Perl 6 - the future is here, just unevenly distributed

IRC log for #perl6, 2016-10-26

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 geekosaur joined #perl6
00:15 pierre_ joined #perl6
00:23 geraud joined #perl6
00:40 inra joined #perl6
00:42 colomon m: say 0.msb
00:42 camelia rakudo-moar 4829f1: OUTPUT«Nil␤»
00:46 k-man joined #perl6
01:01 aborazmeh joined #perl6
01:01 aborazmeh joined #perl6
01:03 Jesmaybe joined #perl6
01:05 kurahaupo joined #perl6
01:17 Jesmaybe joined #perl6
01:19 xfix joined #perl6
01:23 notbenh_ joined #perl6
01:23 isacloud joined #perl6
01:23 avalenn_ joined #perl6
01:24 w4and0er96 joined #perl6
01:29 Ben_Goldberg joined #perl6
01:34 avalenn joined #perl6
01:35 pierre_ joined #perl6
01:39 FROGGS_ joined #perl6
01:39 Jesmaybe joined #perl6
01:41 avalenn joined #perl6
01:48 ilbot3 joined #perl6
01:48 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!
02:01 cibs joined #perl6
02:14 matiaslina joined #perl6
02:31 Jesmaybe joined #perl6
02:49 noganex joined #perl6
02:51 akchli joined #perl6
02:56 akchli left #perl6
02:58 matiaslina joined #perl6
03:01 Herby__ joined #perl6
03:01 Herby__ o/
03:09 remmie left #perl6
03:21 pierre_ joined #perl6
03:24 Herby__ joined #perl6
03:25 pierre_ joined #perl6
03:29 aborazmeh joined #perl6
03:29 aborazmeh joined #perl6
03:41 imcsk8_P1O joined #perl6
04:09 cyphase joined #perl6
04:10 pierre_ joined #perl6
04:31 zacts joined #perl6
04:31 Cabanossi joined #perl6
04:32 TEttinger joined #perl6
04:38 nadim joined #perl6
04:38 canopus joined #perl6
04:41 cognominal joined #perl6
04:52 wamba joined #perl6
05:07 pierre_ joined #perl6
05:08 bjz joined #perl6
05:28 bjz joined #perl6
05:36 cog_ joined #perl6
05:50 RabidGravy joined #perl6
06:08 e1 joined #perl6
06:11 neuraload joined #perl6
06:13 pierre_ joined #perl6
06:22 neuraload joined #perl6
06:24 firstdayonthejob joined #perl6
06:25 darutoko joined #perl6
06:26 itaipu joined #perl6
06:37 domidumont joined #perl6
06:42 domidumont joined #perl6
06:44 pierre_ joined #perl6
06:45 labster joined #perl6
06:46 amalia_ joined #perl6
06:57 pierre_ joined #perl6
06:59 ufobat joined #perl6
07:05 wamba joined #perl6
07:08 rurban joined #perl6
07:11 cooper_ joined #perl6
07:11 FROGGS_ o/
07:14 rindolf joined #perl6
07:24 TheovdH joined #perl6
07:30 bjz joined #perl6
07:34 pdcawley joined #perl6
07:57 pierre_ joined #perl6
07:59 andrzejku_ FROGGS: hey :)
08:01 jonas2 joined #perl6
08:04 g4 joined #perl6
08:04 g4 joined #perl6
08:11 cowens joined #perl6
08:14 TheovdH Hi, if a grammar match fails is there a variable that holds the maximum position of the parsing process?
08:15 cowens The string "\x[09aa]\x[09cd]\x[09b0]" appears to be one character (I think it is the pr ligature in Bengali), the reason "\x[09aa]\x[09cd]\x[09b0]”.chars returns 2 is that “\x[09aa]\x[09cd]” and “\x[09b0]” are the graphemes, and the fact that the font shows প্র instead of প্ and র is just a font level ligature (like fi can become in some fonts), right?
08:16 zakharyas joined #perl6
08:16 dakkar joined #perl6
08:16 TheovdH I did a lot of parsing and such a feature has always been very helpful
08:18 pdcawley joined #perl6
08:22 FROGGS TheovdH: rakudo does that internally, but .match does not do that... but we had several requests already
08:23 FROGGS TheovdH: what I do in such a situation is to use .submatch instead
08:24 pdcawley joined #perl6
08:25 FROGGS m: dd "abc" ~~ /abd/
08:25 camelia rakudo-moar 4829f1: OUTPUT«Nil␤»
08:25 FROGGS m: dd "abc" ~~ /abd/; dd $/
08:25 camelia rakudo-moar 4829f1: OUTPUT«Nil␤Nil $/ = Nil␤»
08:25 FROGGS m: dd "abc".match(/abd/);
08:25 TheovdH FROGGS: thanks. I didn't know about that. Its not in the index. I will see if I get it to work.
08:25 camelia rakudo-moar 4829f1: OUTPUT«Nil␤»
08:25 FROGGS m: dd "abc".submatch(/abd/);
08:25 camelia rakudo-moar 4829f1: OUTPUT«No such method 'submatch' for invocant of type 'Str'␤  in block <unit> at <tmp> line 1␤␤»
08:25 TheovdH right
08:26 FROGGS m: grammar G { token TOP { a+ } }; say G.parse("aaaaaabbbbbb");
08:26 camelia rakudo-moar 4829f1: OUTPUT«Nil␤»
08:26 FROGGS m: grammar G { token TOP { a+ } }; say G.subparse("aaaaaabbbbbb");
08:26 camelia rakudo-moar 4829f1: OUTPUT«「aaaaaa」␤»
08:26 FROGGS ahh, subparse, not submatch
08:26 rindolf joined #perl6
08:27 cibs joined #perl6
08:27 FROGGS m: grammar G { token TOP { a+ } }; dd G.parse("aaaaaabbbbbb");
08:27 camelia rakudo-moar 4829f1: OUTPUT«Nil␤»
08:27 TheovdH ok. thanks
08:28 FROGGS sad that a failed .parse does not through, otherwise we could sneak the highwater mark in there
08:29 TheovdH that would be ideal
08:32 rurban joined #perl6
08:48 zakharyas joined #perl6
08:49 gregf_ joined #perl6
08:59 FROGGS s/through/throw/
09:02 bjz joined #perl6
09:03 pierre_ joined #perl6
09:08 remmie joined #perl6
09:09 remmie left #perl6
09:11 jkramer m: say [Z+] { foo => [1, 2, 3], bar => [2, 3, 4] }.values
09:11 camelia rakudo-moar 4829f1: OUTPUT«(3 5 7)␤»
09:11 jkramer m: say [Z+] { foo => [1, 2, 3], bar => [2, 3, 4] }.values
09:11 camelia rakudo-moar 4829f1: OUTPUT«(3 5 7)␤»
09:11 jkramer m: say [Z+] { foo => [1, 2, 3] }.values
09:11 camelia rakudo-moar 4829f1: OUTPUT«(6)␤»
09:11 jkramer Is there a way to make this work in the edge case where there's only one pair in the hash?
09:13 FROGGS m: say [Z+] { foo => [1, 2, 3], }.values
09:13 camelia rakudo-moar 4829f1: OUTPUT«(6)␤»
09:14 cgfbee joined #perl6
09:14 FROGGS no idea
09:18 moritz m: say [Z+] { foo => [1, 2, 3] }.values, [0 xx *]
09:18 camelia rakudo-moar 4829f1: OUTPUT«(3)␤»
09:27 Averna joined #perl6
09:29 eliasr joined #perl6
09:32 jkramer Hmm, not exactly what I was hoping for, but it'll do :)
09:32 jkramer Thanks
09:40 telex joined #perl6
09:47 aborazmeh joined #perl6
09:47 aborazmeh joined #perl6
09:49 dalek doc: b073a44 | (Vynce Montgomery)++ | doc/Language/about.pod6:
09:49 dalek doc: fix typo in doc about how to fix typo in doc, doc
09:49 dalek doc:
09:49 dalek doc: the first example "=head2" block was spelled as "d2" instead of "=head2" which probably did actually blow some minds... it at least caused a misfire in mine.
