Perl 6 - the future is here, just unevenly distributed

IRC log for #perl6, 2014-07-20

Perl 6 | Reference Documentation | Rakudo

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

All times shown according to UTC.

Time Nick Message
00:06 thou joined #perl6
00:32 ivanshmakov joined #perl6
00:34 timotimo yes, compose is nyi
00:56 cognominal m:   my $ab = 'a b';  say @( << $ab c d $ab >> ).perl
00:56 camelia rakudo-moar 866817: OUTPUT«("a", "b", ("c", "d"), "a", "b").list␤»
00:57 cognominal how do I flatten that?
00:58 cognominal m:   my $ab = 'a b';  say (flat << $ab c d $ab >> ).perl
00:58 camelia rakudo-moar 866817: OUTPUT«("a", "b", "c", "d", "a", "b").list␤»
00:58 cognominal ok
01:00 cognominal "Normal flat list context ignores parcel boundaries and flattens the list. Slice context turns any parcel objects into C<Seq> objects." Not sure what is a "normal flat context" mentionned in S04 though
01:00 jnap joined #perl6
01:02 oJustDust joined #perl6
01:09 ivanshmakov joined #perl6
01:09 silug joined #perl6
01:20 FROGGS_ joined #perl6
01:25 bjz joined #perl6
01:34 Akagi201 joined #perl6
01:51 klapperl_ joined #perl6
01:55 thou joined #perl6
01:56 xenoterracide_ joined #perl6
01:59 vendethiel joined #perl6
02:06 xenoterracide_ joined #perl6
02:13 noganex joined #perl6
02:31 nbrown joined #perl6
02:40 kurahaupo joined #perl6
02:59 rindolf joined #perl6
03:03 kaare_ joined #perl6
03:22 xenoterracide_ joined #perl6
03:43 thou joined #perl6
03:49 nbrown joined #perl6
04:43 atroxaper joined #perl6
05:04 rindolf left #perl6
05:31 xragnar joined #perl6
05:31 thou joined #perl6
06:13 atroxaper joined #perl6
06:22 zengargoyle joined #perl6
07:20 thou joined #perl6
07:31 Isp-sec joined #perl6
07:39 brrt joined #perl6
07:39 kweetal joined #perl6
07:43 atroxaper joined #perl6
08:07 dalek roast: c73bd06 | (Pepe Schwarz)++ | S32-exceptions/misc.t:
08:07 dalek roast: Add corresponding ticket for two tests.
08:07 dalek roast: review: https://github.com/perl6/roast/commit/c73bd06b23
08:07 dalek roast: feeea52 | (Pepe Schwarz)++ | S32-exceptions/misc.t:
08:07 dalek roast: Add test for RT #75320.
08:07 dalek roast: review: https://github.com/perl6/roast/commit/feeea52346
08:07 synopsebot Link: https://rt.perl.org/rt3//Public/Bug/Display.html?id=75320
08:07 psch ...with that, greetings #perl6 :)
08:10 vendethiel o/
08:11 Ven joined #perl6
08:28 molaf joined #perl6
08:33 darutoko joined #perl6
08:40 atroxaper joined #perl6
08:45 virtualsue joined #perl6
08:51 lizmat joined #perl6
08:56 masak good antenoon, #perl6
08:56 psch masak o/
08:57 woolfy joined #perl6
09:08 thou joined #perl6
09:13 PotatoGim joined #perl6
09:18 kivutar joined #perl6
09:19 Ven joined #perl6
09:20 Ven mmh, also, flip flop ranges don't get much description in the synopses
09:22 masak they don't.
09:22 masak the spectests are a much better source in that case.
09:24 dakkar joined #perl6
09:24 psch ff (and fff) are what .. did in scalar context in perl5, right?
09:27 masak ff is what .. did in scalar context.
09:27 masak fff is what ... did in scalar context.
09:27 masak from a Perl 6 perspective, it's quite crazy that infix:<...> was wasted on infix:<fff> of all things :P
09:28 dmol joined #perl6
09:29 psch right, i unintentionally omitted "(and ...)", thanks
09:30 psch it's crazy that perl5 uses ... as infix:<fff>?
09:32 masak it's crazy to waste such an accessible operator on such an unusual semantics.
09:32 masak compare how versatile infix:<...> is in Perl 6.
09:35 rindolf joined #perl6
09:35 FROGGS_ I... don't understand p6bench :/
09:36 masak FROGGS_: are you able to condense something into a question?
09:37 FROGGS_ masak: I run a single microbenchmark, get a json output file with the timings, but the output html does not show the graph
09:38 FROGGS_ and also, the benchmark gets an SCALE argument as input, which results in a string length... the longer the string the slower it gets
09:39 FROGGS_ but my visible graph for one test looks like as is it is the slower the smaller the string is
09:42 spider-mario joined #perl6
09:47 Psyche^ joined #perl6
09:49 beastd joined #perl6
09:50 FROGGS_ hmmm, these plots also start at 0/0... http://jnthn.net/perl6/bench/2014-07-15.html
09:50 FROGGS_ but why?
09:50 spider-mario joined #perl6
09:54 atroxaper joined #perl6
09:59 FROGGS_ ahh, it ignored the compile time, and that is what I wanna bench
10:00 masak oh!
10:01 FROGGS_ and the bench script offers --ignore-compile, so I guessed the default wrong
10:12 rindolf joined #perl6
10:15 vendethiel masak: it's versatile, but I found out that it completly depends on what I'm doing. Sometimes I'll use it many times in one project, other times I won't use it at all
10:19 vendethiel m: say if 'B' ff 'B' for <A B C B A>
10:19 camelia rakudo-moar 866817: OUTPUT«␤␤␤␤␤»
10:20 vendethiel m: .say if 'B' ff 'B' for <A B C B A> # erm
10:20 camelia rakudo-moar 866817: OUTPUT«A␤B␤C␤B␤A␤»
10:21 vendethiel so, basically, ff checks its rhs even though it just checked its lhs ?
10:25 masak vendethiel: oh, for sure.
10:28 vendethiel then, ... is probably what I'll be using most of the time
10:29 psch m: .say if 'B' fff 'B' for <A B C B A>
10:29 camelia rakudo-moar 866817: OUTPUT«A␤B␤C␤B␤A␤»
10:29 psch i'm not really sure i understand those fully, anyway :)
10:30 vendethiel the behavior on camelia is incorrect, fwiw
10:32 brrt joined #perl6
10:35 CuriousSpectator joined #perl6
10:37 molaf joined #perl6
10:40 masak vendethiel: this is my favorite use of infix:<...> ever: https://github.com/masak/dlx-simple/blob/b430a62808b86224f2bbf38ff2215877e7566ed7/perl6/lib/Solver/DancingLinks.pm#L57
10:41 masak vendethiel: re 'behavior on camelia is incorrect' -- you're assuming smartmatch semantics.
10:41 masak vendethiel: (which is not the case in Perl 6)
10:43 spider-mario joined #perl6
10:44 psch S03:2929
10:44 synopsebot Link: http://perlcabal.org/syn/S03.html#line_2929
10:44 denis_boyun_ joined #perl6
10:44 anaeem1 joined #perl6
10:44 psch masak: that says "smartmatch against $_"
10:45 masak huh -- so it does.
