Perl 6 - the future is here, just unevenly distributed

IRC log for #perl6-dev, 2016-07-18

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

All times shown according to UTC.

Time Nick Message
00:27 tadzik joined #perl6-dev
01:27 dalek roast: 7f9045e | (Zoffix Znet)++ | S03-operators/precedence.t:
01:27 dalek roast: [TODO FUDGE] exported multi has correct precedence
01:27 dalek roast:
01:27 dalek roast: RT#128042
01:27 dalek roast: review: https://github.com/perl6/roast/commit/7f9045eb13
01:27 dalek roast: 2d4d315 | (Zoffix Znet)++ | S03-operators/precedence.t:
01:27 synopsebot6 Link:  https://rt.perl.org/rt3//Public/Bug/Display.html?id=128042
01:27 dalek roast: Remove trailing whitespace
01:27 dalek roast: review: https://github.com/perl6/roast/commit/2d4d315d3e
01:46 dalek roast: e1e9617 | (Zoffix Znet)++ | S02-types/array.t:
01:46 dalek roast: [TODO FUDGE] [*-0] index references correct element when autovivifying
01:46 dalek roast:
01:46 dalek roast: RT#128005
01:46 dalek roast: review: https://github.com/perl6/roast/commit/e1e96176cd
01:46 synopsebot6 Link:  https://rt.perl.org/rt3//Public/Bug/Display.html?id=128005
01:46 dalek roast: f39436f | (Zoffix Znet)++ | S02-types/array.t:
01:46 dalek roast: Remove trailing whitespace
01:46 dalek roast: review: https://github.com/perl6/roast/commit/f39436f2a3
01:48 ilbot3 joined #perl6-dev
01:48 Topic for #perl6-dev is now Perl 6 language and compiler development 2.0 | Logs at http://irclog.perlgeek.de/perl6-dev/today
02:15 * Zoffix lols
02:15 Zoffix Why do we have method "fap" in IO::Path? XD
02:16 mst Scala: fmap fmap fmap boom
02:16 mst perl6: fap fap fap spurt
02:16 camelia rakudo-jvm cd19db: OUTPUT«===SORRY!=== Error while compiling <tmp>␤Undeclared routine:␤    fap used at line 1. Did you mean 'map'?␤␤»
02:16 camelia ..rakudo-moar 242baf: OUTPUT«===SORRY!=== Error while compiling <tmp>␤Undeclared routine:␤    fap used at line 1. Did you mean 'map'?␤␤»
02:16 mst (with apologies to sexylosers.com)
02:16 mst (also the bots ;)
02:16 Zoffix :)
02:55 gfldex it's all good. It's a private method.
02:56 gfldex and the name makes perfect sense :D
02:58 Zoffix S17-procasync/no-runaway-file-limit.t hangs on my box :/ occasionally
03:04 dalek rakudo/nom: 9d3b6d0 | (Zoffix Znet)++ | src/core/IO/Path.pm:
03:04 dalek rakudo/nom: Improve IO::Path.new error message when given wrong arguments
03:04 dalek rakudo/nom:
03:04 dalek rakudo/nom: It can take positionals, so its error message must not claim it doesn't
03:04 dalek rakudo/nom:
03:04 dalek rakudo/nom: Fixes RT#127989
03:04 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/9d3b6d0731
03:04 synopsebot6 Link:  https://rt.perl.org/rt3//Public/Bug/Display.html?id=127989
03:04 dalek roast: 5f54695 | (Zoffix Znet)++ | S32-io/io-path.t:
03:04 dalek roast: IO::Path.new with wrong args must not claim it only takes named ones
03:04 dalek roast:
03:04 dalek roast: RT#127989
03:05 Zoffix dalek, bruh... Where's the link to roast commit! What are we paying you for?
03:39 AlexDaniel joined #perl6-dev
04:56 dalek star: 50b6e24 | (Naoum Hankache)++ | docs/perl6intro.pdf:
04:56 dalek star: Update perl6intro.pdf
04:56 dalek star: review: https://github.com/rakudo/star/commit/50b6e244e7
04:56 dalek star: 3946ee2 | moritz++ | docs/perl6intro.pdf:
04:56 dalek star: Merge pull request #72 from hankache/master
04:56 dalek star:
04:56 dalek star: Update perl6intro.pdf
04:56 dalek star: review: https://github.com/rakudo/star/commit/3946ee2506
05:54 RabidGravy joined #perl6-dev
06:30 ugexe joined #perl6-dev
07:39 dalek rakudo/nom: 8aaffa8 | lizmat++ | src/core/ (2 files):
07:39 dalek rakudo/nom: Make Str.indices about 10% faster
07:39 dalek rakudo/nom:
07:39 dalek rakudo/nom: - by giving it a dedicated Str candidate, and have Cool refer to it
07:39 dalek rakudo/nom: - also does 2 allocations less per call, so more GC friendly
07:39 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/8aaffa8b12
08:45 sno joined #perl6-dev
10:09 psch r: sub f { return sub g { say $++ } }; f()(); f()();f()()
10:09 camelia rakudo-moar 8aaffa, rakudo-jvm cd19db: OUTPUT«0␤0␤0␤»
10:09 psch which means moar and jvm somehow differentiate between Blocks and Routines in a way they shouldn't
10:10 psch 'cause i'm inclined to agree with raiph++'s interpretation, each time the loop Block is hit it's effectively a new declaration
10:13 jnthn To moar they're all just frames, it doesn't care for routine vs block. That's a higher-level thing.
