oO( Who's Avi Ranawat? )
19:46 [Coke] brokenchicken: t/fudgeandrun makes that simpler.
19:46 * brokenchicken is being paranoid again, I guess.
19:46 brokenchicken [Coke]: yeah, but mr_ron didn't want to have Rakudo repo checked out, from what I understood
19:47 [Coke] note that there's no guarantee that the file you get is a .rakudo - no fudges == .t
19:47 toolforger joined #perl6
19:47 [Coke] (right, maybe it makes sense to put this helper (which someone probably stole from pugs) into roast.
19:47 mr_ron I cloned roast and started adding some test files.  I was trying to figure out how to run the new tests.  I saw the fudge program in the roast readme.
19:48 toolforger Heh. I had the exactly same problem this morning
19:48 toolforger I have roast, I have moarvm, nqp, and rakudo installen - but how am I supposed to run the tests?
19:48 toolforger I feel so dumb...
19:50 brokenchicken Well, I someone wanted to make contributing.perl6.org
19:50 brokenchicken Or Contribute.Perl6.Org
19:50 brokenchicken https://github.com/perl6/contribute/blob/master/DESIGN.md
19:52 girafe joined #perl6
19:53 brokenchicken But I kinda lost steam on that one with the "when in doubt; merge it" comment
19:54 brokenchicken https://irclog.perlgeek.de/perl6-dev/2016-12-12#i_13725404
19:54 brokenchicken 'cause I strongly disagree with that and I don't want to open flodgates of crappy PRs that get merged just because they're there :/
19:59 mr_ron Coke: fudgeandrun depends on the rakudo directory structure but with just a few changes it works for roast.  Ok to make roast issue and try to add changed fudgeandrun?
20:00 * brokenchicken would rather not touch what ain't broken.
20:00 brokenchicken This is build stuff that's prolly used downstream.
20:02 moritz brokenchicken: I see no evidence for that in perl6-all-modules
20:03 moritz moritz@hack:~/p6/perl6-all-modules$ git grep fudgeandrun|wc -l
20:03 moritz 0
20:03 [Coke] "when in doubt merge it" is the wrong approach in most of our repos, yes.
20:03 [Coke] mr_ron: that seems reasonable to me,
20:04 brokenchicken moritz: that's just modules. I mean people who distribute rakudo. Including our own stuff: Rakudo Star and NeuralAnomaly release bot.
20:04 brokenchicken I mean that if we're *moving* fudgeandrun to roast repo
20:04 moritz brokenchicken: doesn't seem to be used in star either
20:05 toolforger brokenchicken, did you read the follow-up messages? seems like the "when in doubt, merge it" philosophy isn't a thing
20:05 mr_ron thanks ...
20:05 brokenchicken toolforger: I don't see that.
20:06 toolforger well, the idea gets some flak, then the discussion moves on
20:06 toolforger most responses are sarcasm and/or irony I think
20:06 brokenchicken I see two people joking around about the idea then *me* saying it's a bad idea and the discussion moves on.
20:06 brokenchicken And the discussion started when a broken PR got merged.
20:07 toolforger erm... which of the participants was you?
20:07 brokenchicken And it happened more than once that a rakudo/perl6 repo had comments on it, proposing improvements, but it silently got merged by another contributor.
20:08 brokenchicken by silently I mean ignoring or missing those comments.
20:08 toolforger somehow Perl6 isn't utterly broken though
20:09 toolforger not that I'm particularly fond of such a philosophy myself, but then that's just me
20:10 brokenchicken Utterly broken isn't the outcome of trigger-happy merging. Warts and bugs and half-assed solutions are.
20:12 toolforger Until somebody comes along and fixes that. Linux works that way, too, and it's not really worse than most other OSes when it comes to having half-assed solutions. I guess it's a way to speed up development, in the knowledge that if it's really bad than that will scratch somebody's itch and eventually get solved.