10:46 masak and it's been that way since May last year. :)
10:46 masak TimToady++
10:56 thou joined #perl6
11:02 denis_boyun_ joined #perl6
11:08 Akagi201 joined #perl6
11:12 salv0 joined #perl6
11:29 SamuraiJack joined #perl6
11:38 telex joined #perl6
11:42 colomon if I want to parse an HTML file and extrace bits from it, is there a module for that?
11:43 FROGGS_ there is an XML parser
11:44 atroxaper joined #perl6
11:44 masak https://github.com/tony-o/perl6-html-parser-xml/
11:44 FROGGS_ oi, nice
11:45 colomon I don't want an XML document at the end, though.  I just want to be be able to scrape this web site.
11:46 masak ok, could you define "extract bits" a bit more clearly?
11:46 _sri Mojo::DOM might be rather easy to port to perl6
11:47 colomon masak: Here's (one of) the pages: http://thesession.org/recordings/387
11:47 colomon I just need to get at the track listing so I can tag MP3 files with it.
11:48 masak well, I see three paths:
11:48 colomon I guess I could build a simple parser myself
11:48 masak (a) hack -- forget HTML parsing, and just read line-by-line, pulling things out by regex
11:48 masak (b) DOM -- parse the whole document, then access what you want with an XPath-like expression
11:49 _sri with perl5 ;p
11:49 _sri perl -Mojo -E 'g("http://thesession.org/recordings/387")->dom->find(".manifest-item-title a")->text->join("\n")->encode->say'
11:49 masak (c) SAX -- find or build a parser that emits events on opening and closing tags, and use it to build a small state machine that picks up the content that you want.
11:49 masak I guess what _sri++ just posted (which is awesome, by the way) is an instance of (b).
11:49 masak we should totally have that in Perl 6.
11:50 * masak is jealous, in a good way :)
11:50 _sri :D
11:53 masak Perl 5 has a number of things where it can brag superiority to Perl 6. Mojo is one of them.
11:53 _sri the command line testing tool can do it too actually "mojo get http://thesession.org/recordings/387 ".manifest-item-title a" text"
11:53 masak wow.
11:55 * colomon is hacking up a simple grammar
11:57 carlin joined #perl6
12:02 carlin panda doesn't seem to want to build on Windows, as JSON::Tiny is failing tests with; No such method 'package' for invocant of type 'NQPAttribute'
12:03 masak o.O
12:03 masak carlin: could you catch the relevant output from that and put it up as a gist?
12:10 carlin when I run them manually they work fine, but during panda bootstrap they fail
12:10 carlin https://gist.github.com/carbin/0b3de07de9fa3b93b61a
12:11 carlin I'll try compiling them and running the tests again in a bit
12:11 carlin to see if that's the problem
12:11 masak carlin++
12:12 masak could be a precompilation problem, yes.
12:25 jnap joined #perl6
12:26 FROGGS_ benchmark for the charrange fixes: http://froggs.de/perl6/charrange.html
12:31 timotimo wowza.
12:31 timotimo it's very hard to read, as all pre-charrange implementations just abort to the very left %)
12:34 dalek nqp: a0842ef | (Tobias Leich)++ | / (9 files):
12:34 dalek nqp: RT #67122 fix slowdown in large character ranges in regexes
12:34 dalek nqp:
12:34 dalek nqp: On all platforms it built a string with every codepoint of the given range,
12:34 dalek nqp: and when :i was in effect it built that string with uppercase and lowercase
12:34 dalek nqp: codepoints to do a "cheap" character lookup later. Building this string is
12:34 dalek nqp: very expensive and might even crash on long ranges.
12:34 synopsebot Link: https://rt.perl.org/rt3//Public/Bug/Display.html?id=67122
12:34 dalek nqp: Now it just checks if a char is within the range, (lower <= char <= upper).
12:34 dalek nqp:
12:34 dalek nqp: Benchmark: http://froggs.de/perl6/charrange.html
12:34 dalek nqp: review: https://github.com/perl6/nqp/commit/a0842ef500
12:34 FROGGS_ timotimo: true, because they failed to turned invalid codepoints into chars :o)
12:35 FROGGS_ but one can see that there is a critical point, where the performance drops
12:36 timotimo i can't really see anything there :)
12:36 FROGGS_ O.o
12:36 timotimo i mean
12:36 FROGGS_ only look at 2014.07, and you'll see that it looks like /`
12:36 timotimo ... i don't know
12:37 timotimo yeah
12:37 timotimo but the "angle" seems the same as in the charrange version, just further down
12:37 FROGGS_ on the last shown dot the moar/2014.07 is three times slower than moar/charrange
12:38 timotimo ah, ok
12:38 FROGGS_ for the case insensitive version it is six times slower, which is kinda predictable :o)
12:39 timotimo ya :)
12:39 FROGGS_ also, the charrange version has a constant compile+execution time of 0.055s, the 2014.07 takes about 0.3s on the longest range testable
12:42 FROGGS_ also:
12:42 FROGGS_ /usr/bin/time components/nqp-moar/2014.07/install/bin/nqp-m -e '"\c[50000]" ~~ /:i<[\c0..\c[50000]]>/'
12:42 FROGGS_ 0.24user 0.06system 0:00.30elapsed 99%CPU (0avgtext+0avgdata 219804maxresident)k
12:42 FROGGS_ compared to:
12:42 FROGGS_ /usr/bin/time components/nqp-moar/charrange/install/bin/nqp-m -e '"\c[50000]" ~~ /:i<[\c0..\c[50000]]>/'
12:42 FROGGS_ 0.04user 0.01system 0:00.06elapsed 98%CPU (0avgtext+0avgdata 23740maxresident)k
12:42 FROGGS_ the maxresident part
12:44 dalek rakudo/nom: bbdcfd6 | (Tobias Leich)++ | tools/build/NQP_REVISION:
12:44 dalek rakudo/nom: bump nqp rev to get charrange improvements
12:44 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/bbdcfd668e
12:44 thou joined #perl6
12:45 BenGoldberg joined #perl6
12:45 atroxaper joined #perl6
12:46 psch https://github.com/rakudo/rakudo/pull/292 # flip-flop.t passes as before, with a bunch of "unitialized value of type Any in string context", but i think the tests actually rely on that
12:46 timotimo oooooh
12:48 timotimo psch: now i can't ~~ a different variable against a flipflop any more?
12:48 psch timotimo: i haven't tested that.  do you have a snippet handy?
12:49 psch timotimo: fwiw, ff and fff should smartmatch against $_, according to S03:2929
12:49 synopsebot Link: http://perlcabal.org/syn/S03.html#line_2929
12:50 psch although the wording seems a bit unclear to me, "against $_" seems to imply $_ as rhs to ~~, which doesn't really make sense in my opinion
12:52 timotimo oh
12:52 timotimo so it's a spec question really
12:52 timotimo can i $foo ~~ "A" ff "B"?
12:52 kivutar joined #perl6
12:52 masak interesting question.
12:53 psch ~~ sets lhs as $_
12:53 psch so that should still work
12:53 masak I think that would end up involving two smartmatches, semantically.
12:53 masak one from the ~~ and one from the ff
12:54 psch $foo ~~ "A" ff "B" binds $_ to $foo for the ff and "A" and "B" get smartmatched against that
12:54 psch that's how i understand it at least, from poking around in make_smartmatch for tr///
12:54 timotimo hm, ok
12:56 masak precedence problem.