09:49 dalek doc: review: https://github.com/perl6/doc/commit/b073a44fd3
09:49 dalek doc: 2c41617 | Altai-man++ | doc/Language/about.pod6:
09:49 dalek doc: Merge pull request #987 from VynceMontgomery/patch-1
09:49 dalek doc:
09:49 dalek doc: fix typo in doc about how to fix typo in doc, doc
09:49 dalek doc: review: https://github.com/perl6/doc/commit/2c41617068
09:57 ocbtec joined #perl6
09:58 ocbtec joined #perl6
09:58 dalek doc/Altai-man-patch-1: fbed1d3 | Altai-man++ | doc/Type/Cool.pod6:
09:58 dalek doc/Altai-man-patch-1: Fix of the example code
09:58 dalek doc/Altai-man-patch-1:
09:58 dalek doc/Altai-man-patch-1: The example was obviously corrupted with redundand whitespace cleanup commit.
09:58 dalek doc/Altai-man-patch-1: VynceMontgomery++
09:58 dalek doc/Altai-man-patch-1: review: https://github.com/perl6/doc/commit/fbed1d37b9
10:00 dalek doc: 5fccce4 | (Vynce Montgomery)++ | doc/Language/classtut.pod6:
10:00 dalek doc: minor mistake fix
10:00 dalek doc:
10:00 dalek doc: This method was referred to as the second method, which was confusing since the method before it was also referred to as the second method. Removed the ordinal; just use the name, for future-proofing.
10:00 dalek doc: review: https://github.com/perl6/doc/commit/5fccce4eea
10:00 dalek doc: 283692a | Altai-man++ | doc/Language/classtut.pod6:
10:00 dalek doc: Merge pull request #988 from VynceMontgomery/patch-2
10:00 dalek doc:
10:00 dalek doc: minor wording fix
10:00 dalek doc: review: https://github.com/perl6/doc/commit/283692abaa
10:08 zakharyas joined #perl6
10:18 travis-ci joined #perl6
10:18 travis-ci Doc build passed. Altai-man 'Fix of the example code
10:18 travis-ci https://travis-ci.org/perl6/doc/builds/170724977 https://github.com/perl6/doc/commit/fbed1d37b981
10:18 travis-ci left #perl6
10:26 pierre_ joined #perl6
10:36 bjz joined #perl6
10:43 CIAvash joined #perl6
10:48 wamba joined #perl6
11:09 wamba joined #perl6
11:10 rurban joined #perl6
11:11 grondilu joined #perl6
11:20 * grondilu is amazed by the output of $ perl6 --target ast -e ''
11:20 thib joined #perl6
11:21 moritz it's pretty boring housekeeping, really
11:24 * lizmat is more surprised by "$ perl6 --target ast -e 'say "foo"'"   which seems to repeat itself ?
11:24 lizmat at the end
11:26 Alikzus joined #perl6
11:44 rurban joined #perl6
12:04 cschwenz joined #perl6
12:05 bhm joined #perl6
12:05 cschwenz left #perl6
12:05 cschwenz joined #perl6
12:06 pierre_ joined #perl6
12:08 FROGGS lizmat: that's about sink vs no-sink context
12:09 zakharyas joined #perl6
12:10 lizmat ah, yes, in this case it could be determined easily that it is a sink context, but generally it's a lot more involved, right ?
12:11 FROGGS I'm not qualified to answer :/
12:13 dalek modules.perl6.org: 9fa5ea3 | (Zoffix Znet)++ | lib/ModulesPerl6/DbBuilder/Dist​/PostProcessor/METAChecker.pm:
12:13 dalek modules.perl6.org: Redesign dist source-url META checker
12:13 dalek modules.perl6.org:
12:13 dalek modules.perl6.org: The current check gives only 1 hour window of opportunity to notice any issues
12:13 dalek modules.perl6.org: with the source-url in META file, as the checker does not run on dists without
12:13 dalek modules.perl6.org: fresh commits.
12:13 dalek modules.perl6.org:
12:13 dalek modules.perl6.org: Perform the check on all dists regardless whether we have new commits, but do
12:13 dalek modules.perl6.org: not do the check if the source-url matches the repo we fetched the META file
12:13 dalek modules.perl6.org: from; since this already guarantees the source-url is correct. This will give
12:13 dalek modules.perl6.org: us even better reduction of HTTP requests than doing the checks for dists
12:13 dalek modules.perl6.org: with fresh commits only and any issues will stay in the build log until fixed.
12:13 dalek modules.perl6.org: review: https://github.com/perl6/modul​es.perl6.org/commit/9fa5ea3257
12:50 CIAvash joined #perl6
13:11 cdg joined #perl6
13:19 sena_kun joined #perl6
13:23 nicq20 joined #perl6
13:24 dalek doc: fbed1d3 | Altai-man++ | doc/Type/Cool.pod6:
13:24 dalek doc: Fix of the example code
13:24 dalek doc:
13:24 dalek doc: The example was obviously corrupted with redundand whitespace cleanup commit.
13:24 dalek doc: VynceMontgomery++
13:24 dalek doc: review: https://github.com/perl6/doc/commit/fbed1d37b9
13:24 dalek doc: 4ec4dbd | Altai-man++ | doc/Type/Cool.pod6:
13:24 dalek doc: Merge pull request #989 from perl6/Altai-man-patch-1
13:24 dalek doc:
13:24 dalek doc: Fix of the example code
13:24 dalek doc: review: https://github.com/perl6/doc/commit/4ec4dbd95d
13:30 skids joined #perl6
13:40 neuraload joined #perl6
13:48 wamba joined #perl6
13:56 dalek doc: 42c2cd7 | coke++ | / (3 files):
13:56 dalek doc: inline predefined-search-items
13:56 dalek doc: review: https://github.com/perl6/doc/commit/42c2cd782a
13:56 dalek doc: 517b4ad | (Will Coleda)++ | / (3 files):
13:56 dalek doc: Merge pull request #980 from perl6/remove-predefined
13:56 dalek doc:
13:56 dalek doc: inline predefined-search-items
13:56 dalek doc: review: https://github.com/perl6/doc/commit/517b4ad640
13:58 acrussell joined #perl6
14:01 dalek doc: 7193e12 | coke++ | doc/Language/modules.pod6:
14:01 dalek doc: remove trailing whitespace
14:01 dalek doc: review: https://github.com/perl6/doc/commit/7193e128d5
14:03 kurahaupo__ joined #perl6
14:06 [Coke] er...
14:06 [Coke] ah, no, 980 is fine.
14:06 [Coke] I thought it pulled in one of the other two I just checked manually.
14:07 itaipu joined #perl6
14:07 dmaestro joined #perl6
14:08 [Coke] doc: `git branch -a --merged` could probably remove most of those branches.
14:17 rurban joined #perl6
14:18 wamba joined #perl6
14:20 khw joined #perl6
14:25 cyphase joined #perl6
14:26 nicq20 If anyone has a minute, I'm having trouble with 'is-deeply'. :/
14:26 masak m: say "abc def abc".comb(/a<(bc)>/)
14:26 camelia rakudo-moar 23faf6: OUTPUT«(abc abc)␤»
14:26 masak I expected "(bc bc)" here
14:27 * masak submits rakudobug
14:27 masak also, that used to work -- it's a recent regression
14:27 masak will investigate/bisect later
14:28 moritz m: 'abc def' ~~ /a<(bc)>/ and say $/
14:28 camelia rakudo-moar 23faf6: OUTPUT«「bc」␤»
14:28 raiph joined #perl6
14:28 moritz curious
14:29 masak oh, here's a better example for the RT ticket:
14:29 masak m: say "abc xbc".comb(/a<(bc)>/)
14:29 camelia rakudo-moar 23faf6: OUTPUT«(abc)␤»
14:29 nicq20 moritz: I thought that comb was supposed to return what it matches?