20:12 toolforger Of course, over time, such an approach can kill a code base
20:12 toolforger OTOH both Linux and Perl6 are famous for scrapping entire codebases, but salvaging the experience gained and building somthing better on top of that experience
20:13 brokenchicken ouch
20:13 [Coke] toolforger: ... we may have done that historically, it's not something we're planning on repeating. :)
20:14 toolforger you *never* plan on scrapping a codebase :-)
20:14 jayk joined #perl6
20:15 jayk are there any perl coders 4 hire?
20:16 moritz jayk: my experience in Germany shows that there are good ones for hire, but not cheap
20:16 jeek jayk: Try jobs.perl.org
20:16 jayk i have a small problem probably take 15 minutes
20:16 brokenchicken toolforger: I'm not convinced at all by "speed up development" argument in this case. You get first half-assed solution faster, but you lose all the saved time on debugging and rewriting good solution. It took me over an hour to debug this hack before I realized it lets execute arbitrary code within pod and works only half the time for the purpose it was meant to do:
20:17 brokenchicken https://github.com/rakudo/rakudo/blob/nom/src/Perl6/Pod.nqp#L131-L145
20:17 brokenchicken toolforger: and on top of that lost debugging time, the feature is back on the drawing board, where we try to remember if it ever meant to let run arbitrary code in the first place: https://rt.perl.org/Ticket/Display.html?id=130477#ticket-history
20:17 brokenchicken And once that design debate is over, the feature will have to be reimplemented.
20:18 toolforger Sure.
20:18 * toolforger has reimplemented more things in his life than he cares to remember
20:18 brokenchicken toolforger: and since it half-works now, there's no motivation to actually fix it. And so, the half-assed wart lives on :)
20:18 * toolforger also remembers a lot of cases where he couldn't have done the good design before trying the straighforward (but bad one) first
20:18 Cabanoss- joined #perl6
20:19 toolforger "Experience is the sum of mistakes made." Not entirely true, but more true than many people would like.
20:19 brokenchicken jayk: so why not learn to do it yourself?
20:19 flexibea` joined #perl6
20:20 jayk i've tried for a month
20:20 jayk i give up
20:20 brokenchicken hah
20:20 jayk the modules im using are broken
20:20 brokenchicken jayk: so how do you know it'd take 15 minutes? :)
20:20 bjz joined #perl6
20:20 jayk or they don't work right
20:20 brokenchicken jayk: what is it anyway?
20:21 jayk async web/proxy checker
20:21 jayk with mojo it is too slow
20:21 brokenchicken jayk: which mojo? mojolicious.org?
20:21 jayk anyevent::http or net::async::http might work, but im having issues
20:21 jayk yes mojolicous
20:22 brokenchicken jayk: ah, you're in the wrong channel :) You want #perl  This is a related but different language Perl 6.
20:22 moritz IME, nothing with async in p5 ever takes only 15 minutes to develop :-)
20:22 brokenchicken jayk: and I highly doubt we have any easy solutions that would be faster than mojo
20:22 jayk mojo didnt work well with async requests
20:23 brokenchicken weird, I had the impression it rocked with them
20:23 jayk i want to process 500 URLs asyncronously, wait 3 seconds, and process another 500
20:23 jayk it doesn work well with the URLs are broken or don't exist
20:23 brokenchicken What you need this stuff for anyway?
20:24 jayk its a proxy checker
20:24 jayk for a lot of hings
20:24 Tonik joined #perl6
20:24 brokenchicken Well, all good, but as I've said, Perl 6 is a different language.
20:24 brokenchicken Did you want to install it or.. ?
20:25 jayk its written to use a module and it'll work if its not done asyncronously
20:25 zakharyas joined #perl6
20:25 Actualeyes joined #perl6
20:25 jayk the biggest problem im running into now its a broken module
20:25 jayk the module itself needs to be fixed
20:26 jayk it'll run through just fine until it hits a bad URL, and the whole program quits
20:26 pyrimidine joined #perl6
20:26 brokenchicken heh
20:27 brokenchicken How much you're paying anyway?
