phatbot m: my %h = 42 => Mu 00:08
camelia ( no output )
phatbot m: my %h{Int} = 42 => Mu
camelia rakudo-moar 7b84ad: OUTPUT«Type check failed in binding to x; expected Any but got Mu (Mu)␤ in block <unit> at <tmp> line 1␤␤»
phatbot Kinda odd normal hash defaults to Mu values, but parametarizing keys make it default to Any
dalek kudo/nom: 003e654 | (Zoffix Znet)++ | src/core/Hash.pm:
Fix broken .perl of parametarized hash with no keys

Currently, .perl output of a parametarized hash with no keys leaves a stray `=` at the end. Fix by checking whether hash contains stuff.		 00:13
ast: d7d0cef | (Zoffix Znet)++ | S32-hash/perl.t:
Test .perl.EVAL roundtripness of parametarized hashes

that have no values in em
Rakudo fix: github.com/rakudo/rakudo/commit/003e654dd6		 00:21
kudo/nom: a5dc98a | (Zoffix Znet)++ | src/core/Proc/Async.pm:
Minor source formatting fix
 00:59
ast: 3b9eae1 | samcv++ | S15-unicode-information/uniprop.t:
uniprop: add tests for Emoji property
 01:16
ast: 187fb30 | samcv++ | S15-unicode-information/uniprop.t:
uniprop: add tests for Emoji_Modifier, Emoji_All and Emoji_Modifier
 02:08
MasterDuke .tell jnthn explicitly putting a QAST::Op with a say in the while body prints (what i gave the say) for every line of the input file. but 'say "hi"' in the -ne doesn't ever print (not does .say) 03:08
yoleaux2 MasterDuke: I'll pass your message to jnthn.
samcv .tell yoleaux2 hi 03:47
yoleaux2 samcv: Thanks for the message.
samcv you're welcome
MasterDuke .tell jnthn and -p doesn't print the file contents 04:44
yoleaux2 MasterDuke: I'll pass your message to jnthn.
lizmat Files=1155, Tests=53918, 200 wallclock secs (11.19 usr 4.44 sys + 1184.84 cusr 127.90 csys = 1328.37 CPU) 07:26
[Tux] This is Rakudo version 2016.11-233-ga5dc98abc built on MoarVM version 2016.11-44-g8e24145d 07:32
csv-ip5xs 3.464
test 15.716
test-t 6.389
csv-parser 14.471
dalek kudo/nom: 2a8ec40 | samcv++ | / (2 files):
uniprop: Return correct values for na, uc, tc, lc properties

Name, Uppercase_Mapping, Titlecase_Mapping, Lowercase_Mapping		 08:15
kudo/nom: 9ff5b7e | samcv++ | / (2 files):
uniprop: have 'Numeric_Value' return the numeric value

as a number not a string. Call unival routine when the 'Numeric_Value' is requested
rakudo/nom: 7b09bee | lizmat++ | / (2 files):
rakudo/nom: Merge pull request #953 from samcv/uniprop-na-lc-uc-tc
rakudo/nom:
ast: bc74e18 | lizmat++ | S15-unicode-information/uniprop.t:
Unfudge now passing tests
 08:31
lizmat afk&
dalek p: 8c1c322 | (Pawel Murias)++ | src/vm/js/ (7 files):
[js] Repossess STables.
 11:12
p: ce84ef5 | (Pawel Murias)++ | src/vm/js/nqp-runtime/serialization.js:
[js] Fix stuff found by make js-lint.
kudo/nom: 8eef234 | (Zoffix Znet)++ | src/core/Slip.pm:
Fix broken .perl for itemized Slip

The output of Slip.perl is nothing more than `slip` prepended to output of List.perl. This creates a problem where an itemized Slip's
  .perl ends up being `slip$("a", "b")`, which fails to .EVAL
Fix by checking whether output of List.perl is itemized and outputting an itemized slip `$(slip("a", "b"))` in that case.		 14:51
ast: 7eea834 | (Zoffix Znet)++ | S07-slip/slip.t:
Test itemized Slips roundtrip .perl.EVAL

Rakudo fix: github.com/rakudo/rakudo/commit/8eef234a5b441a		 14:55
lucasb m: say slip().item.perl 15:12
camelia rakudo-moar 8eef23: OUTPUT«$(slip())␤»
lucasb Zoffix++ fixed
RT #127224 15:13
synopsebot6 Link: rt.perl.org/rt3//Public/Bug/Displa...?id=127224
lucasb ^^ can be closed
|Tux| notes RT#122892 RT#123597 RT#123980 RT#61602 RT#127358 are all still "red" in my test environment 15:43
dalek kudo/nom: a78d330 | jnthn++ | src/vm/moar/ops/container.c:
Missing MVMROOT in late-bound type check.