14:30 masak it's correct in that it matches only on the first word
14:30 masak it's incorrect in that it includes the "a", when the <( )> tell it to exclude it
14:30 [Coke] doc (huh, my git branch -a shows a lot more branches than github does.)
14:30 masak nicq20: does that answer your question?
14:31 jnthn [Coke]: git remote prune origin
14:31 masak (also, I am not moritz) :P
14:31 moritz [Coke]: git fetch --prune
14:31 jnthn moritz++ # shorter solution :)
14:31 [Coke] moritz: heh. just found that in the docs. have never had to use that before.
14:32 moritz [Coke]: you can also run git config --global fetch.prune 1
14:32 viki nicq20: what's the issue with is-deeply?
14:32 masak bug submitted
14:32 masak &
14:33 moritz [Coke]: then it cleans up automatically on each fetch
14:33 nicq20 masak: Yeah, kind of. I was under the impression that it returns what watches the regex, not just any sub-match (the <(bc)> part in this case). If that is true it works fine.
14:33 viki nicq20: it's supposed to, yes
14:33 raiph hi #perl6, fyi, doc.perl6.org search is broken for me
14:34 viki bisectable6: m: say "abc xbc".comb(/a<(bc)>/)
14:34 bisectable6 viki, On both starting points (old=2015.12 new=b7201a8) the exit code is 0 and the output is identical as well
14:34 bisectable6 viki, Output on both points: (bc)
14:34 nicq20 masak: Oops! Sorry, about the name confusion.
14:34 viki nicq20: that may be due to lizmat++'s recent changes. Bisectable shows the right result and I'm guessing it just hasn't built the most recent commits.
14:35 viki nicq20: what's the problem with is-deeply?
14:35 DrForr raiph: Yep, confirmed.
14:35 sena_kun raiph, syntax error, indeed.
14:36 nicq20 viki: I probably just don't understand it fully, but it seems to not like comparing some hashes in my code. I'll see if I can golf it any.
14:36 [Coke] moritz: probably better to do manual cleanup JIC.
14:36 lizmat masak: I'm not sure I agree with you re /a<(bc)>/
14:37 dalek doc: 0b57313 | (Zoffix Znet)++ | template/search_template.js:
14:37 dalek doc: Unbreak search
14:37 dalek doc: review: https://github.com/perl6/doc/commit/0b5731327f
14:37 viki raiph: ^ should work again in about 15 minutes
14:37 [Coke] viki: that was probably a result of my recent merge. apologies.
14:37 [Coke] thanks for cleaning up my mess@!
14:38 viki nicq20: it should work fine with hashes. Perhaps there are object instances that are different?
14:38 viki nicq20: is-deeply is nothing more than eqv with Failure handler under the hood
14:38 lizmat according to the doc, it comb should return matches, *not* captures
14:38 moritz lizmat: <( limits matches
14:39 moritz m: say 'abc' ~~ /<( bc/; say ~$/; say $/.from
14:39 camelia rakudo-moar 23faf6: OUTPUT«「bc」␤bc␤1␤»
14:39 viki m: say 'abc' ~~ / . <( bc/; say ~$/; say $/.from
14:39 camelia rakudo-moar 23faf6: OUTPUT«「bc」␤bc␤1␤»
14:39 viki And the )> does it for the other end
14:39 moritz viki: thanks, my example was pointless :-)
14:39 viki :)
14:40 nicq20 viki: Yeah, is some tests I got it work fine. Seems to be my code.
14:40 raiph DrForr, sena_kun, viki, [Coke], #perl6 <3
14:40 itaipu joined #perl6
14:42 nicq20 viki: Oh, gee whiz. Looks like my code generates a Seq instead of an Array. Well, fiddlesticks. :/
14:42 moritz that's my typical problem with is-deeply. It's very nitpicky
14:44 viki That's my typical joy with is-deeply. It detects issues :) Seq is very different from an Array.
14:44 lizmat m: say "abc".match(/a<(bc)>, :as(Str)/)   # golfed the problem
14:44 camelia rakudo-moar 23faf6: OUTPUT«5===SORRY!5===␤Unrecognized regex metacharacter , (must be quoted to match literally)␤at <tmp>:1␤------> 3say "abc".match(/a<(bc)>7⏏5, :as(Str)/)   # golfed the problem␤Unable to parse regex; couldn't find final '/'␤at <tmp>:1␤------>…»
14:44 lizmat m: say "abc".match(/a<(bc)>/, :as(Str))   # really golfed the problem
14:44 camelia rakudo-moar 23faf6: OUTPUT«abc␤»
14:45 lizmat m: say "abc".match(/a<(bc)>/, :as(Match))   # correct
14:45 camelia rakudo-moar 23faf6: OUTPUT«「bc」␤»
14:45 viki Ahhh
14:45 FROGGS m: say "abc".match(/a<(bc/, :as(Str)) # golfed more :P
14:45 camelia rakudo-moar 23faf6: OUTPUT«abc␤»
14:45 lizmat note this was not picked up by spectests  :-(
14:45 FROGGS :o(
14:47 moritz lizmat: you can't expect the spectets to cover all combinations of features
14:47 lizmat true
14:47 moritz lizmat: if there are tests that check that match objects are constructed correctly, then I expect there to be one test with .comb(regex)
14:47 lizmat but can you expect a developer to try out all of the combinations of features during development ?
14:48 viki Why not?
14:48 nicq20 Does 'Z' always return a Seq?
14:48 viki I'd expect the tests to be comprehensive enough to test all the features. Especially if the roast is THE Perl 6 specification
14:49 FROGGS viki: it will cover more and more and that is certainly the goal
14:49 FROGGS but >99.999% is unlikely to happen
14:49 nicq20 m: dd <a b> Z <1 2>
14:49 camelia rakudo-moar 23faf6: OUTPUT«(("a", IntStr.new(1, "1")), ("b", IntStr.new(2, "2"))).Seq␤»
14:50 viki m: dd ^2 Z ^2+2
14:50 camelia rakudo-moar 23faf6: OUTPUT«((0, 2), (1, 3)).Seq␤»
14:50 moritz viki: because there are too many combinations to cover if you aim for an exponential space
14:50 FROGGS would be nice to have a overage analysis... maybe next gsoc?
14:50 FROGGS coverage*
14:50 viki FROGGS: there's a rought analyzis already on perl6.WTF
14:51 viki FROGGS: well, it's *full* analysys, but the coverage checker isn't perfect yet. And well, it can't tell you whether you're testing a combination of features
14:51 FROGGS ahh, I see
14:51 FROGGS still awesome though
14:51 viki timotimo++ made it
14:51 hwcomcn joined #perl6
14:52 FROGGS timotimo++ then :o)
14:52 viki s: &infix:<Z>, \(^2, ^2+2)
14:52 SourceBaby viki, Sauce is at https://github.com/rakudo/rakudo/b​lob/23faf61/src/core/List.pm#L1622
14:52 nicq20 So is 'Z' incorrect on the docs? According to https://docs.perl6.org/routine/Z it should give a 'List'.
14:53 nicq20 viki: Oh, I see. It uses a give/take. That would generate the Seq.
