Perl 6 - the future is here, just unevenly distributed

IRC log for #perl6, 2015-05-30

Perl 6 | Reference Documentation | Rakudo

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

All times shown according to UTC.

Time Nick Message
00:07 AlexDaniel hoelzro: https://rt.perl.org/Public/​Bug/Display.html?id=125285
00:07 AlexDaniel done :)
00:08 larion joined #perl6
00:09 hoelzro nice!
00:15 vendethiel joined #perl6
00:18 panchiniak joined #perl6
00:26 BenGoldberg joined #perl6
00:27 caddar joined #perl6
00:33 dpk is it possible to implement parsers for Python-style indentation-sensitive recursive syntaxes with Perl 6 rules?
00:34 dpk without needing an initial tokenization stage or input preprocessing
00:34 dpk pure PEGs (and CFGs in general) cannot do this
00:34 dpk but looking at Perl 6's possibilities with code assertions and closures it seems it *should* be possible …
00:48 FROGGS joined #perl6
00:56 tony-o_ dpk: you can implement a grammar that surrounds anything indented more than the previous line with { }
01:13 tony-o_ then in your source all you'd need is use PyPerl; or whatever you call it
01:24 FROGGS joined #perl6
01:25 adu joined #perl6
01:25 aborazmeh joined #perl6
01:25 aborazmeh joined #perl6
01:40 rjbs hoelzro: converted a bunch of code to Buf.push!  One weird case, going to try to sort that out now.
01:45 rjbs how up to date is "m:"'s build?
01:45 rjbs m: my $b = Buf.new; $b.push(1)
01:45 camelia rakudo-moar c2a57e: OUTPUT«Cannot call push(Buf: Int); none of these signatures match:␤    (Any:U \SELF: *@values, *%_)␤  in block <unit> at /tmp/lO1erViuT4:1␤␤»
01:45 rjbs Not up to date neough. :-)
01:45 rjbs Why does this happen?
01:45 rjbs ~$ perl6 -e 'my $b = Buf.new; $b.push(1 xx 1)'
01:45 rjbs Method 'shift' not found for invocant of class 'Parcel'
01:46 rjbs I'm guessing because of this in Buf.pm: +        nqp::push_i(@splicees, @values.shift) while @values;
01:46 panchiniak left #perl6
01:46 rjbs This smells like failure: my $b = Buf.new; $b.push( (1 xx 1).list )
01:59 kurahaupo joined #perl6
01:59 vendethiel joined #perl6
02:00 FROGGS joined #perl6
02:19 b2gills @values is modified but is not marked as being a copy: https://github.com/rakudo/rakud​o/blob/nom/src/core/Buf.pm#L282
02:20 bjz joined #perl6
02:31 ilbelkyr joined #perl6
02:34 rmgk_ joined #perl6
02:43 noganex joined #perl6
02:44 tinyblak joined #perl6
03:02 SevenWolf joined #perl6
03:03 japhb .tell hoelzro I think the Tab Completion example in the README.md for https://github.com/hoelzro/p6-linenoise is incorrect: it uses linenoiseHistoryAdd(), but shouldn't it use linenoiseAddCompletion() instead?
03:03 yoleaux japhb: I'll pass your message to hoelzro.
03:09 japhb .tell hoelzro Same issue in the POD at the bottom of lib/Linenoise.pm
03:09 yoleaux japhb: I'll pass your message to hoelzro.
03:14 vendethiel joined #perl6
03:25 vendethiel dpk: yes
03:25 vendethiel dpk: that's what I did with https://github.com/vendethiel/Nephrite.pl​/blob/master/lib/Text/Nephrite/Grammar.pm, based on masak++'s blog post
03:29 dpk hmm, interesting. can it handle arbitrary/mixed indentation styles? e.g. it doesn't force you to use tabs or two or four spaces, and you can use e.g. two spaces at one level and four at another?
03:30 * dpk reads also https://github.com/masak/text-indent​ed/blob/master/lib/Text/Indented.pm
03:30 vendethiel it does!
03:30 tony-o_ dpk: you can build it however you want!
03:30 vendethiel but just looking for initial space is easy
03:30 tony-o_ the actions is the harder part but still doable
03:31 dpk http://strangelyconsistent.or​g/blog/parsing-indented-text is the post you mean, i assume
03:32 tadzik ingy: ^^ this is relevant to your interests
03:33 dpk "But there are some indent jumps that we're not allowed to make. Anything that indents more than one step at a time, basically." — this implies that there's a predetermined size to 'steps'
03:34 vendethiel in this post, yes.
03:34 dpk indeed, that blog post looks like it implements a parser makes you use 4 spaces
03:35 vendethiel but really, $baseIndent ~~ /^(\s+)/; should work to detect it
03:36 * dpk also looks at https://github.com/arnsholt/snake/
03:36 tadzik I like.to.think that I follow masak++ blog, but it seems like every day someone posts something cool that I have not read yet :)
03:37 tadzik Ohh, a cool new post! Wait, 2014?