20:27 mspo jayk: ask in #mojo
20:27 mspo or #perl
20:27 jayk id pay up to $100 to fix it
20:28 brokenchicken cool
20:28 jayk #perl sent me to jobs.perl..org
20:28 brokenchicken haha
20:28 jayk i think those guys are looking for bigger programs
20:28 mspo can you ask a direct question instead?
20:29 brokenchicken Wasn't there some freelance site just for this sort of stuff?
20:29 jayk my question was to just find someone to fix my stupid proxy checker, or rewrite it to use a module that works
20:29 jayk yeah freelancer
20:29 brokenchicken Amazon Turks or something or other
20:29 jayk i can do that it just seemed like id be easier to bitcoin someone $100 bucks
20:29 brokenchicken Which module is broken?
20:29 jayk net::async::http
20:30 jayk i already submitted a bug request
20:30 jayk there are several  problems with that module
20:30 mr_ron joined #perl6
20:30 jayk i had to use 2 patches to get as far as i did, and then ran into a 3rd problem that doesn't have a fix
20:32 toolforger monster.com used to be a freelancing portal
20:32 toolforger I think there are others nowadays
20:32 brokenchicken .oO( for @urls.rotor(500) { await .map: {say HTTP::UserAgent.new.get($_).status-line}; sleep 3 } ...)
20:32 jayk is that perl6?
20:33 brokenchicken jayk: well, as I've mentioned. Perl 6 is a different language. We don't fix Perl 5 modules. The majority here probably doesn't even know Perl 5, so... you may be more productive looking at other places.
20:33 jayk how long has perl6 been in development?
20:33 brokenchicken jayk: yeah, that's Perl 6 version of doing 500 urls each time, waiting for them to finish and sleeping 3 seconds.
20:34 mspo jayk: 10 years?
20:34 brokenchicken jayk: the first production release was December 2015. That was preceeded by 15 years of designs and redesigns
20:34 jayk wow
20:34 mspo 15 years
20:34 brokenchicken Yeah. If it's worth doing, better do it right ;)
20:35 jayk its a lot like ipv6 then, solving a need that needs to happen but difficult to fully implement?
20:35 mr-foobar joined #perl6
20:35 brokenchicken No idea. I was here only since production release :)
20:35 moritz yes
20:35 jayk heh
20:35 moritz also with a well-working status quo
20:35 moritz the analogy is apt
20:38 diakopter well, with a pretty flexible definition of "implement"
20:38 moritz right; good that we're all mentally flexible :-)
20:41 toolforger Asking again
20:41 toolforger I have rakudo installed; how do I run roast to test it?
20:41 brokenchicken Good thing Perl 7 was easy to implement :D
20:41 brokenchicken toolforger: installed how?
20:42 toolforger locally
20:42 toolforger in a subdirectory
20:42 brokenchicken toolforger: like built from a checkout?
20:42 toolforger I have a shell script that puts it in $PATH :-)
20:42 toolforger yes, fresh from git
20:42 toolforger s/git/github/
20:42 brokenchicken toolforger: run make stresstest in that dir
20:42 brokenchicken toolforger: run TEST_JOBS=8 make stresstest in that dir
20:42 toolforger :-)
20:43 toolforger I see
20:43 brokenchicken where TEST_JOBS = ($num-of-cores * 1.3).Int
20:43 brokenchicken toolforger: also, there's make spectest
20:43 toolforger Do I need a git clone of roast?
20:43 brokenchicken spectest is about 56K of top-level tests and stresstest is about 130K top-level tests
20:43 brokenchicken It'll do it for you
20:43 toolforger Um
20:44 brokenchicken But you can clone it to t/spec if you want (say if you want to work on your fork)
20:44 toolforger I am offline pretty often, so automatic background downloads aren't ideal for me
20:44 brokenchicken OK. Just check it out to there
20:44 brokenchicken It'll try to fetch updates to it tho
20:45 toolforger For now, an extra download is fine though... let me try
20:45 brokenchicken You can probably see what it runs and run your own command that excludes git pull
20:46 toolforger so I run the make command inside the rakudo source dir?