14:54 dalek doc: fd3d7b8 | (Zoffix Znet)++ | doc/Language/operators.pod6:
14:54 dalek doc: Z meta returns a Seq
14:54 dalek doc: review: https://github.com/perl6/doc/commit/fd3d7b8eca
14:54 viki m: sub foo { return (); }; dd foo().WHAT
14:54 camelia rakudo-moar 23faf6: OUTPUT«Cannot look up attributes in a type object␤  in block <unit> at <tmp> line 1␤␤»
14:55 moritz m: sub foo { return (); }; dd foo()
14:55 camelia rakudo-moar 23faf6: OUTPUT«()␤»
14:55 moritz m: sub foo { return (); }; say foo().^name
14:55 camelia rakudo-moar 23faf6: OUTPUT«List␤»
14:55 viki m: sub foo { gather {} }; dd foo().^name
14:55 camelia rakudo-moar 23faf6: OUTPUT«"Seq"␤»
14:57 viki nicq20: well, it returns an empty List if the list of lists got no elements, but that may be a bug/inconsistency
14:57 viki m: dd [] Z []
14:57 camelia rakudo-moar 23faf6: OUTPUT«().Seq␤»
14:58 * viki has no idea how to trigger that return () if lol.elems == 0
14:58 lizmat nicq20: testing a .comb fix now
14:58 viki Ah
14:58 viki m: dd infix:<Z>
14:58 camelia rakudo-moar 23faf6: OUTPUT«()␤»
14:58 viki nicq20: it returns a list when called without args.
14:59 nicq20 viki: Not super useful, but it returns a list! :)
15:00 viki nicq20: not for long... :P
15:03 nightfrog joined #perl6
15:05 nicq20 Well, if a List is really needed, look like i can just use '.eager' to turn it into a List.
15:05 viki nicq20: why not .List?
15:05 nightfrog joined #perl6
15:05 nicq20 m: (<a b> Z <1 2>).eager.WHAT
15:05 camelia rakudo-moar 23faf6: ( no output )
15:05 nicq20 m: say (<a b> Z <1 2>).eager.WHAT
15:05 camelia rakudo-moar 23faf6: OUTPUT«(List)␤»
15:06 viki m: (1, 2 ... *).List[1..10].say
15:06 camelia rakudo-moar 23faf6: OUTPUT«(2 3 4 5 6 7 8 9 10 11)␤»
15:06 viki m: (1, 2 ... *).eager[1..10].say
15:06 viki (hang)
15:06 camelia rakudo-moar 23faf6: OUTPUT«(timeout)»
15:08 nicq20 viki: Huh, looks like '.List' would work better (didn't know that was a thing :D). '.eager' turns it into a list before returning, hence the hang right?
15:08 viki nicq20: eager evaluates it eagerly, making it fetch all the elements, hence the hang, since it's an infinite sequence
15:09 viki There's a .Type for pretty much all of our types
15:09 viki m: DateTime.now.Date.say
15:09 camelia rakudo-moar 23faf6: OUTPUT«2016-10-26␤»
15:09 viki m: Mix.new(1 => 1.243, 5 => 30).Bag.say
15:09 camelia rakudo-moar 23faf6: OUTPUT«bag(1 => 1.243, 5 => 30)␤»
15:10 viki wat
15:10 nicq20 viki: ... Oh.
15:10 viki oh
15:10 viki m: Mix.new-from-pairs(1 => 1.243, 5 => 30).Bag.say
15:10 camelia rakudo-moar 23faf6: OUTPUT«bag(5(30), 1)␤»
15:11 nicq20 viki: Does the .Type thing go off of constructors? How does it work?
15:12 viki nicq20: it's just a regular method. We simply made them DoTheRightThing on most of core types
15:12 viki s: DateTime.now, 'Date', \()
15:12 SourceBaby viki, Sauce is at https://github.com/rakudo/rakudo/blo​b/23faf61/src/core/DateTime.pm#L364
15:13 viki ^in that case, it just a straightforward instantiation
15:13 stevieb9 joined #perl6
15:13 viki s: Mix.new-from-pairs(1 => 1.243, 5 => 30), 'Bag', \()
15:13 SourceBaby viki, Sauce is at https://github.com/rakudo/rakudo/​blob/23faf61/src/core/Mix.pm#L37
15:13 viki In this case ^ we have to .Int the values and filter out the negatives
15:13 viki And basically all objects have some sort of thing like that (or their ancestor classes have some generic handling)
15:14 stevieb9 joined #perl6
15:14 viki And IIRC this is also what the type coercers use
15:14 viki m: sub (Bag(Mix) $x) { dd $x }(Mix.new(1 => 1.243, 5 => 30))
15:14 camelia rakudo-moar 4943ef: OUTPUT«(1 => 1.243=>1,5 => 30=>1).Bag␤»
15:15 lizmat nicq20: https://github.com/rakudo/rakudo/commit/ae85bac528
15:15 viki m: sub (Bag(Mix) $x) { dd $x }(Mix.new-from-pairs(1 => 1.243, 5 => 30))
15:15 camelia rakudo-moar 4943ef: OUTPUT«(5=>30,1=>1).Bag␤»
15:15 viki m: sub (YourMom(Mix) $x) { dd $x }(Mix.new-from-pairs(1 => 1.243, 5 => 30))
15:15 camelia rakudo-moar 4943ef: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Invalid typename 'YourMom' in parameter declaration.␤at <tmp>:1␤------> 3sub (YourMom7⏏5(Mix) $x) { dd $x }(Mix.new-from-pairs(1␤»
15:15 viki heh
15:17 viki nicq20: and that's it. infix:<Z> now always-always returns a Seq
15:17 dalek Net-ZMQ: 2874908 | (Zoffix Znet)++ | META.info:
15:17 dalek Net-ZMQ: Add mandatory "perl" META field
15:17 dalek Net-ZMQ:
15:17 dalek Net-ZMQ: The `perl` field specifies the minimal perl version for which this distribution can be installed and is a mandatory field. The value of `"6.*"` indicates any version suffices.
15:17 dalek Net-ZMQ:
15:17 dalek Net-ZMQ: It is recommended to use [Test::META](https://modules.perl6.org/repo/Test::META) module as an author test, to catch any issues with the META file.
15:17 dalek Net-ZMQ: review: https://github.com/arnsholt​/Net-ZMQ/commit/28749083ea
15:17 dalek Net-ZMQ: d89c744 | azawawi++ | META.info:
15:17 dalek Net-ZMQ: Merge pull request #8 from zoffixznet/patch-1
15:17 dalek Net-ZMQ:
15:17 dalek Net-ZMQ: Add mandatory "perl" META field
15:17 dalek Net-ZMQ: review: https://github.com/arnsholt​/Net-ZMQ/commit/d89c744c23
15:20 viki m: class YourMom {}; sub (YourMom(Mix) $x) { dd $x }(Mix.new-from-pairs(1 => 1.243, 5 => 30))
15:20 camelia rakudo-moar 4943ef: OUTPUT«No such method 'YourMom' for invocant of type 'Mix'␤  in sub  at <tmp> line 1␤  in block <unit> at <tmp> line 1␤␤»
15:20 viki Yeah, it does
15:22 dalek doc: d123ff2 | MasterDuke17++ | doc/Type/Seq.pod6:
15:22 dalek doc: Fix some awkward grammar
15:22 dalek doc: review: https://github.com/perl6/doc/commit/d123ff2296
15:26 stevie_b joined #perl6
15:27 Actualeyes joined #perl6
15:32 nicq20 Woo, consistancy!