03:37 dpk hmph, https://github.com/arnsholt/snake/​blob/master/src/Snake/Grammar.nqp tokenizes first (unsurprising, since that's what the Python specification says to do)
03:38 dpk https://github.com/arnsholt/snake/bl​ob/master/src/Snake/Grammar.nqp#L175 is relevant nonetheless
03:43 dpk it's rapidly becoming clear that i don't at all understand how closure state is managed while parsing with grammars
03:45 Sqirrel joined #perl6
03:48 ShimmerFairy joined #perl6
04:24 molaf_ joined #perl6
04:37 FROGGS joined #perl6
04:39 tinyblak joined #perl6
04:54 quester joined #perl6
04:56 FROGGS joined #perl6
05:02 xargs joined #perl6
05:10 kurahaupo1 joined #perl6
05:26 amurf joined #perl6
05:32 bjz joined #perl6
05:39 zostay m: my %h = a => [a => 1], b => [b => 1, b => 2]; %h.perl.say; %h.perl.EVAL;
05:39 camelia rakudo-moar c2a57e: OUTPUT«{:a([:a(1)]), :b([:b(1), :b(2)])}<>␤Type check failed in binding op; expected 'Any' but got 'Mu'␤  in block <unit> at EVAL_0:1␤  in any <unit-outer> at EVAL_0:1␤␤»
05:41 cognominal joined #perl6
05:45 lizmat good *, #perl6!
05:45 lizmat m: {:a([:a(1)])}  # shorter golf of zostay++ problem
05:45 camelia rakudo-moar c2a57e: OUTPUT«Type check failed in binding op; expected 'Any' but got 'Mu'␤  in block <unit> at /tmp/2Ry1j8d1C3:1␤␤»
05:46 lizmat m: {:a([:a(1).])}.perl.say  # this *does* work correctly
05:46 camelia rakudo-moar c2a57e: OUTPUT«5===SORRY!5=== Error while compiling /tmp/1ERHiVXyz5␤Unsupported use of . to concatenate strings; in Perl 6 please use ~␤at /tmp/1ERHiVXyz5:1␤------> 3{:a([:a(1).]7⏏5)}.perl.say  # this *does* work correctl␤»
05:46 lizmat m: {:a([:a(1),])}.perl.say  # oops, I meant comma
05:46 camelia rakudo-moar c2a57e: OUTPUT«{:a([:a(1)])}␤»
05:47 lizmat m: [:a(1)]  # even shorterer golf
05:47 camelia rakudo-moar c2a57e: OUTPUT«Type check failed in binding op; expected 'Any' but got 'Mu'␤  in block <unit> at /tmp/hbNia98HAY:1␤␤»
05:48 lizmat zostay: care to rakudobug this ?   At least, the .perl output should be fixed to allow roundtripping again
05:52 kaare_ joined #perl6
05:58 dalek Heuristic branch merge: pushed 55 commits to rakudo/newio by lizmat
06:01 lolisa joined #perl6
06:20 rindolf joined #perl6
06:28 lizmat m: [:a] # shorterest golf
06:28 camelia rakudo-moar c2a57e: OUTPUT«Type check failed in binding op; expected 'Any' but got 'Mu'␤  in block <unit> at /tmp/9yE6mEqesY:1␤␤»
06:29 lizmat m: [:a,].perl.say   # works
06:29 camelia rakudo-moar c2a57e: OUTPUT«[:a]␤»
06:31 lizmat breakfast&
07:00 Sqirrel joined #perl6
07:00 Kachi joined #perl6
07:16 amurf joined #perl6
07:18 RabidGravy joined #perl6
07:20 FROGGS joined #perl6
07:32 masak good whatever-the-heck-time-of-day-this-is, #perl6
07:33 mohij joined #perl6
07:33 lizmat masak o/
07:33 masak m: say [:a,].elems
07:33 camelia rakudo-moar c2a57e: OUTPUT«1␤»
07:33 masak m: say [:a,][0].WHAT
07:33 camelia rakudo-moar c2a57e: OUTPUT«(Pair)␤»
07:33 masak I guess `[:a]` should be an array of one Pair, too.
07:33 masak std: [:a]
07:33 camelia std 28329a7: OUTPUT«ok 00:00 135m␤»
07:34 masak oh, 's just rakudo
07:34 lizmat m: [a=> 1].perl.say   # that was the intent
07:34 camelia rakudo-moar c2a57e: OUTPUT«[:a(1)]␤»
07:34 lizmat m: [a=> True].perl.say   # closer to the thing
07:34 camelia rakudo-moar c2a57e: OUTPUT«[:a]␤»
07:34 lizmat m: [a=> True].perl.EVAL   # not roundtripping
07:34 camelia rakudo-moar c2a57e: OUTPUT«Type check failed in binding op; expected 'Any' but got 'Mu'␤  in block <unit> at EVAL_0:1␤  in any <unit-outer> at EVAL_0:1␤␤»
07:35 lizmat m: [(:a)].perl.say   # does work
07:35 camelia rakudo-moar c2a57e: OUTPUT«[:a]␤»
07:35 masak uhn.