Needed because of mutex acquisition, and thus possible GC trigger, by lazy deserialization of HOW and method cache.		 16:51
lizmat nine: t/invoke_p6_method_from_p5.t, test 3 fails for me atm, which block installing Inline::Perl5 17:53
# got: undef
# expected: 'pos1barbaz'
nine lizmat: that's a MoarVM regression. Either upgrade or downgrade MoarVM please 17:56
lizmat ah, will it be fixed before the release ? 17:57
I can then wait :-)
pyrimidine nine: yep, that was the MoarVM I bisected a few days back 18:00
yoleaux2 9 Dec 2016 12:31Z <nine> pyrimidine: Thanks a lot! Will dig from there.
nine lizmat: I'm actually a little surprised that no one has bumped Moar since we reverted the offending commit 18:01
lizmat well, maybe jnthn isn't comfortable with it yet ?
[Coke] NeuralAnomaly: help
NeuralAnomaly [Coke], stats | blockers
[Coke] NeuralAnomaly: stats 18:02
NeuralAnomaly [Coke], [✘] Next release will be in 2 days and 11 hours. Since last release, there are 50 new still-open tickets (5 unreviewed and 0 blockers) and 55 unreviewed commits. See perl6.fail/release/stats for details
dalek p: d72b104 | coke++ | tools/build/MOAR_REVISION:
bump to get latest moar
 18:03
pyrimidine nine: maybe a bump is on the way? ^^
dalek kudo/nom: be2ed02 | coke++ | tools/build/NQP_REVISION:
bump to get latest nqp/moar
 18:04
[Coke] I mean, with a release coming anyway, want at least a few days to test it out.
nine Coke++ 18:05
lizmat pulls, configures, builds and tests 18:07
pyrimidine nine: installing Inline::Perl5 from latest works for me, though I see a warning 18:13
gist.github.com/cjfields/f0514bae7...5e61aaeee2 18:14
lizmat nine: installed ok the 2nd time I tried 18:16
# Failed test 'at least one destructor ran' 18:17
# at t/p6_object_destructor.t line 1
dalek p: c79ec74 | (Pawel Murias)++ | src/vm/js/nqp-runtime/deserialization.js:
[js] Refactor deserialization a bit by splitting up parts of into methods.
 18:28
p: a33c51c | (Pawel Murias)++ | src/vm/js/ (6 files):
[js] Simple repossesion of objects on native attribute change works.
p: f15aca6 | (Pawel Murias)++ | t/serialization/04-repossession.t:
Test for basic repossession.
p: 539166f | (Pawel Murias)++ | src/vm/js/nqp-runtime/repossession.js:
[js] Fix nqp::scwbdisable/nqp::scwbenable.
nqp: 235b980 | (Pawel Murias)++ | t/serialization/04-repossession.t:
nqp: [js] Test nqp::scwbenable/nqp::scwbdisable.
nine lizmat: that's a flapper :/ 18:29
lizmat nine: yeah, I figured 18:30
nine lizmat: sometimes MoarVM just doesn't want to collect garbage
lizmat nine: not even with nqp::force_gc ?
perhaps an "END nqp::force_gc" might help ? 18:31
samcv [Coke], is it ok if i disable fuzzy search on the docs for now? on my computer it's making the search unusable half the time, not giving me any results or taking like 10 seconds sometimes, and using all my cpu
nine lizmat: the test already does a force_gc
lizmat nine: intriguing 18:32
samcv it's really affecting my productivity wrt the docs
lizmat nine: I don't see a force_gc in t/p6_object_destructor.t ? 18:33
nine ooooooh.....good hint. That's what I get for naming those test files so similarily 18:36
dalek rakudo/nom: 287af6a | (Zoffix Znet)++ | src/core/operators.pm: 18:44
rakudo/nom: Fix andthen → orelse chaining
rakudo/nom:
rakudo/nom: Fixes RT#130034 rt.perl.org/Ticket/Display.html?id=130034
rakudo/nom:
synopsebot6 Link: rt.perl.org/rt3//Public/Bug/Displa...?id=130034
dalek rakudo/nom: Per TimToady[^1], `andthen` returning an Empty is by design, and the
rakudo/nom: behaviour seen in the ticket is due to the RHS `orelse` thunk not being
rakudo/nom: executed but returned as is. Since `orelse` isn't defined as executing any
rakudo/nom: Callables on LHS, we need to somehow figure out when we get the Empty
rakudo/nom: (which vanishes in our parameters when we receive it). So I added a check
rakudo/nom: for whether we got just 1 arg (our RHS thunk) and then eval it if it's
rakudo/nom: a Callable. So the behaviour is that orelse with 1 arg evals its callables.
[Coke] samcv: Ask phatbot, he's the one that merged it. :) Yes, it's fine if you revert those commits for now.
samcv ok thanks :)
dalek ast: 8d6a639 | (Zoffix Znet)++ | S03-operators/orelse.t:
Test andthen chains with orelse