12:56 masak ff binds looser than ~~
12:56 masak so you'd need parens.
12:56 masak (I had to check S03 to be sure)
13:00 colomon Grammar::Debugger still broken.  :(
13:01 carlin that JSON::Tiny problem does only happen when compiled
13:03 carlin and it's only JSON::Tiny itself. JSON::Tiny::Actions and ::Grammar still work
13:03 masak and only on Windows?
13:03 masak which backend? Moar?
13:04 carlin moar, Windows 7 64-bit
13:04 carlin is working fine in Linux 64
13:04 masak it sounds to me like you have the perfect prerequisites for golfing the problem down.
13:05 * masak attempts to build Rakudo/Moar on Windows 7 64-bit
13:10 masak carlin: how did you install Rakudo?
13:12 carlin git clone, strawberry perl, MS VC compiler
13:12 masak ok.
13:13 brrt joined #perl6
13:24 ivanshmakov joined #perl6
13:25 chenryn joined #perl6
13:28 kurahaupo joined #perl6
13:33 ponbiki joined #perl6
13:37 masak building on Windows wasn't completely unproblematic. will continue later.
13:37 masak walk &
13:39 carlin yeah I find getting it to build on Windows isn't as straight-forward as on Unixy OSes
13:39 carlin when I manually compile each module it works... so it must be something specific to what panda does
13:43 carlin and `perl6 -e "use JSON::Tiny"` is enough to get the error
13:51 carlin ah ha
13:51 carlin it's the build order
13:52 carlin panda compiles in the order: JSON::Tiny --> JSON::Tiny::Actions --> JSON::Tiny::Grammar
13:52 carlin when I manually compile in that order it breaks, but if you do JSON::Tiny last it works fine
13:53 carlin now that I've figured out that I can try and golf it :)
13:55 dalek roast: bea8d27 | (Tobias Leich)++ | S05-metasyntax/charset.t:
13:55 dalek roast: RT #67122 unfudge tests but use valid codepoints
13:55 dalek roast: review: https://github.com/perl6/roast/commit/bea8d27bce
13:55 synopsebot Link: https://rt.perl.org/rt3//Public/Bug/Display.html?id=67122
13:59 denis_boyun_ joined #perl6
14:02 guru joined #perl6
14:06 atroxaper joined #perl6
14:06 chenryn joined #perl6
14:08 FROGGS_ .tell jnthn Is that a valid patch about rooting? https://gist.github.com/FROGGS/534f320aa548e12f75c4
14:08 yoleaux FROGGS_: I'll pass your message to jnthn.
14:13 zakharyas joined #perl6
14:18 * colomon has his script working now.  :)
14:20 Akagi201 joined #perl6
14:23 rindolf left #perl6
14:25 chenryn joined #perl6
14:31 brrt joined #perl6
14:33 thou joined #perl6
14:33 carlin masak: here's a golf: https://gist.github.com/carbin/a808d2eb91e1a2ea56ba
14:49 dalek roast: ff5bd05 | (Tobias Leich)++ | S05-metasyntax/regex.t:
14:49 dalek roast: RT #118985 test lexical regexes in classes
14:49 dalek roast: review: https://github.com/perl6/roast/commit/ff5bd050c0
14:49 synopsebot Link: https://rt.perl.org/rt3//Public/Bug/Display.html?id=118985
14:52 FROGGS_ carlin: does it still fail when you add a multi baz also?
14:56 carlin FROGGS_: yes it does
14:57 FROGGS_ hmmm
14:57 FROGGS_ ohh, perhaps because Foo is a module and not also a class?
14:59 carlin changing Foo to a class, the error becomes: No STable at index 5
14:59 FROGGS_ ó.ò
15:00 FROGGS_ you have recompiled both?
15:00 carlin yip
15:03 kivutar joined #perl6
15:18 dalek roast: c6e4b9b | (Tobias Leich)++ | S02-types/hash.t:
15:18 dalek roast: RT #117431 test hash itemization
15:18 dalek roast: review: https://github.com/perl6/roast/commit/c6e4b9bc86
15:18 synopsebot Link: https://rt.perl.org/rt3//Public/Bug/Display.html?id=117431
15:19 molaf joined #perl6
15:24 dalek roast: 2e900e5 | (Tobias Leich)++ | S02-types/set.t:
15:24 dalek roast: RT #117103 test metaop set assign
15:24 dalek roast: review: https://github.com/perl6/roast/commit/2e900e5d4b
15:24 synopsebot Link: https://rt.perl.org/rt3//Public/Bug/Display.html?id=117103
15:30 dalek roast: 4da54d8 | (Tobias Leich)++ | S05-metasyntax/litvar.t:
15:30 dalek roast: RT #117091 test vars in :i in interpolation/assertion
15:30 dalek roast: review: https://github.com/perl6/roast/commit/4da54d891f
15:30 synopsebot Link: https://rt.perl.org/rt3//Public/Bug/Display.html?id=117091
15:43 chenryn joined #perl6
15:46 carlin it's panda's fault. On Windows it compiles the modules in a different order than on Linux
15:47 carlin https://gist.github.com/carbin/d0dff5de558279350be6
15:47 nbrown joined #perl6
15:49 [Coke] masak: aha, but http://host07.perl6.com:8080/report/coke is all green! (of course, that's a pretty teeny single module)
15:52 flussence I'm having a really un-fun day debugging this bit of code; can anyone take a minute to see if the test failures are reproducable (via panda install or whatever)? https://github.com/flussence/Test-Corpus/blob/24465a223e3059b37c9a905e42d630401782a409/lib/Test/Corpus.pm#L33
15:53 flussence (it'll do different things depending on an env var, but right now *all* of them fail/crash/segfault...)
15:54 dalek rakudo-star-daily: 55e661f | coke++ | log/ (14 files):
15:54 dalek rakudo-star-daily: today (automated commit)
15:54 dalek rakudo-star-daily: review: https://github.com/coke/rakudo-star-daily/commit/55e661ffbe
15:54 dalek perl6-roast-data: aba5c7c | coke++ | / (5 files):
15:54 dalek perl6-roast-data: today (automated commit)
15:54 dalek perl6-roast-data: review: https://github.com/coke/perl6-roast-data/commit/aba5c7c989
15:56 [Coke] ugh, I can't believe this quest has taken SIX months: https://questhub.io/realm/perl/quest/52e802c019eee0996000001b
15:56 * [Coke] will finish it today, curses.
16:00 [Coke] .seen petdance
16:00 yoleaux I haven't seen petdance around.
16:02 guru joined #perl6
16:10 [Coke] m: my %h; %h<bar><baz> := 'zoom'
16:10 camelia rakudo-moar bbdcfd: ( no output )
16:11 [Coke] m: my %h; %h<bar><baz> := 'zoom' ; say %h.perl
16:11 camelia rakudo-moar bbdcfd: OUTPUT«("bar" => {"baz" => "zoom"}).hash␤»
16:11 flussence ...do "»." things do real threading now? That'd explain these new failures...