20:46 brokenchicken yeah
20:46 toolforger I want to hack on the jvm port, so that means I'll need to run spectest, and possibly stresstest
20:47 brokenchicken .tell mr_ron now that I remember more specific test files, a lot of them depend on the test file located in t/spec. This is used for auxiliary files locations for some test files, like module loading ones, and the ones that include t/spec/packages
20:47 yoleaux brokenchicken: I'll pass your message to mr_ron.
20:48 brokenchicken toolforger: hehe, well, be sure to pack a lunch :)
20:48 toolforger Oh. It's complaining that Inline::Perl6 is not installed.
20:48 brokenchicken huh
20:48 toolforger s/6/5/
20:48 brokenchicken toolforger: are you sure it's not just a warning about Inline::Perl*5*?
20:48 toolforger sorry
20:48 brokenchicken Yeah, it'll skip Perl 5 compatibility tests if Inline::Perl5 ain't installed
20:49 brokenchicken I just copy-paste step 12 of the release guide to install it: https://github.com/rakudo/rakudo/blob/nom/docs/release_guide.pod
20:49 brokenchicken But there are probably several saner ways to do it
20:49 toolforger When would I need to worry about Inline::Perl5?
20:50 * brokenchicken doesn't follow the question
20:50 toolforger it's skipping Perl5 integration tests
20:50 brokenchicken Right
20:50 toolforger I'm not sure whether I should worry about that
20:50 brokenchicken You can pop open t/spectest.data in rakudo's repo and find the ones marked as perl 5
20:50 brokenchicken And see what they are.
20:50 brokenchicken But they're about Perl 5 compatibility....
20:51 toolforger Aaah... right, there's that compatibility mode
20:51 toolforger I guess I want that
20:52 toolforger I want to tinker with JVM classfile generation, and Perl5 stuff wants an extra jar so I'll want to catch if anything goes missing
20:54 toolforger off into sunset - seeya!
20:55 toolforger left #perl6
21:06 integral joined #perl6
21:06 integral joined #perl6
21:18 TEttinger joined #perl6
21:19 Actualeyes joined #perl6
21:21 raiph m: role A { method m {} }; role B does A {}; class C does A does B {} # golf of bdfoy's recent SO Q
21:21 camelia rakudo-moar 5db0b2: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Method 'm' must be resolved by class C because it exists in multiple roles (B, A)␤at <tmp>:1␤»
21:22 raiph At least LTA; I don't understood this problem yet; I didn't find an RT for it; any comments before I investigate/RT it appreciated
21:28 pyrimidine joined #perl6
21:31 RabidGravy it's actually described in the docs somewhere
21:32 moritz still looks like a bug to me
21:35 lizmat raiph: also RT #130634
21:35 synopsebot6 Link:  https://rt.perl.org/rt3//Public/Bug/Display.html?id=130634
21:36 raiph RabidGravy: is the LTA "exists in multiple roles (B,A)" due to role punning?
21:37 raiph lizmat: thanks
21:37 Sgeo_ joined #perl6
21:39 RabidGravy not sure if it's punning, it's just that as it stands it's too late to know that m came from the same place when the roles are composed into C
21:43 moritz a method that's copied to another role or class could carry a reference to the original to make this detection work
21:43 RabidGravy yes I was wondering whether it could
21:44 lizmat I think a =:= comparison should do the trick, really
21:45 RabidGravy anyway I've fixed and tested all my modules that have a Build.pm so that they work with zef
21:45 RabidGravy which is all good
21:47 bjz joined #perl6
21:54 raiph thanks all
21:59 raiph I want to add a followup comment to rt #130634
21:59 synopsebot6 Link:  https://rt.perl.org/rt3//Public/Bug/Display.html?id=130634
22:00 raiph I know I send an email to rakudobug@perl.org
22:00 raiph Do I just have to put #130634 in the subject to send my email to that bug?
22:00 synopsebot6 Link:  https://rt.perl.org/rt3//Public/Bug/Display.html?id=130634
22:01 raiph Heh, just instinctively typed rt.com to get help for rt...
22:01 samcv what is $*DISTRO on macos?
22:03 lizmat samcv: you want dd or say output  (aka, .perl or .gist) ?