16:10 itaipu joined #perl6
16:12 wamba joined #perl6
16:15 raiph 2009: "jnthn: I've never been entirely clear on what package vs module distinction is ... masak: neither have I." (https://irclog.perlgeek.de/​perl6/2009-11-17#i_1741018)
16:15 raiph 2014: "Ven: well, I still don't really get what's the diff. between package VS module (semantically)" (https://irclog.perlgeek.de/​perl6/2014-07-21#i_9058427)
16:16 * jnthn figured it out somewhere around 2010 :)
16:16 jnthn m: module A { }; module A { }
16:16 camelia rakudo-moar 70a367: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Redeclaration of symbol 'A'␤at <tmp>:1␤------> 3module A { }; module A7⏏5 { }␤    expecting any of:␤        generic role␤»
16:17 jnthn m: package A { }; package A { }
16:17 camelia rakudo-moar 70a367: ( no output )
16:17 jnthn m: package A { }; module A { }
16:17 camelia rakudo-moar 70a367: ( no output )
16:18 jnthn A package is just a stub that can be replaced/stolen by a module/class/role
16:18 jnthn Of note
16:18 jnthn m: class A::B { }; say A::B.HOW.^name; say A.HOW.^name
16:18 camelia rakudo-moar 70a367: OUTPUT«Perl6::Metamodel::ClassHO​W␤Perl6::Metamodel::PackageHOW␤»
16:18 dalek DBIish: 75d2d43 | niner++ | lib/DBDish/Pg/Connection.pm6:
16:18 dalek DBIish: Do not rely on your dependencies to load modules
16:18 dalek DBIish:
16:18 dalek DBIish: Fixes DBDish::Pg::Connection failing to compile when loading a module properly
16:18 dalek DBIish: imports symbols lexically instead of as globals.
16:18 dalek DBIish: review: https://github.com/perl6/DBIish/commit/75d2d43550
16:18 dalek DBIish: 4ce29ec | niner++ | t/0 (2 files):
16:18 dalek DBIish: Prepare tests for lexical module loading
16:18 dalek DBIish: review: https://github.com/perl6/DBIish/commit/4ce29eca0f
16:18 jnthn That's why I can write:
16:18 jnthn m: class A::B { }; class A { }
16:18 camelia rakudo-moar 70a367: ( no output )
16:18 jnthn And it is able to nab the A
16:19 nine Nested packages are...tricky
16:19 nine Especially in combination with scoping
16:20 jnthn Indeed
16:20 jnthn When the "nabbing" happens cross-compilation unit it's..."fun"
16:24 nicq20 Huh. Never quite understood the difference myself, but that makes sense.
16:29 raiph m: package A { sub foo is export { say 42} }; module A { sub foo is export { say 42 } }
16:29 camelia rakudo-moar 70a367: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤A symbol '&foo' has already been exported␤at <tmp>:1␤»
16:29 jnthn It's about the A itself; it doesn't imply anything about what's inside of it.
16:34 raiph jnthn: So `package ...` =~= `namespace ...`?
16:36 jnthn raiph: Well, namespace is a bit of an overloaded word, but I guess so
16:37 jnthn raiph: A package/module/class etc. have a Stash, which is what actually holds the symbols (it's the thing you get back from .WHO)
16:37 jnthn So in that sense they're all namespaces
16:38 jnthn They're all different kinds of type, with package being the only kind that doesn't care about redeclaration.
16:39 canopus joined #perl6
16:46 breinbaas joined #perl6
16:54 dalek json: 1b33a57 | (Zoffix Znet)++ | META.info:
16:54 dalek json: Add mandatory "perl" META field
16:54 dalek json:
16:54 dalek json: The `perl` field specifies the minimal perl version for which this distribution can be installed and is a mandatory field. The value of `"6.*"` indicates any version suffices.
16:54 dalek json:
16:54 dalek json: It is recommended to use [Test::META](https://modules.perl6.org/repo/Test::META) module as an author test, to catch any issues with the META file.
16:54 dalek json: review: https://github.com/moritz/json/commit/1b33a570ef
16:56 * moritz guesses that Zoffix++ won't have problems getting his 4 PRs for the hacktoberfest challenge :-)
16:56 timotimo yeah
16:57 viki "299 contributions on Oct 26, 2016" https://github.com/zoffixznet/
16:57 timotimo "created 102 repositories"
16:58 timotimo fantastic
16:58 timotimo "hot! created a pull request in rust-lang/rust that received 3 comments"
16:58 viki just a typo fix
16:59 aries_liuxueyang joined #perl6
16:59 timotimo i saw %)
17:01 viki .tell masak some of your modules have unmerged PRs for META file issues going back over a year ago. The issues they fix currently render the modules uninstallable (provides section). You also got a few more fixes today. All of these repos have *at least* one META PR: https://gist.github.com/zoffixznet​/66edf838cca4f6ccb9baa1633ef383aa
17:01 yoleaux viki: I'll pass your message to masak.
17:12 raiph m: package A { sub foo { say 11 }; sub bar { say 12 } }; package A {  sub bar { foo }; sub foo { say 22 } }; say A::&foo.signature
17:12 camelia rakudo-moar ab9469: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Malformed lookup of ::&foo; please use ::('&foo'), ::{'&foo'}, or ::<&foo>␤at <tmp>:1␤------> 3bar { foo }; sub foo { say 22 } }; say A7⏏5::&foo.signature␤»
17:12 raiph oops, sorry for noise
17:16 sufrostico joined #perl6
17:18 stevieb9 joined #perl6
17:21 firstdayonthejob joined #perl6
17:24 grondilu guys, do you keep an eye on webassembly?  The more I read about it, the more I'm looking forward to Perl 6 targetting it.
17:25 viki It is asmjs? The one that lets you boot up Windows 98 in your browser XD
17:25 viki s/It is/Is it/;
17:25 grondilu not quite.
17:26 grondilu asm.js is a subset of js that can be compiled into somewhat native code.
17:26 viki Seenms to be similar "The initial implementation of WebAssembly support in browsers will be based on asm.js" https://en.wikipedia.org/wiki/WebAssembly
17:27 grondilu yes but no
17:27 grondilu it's more complicated than that.
17:27 viki OK.
17:27 pdcawley joined #perl6
17:28 viki Well, as with most any of stuff like this: it's all a matter of available volunteers.
17:28 grondilu there is already a C Api that allows you to build wasm code without passing through any javascript stage
17:28 grondilu https://github.com/WebAssembly/binaryen
17:28 viki As in, "keeping an eye on" doesn't matter much if there's no time to do anything.
17:29 pdcawley joined #perl6
17:29 * viki reads http://perl6.party/post/Open-Source​-Projects--When-Ideas-Meet-Reality
17:30 grondilu some stuff can be done I believe.  at least one programming language already has an experimental compiler to wasm:  https://github.com/brson/mir2wasm/
17:30 pdcawley joined #perl6
17:30 grondilu (that language being Rust)
17:30 viki \o/
17:30 * viki is learning Rust ATM
17:30 viki grondilu: can we use that C API through our NativeCall?
17:31 viki Or is that... compiles C to wasm without JS intermediarry
17:31 grondilu of course, but that won't be much useful.
17:31 viki Ah
17:32 espadrine joined #perl6
17:32 grondilu the C API is very low level.  It just serializes a wasm AST if I understand correctly.
17:33 grondilu so you'd still need to do quite a bit of translation between different kinds of AST.
17:33 grondilu or translate Perl 6's 6model into something that can directly be written into wasm code.  Or something like that.
17:35 grondilu if Perl 6 were targetting wasm, we could write code in Perl 6 that would run in the browser, and access all DOM stuff.  That'd be insanely cool.
17:36 viki pmurias++ is currently working on the JS backend. Maybe you should touch base with him on wasm
17:37 stevieb joined #perl6
17:38 grondilu pmurias seems to struggle quite a bit with the JS backend.  I'm starting to lose hope it will ever be done.
17:38 viki :/
17:38 grondilu and wasm is the future anyway.
17:39 viki ehehe
17:39 viki I think I recall someone telling me the same thing about Ruby on Rails :)
17:42 [Coke] pmurias++ is making progress.
17:42 [Coke] if you have specific areas of concern, please let he or I know.
17:42 [Coke] If you'd like to donate cycles, also, same.
17:43 grondilu "donate cycles"?
17:43 perlpilot do work.
17:44 viki Oh...