07:35 masak yeah, something is wrong there.
07:36 masak rakudobug filed yet?
07:36 lizmat not afaik
07:36 * masak submits rakudobug
07:36 lizmat note that the error occurs in from src/gen/m-CORE.setting:23160  (/Users/liz/Github/rakudo.moar/inst​all/share/perl6/runtime/./CORE.sett​ing.moarvm:METAOP_REDUCE_LEFT:17)
07:36 lizmat METAOP_REDUCE_LEFT ???
07:37 lizmat afk for a bit&
07:38 masak METAOP_REDUCE_LEFT could be because someone did [~] somewhere
07:38 darutoko joined #perl6
07:40 rindolf joined #perl6
07:41 diana_olhovik_ joined #perl6
07:49 RabidGravy for $*KERNEL on Linux which of "4.0.4-201.fc21.x86_64" or "#1 SMP Thu May 21 15:58:47 UTC 2015" would you say is the version and which the release?
07:49 RabidGravy currently they are both the latter
07:49 masak lizmat: https://rt.perl.org/Ticket/Display.html?id=125289
07:54 gfldex joined #perl6
07:56 RabidGravy I think the first is closer to what I think of as "version"
08:11 lizmat afk for some more&
08:16 masak gone for the day &
08:23 quester left #perl6
08:28 jgoff joined #perl6
08:40 mohij joined #perl6
08:55 jercos joined #perl6
09:04 lizmat_ joined #perl6
09:19 vendethiel joined #perl6
09:22 cognominal Reading Monitors.pm, I see               build => -> | { Lock.new }
09:22 cognominal what the '|' accomplishes here?
09:23 cognominal in https://github.com/jnthn/oo-monitors​/blob/master/lib/OO/Monitors.pm#L16
09:24 vendethiel cognominal: signature binding
09:25 vendethiel m: sub foo(|c) { say c; }; say foo(1, 2, 3, :a(5));
09:25 camelia rakudo-moar c2a57e: OUTPUT«\(1, 2, 3, :a(5))␤True␤»
09:25 vendethiel m: sub foo(|c) { say c.WHAT.perl; }; say foo(1, 2, 3, :a(5));
09:25 camelia rakudo-moar c2a57e: OUTPUT«Capture␤True␤»
09:25 vendethiel capture*, sorry :)
09:27 cognominal but, in the said example, the pipe char is not in a signature
09:29 cognominal oops, I was confused
09:32 cognominal so difficult to unlearn that signature are not always within parentheses that in marginal cases like this one, parameterless signature, I don't recognize it.
09:32 * vendethiel got used to it with Ruby first
09:32 cognominal learning is difficult, unlearning, even harder :(
09:33 AlexDaniel joined #perl6
09:33 cognominal thx vendethiel++
09:37 cognominal vendethiel, you once said you don't like OO. But with metaOO, it becomes brilliant
09:37 vendethiel cognominal: I still prefer an AST with an eval function :)
10:04 koo6 joined #perl6
10:07 panchiniak joined #perl6
10:08 FROGGS joined #perl6
10:12 zakharyas joined #perl6
10:22 zengargoyle i'd expect $*KERNEL.release to be '4.0.0-1-amd64' and $*KERNEL.version to be '#1 SMP Debian 4.0.2-1 (2015-05-11)' so that they match up with `uname --kernel-release` and `uname --kernel-version`
10:23 zengargoyle though they do seem to be sorta backwards namewise... maybe historical?
10:25 zengargoyle same on Solaris, release -> x.y, version -> some odd string
10:26 zakharyas joined #perl6
10:26 zengargoyle funkily enough $*KERNEL.version seems to be munged into oblivion -> v1.SMP.Debian.4.0.2.1.2015.5.11
10:37 RabidGravy I've just done https://github.com/rakudo/rakudo/pull/434 to fix it slightly so they are at least different
10:37 RabidGravy anyway off out for a bit
10:41 pmurias joined #perl6
10:42 itz2 joined #perl6
10:42 itz2 left #perl6
10:43 itz2 joined #perl6
10:44 telex joined #perl6
10:53 amurf joined #perl6
10:57 spider-mario joined #perl6
11:08 darutoko- joined #perl6
11:25 yqt joined #perl6
11:27 grondilu joined #perl6
12:03 cognominal Somehow, I think that Perl 6 slogan could be  "embrace complexity", being the next conceptual phase after dcomplecting, but go explain that to people that bow to homoiconicty  https://mobile.twitter.com/bo​dil/status/522479098783072256
12:04 Sqirrel joined #perl6
12:16 DrForr Debugger::UI::CommandLine triggers an unhandled exception after a fresh build.
