IOninja .ask lizmat any ideas on how to address the `eager` being ignored here? irclog.perlgeek.de/moarvm/2017-03-18#i_14284447 00:17
yoleaux2 IOninja: I'll pass your message to lizmat.
IOninja .ask nine would you be able to update 6.c-errata/S11-modules/require.t to pass with all the latest changes? I figure you know more about what needs to be changed instead of me trying to cherry-pick commits 00:21
yoleaux2 IOninja: I'll pass your message to nine.
IOninja The new 5 tests with `ligature st` in nqp fail on JVM: gist.github.com/zoffixznet/043877b...ac64e97c6f 00:39
Added in github.com/perl6/nqp/commit/9ab92d...54fed912d9 and github.com/perl6/nqp/commit/ccb549...4bfa242c35
Gonna remove them during release and then re-add afterwards.
If I don't hear from lizmat by the time I cut the release, gonna exclude the github.com/rakudo/rakudo/commit/1754dc5f7e from release. 00:40
And if I don't hear from nine, gonna try to cherry-pick the master commits to S11-modules/require.t into 6.c-errata 00:41
And will be cutting the release in about 14 hours from now.
IOninja & for the night
samcv we've removed pugs references from roast, can we remove them from nqp tests too? 01:24
IOninja I chalked up upgrade instructions for the lexical require thing. I hope I got the explanations right. If not, please edit: rakudo.org/2017/03/18/lexical-requi...rade-info/ 01:30
samcv it's way easier to code in nqp in nqp than in perl 6 lol 01:35
can actually code like a normal person
Geth nqp: 719944a1fc | (Samantha McVey)++ | t/moar/08-indexic.t
[moar] Add tests for new indexic op
 01:55
roast: FCO++ created pull request #256:
Add tests for #130940		 04:00
synopsebot6 Link: rt.perl.org/rt3//Public/Bug/Displa...?id=130940
Geth rakudo: FCO++ created pull request #1042:
Fix infinite loop #130940
synopsebot6 Link: rt.perl.org/rt3//Public/Bug/Displa...?id=130940
samcv NeuralAnomaly, status 04:09
NeuralAnomaly samcv, [✘] Next release is today. Since last release, there are 60 new still-open tickets (0 unreviewed and 0 blockers) and 13 unreviewed commits. See perl6.fail/release/stats for details
samcv today. that is so vague 04:23
before midnight UTC? 04:24
geekosaur [18 00:41:41] <IOninja> And will be cutting the release in about 14 hours from now. 06:22
won't be the first time a scheduled release slipped by a day
although, UTC, it is probably correct for that schedule :)
samcv thanks geekosaur 07:08
gonna watch Iron Fist, just came out on netflix 07:10
nine IOninja: on it 08:14
yoleaux2 00:21Z <IOninja> nine: would you be able to update 6.c-errata/S11-modules/require.t to pass with all the latest changes? I figure you know more about what needs to be changed instead of me trying to cherry-pick commits
Geth roast/6.c-errata: 4dcbbb9097 | (Stefan Seifert)++ | S11-modules/require.t
Bring lexical module loading to "require"

Like "use" and "need", "require" should put the loaded packages into lexical scope. Otherwise we could never load multiple versions of a module in the same program.
This means that we cannot transitively expose require'd symbols. ... (12 more lines)		 09:08
nine IOninja: excellent upgrade instructions! The only part I'm not 100 % sure about is the test for Nil in the last example. I'm not sure if that's specced or just an implementation detail. I don't think we have a test for the Nil result in roast. 09:34
RabidGravy I guess I'd better get testing stuff again 09:36
nine IOninja: ah, never mind. The Nil is the result of the try and as such should be well specced. 09:40
lizmat IOninja: am available now, looking 09:54
yoleaux2 00:17Z <IOninja> lizmat: any ideas on how to address the `eager` being ignored here? irclog.perlgeek.de/moarvm/2017-03-18#i_14284447
[Tux] This is Rakudo version 2017.02-261-g16ef21c16 built on MoarVM version 2017.02-39-gd7caeba3 09:56
csv-ip5xs 3.098
test 12.545
test-t 5.028 - 5.471
csv-parser 12.978
Geth rakudo/nom: 3041072578 | (Elizabeth Mattijsen)++ | src/core/List.pm
Revert "Make thunk xx 42 use the new iterator"