10:13 timotimo i think the difference is with using nqp::clone or not on the block before invoking it?
10:14 psch right, "jvm - in contrast to moar - " or so would've been the better phrasing
10:14 psch r: sub f { loop { say $++; last } }; f;f;f;f;
10:14 camelia rakudo-moar 8aaffa: OUTPUT«0␤0␤0␤0␤»
10:14 camelia ..rakudo-jvm cd19db: OUTPUT«0␤1␤2␤3␤»
10:14 psch jnthn: just to verify, moar is correct there?
10:18 jnthn Well, Moar's output is consistent with how I understand things to work, whereas I'm a bit confused how the JVM output happens :)
10:18 psch aw, shucks, and here i was hoping for a language design decision :)
10:18 jnthn Then, we know that the JVM lexical scope handling has some inconsistencies vs. on Moar (which is the source of some bugs)
10:19 jnthn Well, I can see the lang design argument for wanting it the other way, per the "when is a block a closure"...trouble is that the moment you make a loop lazy then the block really *does* become a closure, so... :)
10:20 jnthn I'd be interested to hear TimToady's take.
10:20 jnthn 'cus while I suspect it'll come down on the "torture the implementer" side, I'm not sure where we draw the line exactly :)
10:20 psch well, we do have the jvm backend :P
10:21 psch anyway, yeah.  i suppose i'll ticket the difference, link the ticket i know is related and do something else
10:38 Zoffix stmuk, yeah, I can cut a 2016.07.01 release in 13 hours.
10:38 Zoffix (to everyone else) It's fine to do that, right? We don't need to sacrifice anything to appease some god?
10:38 dalek rakudo/nom: 328c2d6 | lizmat++ | src/core/Str.pm:
10:38 dalek rakudo/nom: Make Str.samecase 5x as fast
10:38 dalek rakudo/nom:
10:38 dalek rakudo/nom: - completely re-imagine in nqp
10:38 dalek rakudo/nom: - with smarter way to deal with lc/uc/non-alpha
10:38 dalek rakudo/nom: - with way fewer allocations, so easier on the GC
10:38 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/328c2d6d80
10:39 DrForr 13 hours? Lucky you... oh, wait :)
10:39 lizmat Zoffix: not against another release, but what are we fixing ?
10:39 Zoffix Reasoning is because currently 2016.07 apparently has an issue with DESTDIR for packagers, which was fixed in HEAD. And we want to cut a R* release this month: https://rt.perl.org/Ticket/Display.html?id=128652
10:40 lizmat Zoffix: perhaps a new release should also include 242baf256a210f1cb643cb63
10:41 Zoffix sure
10:41 lizmat unless jnthn has another idea on fixing RT #128655
10:41 synopsebot6 Link:  https://rt.perl.org/rt3//Public/Bug/Display.html?id=128655
10:43 jnthn Well, debug the optimizer and figure out why it seems to be picking the wrong candidate, maybe... :)
10:44 stmuk Zoffix++
10:44 yoleaux2 17 Jul 2016 19:00Z <Zoffix> stmuk: not sure if it affects R* release, but FYI the 2016.07 release seems to have issues with custom DESTDIR that I think are fixed in HEAD: https://rt.perl.org/Ticket/Display.html?id=128652
10:45 lizmat jnthn: a more focused suggestion than "debug the optimizer" would be appreciated  :-)
10:46 jnthn lizmat: Well, it could either be a static inlining failure, or a failure in the compile-time multi dispatch resolution.
10:47 lizmat oddity: Str.samecase("") just returns self, whereas Str.samemark("") returns a Failure (and roast expects this both these behaviours)
10:47 jnthn lizmat: Could try to dump out the signature of the candidate it chooses if some env var is turned on, to see which side the problem is on, I guess
10:47 lizmat this feels inconsistent to me
10:50 Zoffix r: sub f { state $x; for ^10 { state $y; say [$x++, $y++]; last if $_ > 1 } }; f;f;f;f;
10:50 camelia rakudo-moar 8aaffa, rakudo-jvm cd19db: OUTPUT«[0 0]␤[1 1]␤[2 2]␤[3 0]␤[4 1]␤[5 2]␤[6 0]␤[7 1]␤[8 2]␤[9 0]␤[10 1]␤[11 2]␤»
10:51 Zoffix jnthn, is this also fine on MoarVM? I'd expect state to keep the $y around. It's what Perl 5 does too
10:52 Zoffix huh... jvm had different output last night
10:53 Zoffix (Perl 5's output: https://gist.github.com/zoffixznet/9f5010d69b6ccc41024c9a4eb50e75b0 )
10:54 jnthn Zoffix: for de-sugars to map, the block is just a closure passed to map, and state varaibles are per closure
10:54 jnthn Zoffix: In Perl 5, "for de-sugars to map" is not the case, afaik
10:55 jnthn Which would account for the discrepancy
10:55 Zoffix I see
10:59 jnthn m: sub f() { do for ^10 { ++(state $x) } }; my @a = lazy f; my @b = lazy f; say @a[^5]; say @b[^5];
10:59 camelia rakudo-moar 328c2d: OUTPUT«(1 2 3 4 5)␤(1 2 3 4 5)␤»
11:00 jnthn If we were to make it magically scope the state variable's lifetime to the routine, we'd break this.