12:22 domidumont joined #perl6
12:27 rindolf joined #perl6
12:28 domidumont joined #perl6
12:50 aborazmeh joined #perl6
12:50 aborazmeh joined #perl6
13:20 timotimo http://blog.printf.net/articles/2015/05/29/a​nnouncing-gittorrent-a-decentralized-github/ - people could like this
13:21 DrForr Ah, someone's around again :) Got an interesting inconsistency in library paths/include statements.
13:22 DrForr (well, use statements, not include.)
13:24 DrForr Let me narrow it down a bit, but running perl6 with multiple -Ilib statements triggers a syntax error in my Readline module that's not present with a single -Ilib argument.
13:27 timotimo uh oh
13:27 timotimo that stuff is a bit new to me, all in all
13:28 timotimo since changes to it have been so recent
13:28 DrForr Okay, so minefield territory.
13:28 DrForr It's also a bit awkward because I'm replacing Debugger::UI::CommandLine.
13:31 Akagi201 joined #perl6
13:36 timotimo "replacing"? :)
13:38 DrForr Well, replacing the prompt mechanism with Readline so we can get history in the debugger.
13:39 timotimo ah
13:40 timotimo i think you can tell perl6-debug-m with a flag what exact UI module it's supposed to look for
13:40 itz joined #perl6
13:41 DrForr I noticed that it had some CLI parameters, but I think the problem I found is more interesting :)
13:42 timotimo probably
13:44 DrForr #125290
13:44 synbot6 Link:  https://rt.perl.org/rt3/Publi​c/Bug/Display.html?id=125290
13:47 DrForr Truth be told, I'm not sure whether it's the include paths, the repeated use of NativeCall (shows up in both the Debugger::UI and Readline modules) or just the paths.
13:48 DrForr It just felt like action at a distance, but in a bad way :/
13:57 Erendis42d joined #perl6
13:57 Erendis42d <3
14:00 Erendis42d left #perl6
14:02 RabidGravy joined #perl6
14:08 dalek rakudo-star-daily: 14906f6 | coke++ | log/ (2 files):
14:08 dalek rakudo-star-daily: today (automated commit)
14:08 dalek rakudo-star-daily: review: https://github.com/coke/rakudo​-star-daily/commit/14906f6049
14:08 dalek perl6-roast-data: 1613a57 | coke++ | / (9 files):
14:08 dalek perl6-roast-data: today (automated commit)
14:08 dalek perl6-roast-data: review: https://github.com/coke/perl6​-roast-data/commit/1613a57487
14:10 raiph joined #perl6
14:31 amurf joined #perl6
14:31 xfix joined #perl6
14:33 llfourn joined #perl6
14:35 dalek doc: 93f4bc1 | paultcochrane++ | lib/Language/variables.pod:
14:35 dalek doc: Document the $*PROGRAM variable
14:35 dalek doc: review: https://github.com/perl6/doc/commit/93f4bc1c0a
14:37 ingy tadzik: thanks
14:39 pecastro joined #perl6
14:42 vendethiel joined #perl6
14:44 llfourn hey I was just messing around with opreator overloading -- I was wondering if it's possible to add an infix operator that worked like '.'. Eg $obj->method instead of $obj.method
14:44 llfourn I can do $obj->"method" but ofc I get error without the ""
14:45 llfourn using a multi sub infix
14:46 skids joined #perl6
14:51 timotimo right, . is not really an operator, because the RHS isn't a "thing"
14:51 timotimo it's not an expression, that is
14:52 timotimo so you'll have to build a slang instead of just adding an operator
14:52 timotimo https://github.com/FROGGS/p6-Slang-Piersing/ - check this for an example
14:52 llfourn ok thanks!
14:54 kst joined #perl6
14:54 Zoffix joined #perl6
14:54 timotimo you're welcome :)
14:54 llfourn so slang = hacking the grammar?
14:55 timotimo that's right
14:55 timotimo you're going to have to know a bit about the structure of Perl6::Grammar, unfortunately
14:56 llfourn well I'll give it a shot
14:56 llfourn cheers
14:57 timotimo good luck! :)
14:58 [ptc] moritz++   # nitpicking :-)
14:59 * grondilu thought rakudo was not quite ready for slangs yet.
14:59 timotimo rakudo is quite ready
14:59 hoelzro rjbs: thanks for trying that out; I'll have a look
14:59 yoleaux 03:03Z <japhb> hoelzro: I think the Tab Completion example in the README.md for https://github.com/hoelzro/p6-linenoise is incorrect: it uses linenoiseHistoryAdd(), but shouldn't it use linenoiseAddCompletion() instead?
14:59 yoleaux 03:09Z <japhb> hoelzro: Same issue in the POD at the bottom of lib/Linenoise.pm
14:59 grondilu is there a simple example of a slang somewhere?
15:00 timotimo froggs' v5 has been properly slangified a few months ago and we have a simple SQL slang and Tuxic, which allows [Tux] to adhere to his coding style properly
15:00 timotimo i think Piersing is a very simple slang that might serve as a good example, no?