This reverts commit 1754dc5f7e2dc29d18653b131dde6056f16c21fc.
Also causing problems with eagerness, as shown in:
   irclog.perlgeek.de/moarvm/2017-03-18#i_14284447
So put past the 2017.03 release.		 09:58
lizmat IOninja: ^^^
samcv cool fixed the nqp::indexic todo's just now. Not important to release because we don't use it yet. But that's nice it's passing all the tests i've written 10:24
Geth roast: 6e56c434f4 | (Elizabeth Mattijsen)++ | 2 files
Unfudge now again passing tests: RT #130980
 10:30
synopsebot6 Link: rt.perl.org/rt3//Public/Bug/Displa...?id=130980
samcv need to figure out what code needs to actually change though. somewhere in QASTRegexCompilierMAST.nqp maybe.. hm 10:34
Geth roast: 289f08736d | usev6++ | S06-advanced/wrap.t
[JVM] Add ticket number to fudged test
 10:56
gfldex m: multi sub dir($path = '.', :$some-named){}; say dir 13:29
camelia ===SORRY!===
Circularity detected in multi sub types for &dir
gfldex m: multi sub dir($path, :$some-named){}; say dir
camelia (".cpanm".IO ".local".IO ".npm".IO ".perl6".IO ".perlbrew".IO ".rcc".IO ".ssh".IO "Perlito".IO "evalbot".IO "log".IO "nqp-js".IO "p1".IO "p2".IO "perl5".IO "std".IO ".bash_history".IO ".bashrc".IO "mbox".IO ".lesshst".IO "evalbot.log".IO ".cpan".IO "dale…
gfldex that makes overloading &dir quite difficult 13:30
IOninja cool didn't know we had any sort of circularity detection 13:34
timotimo must be something IOninja built
IOninja Nope
gfldex and IO::Path.dir is an only, so no way to augment 13:35
IOninja samcv: I think NeuralAnomaly thinks release happens on Saturday 00:00:01 EST. Normal time is around 11am EST. 13:41
gfldex is gather supposed to work inside a wrapper? 13:58
timotimo wrapper as in &func.wrap? 14:00
gfldex yes
timotimo i don't immediately see a reason against it
gfldex it's actually Method.wrap
I shall golf later.
timotimo what kind of explosion do you observe? 14:01
gfldex I get a whole type object back 14:03
timotimo oh, so not exploded 14:04
gfldex i do `gather for callsame { .take }` and no matter how the list is that is returned by callsame, I only ever get a single type object 14:05
*how long
timotimo interesting that that works without () after callsame; is that a term?
gfldex luckily there is .map and .Slip, so I got my recursive &dir working now :)
it does need the (), I just forgot 14:06
timotimo OK
MasterDuke is this correct? github.com/rakudo/rakudo/blob/nom/...uf.pm#L712 14:17
`my int $mask = 0xFFFFFFFFFFFFFFFF;`
timotimo given that blobs can have different sizes assigned to them ... maybe not? 14:18
probably depends on how assigning back into the blob works
MasterDuke well, 0xFFFFFFFFFFFFFFFF is bigger than an int can hold 14:19
timotimo oops :)
my eyes just glazed over the exact amount of Fs
.o( look at all the Fs i give )
MasterDuke so is it better to make that smaller? or just make it a uint? 14:20
Geth rakudo/nom: dd3f22875c | (Zoffix Znet)++ | docs/ChangeLog
Log all commits to date

Documents commits: e1ebb50 f94a2c7 3de5fb2 79bb179 65b0040 10f5f74 9644fc3		 14:22
rakudo/nom: 53b38789ee | (Zoffix Znet)++ | CREDITS
Remove duplicate entry
 14:30
rakudo: FCO++ created pull request #1043:
Fix #128977 and #130954		 14:33
synopsebot6 Link: rt.perl.org/rt3//Public/Bug/Displa...?id=128977
Link: rt.perl.org/rt3//Public/Bug/Displa...?id=130954
Geth roast: FCO++ created pull request #257:
Tests for #128977 and #130954
synopsebot6 Link: rt.perl.org/rt3//Public/Bug/Displa...?id=128977
Link: rt.perl.org/rt3//Public/Bug/Displa...?id=130954
IOninja Let's wait 'till past release before merging anything else. I'm about to start. 14:34
SmokeMachine IOninja: should I stop PRing or can I PR and it gets merged (if it's the case) after the release? 14:38
IOninja SmokeMachine: you can PR 14:39
SmokeMachine IOninja: ok! 14:40
Geth nqp: 6607ecbdc9 | (Zoffix Znet)++ | t/qregex/rx_modifiers
Temporarily remove new tests