11:21 Frameless yeah, makes sense to me now. raiph explained it last night too, but it didn't click for me then. Mostly 'cause I confused myself thinking it was /last/next that reset state in the examples, but I see it's not actually the case now
11:40 dalek rakudo/nom: 9c796f2 | (David Warring)++ | src/core/ (2 files):
11:40 dalek rakudo/nom: fix Blob/bytecomp SHA-1 checksums
11:40 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/9c796f2e88
11:40 dalek rakudo/nom: 21cfe2a | niner++ | src/core/ (2 files):
11:40 dalek rakudo/nom: Merge pull request #811 from dwarring/sha1-checksums
11:40 dalek rakudo/nom:
11:40 dalek rakudo/nom: fix Blob/precomp SHA-1 check-sums
11:40 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/21cfe2a8a2
11:48 tbrowder joined #perl6-dev
11:50 tbrowder n: my $s := ''; my $b = nqp::isstr($s); say($b);
11:50 camelia niecza : OUTPUT«No such file or directory»
11:51 camelia niecza : OUTPUT«(timeout)cat: /home/camelia/niecza/VERSION: No such file or directory␤»
11:52 tbrowder n: my $s := ''; my $b := nqp::isstr($s); my say($b);
11:52 camelia niecza : OUTPUT«No such file or directory»
11:52 tbrowder n: my $s := ''; my $b := nqp::isstr($s); say($b);
11:52 camelia niecza : OUTPUT«No such file or directory»
11:52 camelia niecza : OUTPUT«(timeout)cat: /home/camelia/niecza/VERSION: No such file or directory␤#perl6-dev <tbrowder> niecza: my $s := ''; my $b := nqp::isstr($s); say($b);␤»
11:53 psch tbrowder: i don't think you can convince camelia to fix niecza :)
11:54 tbrowder hm, i thought there was way to use nqp here like rakudo
11:54 psch nqp: my $s := ''; my $b := nqp::isstr($s); say($b);
11:54 camelia nqp-moarvm: OUTPUT«1␤»
11:54 psch camelia: help
11:54 camelia psch: Usage: <(rakudo-jvm|p5-to-p6|nqp-js|nqp-moarvm|star-m|nqp-parrot|star-j|std|debug-cat|prof-m|rakudo-MOAR|niecza|rakudo-moar|nqp-jvm|pugs|rPn|rm|r-jvm|sj|nPr|nrP|n|r|perl6|j|p56|nqp-q|Pnr|nom|star|rn|rj|P|Prn|sm|nqp|r-j|rnP|m|nqp-m|nr|r-m|nqp-mvm|p6|M|rakudo)(?^::\s(?!OUTPUT)) $perl6_program>
11:54 psch ...should probably prune that somewhat eventually
11:55 tbrowder psch: thanks!
11:55 tbrowder nqp-moarvm: my $s := ''; my $b := nqp::isstr($s); say($b);
11:55 camelia nqp-moarvm: OUTPUT«1␤»
11:56 tbrowder cool--as expected
12:02 Frameless p5-to-p6: for (1..10) { say $_ }
12:02 camelia p5-to-p6 : OUTPUT«# Do not edit this file - Generated by Perlito5 9.021␤␤for 1 .. 10 -> $_ {␤    say($_)␤}␤»
12:10 Frameless [2016.07.01 release] I lied about in 13 hours. I'll start doing it now
12:11 timotimo oh wow
12:11 timotimo that's quite a bit earlier
12:11 pmurias joined #perl6-dev
12:11 Frameless Why? Is that a problem?
12:12 dalek nqp: a9aaaf9 | (Pawel Murias)++ | src/vm/js/nqp-runtime/package.json:
12:12 dalek nqp: [js] Update the npm package issue tracker url.
12:12 dalek nqp: review: https://github.com/perl6/nqp/commit/a9aaaf938f
12:12 dalek nqp: e8e0146 | (Pawel Murias)++ | src/vm/js/HLL/Backend.nqp:
12:12 dalek nqp: [js] Make run_profiled use v8-profiler.
12:12 dalek nqp:
12:12 dalek nqp: This should make --profile and --profile-compile actually write CPU profiling info.
12:12 dalek nqp: review: https://github.com/perl6/nqp/commit/e8e0146ab0
12:12 dalek nqp: 87733e5 | (Pawel Murias)++ | src/vm/js/HLL/Backend.nqp:
12:12 dalek nqp: [js] Delete the CPU profile after it has been writen.