15:01 itz joined #perl6
15:01 timotimo um ... wait what
15:01 * grondilu has no idea what Piersing is.
15:01 timotimo allows ? and ! at the end of identifiers
15:01 timotimo froggs forgot to put "Piersing" back into the code when he copy-pasted it from Tuxic
15:01 timotimo https://github.com/FROGGS/p6-Slang-Piersing/co​mmit/d78041daee398ec5f3beb32a615eafa72689c14e
15:01 * grondilu notices the link above https://github.com/FROGGS/p6-Slang-Piersing/
15:02 * grondilu checks it out
15:02 timotimo https://github.com/FROGGS/p6-Slang-Piersing/pull/1 :)
15:03 grondilu so nqp code is required, apparently.
15:03 timotimo no, just nqp ops
15:03 grondilu well, that qualifies at nqp code in my book
15:03 timotimo though "use nqp" kind of sounds like "use the nqp slang" or "this is nqp code"
15:04 timotimo i respectfully disagree
15:04 timotimo but then i come at this with a guts-programmer mindset
15:04 grondilu will it always be necessary to invoke nqp ops to define slangs?
15:04 moritz I hope not
15:05 grondilu then that's what I meant when I wrote that I thought rakudo was not quite ready for slangs.
15:05 moritz in the past, we've provided a way to avoid nqp:: ops in the MOP by having Perl6::Metamodel::Primitives
15:05 moritz maybe we need the same for Perl6::Compiler::Primitives or so
15:06 hoelzro japhb: thanks for pointing that out, fixed!
15:08 timotimo i don't remember the primitives thing
15:08 timotimo where can i find that?
15:10 pecastro joined #perl6
15:11 dalek rakudo/nom: ff727fd | hoelzro++ | src/core/Buf.pm:
15:11 dalek rakudo/nom: Copy incoming @values for Buf.push
15:11 dalek rakudo/nom:
15:11 dalek rakudo/nom: Otherwise this doesn't work properly for Parcels
15:11 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/ff727fdad5
15:11 dalek roast: 011af40 | hoelzro++ | S03-operators/buf.t:
15:11 dalek roast: Add a test for pushing Parcels onto Bufs
15:11 dalek roast:
15:11 dalek roast: rjbs++ for finding this
15:11 dalek roast: review: https://github.com/perl6/roast/commit/011af40830
15:11 dalek roast: 4806305 | hoelzro++ | S03-operators/buf.t:
15:11 dalek roast: Fix bad grammar in Buf.push test descriptions
15:11 dalek roast: review: https://github.com/perl6/roast/commit/4806305917
15:14 hoelzro is jpve still active in the community? Pod::To::Markdown has some opening PRs to address some newer warnings
15:16 moritz try shooting 'em an email?
15:16 moritz if there's no reaction, or a negative reaction, we can clone into the perl6-community-modules org
15:17 hoelzro moritz: the PRs should send jpve an email, right? I figure we can clone if they PRs aren't addressed for quite some time
15:17 hoelzro maybe another few months
15:20 timotimo the PRs will only send an email if it's not deactivated in the account, or the repository is muted ... can you mute your own repositories?
15:20 moritz hoelzro, timotimo: iirc one can disable email notifications from github altogether
15:20 hoelzro ah, good point
15:20 hoelzro timotimo: I think you unwatch it
15:21 moritz and a mail from a human asking about several pull requests is quite a different thing than a few automated mails
15:22 timotimo aye
15:22 timotimo on the other hand, mails that are automated, but made to look like they're from a person ... those are aggravating :S
15:24 JimmyZ hoelzro: any reason why not https://github.com/MoarVM/linenoise?
15:25 hoelzro JimmyZ: ignorance of the existence of that repo, mainly =)
15:25 JimmyZ hoelzro: which supports windows :)
15:25 JimmyZ ha
15:28 timotimo but MoarVM/linenoise is just a fork of the upstream linenoise repository with a tiny amount of patches, if any
15:28 timotimo AFK for a bit
15:30 llfourn can I do something like 'use STD' so I can do STD.parse( 'my $foo = "bar"' )
15:30 llfourn for STD.p6
15:31 llfourn STD.pm6*
15:31 llfourn im trying to sort of figure out what it calls each of the things so I can make a slang
15:33 hoelzro it's quite difficult to find this guy's e-mail
15:34 JimmyZ timotimo: yeah, for supporting windows
15:34 hoelzro this is making me wish GH hadn't gotten rid of messaging =/
15:37 mohij joined #perl6
15:47 brrt joined #perl6
15:48 hoelzro I just came across a segfault when installing JSON::Tiny with rakudo from git; does anyone else see this?
15:50 DrForr Yep, just confirmed with 'panda install JSON::Tiny'.