They fail on JVM. Will bring back after release.		 14:42
IOninja NeuralAnomaly: cut the release 14:50
NeuralAnomaly IOninja, Will do! If you're feeling particularly naughty, you can watch me at perl6.fail/release/progress or go look at some cats icanhas.cheezburger.com/lolcats
IOninja, ☠☠☠☠☠☠☠☠☠☠ ABNORMAL EXIT!
IOninja forgot to tell it the new IP for the VM 14:51
NeuralAnomaly: cut the release
NeuralAnomaly IOninja, Will do! If you're feeling particularly naughty, you can watch me at perl6.fail/release/progress or go look at some cats icanhas.cheezburger.com/lolcats
IOninja, ♥♥♥♥♥♥ Prep done
Geth nqp: 97e82070bf | (Zoffix Znet)++ | tools/build/MOAR_REVISION
bump MoarVM version to 2017.03
 14:52
nqp: version bump brought these changes: github.com/MoarVM/MoarVM/compare/2......2017.03
f6bd03b4b7 | (Zoffix Znet)++ | VERSION
¦ nqp: version bump brought these changes: github.com/MoarVM/MoarVM/compare/2......2017.03
IOninja And yeah, it'll still point to rakudo.org for download. I'll make the rakudo.perl6.org "pretty" by April's release and will update instructions to point there (and add redirects to rakudo's server) 14:54
travis-ci Rakudo build failed. Zoffix Znet 'Log all commits to date 14:57
travis-ci.org/rakudo/rakudo/builds/212455146 github.com/rakudo/rakudo/compare/3...3f22875c8f
buggable [travis build above] ✓ All failures are due to timeout (0), missing build log (0), GitHub connectivity (0), or failed make test (1). Across all jobs, only 04-nativecall/08-callbacks.t test file failed.
IOninja Looks like a single travis link now makes buggable use 2GB+ RAM instead of ~300MB 15:00
Ever since the update to report failed tests.
Geth rakudo: MasterDuke17++ created pull request #1044:
Use uint64 for mask in Buf's `~^` operator
NeuralAnomaly IOninja, ♥♥♥♥♥♥ nqp tests OK 15:01
IOninja, ♥♥♥♥♥♥ nqp release tarball tests OK 15:10
IOninja, ♥♥♥♥♥♥ nqp release DONE 15:11
Geth rakudo/nom: 1edb1d3470 | (Zoffix Znet)++ | docs/announce/2017.03.md
Generate release announcement for 2017.03
rakudo/nom: version bump brought these changes: github.com/perl6/nqp/compare/2017.......2017.03
c269bffbfc | (Zoffix Znet)++ | tools/build/NQP_REVISION
rakudo/nom: version bump brought these changes: github.com/perl6/nqp/compare/2017.......2017.03
a0a28432f5 | (Zoffix Znet)++ | VERSION
rakudo/nom: version bump brought these changes: github.com/perl6/nqp/compare/2017.......2017.03
bfc5cb524f | (Zoffix Znet)++ | docs/release_guide.pod
IOninja Well, I repro-ed buggable's RAM nomming with this snippet: gist.github.com/zoffixznet/9d9783a...1daee7e813 15:19
Which may not be a leak :(
IOninja gonna try to loop it after release
NeuralAnomaly IOninja, ♥♥♥♥♥♥ Rakudo stresstest (master) OK
timotimo that's the json-fast related one?
IOninja No idea. 15:20
timotimo hm, there's no loop in there
IOninja * │ IOninja gonna try to loop it after release
timotimo i wonder if it'd be enough to just parse the same text over and over
oh, i read "stop" where you wrote "loop"
IOninja :D 15:21
NeuralAnomaly IOninja, ♥♥♥♥♥♥ Rakudo stresstest (6.c-errata) OK
IOninja, ♥♥♥♥♥♥ Rakudo release DONE 15:29
IOninja, ♥♥♥♥♥♥ Post: upload tarballs to rakudo.org and rakudo.perl6.org
IOninja, 🎺🎺🎺📯📯📯📯📯📯🌈🌈🌈📦📦📦
IOninja, The release of **Rakudo #109 2017.03** has now been completed
IOninja, 🎺🎺🎺📯📯📯📯📯📯🌈🌈🌈📦📦📦
NeuralAnomaly celebrates with an appropriate amount of fun
timotimo cool 15:30
IOninja NeuralAnomaly: thanks, baby 15:31
NeuralAnomaly IOninja, any time, buddy!
timotimo "the past is far behind us, the future doesn't exist!"
Geth nqp: 7ca9d1779b | (Zoffix Znet)++ | t/qregex/rx_modifiers
Revert "Temporarily remove new tests"