17:44 perlpilot I'd be interested in hearing more about this "struggle quite a bit" in some detail.
17:44 * viki puts away his bicycle
17:45 viki I saw a ton of commits fly in recently. Didn't look like a struggle to me.
17:45 grondilu if you want me to admit that behind my "struggle quite a bit" there is only a "it's not there yet", I will.
17:45 [Coke] I think that's a more reasonably stated concern, yes. :)
17:46 viki :D
17:46 CIAvash joined #perl6
18:09 hobbs joined #perl6
18:10 * [Coke] adds https://github.com/rakudo/rakudo/wiki/dev-env-vars
18:11 pmurias joined #perl6
18:11 viki FWIW RAKUDO_EXCEPTIONS_HANDLER is a user-facing feature, not just dev
18:12 viki There's also RAKUDO_OPTIMIZER_DEBUG and RAKUDO_MODULE_DEBUG
18:12 [Coke] viki: the word dev is confusingly used there.
18:13 [Coke] viki: note that some of these are already listed in doc's 00-running and need to be moved to the wiki.
18:13 timotimo do we want to have MVM env vars in there, too?
18:13 [Coke] more like dev as in "foo-dev", "I'm going to do dev work with foo", not "I am a dev of foo"
18:13 viki [Coke]: what are we moving? Just the ones useful for rakudo dev only?
18:13 pmurias grondilu: wasm is currently mostly a more concise form of asm.js
18:14 [Coke] timotimo: sure. I was just finding a place to hang them that wasn't in doc's, based on previous chat about rakudo-specific things not being in there.
18:14 [Coke] viki: anything RAKUDO specific, likely.
18:14 viki fair nuff
18:18 dalek doc: ce19814 | coke++ | doc/Programs/00-running.pod6:
18:18 dalek doc: link to rakudo-specific vars documented elsewhere
18:18 dalek doc: review: https://github.com/perl6/doc/commit/ce198142ed
18:18 rurban joined #perl6
18:21 [Coke] timotimo: added MVM evn vars.
18:21 RabidGravy joined #perl6
18:22 timotimo nice
18:22 viki [Coke]++
18:23 [Coke] that leaves https://github.com/perl6/doc/issues/990 which I just opened.
18:23 [Coke] (basically the inverse of my original ticket, which was to put all the RAKUDO_ ones in there; alex pointed out my wrong-thinking on that one.)
18:24 [Coke] (mainly because we don't have a styleguide/actual style on dealing with rakudo-specific things that aren't really spec.)
18:30 Sqirrel joined #perl6
18:31 pdcawley joined #perl6
18:33 avuserow hi all. I saw a link about JSON parsers on Hacker News that had a test suite, so I forked it and added Perl 6's JSON::Tiny. My fork is at https://github.com/avuserow/JSONTestSuite , with results in the results directory and the implementation in parsers/test_json_tiny.p6
18:33 avuserow I'm still looking over the results gradually (since I'm at $dayjob) but thought some people here might be interested
18:34 avuserow especially since it's easy to add more parser implementations and it seems to have found some crashes at various levels
18:34 avuserow oh, and the original writeup is at http://seriot.ch/parsing_json.html
18:37 viki avuserow: are there prepared results or do we need to run test_json_tiny?
18:37 timotimo JSON::Fast probably wants to go there, too
18:37 timotimo it's not ... the most robust implementation :)
18:38 FROGGS viki: look at https://github.com/avuserow/JSONTes​tSuite/blob/master/results/logs.txt
18:38 vendethiel joined #perl6
18:38 avuserow viki, there are visual results in `results/parsing.html` if you want. (clone and open locally is probably easiest)
18:38 viki thanks
18:39 avuserow if you want to replicate it locally, run `python3 run_tests.py` in the root directory. Looks like it should work on OSX and Linux.
18:40 viki Yeah, it's gonna take forever to clone on my crappy network with 38 KiB/s :}
18:41 timotimo ugh
18:41 timotimo poor you
18:42 avuserow :( might be faster to just download the result file(s) you care about
18:42 timotimo github ought to let you grab it via the raw link
18:42 viki Yeah, I grabbed results and style.css and looks like it worked
18:42 mst viki: 'git clone --depth 2 ...'
18:43 viki star: use JSON::Tiny; dd from-json "-1e+9999"
18:43 camelia star-m 2016.04: OUTPUT«-Inf␤»
18:44 viki mst: I don't think that'll work due to this being a work. I get "Counting objects: 1047" normally and "Counting objects: 1023" with --depth... Still slow :)
18:44 viki a fork, not a work :|
18:44 mst viki: oh, well, yes, if it's basically all a single commit that won't help
18:44 mst tis a useful thing to know exists though, because -usually- it does
18:45 viki Yeah, thanks. mst++
18:45 girafe joined #perl6
18:46 viki star: use JSON::Tiny; dd from-json Q|["\uD800\u"]|
18:46 camelia star-m 2016.04: OUTPUT«Malformed UTF-16; incomplete surrogate pair␤  in method str_escape at /home/camelia/star-2016.04/share/perl6/site/so​urces/E796CB0CB7DEAEAFA56E8558AD18241417AC328A (JSON::Tiny::Actions) line 51␤  in regex str_escape at /home/camelia/star-2016.04/share/perl6…»
18:46 viki hehe
18:49 viki Looks like a lot of those "parser crashed" are just Unicode encoding stuff
18:49 avuserow I tried to make it so anything but a properly typed exception from JSON::Tiny counted as a crash, so some crashes might just be uncaught exceptions that should be rewrapped
18:51 viki star: use JSON::Tiny; dd from-json Q|["\uD800\u"]|; CATCH { default { .^name.say } }
18:51 camelia star-m 2016.04: OUTPUT«X::AdHoc␤»
18:51 viki aye
18:51 viki star: use JSON::Fast; dd from-json Q|["\uD800\u"]|;
18:51 camelia star-m 2016.04: OUTPUT«expected exactly four alnum digits after \u␤  in block  at sources/11ED005DCC03AA42CAA87A439556DB9B6B2C9E59 (JSON::Fast) line 102␤  in sub parse-string at sources/11ED005DCC03AA42CAA87A439556DB9B6B2C9E59 (JSON::Fast) line 76␤  in block  at sources/11ED00…»
18:52 viki .oO( we pack two JSON encoders with Star? )
18:52 timotimo i think panda uses JSON::Fast
18:52 viki Ah
18:54 timotimo JSON::Fast doesn't do any typed exceptions at all
18:55 avuserow viki, I re-ran it without treating untyped exceptions as crashes (just fails), the affected reports are at https://gist.github.com/avuserow/​3805b294df824b0f9cb2bf2df98f3e31
18:57 viki star: use JSON::Tiny; dd from-json Q|[1,]|
18:57 camelia star-m 2016.04: OUTPUT«Input (4 characters) is not a valid JSON string␤  in sub from-json at sources/9B467EEF9267A777BB53BAA2F19BE2C9D756BEED (JSON::Tiny) line 59␤  in block <unit> at <tmp> line 1␤␤»
18:57 pochi joined #perl6
18:57 viki Oh, I'm looking at wrong thing
18:58 timotimo Perl6 JSON::TinyCRASHn_structure_​100000_opening_arrays.json - stack exhaustion?