15:51 caddar joined #perl6
15:52 hoelzro =/
15:53 hoelzro I'm going to dig in, see if this happens with 2015.05 and 2015.04
15:54 hoelzro ok, doesn't happen with 2015.05
15:55 timotimo oh
15:55 timotimo ... oh god, what have i done ...
15:58 llfourn which perl6
15:58 llfourn oops ^^ nvm that
16:00 btyler joined #perl6
16:06 hoelzro ok, just reproduced it on 2015.05 =o_O
16:06 hoelzro it just didn't happen with rakudobrew'd 2015.05?
16:06 hoelzro this is getting weird
16:07 kaare_ joined #perl6
16:11 AlexDaniel Hm, I'm wondering: if you can change perl6 syntax so easily, how the hell text editors are supposed to do syntax highlighting and other stuff?
16:11 llfourn ==> Successfully installed JSON::Tiny with my rakudobrew 2015.05
16:12 vendethiel joined #perl6
16:12 timotimo AlexDaniel: there's always the problem of trust, but the syntax highlighter could use a perl6 grammar and load the slang as well; sounds very flaky, though, doesn't it?
16:13 llfourn speaking of loading the perl6 grammar -- can you actually do that atm with rakudo? ie use Perl6::Grammar; Perl6::Grammar.parse(...)
16:14 llfourn I've been trying to figure this out for last 20 mins :)
16:18 moritz I don't think it's possible (or easy)
16:19 moritz https://github.com/cedric-vince​nt/kains/blob/master/META.info is it normal/correct to not have a 'source-url' or 'repo-url', but rather a 'support'/'url'?
16:20 cognominal joined #perl6
16:20 timotimo you can nqp::getcomp('Perl6').parse( ... )
16:20 timotimo with getcomp you get the Perl6::Compiler object
16:21 llfourn moritz: hmm shame -- i thought my Mu $MAIN-grammar := nqp::atkey(%*LANG, 'MAIN'); make give aceess to it
16:21 timotimo hm, it might
16:22 llfourn nqp::getcomp('Perl6').parse( 'my $a = "foo";' ); Cannot call method 'parse' on a null object
16:22 kaare__ joined #perl6
16:23 moritz try it with lower case?
16:23 moritz llfourn: I don't know to which degree the grammar depends on the actions being there
16:23 moritz m: use nqp; nqp::getcomp('perl6').parse( 'my $a = "foo";' );
16:23 camelia rakudo-moar c2a57e: ( no output )
16:23 moritz m: use nqp; say nqp::getcomp('perl6').parse( 'my $a = "foo";' );
16:23 camelia rakudo-moar c2a57e: OUTPUT«Method 'gist' not found for invocant of class 'NQPMatch'␤  in block <unit> at /tmp/oRUgM93U_b:1␤␤»
16:24 mr-foobar joined #perl6
16:26 llfourn cool thanks. I wonder how easy NQPMatch can be used.
16:28 FROGGS joined #perl6
16:29 llfourn not easily I guess. Ok I'll scratch the parse perl6 with perl6 to figure out what tokens to override to make a slang.
16:29 llfourn for now :)
16:36 hoelzro ok, this is just wacky
16:37 hoelzro rakudobuild build moar nom and panda install JSON::Tiny works
16:39 tony-o_ llfourn: what kind of slang?
16:40 llfourn tony-o_: I was simply trying to add -> operator which to start off with was just going to be the same as '.' method call
16:41 tony-o_ add that operator to what?
16:41 tony-o_ or to mean what?
16:41 llfourn like $obj->method as well as $obj.method
16:41 tony-o_ ah!, that should be fun :-)
16:41 llfourn I hope so :)
16:42 tony-o_ there are a couples slangs out there now that can be used as examples
16:42 tony-o_ slang::tuxic
16:42 tony-o_ slang::sql
16:43 tony-o_ slang::piersing is out there now too, not sure what the source for that one looks like
16:43 llfourn https://github.com/FROGGS/p6-Slang-Pier​sing/blob/master/lib/Slang/Piersing.pm
16:45 tony-o_ yea i mean i haven't studied it, i wrote slang-sql, i read slang-tuxic quite a bit
16:45 llfourn hmmm thanks
16:45 llfourn yeah methodop might be the one I need to look at
16:50 llfourn wait... maybe I can just do autoload so $obj->meth doesn't give undecl routine and just return string from there
16:51 rjbs general #perl6 reminder that you can throw questions at me via email or whatever for https://github.com/rjbs/QA-LarryWall/
16:51 rjbs I assume it'll be YouTubed.