This reverts commit 6607ecbdc9faaf9621a4254f4b426b50a0039509.		 15:42
IOninja Any more comments for github.com/rakudo/rakudo/pull/1020 ? 16:15
Geth rakudo/nom: b1ff41766f | (Zoffix Znet)++ | CONTRIBUTING.md
Include link to info on good commit content
 16:24
dogbert17 IOninja: the gist you posted uses massive amounts of memory 16:36
everything after 'my $job = ua-get-json 'api.travis-ci.org/jobs/212455149';' can be removed, the massive memory use is still present 16:37
IOninja dogbert17: yes, that's the problem with it. 16:39
Good. That makes me hopeful it's just the JSON module that causes that instead of Perl 6 itself 16:40
MasterDuke some other metrics about the gist. a lot of the memory is allocated in MVM_string_join 16:41
twice there's a comment `/* XXX Can special-case 8-bit NFG and ASCII here too. */`
8-bit NFG is about one third of the cases when running your gist 16:42
but i don't know if that would help with memory
dogbert17 but how can it use so much memory in the first place, the downloaded content isn't even 200k 16:49
Geth rakudo/nom: 4605d5219e | (Fernando Correa de Oliveira)++ | src/core/Capture.pm
Fix #128977 and #130954
 16:57
rakudo/nom: 0de1daabd7 | (Zoffix Znet)++ | src/core/Capture.pm
Merge pull request #1043 from FCO/fix-128977

Fix #128977 and #130954
synopsebot6 Link: rt.perl.org/rt3//Public/Bug/Displa...?id=128977
Link: rt.perl.org/rt3//Public/Bug/Displa...?id=130954
synopsebot6 Link: rt.perl.org/rt3//Public/Bug/Displa...?id=128977
Link: rt.perl.org/rt3//Public/Bug/Displa...?id=130954
japhb .tell jnthn 6guts.wordpress.com/2017/03/16/con...semantics/ keeps asserting incorrect arithmetic (6+5=12,9+5=15,9+5=16). 16:58
yoleaux2 japhb: I'll pass your message to jnthn.
Geth rakudo/nom: 27539320bd | (Daniel Green)++ | src/core/Buf.pm
Use uint64 for mask in Buf's `~^` operator

0xFFFFFFFFFFFFFFFF == 2**64-1, which is too big a value to assign to signed natives.		 17:00
rakudo/nom: 88aa78cbe0 | (Zoffix Znet)++ | src/core/Buf.pm
Merge pull request #1044 from MasterDuke17/use_unsigned_native_for_Buf_mask

Use uint64 for mask in Buf's `~^` operator
IOninja m: my $x = 42; dd \(42, $x).WHICH.^nam 17:06
camelia No such method 'nam' for invocant of type 'Perl6::Metamodel::ClassHOW'
in block <unit> at <tmp> line 1
IOninja m: my $x = 42; dd \(42, $x).WHICH.^name
camelia "Str"
IOninja m: my $x = 42; dd $x.WHICH.^name
camelia "ObjAt"
IOninja Is that normal? That Capture.WHICH gives a Str instead of ObjAt?
m: dd Capture.WHICH.^name 17:07
camelia "ObjAt"
IOninja m: my $x = 42; my %h := Map.new: ($x => 42); dd %h.WHICH.^name 17:08
camelia "Str"
IOninja guess it is...
Geth roast: 7cf30b225a | (Fernando Correa de Oliveira)++ | S02-types/capture.t
Tests for #128977 and #130954 (#257)
 17:09
synopsebot6 Link: rt.perl.org/rt3//Public/Bug/Displa...?id=128977
Link: rt.perl.org/rt3//Public/Bug/Displa...?id=130954
travis-ci Rakudo build passed. Zoffix Znet '[release] bump VERSION to 2017.03' 17:26
travis-ci.org/rakudo/rakudo/builds/212464673 github.com/rakudo/rakudo/compare/5...a28432f54d
Geth rakudo/nom: 8c88b0cc62 | (Fernando Correa)++ | 2 files
Fix infinite loop #130940
 19:48
rakudo/nom: bfbe4298a7 | (Zoffix Znet)++ | 2 files
Merge pull request #1042 from FCO/fix-130940-sd-try