12:12 dalek nqp: review: https://github.com/perl6/nqp/commit/87733e5553
12:14 timotimo not a problem at all!
12:14 timotimo go right ahead :)
12:14 Frameless \o/
12:16 perlpilot joined #perl6-dev
12:22 Frameless Another ticket where I've seen a reply on the mailing list, but the reply did not make it to the ticket itself (and potentially neither to the reporter) https://rt.perl.org/Ticket/Display.html?id=128632
12:37 travis-ci joined #perl6-dev
12:37 travis-ci Rakudo build errored. niner 'Merge pull request #811 from dwarring/sha1-checksums
12:37 travis-ci https://travis-ci.org/rakudo/rakudo/builds/145516459 https://github.com/rakudo/rakudo/compare/328c2d6d80d6...21cfe2a8a27c
12:37 travis-ci left #perl6-dev
12:39 dalek rakudo/nom: c07b2d7 | (Zoffix Znet)++ | docs/ChangeLog:
12:39 dalek rakudo/nom: Populate ChangeLog for 2016.07.01 release
12:39 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/c07b2d7903
12:39 timotimo ^- travis failure seems meaningless
12:41 pmurias_ joined #perl6-dev
12:42 Frameless Zoffix-- # forgot to git push changes to the Rakudo Rele6sr app :(
12:47 perlpilot What is the "Rakudo Rele6sr app"?
12:48 Frameless perlpilot: a helper app to release Rakudo: http://i.imgur.com/eQJ7S8o.png    https://github.com/zoffixznet/Rele6sr
12:49 Frameless perlpilot: right now, aside from the RT queue reviewer, not much more than a set of instructions you can check off as completed, but I plan to automate as much stuff as possible. Including spinning up Google Compute Engine instances to do stresstesting and stuff :)
12:49 perlpilot very neat
12:50 nine Frameless: FWIW the point release could have been avoided if the release guide included a make DESTDIR test
12:50 gfldex lizmat: that was the week that was for the docs https://gist.github.com/gfldex/9acb0265491702c5eecf31ad8766ef3a
12:50 nine Of course, once we add such a test, it probably won't ever be needed anymore...
12:51 lizmat gfldex++
12:51 Frameless nine: I actually did a test by running make DESTDIR and it seemed to work fine on my Debian ¯\_(ツ)_/¯
12:51 lizmat gfldex: "2 new traps" , what does that mean ?
12:52 Frameless I'll look into cooking up some sort of automated test for that area this month
12:52 gfldex https://docs.perl6.org/language/traps
12:52 nine Frameless: so it escaped both our tests :/
12:53 lizmat gfldex: ah, ok  :-)
12:53 gfldex lizmat: it would be nice if you could link to that section
12:53 lizmat yup, will do
12:54 gfldex lizmat: and to https://github.com/perl6/doc/blob/master/CONTRIBUTING.md
12:55 gfldex lizmat: also, i had 2 blögpost last week
12:56 lizmat gfldex: ah?    urls?
12:57 Frameless gfldex: if you have an Atom/RSS feed, may be worth adding your blog to perl6.org recent blogs panel. Just add the metadata here: https://github.com/stmuk/pl6anet.org/blob/master/perlanetrc#L77
12:58 timotimo i wonder if we should teach pl6anet to have a little checkbox for "hide perl6weekly", as it tends to drown out all slower blogs
12:58 gfldex please don't
12:58 timotimo OK
13:00 dalek rakudo/nom: b9e9743 | (Zoffix Znet)++ | docs/release_guide.pod:
13:00 dalek rakudo/nom: Clarify conditions for running tools/contributors.p6
13:00 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/b9e9743271
13:01 Frameless timotimo: seems an overkill, considering "view more" button is just a click away.
13:01 timotimo and how about a clearer division between individual posts, and perhaps a little help navigating between posts?
13:02 lizmat what I would like to know is how do I get it to only post a summary in pl6anet.org, instead of the whole article ?
13:03 Frameless TBH, I rather http://pl6anet.org/ did not show entire bodies of blogs. I'd imagine some folks write for ad revenue or at least would like comments on their blogs. This is essentially equivalent to stealing their content :/
13:04 Frameless OTOH, that seems to be a feature of the blogs themselves. I don't see bodies of my blogs.
13:04 lizmat Frameless: so how do you do it for your blogs?  Did you manipulate the RSS feed for that ?
13:05 timotimo right, it just comes down to what the blog software puts into the rss feed
13:06 Frameless lizmat: I wrote the feed by hand: https://github.com/zoffixznet/perl6.party/blob/master/app.pl#L65 https://github.com/zoffixznet/perl6.party/blob/master/templates/feed.xml.ep
13:06 Frameless And I guess I just did not include whatever wordpress includes to put all of content in. http://perl6.party/feed
13:06 lizmat hmmm...  ok, lemme see if I can do something like that in wordpress
13:07 gfldex https://codex.wordpress.org/WordPress_Feeds#Finding_Your_Feed_URL
13:11 gfldex lizmat: it's 3 traps now
13:12 dalek rakudo/nom: 3ae7477 | (Zoffix Znet)++ | tools/create-release-announcement.pl:
13:12 dalek rakudo/nom: Remove profanity in error message.