16:11 flussence "Internal error: zeroed target thread ID in work pass"
16:12 btyler joined #perl6
16:13 timotimo no, they do not
16:14 ajr joined #perl6
16:14 dmol joined #perl6
16:14 dalek roast: b77befe | coke++ | S02-types/hash.t:
16:14 dalek roast: mention RT in todo
16:14 dalek roast: review: https://github.com/perl6/roast/commit/b77befe573
16:19 dalek roast: e5dbea7 | coke++ | TODO:
16:19 dalek roast: Add a note about RTs.
16:19 dalek roast: review: https://github.com/perl6/roast/commit/e5dbea7826
16:20 carlin I found the panda problem!
16:20 carlin https://github.com/tadzik/panda/blob/master/lib/Panda/Builder.pm#L28
16:20 carlin that doesn't account for the paths containing backslashes
16:21 carlin so the it tries to find deps for things like "lib\JSON\Tiny" instead of "JSON::Tiny" and ends up compiling the modules in the wrong order
16:21 thou joined #perl6
16:22 [Coke] m: my %hash = :a(1), :b(2), :c(3), :d(4); my @slice := %hash<b c>; say @slice.perl
16:22 camelia rakudo-moar bbdcfd: OUTPUT«(2, 3)␤»
16:25 [Coke] m: my %hash = :a(1), :b(2), :c(3), :d(4); my @slice := %hash<b c>; say ((@slice,*) = <A B C D>).perl;
16:25 camelia rakudo-moar bbdcfd: OUTPUT«(("A", "B"), *)␤»
16:26 [Coke] ^^ there's a test that expects that to be just "A", "B"; I suspect whatever handling in this case might have mutated since the test was written.
16:26 * [Coke] updates the test to match camelia, since that seems reasonable here.
16:26 [Coke] m: my %hash = :a(1), :b(2), :c(3), :d(4); my @slice := %hash<b c>; say ((@slice,*) = <A B C D>).perl; say @slice.perl;
16:26 camelia rakudo-moar bbdcfd: OUTPUT«(("A", "B"), *)␤("A", "B")␤»
16:33 dalek roast: 28f9dba | coke++ | S32-hash/slice.t:
16:33 dalek roast: Fix test - validating @slice, not @slice && *
16:33 dalek roast: review: https://github.com/perl6/roast/commit/28f9dba94c
16:35 * [Coke] wonders where this is defined:
16:35 [Coke] m: import PROCESS <%ENV>;
16:35 camelia rakudo-moar bbdcfd: ( no output )
16:36 [Coke] r: import PROCESS <%ENV>;
16:36 camelia rakudo-{parrot,jvm,moar} bbdcfd: ( no output )
16:36 [Coke] (er, the import works, but there is no "%ENV" after import.
16:42 bjz joined #perl6
16:43 [Coke] m: my $पहला = 1;
16:43 camelia rakudo-moar bbdcfd: OUTPUT«[31m===[0mSORRY![31m===[0m Error while compiling /tmp/Txc2zns0pXâ�¤Two terms in a rowâ�¤at /tmp/Txc2zns0pX:1â�¤------> [32mmy $पहल[33mâ��[31mा = 1;[0mâ�¤    expecting any of:â�¤        scoped declaratorâ�¤        constraintâ�¤        postfi…»
16:47 [Coke] m: say uniname(0x2029)
16:47 camelia rakudo-moar bbdcfd: OUTPUT«PARAGRAPH SEPARATOR␤»
16:49 dalek roast: d4a402b | coke++ | S02- (2 files):
16:49 dalek roast: Refer to newly-created RTs
16:49 dalek roast: review: https://github.com/perl6/roast/commit/d4a402b5eb
16:51 [Coke] m: $_ = "aaaa"; .++.say while 1
16:51 camelia rakudo-moar bbdcfd: OUTPUT«(timeout)aaaa␤aaab␤aaac␤aaad␤aaae␤aaaf␤aaag␤aaah␤aaai␤aaaj␤aaak␤aaal␤aaam␤aaan␤aaao␤aaap␤aaaq␤aaar␤aaas␤aaat␤aaau␤aaav␤aaaw␤aaax␤aaay␤aaaz␤aaba␤aabb␤aabc␤aabd␤aabe␤aabf␤aabg␤aabh␤aabi␤aa…»
16:52 [Coke] m: my @array = 1,2,3; @array>>\      .++ ; say @array
16:52 camelia rakudo-moar bbdcfd: OUTPUT«Cannot find method 'postcircumfix:<( )>'␤  in sub METAOP_HYPER_CALL at src/gen/m-CORE.setting:17538␤  in block  at /tmp/1qE1RsIlEN:1␤␤»
16:52 [Coke] m: my @array = 1,2,3; @array>>.++ ; say @array
16:52 camelia rakudo-moar bbdcfd: OUTPUT«Cannot find method 'postcircumfix:<( )>'␤  in sub METAOP_HYPER_CALL at src/gen/m-CORE.setting:17538␤  in block  at /tmp/MYp0yFPj5U:1␤␤»
16:52 [Coke] m: my @array = 1,2,3; @array>>.sqrt ; say @array
16:52 camelia rakudo-moar bbdcfd: OUTPUT«1 2 3␤»
16:55 SamuraiJack joined #perl6
16:57 dalek roast: 657c54a | coke++ | S0 (2 files):
16:57 dalek roast: Link to RT #122342
16:57 dalek roast: review: https://github.com/perl6/roast/commit/657c54ac56
16:57 synopsebot Link: https://rt.perl.org/rt3//Public/Bug/Display.html?id=122342
16:59 masak carlin: nice golf!
17:00 masak carlin++
17:02 masak carlin: heh, and I saw the different build order on Windows, and thought "it's probably because of the backslashes" :)
17:02 masak carlin: well debugged!
17:03 masak carlin, tadzik: that line, https://github.com/tadzik/panda/blob/master/lib/Panda/Builder.pm#L28 -- is there a way to have IO::Path handle some of that, for great compatibility?
17:04 masak haha *sob* -- and that subroutine is straight out of ufo, so it's ultimately "my fault" ;(
17:04 [Coke] m: use MONKEY_TYPING: augment class Code{method xyzzy(Code $x: *@y) { $x.(@y) }}; say (xyzzy { @^x }: 1, 2, 3)
17:04 camelia rakudo-moar bbdcfd: OUTPUT«[31m===[0mSORRY![31m===[0m Error while compiling /tmp/ON_0uohFWkâ�¤Confusedâ�¤at /tmp/ON_0uohFWk:1â�¤------> [32muse MONKEY_TYPING[33mâ��[31m: augment class Code{method xyzzy(Code $[0mâ�¤    expecting any of:â�¤        statement listâ�¤Â»
17:04 [Coke] m: use MONKEY_TYPING; augment class Code{method xyzzy(Code $x: *@y) { $x.(@y) }}; say (xyzzy { @^x }: 1, 2, 3)
17:04 camelia rakudo-moar bbdcfd: OUTPUT«[31m===[0mSORRY![31m===[0m Error while compiling /tmp/NyubD3Pthwâ�¤Two terms in a rowâ�¤at /tmp/NyubD3Pthw:1â�¤------> [32mx: *@y) { $x.(@y) }}; say (xyzzy { @^x }[33mâ��[31m: 1, 2, 3)[0mâ�¤    expecting any of:â�¤        postfixâ�¤        statemen…»
17:05 [Coke] ^^^ does that look like it should work?