22:03 samcv i want to programmatically determine if running on macos
22:04 samcv that's all i need
22:04 [Coke] samcv: $ perl6 -e 'say $*DISTRO'
22:04 [Coke] macosx (10.12.2)
22:04 lizmat $ 6 'say $*DISTRO eq "macosx"'
22:04 lizmat True
22:04 cdg joined #perl6
22:05 brokenchicken Would be handy to have the fullish list of possible values, unless we got it in docs alrteady.
22:05 cdg joined #perl6
22:05 samcv it's determined by the running os though?
22:05 samcv idk for mine  say $*DISTRO eq 'arch' #> True
22:06 brokenchicken for mine say $*DISTRO eq "ubuntu" #> True
22:06 brokenchicken There's bunch of system calls in https://github.com/rakudo/rakudo/blob/nom/src/core/Distro.pm
22:08 RabidGravy yeah, it rather annoyingly uses "uname" on Unix-like system
22:10 lizmat raiph: my attempt so far: https://gist.github.com/lizmat/1badd87639f599e73fdddfecf6ed1ca5
22:11 samcv well it's called $*DISTRO :P
22:11 samcv it tells you the distro ;)
22:16 Cabanoss- joined #perl6
22:16 jnthn The role collision problem some of you are looking at/discussing has an important subtlety
22:16 jnthn That any solution must take into account
22:16 cdg_ joined #perl6
22:17 jnthn Roles are generic and may take extra parameters
22:17 brokenchicken hmmm... a PayPal phishing email arrived right about the time I paid by PayPal on leanpub...
22:17 jnthn Consider `role Foo[::T] { method bar(T $x) { } }`
22:17 jnthn We concretize methods at composition time.
22:17 jnthn This explains why =:= isn't going to work
22:18 jnthn The method that ends up in the class is a concrete rather than a generic
22:18 jnthn Since all methods are generic on ::?CLASS, and we have to instantiate that type variable when we compose. 6 years back when we didn't do that, we got various multi-dispatches really wrong.
22:19 Geth doc/master: 15 commits pushed by coke++
22:19 Geth doc/master: review: https://github.com/perl6/doc/compare/cfcfe5dce4...4516551d37
22:19 Actualeyes joined #perl6
22:19 jnthn Clearly, it's fine if a method that's *only* generic on the invocant is considered non-colliding
22:19 jnthn A method that's got a signature that's otherwise generic, less so.
22:20 jnthn So we'll need to consider the signature of the generic method and see if it contains any genericity besides on the usual ::?CLASS
22:23 jnthn It's tempting to do it somehow use .static_id on a code object, but I fear that will spoil somebody's meta-programming day in the future...
22:23 jnthn And it can only be done in combination with the generics checks anyway.
22:25 Geth doc: 0dfdcc2556 | (Will "Coke" Coleda)++ | doc/Programs/00-running.pod6
22:25 Geth doc: use nbsp
22:25 Geth doc: review: https://github.com/perl6/doc/commit/0dfdcc2556
22:25 cdg joined #perl6
22:29 ChoHag joined #perl6
22:43 kurahaupo__ joined #perl6
22:45 ChoHag joined #perl6
23:05 cdg_ joined #perl6
23:12 Cabanossi joined #perl6
23:19 cdg joined #perl6
23:30 felher joined #perl6
23:31 Geth doc: a73747f62e | (Will "Coke" Coleda)++ | doc/Language/modules.pod6
23:31 Geth doc: clarify toolchain reference
23:31 Geth doc:
23:31 Geth doc: Closes #1136
23:31 Geth doc: review: https://github.com/perl6/doc/commit/a73747f62e
23:34 dataf3l joined #perl6
23:42 cdg joined #perl6
23:43 zdenek joined #perl6
23:47 Actualeyes left #perl6
23:56 jdv79 Cannot call method 'package_at_key' on a null object
23:56 jdv79 any ideas on what that means?
23:58 brokenchicken something with EXPORT sub?

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

Perl 6 | Reference Documentation | Rakudo