RT#130034: rt.perl.org/Ticket/Display.html?id=130034 Rakudo fix: github.com/rakudo/rakudo/commit/287af6a37e00ad		 18:45
synopsebot6 Link: rt.perl.org/rt3//Public/Bug/Displa...?id=130034
kudo/nom: e327f30 | (Zoffix Znet)++ | t/05-messages/01-errors.t:
Add t/05-messages test directory

I regularly come across some tickets where we need a test (to ensure we fixed the problem), yet that test should not be part of the Perl 6 Specification because that would control definition of what exact wording of a message should be, for example.
So I started this directory in Rakudo's test suite for such tests.
N.B.: I realize this isn't being run by make test yet, will fix soon.		 18:53
ast: 53091aa | (Zoffix Znet)++ | integration/error-reporting.t:
Remove curly/smart quotes test

This test was added 2 days ago as part of a fix for
  rt.perl.org/Ticket/Display.html?id=130260
IMO Perl 6 language should not define such minute details as whether curly quotes are called curly in displayed user messages.
Moving this test to Rakudo's t/05-messages/01-errors.t		 18:58
kudo/nom: 3df4310 | (Zoffix Znet)++ | t/05-messages/01-errors.t:
Test we call curly quotes curly and not smart

Moved from roast.
Roast entry:
  github.com/perl6/roast/commit/f305...abdb1d78e8
Roast exit:
  github.com/perl6/roast/commit/5309...4155f57523
JimmyZ jnthn: do we need to add misssing MVMROOT to moarvm where uses MVM_6model_find_method_cache_only too? 19:33
lizmat gives in to the flu again 19:34
afk&
masak lizmat: get well soon! 20:16
jnthn JimmyZ: Yes, would be wise. 20:38
dalek kudo/nom: 7345e08 | (Zoffix Znet)++ | src/Perl6/ (2 files):
Fix heredoc trimming warnings referencing guts

Fixes RT#129838: rt.perl.org/Ticket/Display.html?id=129838		 20:52
synopsebot6 Link: rt.perl.org/rt3//Public/Bug/Displa...?id=129838
dalek ast: 171d5c6 | (Zoffix Znet)++ | S02-literals/heredocs.t:
Test heredoc trimming warnings don't reference guts

RT#129838: rt.perl.org/Ticket/Display.html?id=129838 Rakudo fix: github.com/rakudo/rakudo/commit/7345e08c9b2fad
synopsebot6 Link: rt.perl.org/rt3//Public/Bug/Displa...?id=129838
jnthn m: $*SCHEDULER.cue: :in(1), { say now }; sleep 3; 21:16
camelia rakudo-moar 7345e0: OUTPUT«Instant:1481750209.871443␤»
jnthn m: $*SCHEDULER.cue: :in(1), :times(1), { say now }; sleep 3;
camelia rakudo-moar 7345e0: OUTPUT«Instant:1481750219.818066␤»
jnthn m: $*SCHEDULER.cue: :in(1), :times(2), { say now }; sleep 3;
camelia rakudo-moar 7345e0: OUTPUT«Instant:1481750228.830035␤Instant:1481750229.831004␤»
jnthn .ask lizmat Is it accidental that `$*SCHEDULER.cue: :in(1), :times(2), { say now }` only applies the "in" to the first time? (It's actually setting the delay to 0 after scheduling the first one...) 21:17
yoleaux2 jnthn: I'll pass your message to lizmat.

Powered by Cro and the Raku® Programming Language.

Please report any issues / comments / feature requests as an issue on App::Raku::Log.

Thank you!