13:12 dalek rakudo/nom:
13:12 dalek rakudo/nom: Seems unprofessional :/
13:12 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/3ae7477fb9
13:13 gfldex `fail "shit"` actually made me giggle
13:13 dalek rakudo/nom: 9ef6f6e | coke++ | docs/release_guide.pod:
13:13 dalek rakudo/nom: fix typo
13:13 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/9ef6f6e34c
13:13 mst gfldex: same
13:14 mst however, one can make error messages funny, non-profane, *and* informative
13:14 mst fail "Couldn't find \d\d\d\d-\d\d-\d\d release date - what the flying fornication am I supposed to do here?"
13:14 mst Zoffix: ^^ proposed improvement :P
13:19 skids joined #perl6-dev
13:32 dalek rakudo/nom: 53ff95a | (Zoffix Znet)++ | docs/announce/2016.07.1.md:
13:32 dalek rakudo/nom: Prepare 2016.07.1 point release announcement
13:32 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/53ff95aa1a
13:32 dalek rakudo/nom: 2595ff3 | (Zoffix Znet)++ | docs/ChangeLog:
13:32 dalek rakudo/nom: Remove extraneous zero from 2016.07.1 release
13:32 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/2595ff3f12
13:33 travis-ci joined #perl6-dev
13:33 travis-ci Rakudo build passed. Zoffix Znet 'Populate ChangeLog for 2016.07.01 release'
13:33 travis-ci https://travis-ci.org/rakudo/rakudo/builds/145528004 https://github.com/rakudo/rakudo/compare/21cfe2a8a27c...c07b2d790312
13:33 travis-ci left #perl6-dev
13:33 Frameless 07.md is sorted later than 07.01.md on GitHub, which now irks me :(
13:35 timotimo urgh
13:37 AlexDaniel joined #perl6-dev
13:46 dalek rakudo/nom: f97e227 | (Zoffix Znet)++ | docs/release_guide.pod:
13:46 dalek rakudo/nom: List 2016.07.1 release in past releases list
13:46 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/f97e227ff2
13:47 dalek rakudo/nom: 04933fd | (Zoffix Znet)++ | VERSION:
13:47 dalek rakudo/nom: [release] bump VERSION
13:47 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/04933fd86d
13:49 benchable joined #perl6-dev
14:00 ilbot3 joined #perl6-dev
14:00 Topic for #perl6-dev is now Perl 6 language and compiler development 2.0 | Logs at http://irclog.perlgeek.de/perl6-dev/today
14:02 dalek rakudo/nom: 0e20a34 | (Zoffix Znet)++ | docs/release_guide.pod:
14:02 dalek rakudo/nom: Use https instead of git URL in instructions
14:02 dalek rakudo/nom:
14:02 dalek rakudo/nom: git URL fails on half of my boxes, but https always works
14:02 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/0e20a344ee
14:02 dalek rakudo/nom: 477b641 | (Zoffix Znet)++ | docs/release_guide.pod:
14:02 dalek rakudo/nom: Clarify stresstest instructions assume t/spec already has roast
14:02 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/477b6410fa
14:14 travis-ci joined #perl6-dev
14:14 travis-ci Rakudo build passed. Zoffix Znet 'Clarify conditions for running tools/contributors.p6'
14:14 travis-ci https://travis-ci.org/rakudo/rakudo/builds/145533160 https://github.com/rakudo/rakudo/compare/c07b2d790312...b9e974327110
14:14 travis-ci left #perl6-dev
14:15 dalek rakudo/nom: 664af92 | (Zoffix Znet)++ | docs/release_guide.pod:
14:15 dalek rakudo/nom: Convert incorrectly used Markdown markup to Pod
14:15 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/664af92fba
14:28 Frameless How come we package 6.c-errata roast with Rakudo releases instead of latest? HEAD has better coverage, so it'd make sense to include that instead.
14:29 timotimo because that's where we add new features
14:29 timotimo i mean, HEAD is
14:29 Frameless Ah
14:29 timotimo and we're guaranteeing compatibility with 6.c
14:35 lizmat m: my @a is List = 1,2,3; dd @a   # didn't this use to work at one time ?
14:35 camelia rakudo-moar 664af9: OUTPUT«()␤»
14:35 gfldex you could ask bisectable
14:37 lizmat m: my @a is List; @a.push(42)  # I mean, it *is* a List
14:37 camelia rakudo-moar 664af9: OUTPUT«Cannot call 'push' on an immutable 'List'␤  in block <unit> at <tmp> line 1␤␤»
14:37 lizmat afk&
14:37 Frameless bisect: my @a is List = 1,2,3; dd @a
14:37 bisectable Frameless: On both starting points the exit code is 0 and the output is identical as well
14:37 bisectable Frameless: Output on both points: ()
14:38 lizmat well, guess it didn't ever  :-)
14:38 lizmat Frameless++
14:38 timotimo m: my @a is List; @a.ASSIGN((1, 2, 3, 4)); dd @a
14:38 camelia rakudo-moar 664af9: OUTPUT«Method 'ASSIGN' not found for invocant of class 'List'␤  in block <unit> at <tmp> line 1␤␤»
14:38 timotimo what's that spelled again? STORE?