17:05 [Coke] m: use MONKEY_TYPING; augment class Code{method xyzzy(Code $x: *@y) { $x.(@y) }}; say (xyzzy { @^x },: 1, 2, 3)
17:05 camelia rakudo-moar bbdcfd: OUTPUT«[31m===[0mSORRY![31m===[0m Error while compiling /tmp/4vy8mNT5xIâ�¤Two terms in a rowâ�¤at /tmp/4vy8mNT5xI:1â�¤------> [32m: *@y) { $x.(@y) }}; say (xyzzy { @^x },[33mâ��[31m: 1, 2, 3)[0mâ�¤    expecting any of:â�¤        postfixâ�¤        statemen…»
17:05 carlin masak: my suggest fix was going to be this: https://github.com/carbin/panda/commit/ff45e43212ab3862c489e60efb527322656247ed
17:06 masak [Coke]: what, wuh...
17:06 [Coke] S02-lexical-conventions/unspace.t
17:06 masak [Coke]: I don't see what that colon even tries to do there.
17:07 masak carlin: yes, I like that as a stopgap solution.
17:07 [Coke] there's a test for that first one, and a few variations of how to write the colon: ":", "\ :", and " :"
17:07 guru joined #perl6
17:07 [Coke] Looks like it was "we are passing a named argument, here, have a colon".
17:07 masak carlin: but the underlying problem is we're treating an interesting problem (paths on different platforms) with blunt tools (strings and regexes)
17:07 [Coke] oh. test says "indirect method calls"
17:08 masak oh!
17:08 [Coke] We don't have those, yes?
17:08 masak and we're augmenting Code. I see.
17:08 masak no, we don't have those, AFAIR.
17:08 xinming joined #perl6
17:08 masak it's the same as would allow the syntax `say $*ERR: 1, 2, 3`
17:08 [Coke] killing test…
17:09 [Coke] m: say $*ERR: 1, 2, 3
17:09 masak [Coke]: um, as in "NYI".
17:09 camelia rakudo-moar bbdcfd: OUTPUT«[31m===[0mSORRY![31m===[0m Error while compiling /tmp/M1gHkcgZzUâ�¤Two terms in a rowâ�¤at /tmp/M1gHkcgZzU:1â�¤------> [32msay $*ERR[33mâ��[31m: 1, 2, 3[0mâ�¤    expecting any of:â�¤        postfixâ�¤        infix stopperâ�¤        infix or meta-in…»
17:09 masak [Coke]: don't know if they should be killed.
17:09 masak niecza: say $*OUT: 1, 2, 3
17:09 [Coke] masak: indirect method calls appear only once in the spec, and not in a way that makes me think they're coming
17:09 camelia niecza v24-109-g48a8de3: OUTPUT«[31m===[0mSORRY![31m===[0mâ�¤â�¤Invocant handling is NYI at /tmp/Y0_UecSxpU line 1 (EOF):â�¤------> [32msay $*OUT: 1, 2, 3[33mâ��[31m<EOL>[0mâ�¤â�¤Unhandled exception: Check failedâ�¤â�¤  at /home/p6eval/niecza/boot/lib/CORE.setting line 1502 …»
17:10 [Coke] ok. I'll open a ticket and let someone else decide. no worries.
17:10 masak [Coke]: I think I recall TimToady wistfully hoping for that feature.
17:10 masak I do note that it doesn't have its own table row on http://perl6.org/compilers/features though.
17:10 * masak adds one
17:12 dalek roast: fd13f17 | coke++ | S02-lexical-conventions/unspace.t:
17:12 dalek roast: link to RT #122343
17:12 dalek roast: review: https://github.com/perl6/roast/commit/fd13f17f59
17:12 dalek roast: 49d414d | coke++ | S02-lexical-conventions/unspace.t:
17:12 dalek roast: Link to RT #122344
17:12 synopsebot Link: https://rt.perl.org/rt3//Public/Bug/Display.html?id=122343
17:12 dalek roast: review: https://github.com/perl6/roast/commit/49d414d334
17:12 synopsebot Link: https://rt.perl.org/rt3//Public/Bug/Display.html?id=122344
17:13 carlin https://github.com/tadzik/panda/pull/91 # there's a pull request for now
17:13 carlin I'm not sure what a IO::Path solution would be at this point
17:15 dalek features: 9847622 | (Carl Mäsak)++ | features.json:
17:15 dalek features: [features.json] add "Indirect method calls"
17:15 dalek features:
17:15 dalek features: Which currently are Not Yet Implemented on Niecza as well as Rakudo, all backends.
17:15 dalek features: review: https://github.com/perl6/features/commit/9847622183
17:17 masak carlin: me neither.
17:17 dalek roast: 75cd875 | coke++ | S02-literals/char-by-name.t:
17:17 dalek roast: link to RT #117683
17:17 dalek roast: review: https://github.com/perl6/roast/commit/75cd875600
17:17 synopsebot Link: https://rt.perl.org/rt3//Public/Bug/Display.html?id=117683
17:17 masak but I think the long-term right solution is to delegate to IO::Path.
17:17 masak because this is a problem that should be solved centrally, not in userland.
17:17 carlin it would be a much nicer solution
17:18 masak in fact, why not have a 'path-to-module' builtin?
17:19 dmol joined #perl6
17:19 * masak admires the pure evil of RT #122325
17:19 * synopsebot Link: https://rt.perl.org/rt3//Public/Bug/Display.html?id=122325
17:19 masak FROGGS++
17:20 masak oh, maybe FROGGS-- didn't originate it; only reported it.
17:20 dalek roast: aa03827 | coke++ | S02-literals/listquote-whitespace.t:
17:20 dalek roast: Link to RT #122285
17:20 dalek roast:
17:20 dalek roast: (this makes it easier to skip these emit lines when looking for TODOs)
17:20 dalek roast: review: https://github.com/perl6/roast/commit/aa03827897
17:20 synopsebot Link: https://rt.perl.org/rt3//Public/Bug/Display.html?id=122285
17:20 masak sorear++ is the author of that test.
17:21 denis_boyun_ joined #perl6
17:21 masak it also says "Test primarily aimed at Niecza", but it seems to have caught a Parrotbug. nice.
17:21 * [Coke] wonders if he can get the non RT/NYI/unspecced fudges down under 800 today.
17:21 SamuraiJack joined #perl6
17:21 guru joined #perl6
17:22 masak r: for 1, 2, 3 { sub foo($x = next) { $x }; say foo }; say "alive"
17:22 camelia rakudo-{jvm,moar} bbdcfd: OUTPUT«alive␤»
17:22 camelia ..rakudo-parrot bbdcfd: OUTPUT«alive␤Nominal type check failed for parameter '$x'; expected Any but got Mu instead␤»
17:23 masak haha, Parrot, you silly.
17:23 masak r: sub { sub foo($x = return) { $x }; say foo }(); say "alive"
17:23 camelia rakudo-jvm bbdcfd: OUTPUT«alive␤»
17:23 camelia ..rakudo-moar bbdcfd: OUTPUT«Attempt to return outside of any Routine␤  in block  at src/gen/m-CORE.setting:556␤  in sub foo at /tmp/tmpfile:1␤  in sub  at /tmp/tmpfile:1␤  in block  at /tmp/tmpfile:1␤␤»
17:23 camelia ..rakudo-parrot bbdcfd: OUTPUT«alive␤Nominal type check failed for parameter '$x'; expected Any but got Mu instead␤»
17:23 masak same trick works in many shapes.