18:59 timotimo avuserow: could you be so lovely and also run it with JSON::Fast? :)
18:59 stmuk_ http://pl6anet.org/drop/rakudo-star-2016.10.1.dmg
19:00 stmuk_ can anyone confirm that works on 10.11 and 10.12
19:00 labster joined #perl6
19:01 ufobat joined #perl6
19:01 timotimo i know JSON::Fast does absolutely nothing about surrogates and such
19:01 avuserow timotimo: I just added it, let me push
19:01 timotimo fantastic
19:01 timotimo i expect much more fails :)
19:02 timotimo it's JSON::Fast, not JSON::Correct, after all ;)
19:02 timotimo basically it should be called JSON::Fast::GoodEnough
19:02 RabidGravy xl
19:02 avuserow timotimo: pushed, including results. please take a look at the test script too to make sure it's not insane
19:02 avuserow lunch &
19:03 timotimo i'm only looking at logs.txt, and i don't see the update there :P
19:03 avuserow (I put it in the repo, not the gist)
19:03 RabidGravy we're easily pleased
19:03 timotimo OK
19:04 FROGGS JSON::Fast is better at least
19:04 dha joined #perl6
19:05 timotimo better than what?
19:05 timotimo anyway, i'll go grocery shopping so we can have dinner
19:06 FROGGS timotimo: better than JSON::Tiny
19:06 RabidGravy cool, can you get some of that really fine spaghetti as we're all out now
19:13 cognominal joined #perl6
19:17 RabidGravy speaking of JSON I would like a thing like XSLT but for JSON (and in JSON,) someone be a love and write it for me
19:22 amalia_ joined #perl6
19:27 buggable joined #perl6
19:27 viki buggable: eco
19:28 buggable viki, Out of 739 Ecosystem dists, 323 have warnings and 0 have errors. See https://modules.perl6.org/update.log for details
19:28 viki \o/
19:28 viki The front page says we have 736 dists... One missing is the test module Foo, one more may be another test one, but third one is missing somehow
19:29 RabidGravy "Required `perl` field is missing" seems to be the favourite
19:29 moritz maybe one of them counts multiple versions of the same dist as as separate?
19:30 viki Yeah, it does. And now that you mention it, I recall tbrowder uploading two different versions, so I guess that accounts for the discrepancy.
19:30 viki RabidGravy: not for long, hopefully. I'm about half-way through submitting PRs to fix that :)
19:31 FROGGS viki: 731 and 732 look identical...
19:31 viki Yeah, that's tbrowder's two different versions (sha1s are different)
19:32 RabidGravy there's one missing description and a handful of missing 'provides'
19:32 viki :o
19:32 viki the latter are uninstallable
19:32 RabidGravy viki, are you just doing the 'perl' ones - I'll do the rest over the next few days
19:33 viki Well, unless they're scripts. I think those would still install.
19:33 viki (and IIRC the site doesn't check to see whether there are modules in the distro)
19:33 viki RabidGravy: yeah, just the perl ones
19:33 RabidGravy $ grep '\[warn\]' update.log | cut -c35- | sort | uniq -c
19:33 RabidGravy 1 Required `description` field is missing
19:33 RabidGravy 322 Required `perl` field is missing
19:33 RabidGravy 11 Required `provides` field is missing
19:33 viki RabidGravy: note that I spotted a couple of masak's distros without provides but with PRs adding those provides
19:34 viki your bash-foo impresses me :)
19:34 RabidGravy the missing provides means they won't work when they're installed doesn't it?
19:34 viki RabidGravy: right. The exception being binaries, as those do not need to be listed in provides. Though, I'm of the opinion such dists need to include an empty provides section.
19:35 viki Since it's a mandatory section.
19:35 RabidGravy I just impressed myself by guessing the first time the value for cut
19:35 RabidGravy I'll start doing them then :0)
19:35 viki RabidGravy++
19:37 andrzejku joined #perl6
19:48 viki Do we have a way to specify in the `perl` field version PRIOR to some version, rather than a minimum? Looking for answer for https://github.com/labster/p6-Lingua-​Number/pull/4#issuecomment-256449491
19:51 cdg joined #perl6
19:52 masak moritz: I just saw http://seriot.ch/parsing_json.html and thought about you for some reason :)
19:52 yoleaux 17:01Z <viki> masak: some of your modules have unmerged PRs for META file issues going back over a year ago. The issues they fix currently render the modules uninstallable (provides section). You also got a few more fixes today. All of these repos have *at least* one META PR: https://gist.github.com/zoffixznet​/66edf838cca4f6ccb9baa1633ef383aa
19:52 * viki & relocating
19:56 moritz masak: :-)
19:56 moritz masak: it's an interesting exercise, but I won't start rejecting JSON files with non-ASCII white space, for example
19:57 pdcawley joined #perl6
19:59 masak why do people insist on writing "white spaces" (countable) instead of the much more intuitive "white space" (uncountable)?
20:00 masak "space" is a mass noun; so is "white space"
20:00 * lucs likes "whitespace"
20:01 geekosaur nobody understands mass nouns any more
20:01 rindolf joined #perl6
20:02 harmil_wk I want to write a function that takes two functions as parameters and returns a function that has the union of their signatures and calls both of them with the appropriate parameters. Is that possible?
20:02 masak harmil_wk: "union of their signatures"? could you provide an example?
20:02 vendethiel masak: just for the anecdote... In french, if you're talking about "space" as in "void", it's "un espace". But for the character, it's "une escape" (fem.)
20:03 vendethiel une espace, rather.
20:03 masak to me, "union" is something like set("A", "B") ∪ set("A", "C") -> set("A", "B", "C")
20:03 vendethiel sounds about right
20:03 harmil_wk e.g. I have "sub foo($a, :$b) {...}" and "sub bar($c, :$b) {...}" and I want to take those as parameters and return something like "-> $a, $c, :$b {...}'
20:03 masak aha
20:03 masak right, so "union" on the nameds side of things
20:03 vendethiel wouldn't it be easier if everything was named there?
20:03 harmil_wk yes
20:03 masak and concatenation on the positionals side of things
20:04 jnthn I'd just accept everything into a slurpy and a cache, introspect the sigs, and slice
20:04 jnthn wtf
20:04 harmil_wk I guess for positionals, the naming won't get joined in a useful way, but that's okay
20:04 jnthn I'd just accept everything into a slurpy array and hash, introspect the sigs, and slice
20:04 leego joined #perl6
20:04 vendethiel jnthn: are you doing cache-related stuff on moarvm? :P
20:05 harmil_wk I need to return a function that introspects correctly (I'm trying to define MAIN programmatically, so I actually have to have it have the right signature)
20:05 jnthn heh, no, though I guess I was yesterday at dayjob and I will be tomorrow :P
20:07 harmil_wk I'm trying to write a module that gives MAIN defaults from a config file, but otherwise does the normal MAIN processing on args.
20:07 * vendethiel has been doing Perl 5 for work recently, and fell even more in love with Perl 6 doing so..
20:07 vendethiel The fact it was written by people several years ago who barely knew Perl 5 doesn't help; certainly.
20:07 timotimo FROGGS: i'm not aware of doing anything right with JSON::Fast :)
20:07 vendethiel .oO( At least I got all my colleagues to install rakudo on their machine to get some tools running )
20:08 harmil_wk vendethiel: I highly recommend looking into http://modernperlbooks.com/ it might help
20:08 harmil_wk But so will installing rakudo :)
20:08 vendethiel harmil_wk: I'm not allowed to rewrite that code ;-)
20:08 vendethiel "it could break stuff" I mean, I found bugs. Turns out -- they didn't matter because most of this is dead code anyway...
20:08 harmil_wk Well, adding some "use" statements at the top isn't really "rewriting" ;-)
20:09 vendethiel harmil_wk: having to fix the warnings/errors that ensue however... (it does have strict & warnings, though! Good start.)
20:10 vendethiel No one here likes Perl, partly because of this... (and because they are C++/Java developers and refuse to believe their language isn't the One True Way)
20:11 RabidGravy right, just reviewed all of those "provides" things
20:11 RabidGravy they all have PRs now
20:12 RabidGravy masak, you have a bunch of PRs on various modules updating the META (some are missing 'perl' or 'provides' or both )
20:13 setty1 joined #perl6
20:21 masak RabidGravy: thanks for the heads-up. I have many such notifications turned off, so I need to be told which ones to pay attention to.