14:38 timotimo m: my @a is List; @a.STORE((1, 2, 3, 4)); dd @a
14:38 camelia rakudo-moar 664af9: OUTPUT«()␤»
14:38 timotimo oh, lists are immutable, of course
14:38 lizmat really afk&
14:39 lizmat m: my %h is Map = a => 42; dd %h  # it does work with Hash/Map
14:39 camelia rakudo-moar 664af9: OUTPUT«Map.new((:a(42)))␤»
14:39 lizmat really really afk&
14:44 Frameless Can someone explain what perl Configure.pl --prefix=DIR and make DESTDIR=DIR do? I've used the same dir for both and I got... a mess
14:45 timotimo using the same value for both seems like a recipe for disaster
14:46 timotimo after all, the option wouldn't exist if you could/should just pass the same path
14:47 Frameless Right. I don't what what they do. If I just use DESTDIR, then it creates /home/cpan/CPANPRC/release/rakudo-2017.07.1/ directory in the directory I specify in DESTDIR
14:47 nine Frameless: --prefix is where the files should be once they are installed and ready for use. DESTDIR is the temporary place where generated files are stored for packaging
14:47 nine Frameless: common example would be --prefix=/usr  and DESTDIR=/tmp/buildroot
14:48 Frameless ok, lemme try that
14:49 nine The whole point is to be able to create e.g. an rpm package for a system wide installation with the whole build run by an unpriviledged user
14:54 CQ joined #perl6-dev
14:55 travis-ci joined #perl6-dev
14:55 travis-ci Rakudo build passed. Zoffix Znet 'Remove profanity in error message.
14:55 travis-ci https://travis-ci.org/rakudo/rakudo/builds/145535732 https://github.com/rakudo/rakudo/compare/b9e974327110...3ae7477fb942
14:55 travis-ci left #perl6-dev
14:58 hoelzro whoops =/
14:58 timotimo fucking shit.
14:58 hoelzro that's my mistake - I should really have a commit filter for that kind of thing
14:59 hoelzro Zoffix++ for fixing that
14:59 timotimo :)
14:59 Frameless :)
14:59 timotimo didn't know you're such a potty-mouth! :P
15:00 dalek nqp: 735f56d | (Pawel Murias)++ | src/vm/js/HLL/Backend.nqp:
15:00 dalek nqp: [js] Print out a better error message when v8-profiler is not installed.
15:00 dalek nqp: review: https://github.com/perl6/nqp/commit/735f56d520
15:00 dalek nqp: f882309 | (Pawel Murias)++ | src/vm/js/nqp-runtime/reprs.js:
15:00 dalek nqp: [js] Get rid of leftover debugging code.
15:00 dalek nqp: review: https://github.com/perl6/nqp/commit/f88230979f
15:03 hoelzro I see mention of a LINK phaser in the design docs, but no mention of it in Rakudo itself - is it NYI, or something that doesn't belong in the spec?
15:04 RabidGravy joined #perl6-dev
15:05 Frameless So I'm looking at output listed in https://rt.perl.org/Ticket/Display.html?id=128652 and mine doesn't match it. It doesn't have any mention of the modules. How come? https://gist.github.com/zoffixznet/0876525efa6185cb6747494222af2e50
15:05 Frameless It's probably why I didn't notice anything wrong during 2016.07 release.
15:07 [Coke] hoelzro: that ain't the spec.
15:08 perlpilot hoelzro: That's a really good question ...  for TimToady  (he s/FINAL/LINK/ there back in 2014)
15:08 [Coke] We don't have anyone to answer this, really, but I think a better way to phrase the question is " do we want this in some version of six post c?"
15:08 [Coke] for our RTs on NYI things, someone needs to decide the same thing: is this NYI or an RFC?
15:10 * hoelzro nods
15:10 [Coke] do we have anyone who can cut a new R* for win64?
15:12 Frameless I don't know the process, but I do possess a win64...
15:13 hoelzro anyway, I ask because I made note a while ago to add it to the pygments lexer
15:14 hoelzro and I'm checking those notes against Rakudo and the specs
15:18 nine Frameless: the additional output appears when you set the RAKUDO_LOG_PRECOMP env var
15:18 Frameless Thanks.
15:23 [Coke] anyone here have access to edit the ticket redirects on rakudo.org?
15:30 Frameless [Coke]: yeah
15:31 AlexDaniel commit: 2015.10 my @a is List = 1,2,3; dd @a
15:31 committable AlexDaniel: https://gist.github.com/0cd1e0e63c869e4911173388e06f6362
15:31 * [Coke] opens #128664 for other rakudo.org stuff.