17:23 masak hey, but that's a moarbug right there...
17:23 [Coke] masak: check out https://github.com/perl6/roast/blob/master/S02-names-vars/variables-and-packages.t#L93
17:23 * masak submits moarbug
17:24 masak [Coke]: yeees...?
17:24 [Coke] looks like it's expecting the first call to foo() to be undefined?
17:24 masak because $a is
17:25 masak makes sense to me.
17:25 masak at least that's one way I could've argued it.
17:25 masak (the other way being that foo() returns 0 the first time)
17:25 [Coke] m: say foo; my $a; sub foo() { $a++}
17:25 camelia rakudo-moar bbdcfd: OUTPUT«0␤»
17:26 masak right, that way.
17:26 [Coke] m: say foo, foo, foo; my $a; sub foo() { $a++}
17:26 camelia rakudo-moar bbdcfd: OUTPUT«012␤»
17:26 atroxaper joined #perl6
17:26 [Coke] so if you could argue it either way, what next? :)
17:26 [Coke] spec question?
17:26 [Coke] or let jnthn++ decide?
17:28 [Coke] (there are 3 "nom regressions" in that file which are all basically the same issue)
17:28 FROGGS_ masak: aye, I've only filed RT #122325
17:28 synopsebot Link: https://rt.perl.org/rt3//Public/Bug/Display.html?id=122325
17:29 FROGGS_ eww, t/spec/S02-magicals/args.t fails only in m-spectests here, and is a speshbug :(
17:29 [Coke] masak: opening a ticket on that one.
17:30 masak [Coke]: "spec question" sounds good to me.
17:33 masak [Coke]++ # opening RT tickets for failing tests
17:33 masak I know that can't be fun work, but it's *very useful*.
17:34 dalek roast: 9c096d9 | coke++ | S02-names-vars/variables-and-packages.t:
17:34 dalek roast: link to RT #122346
17:34 dalek roast: review: https://github.com/perl6/roast/commit/9c096d9ebe
17:34 synopsebot Link: https://rt.perl.org/rt3//Public/Bug/Display.html?id=122346
17:35 masak re flapping tests, by the way: http://martinfowler.com/articles/nonDeterminism.html -- required reading. I've always liked this one.
17:38 spider-mario joined #perl6
17:39 Akagi201 joined #perl6
17:39 dalek roast: 5717a50 | coke++ | S02-types/capture.t:
17:39 dalek roast: Reduce scope of skip
17:39 dalek roast: review: https://github.com/perl6/roast/commit/5717a50480
17:40 [Coke] r: my buf8 $buffer = ('A' .. 'Z').map({sprintf('%08d', .ord)}).cat;
17:40 camelia rakudo-{parrot,jvm,moar} bbdcfd: OUTPUT«No such method 'cat' for invocant of type 'List'␤  in block  at /tmp/tmpfile:1␤␤»
17:41 brrt joined #perl6
17:43 [Coke] masak: how do you feel about fudges in files that aren't in t/spectest.data ?
17:44 [Coke] I am leaning towards a "keep it in a branch if you're working on a new feature" policy so that master is "clean". (and I don't spend time trying to fix directives for stuff that is obviously not done.)
17:50 xenoterracide_ joined #perl6
17:51 [Coke] m: my Complex $namcu =1.3; say $namcu.WHAT;
17:51 camelia rakudo-moar bbdcfd: OUTPUT«Type check failed in assignment to '$namcu'; expected 'Complex' but got 'Rat'␤  in block  at /tmp/RwaUOyDQ9D:1␤␤»
17:54 [Coke] m: my Module $a;
17:54 camelia rakudo-moar bbdcfd: OUTPUT«[31m===[0mSORRY![31m===[0mâ�¤Type 'Module' is not declaredâ�¤at /tmp/S0G6RhnFs9:1â�¤------> [32mmy Module[33mâ��[31m $a;[0mâ�¤Malformed myâ�¤at /tmp/S0G6RhnFs9:1â�¤------> [32mmy [33mâ��[31mModule $a;[0mâ�¤    expecting any of:â�¤        scoped…»
17:55 FROGGS_ wth... I closed a bunch of RT tickets and now it grew again :o(
17:55 FROGGS_ eight more tickets
17:56 nbrown joined #perl6
17:56 [Coke] Only about 800 more to go.
17:56 [Coke] (there's lots of stuff with no tickets in roast)
17:57 bowtie joined #perl6
18:01 [Coke] do we have a standard way to refer to github issues? We have "RT #foo" for RTs...
18:01 [Coke] MoarVM #87 ?
18:04 mr-foobar joined #perl6
18:07 denis_boyun joined #perl6
18:09 thou joined #perl6
18:10 Ven joined #perl6
18:11 denis_boyun___ joined #perl6
18:26 Sqirrel joined #perl6
18:31 ventica joined #perl6
18:33 hoverboard joined #perl6
18:40 dalek roast: 9ef729f | coke++ | S02-types/declare.t:
18:40 dalek roast: link to RT; use "NYI"
18:40 dalek roast: review: https://github.com/perl6/roast/commit/9ef729f9f3
18:40 dalek roast: a63adf3 | coke++ | S32-str/lc.t:
18:40 dalek roast: use <system> #<number>
18:40 dalek roast: review: https://github.com/perl6/roast/commit/a63adf3587
18:47 masak [Coke]: keeping master clean is something I'm 100% on board with.
18:47 masak [Coke]: fudges in un-spectest.data'd files -- no particular opinion.
18:48 [Coke] ok. Plenty of other stuff for me to clean in the meantime.
18:49 [Coke] er:
18:49 [Coke] m: "what the" ~~ / (<[a..z]>)+ <ws> (\d+)? <ws> (<[a..z]>)+ / ; say $/;
18:49 camelia rakudo-moar bbdcfd: OUTPUT«No such method 'from' for invocant of type 'Any'␤  in method caps at src/gen/m-CORE.setting:12942␤  in method gist at src/gen/m-CORE.setting:12973␤  in sub say at src/gen/m-CORE.setting:13793␤  in block  at /tmp/KdkIh8lqiA:1␤␤»
18:50 [Coke] the +'s are in the wrong place, but that seems like an odd error.
18:50 [Coke] (same error if you move them inside the ()'s, so that's not it..
18:50 FROGGS_ now that is weird
18:52 Ven joined #perl6
18:52 FROGGS_ m: "what the" ~~ / (<[a..z]>)+ <ws> (\d+)? <ws> (<[a..z]>)+ / ; say $/.perl
18:52 camelia rakudo-moar bbdcfd: OUTPUT«Match.new(orig => "what the", from => 0, to => 8, ast => Any, list => ([Match.new(orig => "what the", from => 0, to => 1, ast => Any, list => ().list, hash => EnumMap.new()), Match.new(orig => "what the", from => 1, to => 2, ast => Any, list => ().list, ha…»
19:01 * [Coke] wonders why S01-perl-5-integration/basic.t is in t/spectest.data - we skip everything but a few simple assignments.
19:02 [Coke] (everything is fudged AND skip_rest'd)
19:03 [Coke] FROGGS_: that looks like it's your file. How to proceed? (leave it alone? remove from the list of things to test..?)