20:27 RabidGravy I can give you a list of the afflicted modules in a tick,
20:32 RabidGravy viki, I just sent a PR for the lone missing "description" one too
20:34 RabidGravy masak, they are tardis, Adventure-Engine, crypt, text-indented :)
20:38 bjz joined #perl6
20:39 timotimo https://twitter.com/tinLoaf​/status/791375287430868992 - someone recognize this language/alphabet/writing system?
20:43 dalek doc: 3813f5e | (David H. Adler)++ | doc/Language/5to6-perlop.pod6:
20:43 dalek doc: Fixed typo ("explains" to "explain" on line 14)
20:43 dalek doc: review: https://github.com/perl6/doc/commit/3813f5e4df
20:44 RabidGravy timotimo, I've seen that one today
20:44 vendethiel joined #perl6
20:46 moritz Armenian
20:46 bjz joined #perl6
20:46 timotimo oh
20:46 moritz I've cheated, and painted two of them on http://shapecatcher.com/
20:46 timotimo thanks :)
20:46 moritz https://en.wikipedia.org/wi​ki/Armenian_(Unicode_block)
20:46 moritz .. and I've answered on twitter
20:47 timotimo noooooo, you stole my thunder
20:47 timotimo that i was going to steal from you
20:47 moritz lol
20:50 sufrostico joined #perl6
20:51 leego joined #perl6
20:52 timotimo nobody cares about my #WeltraumSchlager tweet ;_;
20:52 moritz oh noez.
20:54 RabidGravy shocking
20:55 wamba joined #perl6
20:57 lizmat timotimo: https://www.youtube.com/watch?v=9T3Xz5elbe4  ??
21:03 Zoffix joined #perl6
21:03 RabidGravy my naming strategy for modules is now "pick a single word in English that describes it the use google translate to find a pleasing word" :)
21:05 snowman joined #perl6
21:06 RabidGravy so now I know what the Shona for "go-between" is ;-)
21:09 jonadab RabidGravy: And if you translate that from Shona to Japanese, and the result from Japanese to Spanish, then from Spanish to Finnish, then Finnish back to English, is it still "go-between"?
21:09 RabidGravy Oooh
21:10 RabidGravy well it ends up as "envoy" after a few iterations ;-)
21:11 jonadab That's actually not too bad.  Still same idea.
21:13 dalek doc: 939f369 | (Zoffix Znet)++ | META.info:
21:13 dalek doc: Add mandatory "perl" META field
21:13 dalek doc:
21:13 dalek doc: The `perl` field specifies the minimal perl version for which this distribution can be installed and is a mandatory field. The value of `"6.*"` indicates any version suffices.
21:13 dalek doc:
21:13 dalek doc: It is recommended to use [Test::META](https://modules.perl6.org/repo/Test::META) module as an author test, to catch any issues with the META file.
21:13 dalek doc: review: https://github.com/perl6/doc/commit/939f369b9f
21:19 geraud joined #perl6
21:26 eliasr joined #perl6
21:30 rurban joined #perl6
21:38 Herby__ joined #perl6
21:41 pdcawley joined #perl6
21:41 Herby__ o/
21:41 Zoffix \o
21:42 Herby__ Anyone here use Vim and vim-perl for perl6?  I'm a vim novice and am having a hard time having the plugin recognize .p6 or .pm6 files for correct syntax highlioghting
21:42 Herby__ I have vim-perl plugin installed with Vim Plug
21:42 Herby__ the plugin works properly for .pl files, not so much for .p6 or .pm6
21:42 Herby__ kind of stuck at this point
21:42 ggoebel joined #perl6
21:45 timotimo Herby__: i have a key binding to :set ft=perl6
21:46 timotimo it's <Leader>p6 where i set my leader to ,
21:46 Herby__ hmm
21:47 Herby__ what file extension do you typically use for a simple perl script? .p6?
21:50 nicq20 joined #perl6
21:50 dha unless something changed while I wasn't looking, I think programs are .p6 and modules are .pm6
21:50 Herby__ k, thats what I thought but then saw this on perltricks: https://perltricks.com/article/194/2015/9/22/​Activating-Perl-6-syntax-highlighting-in-Vim/
21:51 Herby__ where it mentions using .pm6
21:52 nicq20 Herby__: As far as I know, tests are just '*.t'
22:00 dha Huh. .p6 seems to work with my vim. But I've got other syntax things installed besides vim-perl, so maybe that's why.
22:03 Herby__ hmm
22:12 bjz joined #perl6
22:28 stevieb joined #perl6
22:35 bjz joined #perl6
22:40 cyphase joined #perl6
22:50 Zoffix Found the last "missing" dist; it's ugexe's Foo. So we have 2 extra foos and 1 extra tbrowder's dist making up the 3 extra dist discrepancy
22:55 Juerd joined #perl6
22:57 Zoffix w00t... finished sending all 382 PRs to fix missing `perl` in META \o/
22:58 Zoffix rofl "698 contributions on Oct 26, 2016" on https://github.com/zoffixznet
22:58 tbrowder Zoffix: is my dist causing problems?
22:58 Zoffix tbrowder, no, but that dist is not installable
22:59 avuserow Zoffix++ # keeping the ecosystem in good shape
23:01 tbrowder Which dist? AFAIK, I only have one in the ecosystem: Net::IP::Lite
23:02 Zoffix tbrowder, this one: https://github.com/perl6/ecosyste​m/blob/master/META.list#L731-L732
23:02 Zoffix tbrowder, I forget what it is... but I recall it wasn't installable with neither zef nor panda; they weren't able to find stuff.
23:03 Zoffix buggable, eco
23:03 buggable Zoffix, Out of 739 Ecosystem dists, 296 have warnings and 0 have errors. See https://modules.perl6.org/update.log for details
23:03 Zoffix w00t, already below 300. Will go down as people merge PRs
23:03 Zoffix My job here is done
23:03 timotimo whoa, did buggable just learn that?
23:03 timotimo i don't think i've seen this before
23:03 Zoffix timotimo, yeah, it's new.
23:03 * Zoffix flicks the cape
23:03 Zoffix left #perl6
23:04 ugexe his module fetches with zef on my end anyway
23:06 tbrowder ref ecosystem, Zoffix: the way to have two or more versions was discussed and i did as instructed. so it looks like i should forget anything but the master branch, or should i refer to a release branch? or?
23:09 vendethiel joined #perl6
23:11 viki ugexe: tbrowder oh yeah, just tried and works for me to. Maybe the time when it didn't work was on my other box with older zef. And what wasn't working is trying to install a particular version, like zef install 'Net::IP::Lite:ver<0.1.4>', but I've just tried and it worked, so never mind.
23:12 pierre_ joined #perl6
23:12 tbrowder ok, so my two versions of one dist are ok?
23:12 viki tbrowder: yup. Just tried both and this time both worked.
23:13 tbrowder thanks, good to hear
23:14 viki And last time was this, but that was on my other box and I don't even know what versions I have installed on it: https://irclog.perlgeek.de/​perl6/2016-10-12#i_13385804
23:14 skids joined #perl6
23:16 ugexe you would have had to of had a 3 day old or newer zef on that day for it to have worked, so that probably explains why
23:16 viki Ah :)
23:18 * geekosaur 's classically trained brain splats at the "of"
23:20 viki .oO( should of, would of, could of... )
23:21 geekosaur 've
23:22 timotimo i wouldn't've written it that way
23:22 geekosaur ("have" after the guillotine got it)
23:29 skink joined #perl6
23:30 ugexe depending on where you are from "would have had to of" becomes one giant slur-word
23:35 pdcawley joined #perl6
23:36 jast joined #perl6
23:53 fredstorm joined #perl6
23:57 fredstorm left #perl6

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

Perl 6 | Reference Documentation | Rakudo