15:31 AlexDaniel commit: 2015.12 my @a is List = 1,2,3; dd @a
15:31 committable AlexDaniel: |«2015.12»: ()
15:31 AlexDaniel bisect: good=2015.10 my @a is List = 1,2,3; dd @a # just for fun
15:32 bisectable AlexDaniel: Exit code on a 'good' revision is 1 (which is bad), bisecting with inverted logic
15:32 bisectable AlexDaniel: bisect log: https://gist.github.com/9e834fe52ef43f78b5d4c9648bbbfc3b
15:32 bisectable AlexDaniel: (2015-11-18) https://github.com/rakudo/rakudo/commit/29cc3c2
15:34 AlexDaniel commit: 2015.11 my @a is List = 1,2,3; dd @a
15:34 committable AlexDaniel: |«2015.11»: List $var = $()
15:34 AlexDaniel bisect: good=2015.11 bad=2015.12 my @a is List = 1,2,3; dd @a # just for fun
15:35 bisectable AlexDaniel: Exit code is 0 on both starting points, bisecting by using the output
15:35 bisectable AlexDaniel: bisect log: https://gist.github.com/b6c7ca5dfb2552dd72325634a18616fa
15:35 bisectable AlexDaniel: (2015-12-23) https://github.com/rakudo/rakudo/commit/9308de3
15:35 Frameless Oh shit
15:35 Frameless Oh.. no shit. Just spurious newline :)
15:36 travis-ci joined #perl6-dev
15:36 travis-ci Rakudo build passed. Will "Coke" Coleda 'fix typo'
15:36 travis-ci https://travis-ci.org/rakudo/rakudo/builds/145536045 https://github.com/rakudo/rakudo/compare/3ae7477fb942...9ef6f6e34cd4
15:36 travis-ci left #perl6-dev
15:36 AlexDaniel lizmat: “didn't this use to work at one time ?” – you are right, look ↑
15:37 Frameless [Coke]: I changed /rt/open-all to Public, but I'm still getting a CSFR error
15:38 [Coke] (see /msg window)
15:39 [Coke] bumped it back to the rt bugadmins. Thanks!
15:40 Frameless [Coke]: I think you have access to edit those redirects too. They're in public_html/.htaccess
15:40 * Frameless celebrates with the appropriate amount of fun.
15:40 [Coke] Frameless: in future, I'd keep the point releases as "the previous release with just the bugfix" rather than "everything that snuck in"
15:41 [Coke] Frameless: I ... might have access to that, then. Thanks.
15:41 Frameless .tell stmuk Rakudo 2016.07.1 shipped out
15:41 yoleaux2 Frameless: I'll pass your message to stmuk.
15:41 Frameless [Coke]: OK. Noted.
15:41 [Coke] (release) ... crap. meant to lead with: "Thank you very much for doing the thankless job of releases!"
15:41 Frameless :)
15:41 [Coke] I, especially, really appreciate you stepping up to do these.
15:46 Frameless Hm. I think all my statistics on running on 24-30 core machines were all wrong. I notice Google is still not charging me ($300 free trial) and in description it says the free trial boxes use at most 8 cores :o
15:49 Frameless Though I see no way to upgrade to non-free version, so maybe I'm wrong that it's using just 8 cores :/
15:57 stmuk Frameless++
15:57 yoleaux2 15:41Z <Frameless> stmuk: Rakudo 2016.07.1 shipped out
16:10 dalek star: ffa0220 | (Steve Mynott)++ | tools/star/Makefile:
16:10 dalek star: bump rakudo version
16:10 dalek star: review: https://github.com/rakudo/star/commit/ffa0220f8d
16:18 travis-ci joined #perl6-dev
16:18 travis-ci Rakudo build passed. Zoffix Znet 'Remove extraneous zero from 2016.07.1 release'
16:18 travis-ci https://travis-ci.org/rakudo/rakudo/builds/145540699 https://github.com/rakudo/rakudo/compare/9ef6f6e34cd4...2595ff3f1268
16:18 travis-ci left #perl6-dev
16:37 dalek star: dd94806 | (Steve Mynott)++ | modules/doc:
16:37 dalek star: bump doc submodule
16:37 dalek star: review: https://github.com/rakudo/star/commit/dd94806141
16:41 dalek rakudo/nom: 18455a5 | (Zoffix Znet)++ | docs/release_guide.pod:
16:41 dalek rakudo/nom: Fix typo
16:41 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/18455a5f23
16:42 dalek star: e02930b | (Steve Mynott)++ | docs/announce/2016.07.md:
16:42 dalek star: more changes in announce
16:42 dalek star: review: https://github.com/rakudo/star/commit/e02930bf33
16:44 Frameless stmuk: s/none-multi/non-multi/; in announcement
16:44 Frameless That's left-over from a typo in Rakudo's ChangeLog
16:44 timotimo thank you :)
16:46 dalek star: a244d87 | (Steve Mynott)++ | docs/announce/2016.07.md:
16:46 dalek star: correct typo to make timotimo happy
16:46 dalek star: review: https://github.com/rakudo/star/commit/a244d87c7e
16:46 timotimo <3
17:07 ilmari joined #perl6-dev
17:17 geekosaur joined #perl6-dev
17:18 tomboy64 joined #perl6-dev
17:20 timotimo joined #perl6-dev
18:14 gfldex Pod::To::HTML has be fixed to render S26 (maybe even correctly). I do not know where, when, how and by whom it a needs to updated. If you do please wave with both hands.