19:04 FROGGS_ I'd rip it out because I that is something fits in star but not the compiler itself
19:05 zakharyas joined #perl6
19:09 Akagi201 joined #perl6
19:11 * Ven wrote a little something about ff vs fff, but still think it might be confusing
19:15 denis_boyun_ joined #perl6
19:16 Ven timotimo: erm, are the `orelse` tests okay now :P ?
19:16 dalek rakudo/nom: a8cfa19 | coke++ | t/spectest.data:
19:16 dalek rakudo/nom: Remove file that runs no tests.
19:16 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/a8cfa19d39
19:18 Ven 600 lines already, uh.
19:23 denis_boyun_ joined #perl6
19:35 avuserow [Coke]++ # unending test improvements
19:36 FROGGS_ masak: I've got a patch for RT #122332
19:36 synopsebot Link: https://rt.perl.org/rt3//Public/Bug/Display.html?id=122332
19:37 [Coke] avuserow: I think they are ending. :)
19:38 Ven It's "fun" to read RTs from 2008, even back then it was really different
19:38 masak FROGGS_: nice!
19:41 dalek roast: f35b982 | (Tobias Leich)++ | S13-overloading/metaoperators.t:
19:41 dalek roast: RT #122332 test for sub as infix op between []
19:41 dalek roast: review: https://github.com/perl6/roast/commit/f35b982ff4
19:41 synopsebot Link: https://rt.perl.org/rt3//Public/Bug/Display.html?id=122332
19:41 Ven (reading stuff from 2002/2003 perl monks posts and seeing now ... well!)
19:44 denis_boyun_ joined #perl6
19:45 masak ooh, perlmonks posts from 2002/2003? what were people saying? :)
19:46 Ven well, I think I linked it back then (I was barely starting on perl 6, and either you or moritz++ pointed out it was from 2003 ..)
19:46 masak ok. :) I don't remember that.
19:46 Ven I can probably find it again, though.
19:46 masak would be interesting, if it's not too onerous.
19:47 Ven (can you filter by year on google ?)
19:47 masak no idea.
19:47 Ven http://www.perlmonks.org/index.pl?node_id=253366 found it :)
19:47 * masak looks
19:48 Ven ha, it still had the diamond :P
19:48 masak oh, I remember using the diamond, I think.
19:48 Ven it had ??::, $x+(-1)
19:48 masak it was cute that =<> looked like a fish.
19:48 Ven damn, that keyboard's enter is not where I think it is.
19:48 masak or a jellyfish or something.
19:49 kaare_ joined #perl6
19:49 Ven but it already had any/none back then ... Other languages are so far behind :P
19:49 dalek rakudo/nom: fc0cbb1 | (Tobias Leich)++ | src/Perl6/ (2 files):
19:49 dalek rakudo/nom: fix sub as infix op like in `2 [&add] 4`
19:49 dalek rakudo/nom:
19:49 dalek rakudo/nom: We have to wrap it in another QAST::Op, so that additional arguments to
19:49 dalek rakudo/nom: the infix sub will be pushed to the added QAST::Op, and not to the QAST::Op
19:49 dalek rakudo/nom: that does the lookup of the &add sub we wanna call.
19:49 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/fc0cbb138c
19:52 Ven also, http://www.perl.com/pub/2003/04/p6pdigest/20030427.html
19:52 avuserow what is the state of moarvm support for star? I'm poking around with a star release and wondering if I can do my testing on moarvm first for speed, and then parrot (and jvm if needed?)
19:52 Ven avuserow: I think moar is in
19:53 avuserow (poking around with trying to *make* a 2014.07 star release)
19:54 zengargoyle joined #perl6
19:55 masak "Of cours"Of course as Perl 6 doesn't exist yet [...]" -- it *did*, but the PR efforts from Parrot's side were nonexistent.
19:57 * masak does a quick translation of that script to Today's Perl 6
19:58 thou joined #perl6
20:08 Ven joined #perl6
20:10 Akagi201 joined #perl6
20:11 dalek rakudo/nom: 9a9f357 | (Pepe Schwarz)++ | src/Perl6/Grammar.nqp:
20:11 dalek rakudo/nom: Added reason "Missing semicolon" for RT #75230.
20:11 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/9a9f357fea
20:11 dalek rakudo/nom: aa15259 | (Tobias Leich)++ | src/Perl6/Grammar.nqp:
20:11 dalek rakudo/nom: Merge pull request #291 from peschwa/rt75230
20:11 synopsebot Link: https://rt.perl.org/rt3//Public/Bug/Display.html?id=75230
20:11 dalek rakudo/nom:
20:11 dalek rakudo/nom: Added reason "Missing semicolon" for RT #75230.
20:11 synopsebot Link: https://rt.perl.org/rt3//Public/Bug/Display.html?id=75230
20:11 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/aa15259244
20:14 dalek roast: 6811e52 | (Pepe Schwarz)++ | S32-exceptions/misc.t:
20:14 dalek roast: Match the reason, compare Rakudo PR #291 and commit aa15259.
20:14 dalek roast: review: https://github.com/perl6/roast/commit/6811e5268e
20:14 psch thanks FROGGS++
20:15 FROGGS_ ahh, I was about to adjust the test :o)
20:15 masak ok, here's my translation: https://gist.github.com/masak/f74ce61b40ea996b9583
20:15 masak haven't tried to run it yet -- trying now :)
20:15 psch FROGGS_: you can update the ticket, if you want :)
20:16 FROGGS_ psch: was about to ask that :o)
20:16 FROGGS_ psch: done
20:19 FROGGS_ .u a
20:19 yoleaux U+0061 LATIN SMALL LETTER A [Ll] (a)
20:21 masak ok, it runs now, a couple of changes later.
20:21 masak it doesn't advance the board, though.
20:22 Ven (why can't we have package foo; like in p5 ?)
20:22 * Ven is reading S10 to explain modules in his learnxiny
20:23 Ven masak: you decided to go with fixed pos, uh ?
20:24 FROGGS_ Ven: package foo; is disallowed to detect P5 early
20:24 Ven (instead of for ^$.dimension->$x ... for ^$.dimension->$y, why not ^$.dimension X ^$.dimension ?)
20:24 masak no real reason.
20:24 masak just staying close to the original on that one.
20:24 Ven FROGGS_: seems like something quite useful, though ? maybe allow it if there's a use v6 ?
20:25 masak Ven: I think X still has a bit of a performance penalty, too. not 100% sure, though.
20:25 Ven masak: fwiw, https://github.com/Nami-Doc/GameOfLife/blob/master/gol.pl, my first perl 6 script
20:25 Ven when I first ran it, it ran at maybe 1 fps :P
20:26 FROGGS_ Ven: you can still use module Foo; or class Foo;
20:26 masak ok, now it works.
20:26 Ven FROGGS_: why have both package and module, then ?
20:26 Ven (it looks like s10 is missing some bits)
20:26 FROGGS_ m: package Foo { }
20:26 camelia rakudo-moar bbdcfd: ( no output )
20:26 FROGGS_ I dunno, really
20:26 Ven uh
20:27 [Coke] I think blaming "the pr efforts from parrot's side" is... not right.
20:27 masak having both package and module always confused me, too.
20:28 Ven so, humm, nobody deep into perl 6 really knows why :P ? *waits for somebody else to pop up with the answer*
20:29 masak I think 'package' is there mostly for side-compat with Perl 5.