17:00 araujo joined #perl6
17:07 itz2 joined #perl6
17:10 llfourn well I got the $obj->meth to be valid perl6 -- now I need to make it do something by overriding an action I think
17:13 dalek rakudo/nom: 57a959f | RabidGravy++ | src/core/Kernel.pm:
17:13 dalek rakudo/nom: On Linux use the counterintuitive -r for $*KERNEL.version
17:13 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/57a959fe7d
17:13 dalek rakudo/nom: 335c1eb | lizmat++ | src/core/Kernel.pm:
17:13 dalek rakudo/nom: Merge pull request #434 from jonathanstowe/nom
17:13 dalek rakudo/nom:
17:13 dalek rakudo/nom: On Linux use the counterintuitive -r for $*KERNEL.version
17:13 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/335c1eb0a7
17:14 gfldex joined #perl6
17:16 Peter_R joined #perl6
17:18 colomon joined #perl6
17:19 jercos joined #perl6
17:19 vendethiel joined #perl6
17:25 yqt joined #perl6
17:32 tinyblak joined #perl6
17:34 lizmat_ .tell hoelzro Looking at Buf.push(*@a), why isn't that just nqp::push_i(self,$_) for @values ?
17:34 yoleaux lizmat_: I'll pass your message to hoelzro.
17:35 itz2 joined #perl6
17:40 AlexDaniel joined #perl6
18:00 dalek rakudo/nom: e9fa8d9 | lizmat++ | src/core/Buf.pm:
18:00 dalek rakudo/nom: Add sanity check for Infinite lists to Buf.push
18:00 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/e9fa8d9d3d
18:08 * [Coke] realizes he can add "&& say done" when doing long build stuff on his mac.
18:09 amurf joined #perl6
18:10 masak ooh, nice
18:10 masak is there something simliar on Linux?
18:11 domidumont joined #perl6
18:13 raydiak sudo apt-get install gnustep-gui-runtime gives you say
18:13 itz2 joined #perl6
18:13 raydiak and most of the rest of gnustep, I'm sure
18:20 [Coke] might be worth adding an env var to my shell that will say something in a shell when I get the prompt back.
18:21 dalek rakudo/nom: f7fb908 | Coke++ | t/spectest.data:
18:21 dalek rakudo/nom: Run this already passing test.
18:21 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/f7fb908698
18:22 llfourn tony-o_: I did it! I got $obj~>method to work the same as $obj.method
18:23 llfourn your SQL one really helped
18:24 vendethiel joined #perl6
18:26 mohij joined #perl6
18:29 larion joined #perl6
18:30 adu joined #perl6
18:30 dalek roast: b811a84 | Coke++ | S05-match/perl.t:
18:30 dalek roast: Fudge for rakudo
18:30 dalek roast: review: https://github.com/perl6/roast/commit/b811a8460a
18:31 mr-foobar joined #perl6
18:31 dalek rakudo/nom: 66f6258 | Coke++ | t/spectest.data:
18:31 dalek rakudo/nom: run fudged test
18:31 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/66f6258863
18:33 [Coke] hoelzro: S26-documentation/block-leading.t isn't being run - there's some failures in there that are tricky to fudge. Can you take a look?
18:33 [Coke] (I ping you since it's got your name on it.)
18:34 [Coke] jnthn: hope your weekend is going well!
18:37 [Coke] is pepe schwarz aroudn?
18:39 dalek roast: 3ae1a36 | Coke++ | S19-command-line-options/06-deprecated.t:
18:39 dalek roast: Per design specs, -E isn't in perl6.
18:39 dalek roast:
18:39 dalek roast: We only issue deprecation warnings for things that existed in 6 and are going
18:39 dalek roast: to be removed.
18:39 dalek roast: review: https://github.com/perl6/roast/commit/3ae1a3623e
18:39 [Coke] ^^
18:40 dalek rakudo/nom: 6fcbe05 | Coke++ | t/spectest.data:
18:40 dalek rakudo/nom: remove reference to non-existant test file
18:40 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/6fcbe05d93
18:47 timotimo [Coke]: "Removed file this was in" - this looks like yodaspeak at first glance :D
18:48 moritz if the force so strong is in master yoda, why proper English he can't speak?
18:53 hoelzro lizmat_: I don't have a good reason, tbh; we can change that
18:53 yoleaux 17:34Z <lizmat_> hoelzro: Looking at Buf.push(*@a), why isn't that just nqp::push_i(self,$_) for @values ?
18:53 hoelzro [Coke]: sure, I'll have a look
18:53 timotimo that could give bad performance because of GLR not-there-ness
18:55 hoelzro timotimo: are you talking about lizmat's suggestion?
18:56 timotimo yeah
19:01 spider-mario joined #perl6
19:02 hoelzro this segfault when one installs JSON::Tiny is driving me nuts
19:02 hoelzro oh, and of course when I run MoarVM through gdb, there's no problem!