Fix infinite loop #130940
synopsebot6 Link: rt.perl.org/rt3//Public/Bug/Displa...?id=130940
IOninja I'm not seeing any huge memory use issues with JSON::Tiny, so I'll just swap buggable to use it 19:52
.tell jnthn recall I kept mentioning buggable leaking? Seems it's just something with JSON::Fast (and it doesn't seem to be leaking but just using lots of RAM). Swapping to JSON::Tiny avoids the issue. 19:53
yoleaux2 IOninja: I'll pass your message to jnthn.
japhb IOninja: At this point is there a functional difference between them? Or just different time/memory tradeoff? 19:55
AlexDaniel IOninja: relevant: github.com/timo/json_fast/issues/21 19:59
I'm happy that I'm not the only one seeing it because I was unable to reproduce it without the whole Whateverable code base…
japhb: should be no functional difference in theory, but there are some unicode corner cases… 20:00
IOninja japhb: JSON::Tiny is non-compilant 20:05
zoffix@VirtualBox:~$ perl6 -MJSON::Tiny -e 'dd from-json to-json([NaN, Inf])'
Input (12 characters) is not a valid JSON string
Generates invalid JSON
ZOFVM: Files=1228, Tests=132863, 126 wallclock secs (21.60 usr 3.45 sys + 2389.41 cusr 279.03 csys = 2693.49 CPU) 20:06
It's barely more than our core META6.json parser. 20:07
timotimo yeah, i don.t think there's a big difference in those at all 20:09
unless someone imported JSON::Fast at some point
Geth roast: 462a86ff9b | (Fernando Correa de Oliveira)++ | 8 files
Add tests for #130940 (#256)
 20:15
synopsebot6 Link: rt.perl.org/rt3//Public/Bug/Displa...?id=130940
Geth roast: 0a39902ba1 | (Zoffix Znet)++ | S32-str/indices.t
Fix type name in test
 20:16
IOninja .ask lizmat is it normal that Array.reverse now also returns a Seq, like List.reverse does? On 2017.02 Array.reverse returns an Array. 20:22
yoleaux2 IOninja: I'll pass your message to lizmat.
lizmat .
yoleaux2 20:22Z <IOninja> lizmat: is it normal that Array.reverse now also returns a Seq, like List.reverse does? On 2017.02 Array.reverse returns an Array.
lizmat good question, but yeah, that's what I did 20:23
pre-GLR speculation mentions it returns an Iterable 20:24
docs.perl6.org/routine/reverse has it as either returning a List:D or a Seq:D or a Supply:D 20:25
there is no documentation / tests for Array.reverse returning an Array, afaik 20:26
IOninja Then, IMO it's fine to leave Array.reverse as is. 20:27
lizmat the whole point of making it a Seq was to prevent needing to copy a reified Array when it it not necessary
IOninja Makes sense.
lizmat FWIW, I plan to give the same treatment to .rotate 20:31
but after the release :-)
IOninja m: say $*PERL.compiler.version 20:32
camelia v2017.03.8.gbfbe.429
IOninja :)
Release happened.
lizmat ah, we're post release already
cool
IOninja++
lizmat goes back to schlepping 20:36
timotimo good schlepp 20:38
samcv good * 21:52
yoleaux2 19:46Z <MasterDuke> samcv: parsing the json from a travis log with JSON::Fast spends 46% of its time here github.com/MoarVM/MoarVM/blob/mast...#L86-L139, which mentions it could be smarter. see anything you could do to optimize it?
IOninja \o 21:53
samcv ooo
MasterDuke thanks :) always good to find new places to focus improvements
o/
46% is a lot. heh. let me look at it 21:54
MasterDuke may want to backlog #moarvm, been a bunch of related chat 21:55
samcv maybe it's all the mallocs?
kk 21:56
Geth nqp: a36cc5a987 | (Samantha McVey)++ | t/moar/08-indexic.t
[moar] Unfudge now passing nqp::indexic tests
 22:25
travis-ci NQP build failed. Samantha McVey '[moar] Unfudge now passing nqp::indexic tests' 22:53
travis-ci.org/perl6/nqp/builds/212559245 github.com/perl6/nqp/compare/7ca9d...6cc5a9877c

Powered by Cro and the Raku® Programming Language.

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

Thank you!