20:29 Ven makes sense. But then, I'll probably avoid explaining it.
20:30 dalek roast: 425f65c | (David Warring david.warring@gmail.com)++ | / (2 files):
20:30 dalek roast: Corrected and unfudged tests. Added proto to multisub example.
20:30 dalek roast: review: https://github.com/perl6/roast/commit/425f65cabc
20:30 masak sounds good to me.
20:33 tgt joined #perl6
20:34 avuserow joined #perl6
20:34 masak Ven: very interesting reading. I read the rest of the thread too.
20:37 dalek nqp: fc99946 | (Tobias Leich)++ | src/QRegex/P6Regex/Actions.nqp:
20:37 dalek nqp: RT #122245 fix \C[...] (negated \c)
20:37 dalek nqp: review: https://github.com/perl6/nqp/commit/fc99946fee
20:37 synopsebot Link: https://rt.perl.org/rt3//Public/Bug/Display.html?id=122245
20:47 virtualsue left #perl6
20:49 virtualsue joined #perl6
20:50 dalek roast: 9e3a259 | (David Warring david.warring@gmail.com)++ | integration/advent2009-day14.t:
20:50 dalek roast: is() => is_deeply() on type tests
20:50 dalek roast: review: https://github.com/perl6/roast/commit/9e3a2594be
20:55 dalek rakudo/nom: 97882c3 | (Tobias Leich)++ | tools/build/NQP_REVISION:
20:55 dalek rakudo/nom: bump nqp rev for <[\C[...]]> fix
20:55 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/97882c3191
20:56 dalek roast: 29abad5 | (Tobias Leich)++ | S05-mass/named-chars.t:
20:56 dalek roast: RT #122245 unfudge now passing \C tests
20:56 dalek roast: review: https://github.com/perl6/roast/commit/29abad5d52
20:56 synopsebot Link: https://rt.perl.org/rt3//Public/Bug/Display.html?id=122245
20:57 Ven ooh, around 66 fudges ! FROGGS_ ++
20:57 psch m: say "abc" ~~ s/c$// # uhm...
20:57 camelia rakudo-moar fc0cbb: OUTPUT«Cannot modify an immutable Str␤  in sub infix:<=> at src/gen/m-CORE.setting:16978␤  in block  at /tmp/BP7SmnVwsQ:1␤␤»
20:57 * psch duhs
20:57 psch m: my $a ="abc"; say $a ~~ s/c$// # uhm...
20:57 camelia rakudo-moar fc0cbb: OUTPUT«True␤»
20:58 Ven psch: well, you clearly assigned to this poor `"abc"`
20:58 psch m: my $a ="abc"; $a ~~ s/c$//; say $a # uhm...
20:58 camelia rakudo-moar fc0cbb: OUTPUT«ab␤»
20:58 psch that's curious
20:58 psch m: my $a ="abc"; $a ~~ s/c$/i/; say $a
20:58 camelia rakudo-moar fc0cbb: OUTPUT«[31m===[0mSORRY![31m===[0m Error while compiling /tmp/VYmw2fyuzcâ�¤Malformed replacement part; couldn't find final /â�¤at /tmp/VYmw2fyuzc:1â�¤------> [32mmy $a ="abc"; $a ~~ s/c$/i/; say $a[33mâ��[31m<EOL>[0mâ�¤    expecting any of:â�¤        postf…»
20:58 psch aha
20:59 psch suddenly $/
20:59 psch m: my $a ="abc"; $a ~~ s/c$ /i/; say $a
20:59 camelia rakudo-moar fc0cbb: OUTPUT«abi␤»
21:01 psch for reference, that came from RT #82108 but clearly has another problem than the one in the ticket
21:01 synopsebot Link: https://rt.perl.org/rt3//Public/Bug/Display.html?id=82108
21:02 dalek roast: 2946712 | (David Warring david.warring@gmail.com)++ | integration/advent2009-day17.t:
21:02 dalek roast: tweak. 'our' no longer needed
21:02 dalek roast: review: https://github.com/perl6/roast/commit/2946712f21
21:03 akaseki joined #perl6
21:08 nbrown joined #perl6
21:19 masak m: my $a = "abc"; say $a ~~ /c$/
21:19 camelia rakudo-moar fc0cbb: OUTPUT«「c」␤␤»
21:19 masak arguably, that's a bit inconsistent.
21:19 masak if // can handle $ appearing right by the end of the pattern, why not s/// ?
21:28 psch m: my $a = 'abc'; $a ~~ s/c$/ /; say $a.perl # this is fine too
21:28 camelia rakudo-moar fc0cbb: OUTPUT«"ab "␤»
21:29 psch masak: i agree, it should work
21:32 masak psch: feel free to submit a rakudobug about it.
21:32 SevenWolf joined #perl6
21:34 psch masak: i'll do that; i'll look at it a bit more, but i doubt i'll get it today; and even if, having a ticket is better than not
21:35 masak aye -- ticket regardless ;)
21:46 thou joined #perl6
21:57 kurahaupo joined #perl6
22:19 masak 'night, #perl6
22:26 hoverboard joined #perl6
22:27 nbrown joined #perl6
22:31 atroxaper joined #perl6
22:39 aoseki joined #perl6
22:46 [Coke] m: 0.1234567891011121314151617181920.Rat.say
22:46 camelia rakudo-moar 97882c: OUTPUT«0.123456789101112129891646418400␤»
22:47 [Coke] m: 0.1234567891011121314151617181920.Rat.WHAT.say
22:47 camelia rakudo-moar 97882c: OUTPUT«(Rat)␤»
22:47 [Coke] m: 0.1234567891011121314151617181920.Rat.perl.say
22:47 camelia rakudo-moar 97882c: OUTPUT«0.123456789101112129891646418400␤»
22:48 Ulti_ what's the difference between str and Str?
22:48 Ulti_ other than one is native :P
22:48 Ulti_ how is str represented differently?
22:49 [Coke] m: my str $hi;
22:49 camelia rakudo-moar 97882c: ( no output )
22:49 [Coke] huh, I was guessing NYI
22:51 Ulti_ http://jnthn.net/perl6/bench/2014-07-13.html it's in some of the code examples in the benchmark of jnthns string magic
22:51 [Coke] in general, natives are immutable and are represented with native data types. (I am not sure if str follows that latter)
22:52 Ulti_ well the value might not be mutable but the variable is getting changed
22:52 Ulti_ for_concat_2_native vs the non native one
22:53 Ulti_ I have some tests that take 35 seconds atm that might get suddenly fast if just switching to 'str' instead of 'Str' works out
22:54 labster joined #perl6
22:54 Ulti_ though the concatenation isn't in a for loop but I can just make it like that
23:06 dalek roast: def56cb | (David Warring david.warring@gmail.com)++ | integration/advent2009-day18.t:
23:06 dalek roast: enable/unfudge parametric role tests
23:06 dalek roast: review: https://github.com/perl6/roast/commit/def56cbf16
23:34 thou joined #perl6
23:36 timotimo oh, huh. we haven't had a Rakudo Star release since 2014.04?
23:45 nbrown_ joined #perl6
23:50 xenoterracide_ joined #perl6
23:51 akaseki joined #perl6
23:57 nbrown joined #perl6

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

Perl 6 | Reference Documentation | Rakudo