19:04 hoelzro hmm, it might just be the repeated running of panda install
19:07 yqt joined #perl6
19:08 pecastro joined #perl6
19:10 zakharyas joined #perl6
19:13 tinyblak joined #perl6
19:16 pecastro joined #perl6
19:20 RabidGravy for the last week or so panda smoke has segfaulted about 2/3 of the way through
19:21 RabidGravy I'm just putting it down to one of those things
19:21 hoelzro RabidGravy: really? I'm guessing that's the issue =(
19:22 FROGGS joined #perl6
19:22 RabidGravy memory probably
19:26 andreoss joined #perl6
19:29 zakharyas joined #perl6
19:44 dalek ecosystem: 868f429 | (Sterling Hanenkamp)++ | META.list:
19:44 dalek ecosystem: Add Hash::MultiValue to the Perl 6 ecosystem
19:44 dalek ecosystem: review: https://github.com/perl6/e​cosystem/commit/868f4299a8
19:45 bin_005 joined #perl6
19:55 vendethiel joined #perl6
19:57 kurahaupo1 joined #perl6
20:34 dalek roast: 89bdd71 | Coke++ | S29-context/evalfile.t:
20:34 dalek roast: Fudge test, add RT.
20:34 dalek roast: review: https://github.com/perl6/roast/commit/89bdd71db5
20:34 dalek rakudo/nom: 8acfd7f | Coke++ | t/spectest.data:
20:34 dalek rakudo/nom: run fudged test
20:34 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/8acfd7f15a
20:39 dalek rakudo/nom: 75605a5 | Coke++ | t/spectest.data:
20:39 dalek rakudo/nom: Run this test that is passing.
20:39 dalek rakudo/nom:
20:39 dalek rakudo/nom: not sure why it was skipped for a release.
20:39 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/75605a5def
20:39 koo6 joined #perl6
20:49 itz2 joined #perl6
20:50 [Coke] S05-nonstrings/basic.t is really old and needs cleanup to modern sixperl.
20:54 dalek roast: 86ed3a6 | Coke++ | S24-testing/4-version_lt.t:
20:54 dalek roast: remove old test file from mu.
20:54 dalek roast:
20:54 dalek roast: This test function no longer exists.
20:54 dalek roast: review: https://github.com/perl6/roast/commit/86ed3a69fb
20:54 dalek rakudo/nom: fffe76c | Coke++ | t/spectest.data:
20:54 dalek rakudo/nom: remove ref to nonexistent file
20:54 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/fffe76ccc7
20:55 dalek rakudo-star-daily: 87cdeef | coke++ | log/ (2 files):
20:55 dalek rakudo-star-daily: today (automated commit)
20:55 dalek rakudo-star-daily: review: https://github.com/coke/rakudo​-star-daily/commit/87cdeefd8a
20:55 dalek perl6-roast-data: fcca801 | coke++ | / (9 files):
20:55 dalek perl6-roast-data: today (automated commit)
20:55 dalek perl6-roast-data: review: https://github.com/coke/perl6​-roast-data/commit/fcca801c46
20:59 lizmat joined #perl6
21:03 andreoss joined #perl6
21:04 lizmat_ joined #perl6
21:11 larion joined #perl6
21:16 Possum joined #perl6
21:30 RabidGravy boo!
21:30 telex joined #perl6
21:31 timotimo boo?
21:32 masak 不!
21:32 masak 'night, #perl6
21:33 lizmat 'night masak
21:36 bin_005 joined #perl6
21:46 amurf joined #perl6
21:47 zakharyas joined #perl6
21:51 vendethiel joined #perl6
22:01 kurahaupo1 joined #perl6
22:04 [Coke] m: (12 X-3,4,6 X-2,3,6).sort.unique
22:04 camelia rakudo-moar c2a57e: ( no output )
22:04 [Coke] m: (12 X-3,4,6 X-2,3,6).sort.unique.say
22:04 camelia rakudo-moar c2a57e: OUTPUT«0 2 3 4 5 6 7␤»
22:04 [Coke] huh. that got me a parcel error locally. must be old.
22:18 dalek rakudo/nom: 5e006b9 | lizmat++ | src/Perl6/Actions.nqp:
22:18 dalek rakudo/nom: Handle $*OS/$*OSVER deprecation at compile time
22:18 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/5e006b9a94
22:18 dalek rakudo/nom: 27499bb | lizmat++ | src/core/ (4 files):
22:18 dalek rakudo/nom: Remove Obsolete class hack
22:18 dalek rakudo/nom:
22:18 dalek rakudo/nom: Since $*OS / $*OSVER are now handled at compile time, we no longer
22:18 dalek rakudo/nom: need the runtime Obsolete class hack.
22:18 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/27499bbdbf
22:19 jferrero joined #perl6
22:19 * vendethiel waves
22:20 * lizmat waves back while wishing everybody good night!
22:22 mattp__ joined #perl6
22:22 jferrero joined #perl6
22:25 jferrero joined #perl6
22:28 jferrero joined #perl6
22:50 TimToady joined #perl6
22:54 bin_005 joined #perl6
22:57 panchiniak joined #perl6
22:59 panchiniak left #perl6
23:01 zeleiadi joined #perl6
23:16 zeleiadi left #perl6
23:29 flussence Re: the thing masak said about 7h ago - most Linux desktop distros give you a notify-send command, not exactly the same mechanism as say but gets you similar results.

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

Perl 6 | Reference Documentation | Rakudo