18:22 stmuk I'm seeing test failures with 07.1
18:23 stmuk t/spec/S17-supply/syntax.t & t/spec/S32-io/IO-Socket-Async.t
18:24 [Coke] stmuk: S17 are flappy
18:24 [Coke] try running them by hand?
18:24 hoelzro I think both of those are flappers
18:25 * gfldex .oO( Heisentest )
18:31 sno joined #perl6-dev
18:32 [Coke] Anyone have suggestions on https://github.com/perl6/doc/pull/714 ? looks like some people need a PERL6LIB set, but others don't.
18:34 stmuk oh they pass on their own
18:37 Frameless gfldex: assuming I've done things right, it now has been updated
18:38 gfldex Frameless: looks like it will need a push to roast to get updated
18:39 Frameless roast? :/
18:39 gfldex or mu
18:41 gfldex Frameless: see https://github.com/perl6/mu/blob/master/util/update-design.perl6.org.sh#L26
18:46 dalek roast: 67794d0 | (Zoffix Znet)++ | S02-types/bool.t:
18:46 dalek roast: Remove trailing whitespace
18:46 dalek roast: review: https://github.com/perl6/roast/commit/67794d0443
18:46 Frameless There
18:51 nine OMG
18:51 * nine frowns at the horrors of code in p6doc
18:54 Frameless gfldex: this means everything worked, right? http://design.perl6.org/S26.html
18:54 lucasb_ joined #perl6-dev
18:54 FROGGS joined #perl6-dev
18:55 lucasb_ Frameless: are you Zoffix, right? thanks for the releases!
18:56 Frameless .oO( The real question is... is Zoffix Zoffix? :o )
18:57 lucasb_ this last roast commit "Remove trailing whitespace", it actually removed an significant plus sign.
18:57 Frameless huh
18:57 lucasb_ it's just a comment, I know, but it was talking about the iteraction with "+"
18:57 FROGGS joined #perl6-dev
18:58 Frameless I guess `find . -name \* -exec sed -i 's/\s+$//g' {} \;` doesn't do what I think it should :P
18:58 gfldex Frameless: looks good
18:58 lucasb_ maybe sed uses '\s\+' ?
18:58 dalek roast: e1a233f | (Zoffix Znet)++ | S02-types/bool.t:
18:58 dalek roast: Revert "Remove trailing whitespace"
18:58 dalek roast:
18:58 dalek roast: This reverts commit 67794d0443eff81a85ecf5275ebc6c24bbbcfbee.
18:58 dalek roast: review: https://github.com/perl6/roast/commit/e1a233ff11
18:59 Frameless gfldex++ sweet. Thanks for fixing Pod::To::HTML
19:00 Frameless lucasb_++ # reviewing commits :)
19:00 lucasb_ Frameless: thank you for fixing :)
19:01 Frameless Zoffix fixed it, actually. I have him locked in the basement. Doing my bidding. Muahaha
19:02 stmuk http://pl6anet.org/drop/rakudo-star-2016.07-RC1.tar.gz
19:09 * stmuk pretends to have nothing to do with p6doc
19:09 Frameless :)
19:12 * nine feels like re-inventing the CURI repository format just to demonstrate how p6doc is full of poking around in internals it should never touch.
19:13 timotimo hehehe
19:13 stmuk re-re-inventing?
19:17 * stmuk wonders if nine has a particular sub in mind? :)
19:19 brrt joined #perl6-dev
19:20 nine old-locate-curli-module, locate-curli-module, list-installed, search-paths
19:22 stmuk oh all of it then! well at least mine were shorter!
19:26 * nine is now implementing CompUnit::Repository::Installation::installed as there seems to be demand for it
19:27 stmuk list-installed() made my head hurt
19:27 stmuk nine++
19:27 nine So why has no one of the involved ever asked for installed() or the rest of the API?
19:28 nine You wouldn't need list-installed for most use cases if you just used .resolve()
19:32 * lizmat starts on the P6W
19:35 [Coke] nine: the htmlify stuff was added basically to have a prayer of finishing a doc build before the heat death of the universe. go easy on us. :)
19:35 [Coke] I'm happy to patch it however you suggest. (up to, and sadly including, ripping it out.)
19:37 stmuk [Coke]: it's p6doc not htmlify and p6doc has had a rather chequered history
19:38 stmuk its a old script which has just been hacked up to continue working
19:38 nine ugexe: which of the Distribution roles/classes is meant for being returned from CURI.installed()? I.e. which will allow for accessing the stored files?
20:29 Ven joined #perl6-dev
21:10 AlexDaniel joined #perl6-dev
21:30 lizmat and another Perl 6 weekly hits the Net: https://p6weekly.wordpress.com/2016/07/18/2016-29-the-zoffix-has-landed/
21:30 hoelzro lizmat++
22:12 ugexe list-installed works on rakudos before .resolve landed
22:25 lizmat good night, #perl6-dev
22:32 Zoffix left #perl6-dev
22:44 skids joined #perl6-dev

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