Perl 6 language and compiler development | Logs at | For toolchain/installation stuff see #perl6-toolchain | For MoarVM see #moarvm
Set by Zoffix on 27 July 2018.
00:39 [Tux] joined 01:24 titsuki left 01:25 titsuki joined
Kaiepi to make an op return null on the jvm do you make it a void method and make the return type $RT_VOID, make it return the null type object from HLL and make the return type $RT_OBJ, or some other way? 02:42
oh it looks like i need to use $RT_VOID and pass :cont 02:47
02:55 leont left 04:26 llfourn left, vrurg left, AlexDaniel` left, gfldex left, epony left, titsuki left, camelia left, dogbert17 left, lgtaube left, sivoais left, zostay left, b2gills left, nebuchadnezzar left, tobs left, AlexDaniel left, squashable6 left, TimToady left, unicodable6 left, greppable6 left, reportable6 left, notable6 left, moritz left, timotimo left, coverable6 left, MasterDuke left, Geth left, [Tux] left, Kaiepi left, tyil left, awwaiid left, masak left, ExtraCrispy left, benchable6 left, bisectable6_ left, tbrowder left, eater left, bartolin left, klapperl left, pyrimidine left, btyler_ left, jkramer left, yoleaux left, undersightable6 left, jnthn left, gugod left, krunen left, Ulti left, go|dfish left, DrForr left, p6lert left, dalek left, statisfiable6 left, quotable6 left, bloatable6 left, hoelzro left, synopsebot left, tailgate left, Util left, nativecallable6 left, shareable6 left, releasable6 left, committable6 left, evalable6 left, chansen_ left, tyil[m] left, SmokeMachine left, scovit left, nine left, SqrtNegInf left, literal left, samcv left, |Tux| left, japhb left, mst left, commavir left, jdv79 left, tony-o left 04:57 dhyan_nataraj joined, titsuki joined, [Tux] joined, Kaiepi joined, MasterDuke joined, AlexDaniel joined, ExtraCrispy joined, epony joined, llfourn joined, squashable6 joined, TimToady joined, vrurg joined, AlexDaniel` joined, gfldex joined, committable6 joined, releasable6 joined, shareable6 joined, nativecallable6 joined, sets mode: +vvvv committable6 releasable6 shareable6 nativecallable6, scovit joined, eater joined, unicodable6 joined, greppable6 joined, reportable6 joined, lgtaube joined, evalable6 joined, bartolin joined, hoelzro joined, zostay joined, chansen_ joined, klapperl joined, b2gills joined, sivoais joined, gugod joined, nebuchadnezzar joined, tyil joined, benchable6 joined, samcv joined, |Tux| joined, camelia joined, commavir joined, pyrimidine joined, dogbert17 joined, awwaiid joined, yoleaux joined, Geth joined, sets mode: +vvvv evalable6 benchable6 camelia Geth, jdv79 joined, p6lert joined, synopsebot joined, dalek joined, nine joined, tobs joined, masak joined, bisectable6_ joined, Ulti joined, notable6 joined, undersightable6 joined, sets mode: +vvvv synopsebot dalek bisectable6_ undersightable6, statisfiable6 joined, quotable6 joined, japhb joined, SqrtNegInf joined, Util joined, jnthn joined, moritz joined, sets mode: +vvvv statisfiable6 quotable6 jnthn moritz, krunen joined, timotimo joined, coverable6 joined, bloatable6 joined, sets mode: +vvvv krunen timotimo coverable6 bloatable6, tyil[m] joined, btyler_ joined, go|dfish joined, DrForr joined, sets mode: +vvvv tyil[m] btyler_ go|dfish DrForr, jkramer joined, literal joined, tbrowder joined, SmokeMachine joined, sets mode: +vvvv jkramer literal tbrowder SmokeMachine, tony-o joined, tailgate joined, mst joined, sets mode: +vvv tony-o tailgate mst 06:31 lizmat joined 07:02 Kaiepi left, Kaiepi joined 07:12 lizmat left 07:44 lizmat joined 07:51 lizmat left
[Tux] Rakudo version 2018.12-338-g9770c1aa1 - MoarVM version 2019.03-1-gce2152df1
csv-ip5xs0.735 - 0.774
csv-ip5xs-206.078 - 6.341
csv-parser22.178 - 23.500
csv-test-xs-200.437 - 0.441
test7.268 - 7.622
test-t1.814 - 1.956
test-t --race0.884 - 0.923
test-t-2030.797 - 31.339
test-t-20 --race10.586 - 10.626
working from home, so that might explain the wider range 08:03
08:11 Kaiepi left 08:15 Kaiepi joined
Kaiepi anyone on familiar with the jvm? i'm running into a weird build error with rakudo trying to add a setsockopt op and i can't figure out what's going wrong 10:04
i'll need about 10 minutes to get the exact error 10:05
running 'nqp -e 'nqp::setsockopt(nqp::socket(0), 0, 0)' throws the error i expect it to, so the op itself is working, but when i try to build rakudo it complains about not being able to unbox a type object 10:18
but i don't unbox anything anywhere in my code?
10:18 pmurias joined
pmurias Kaiepi: what op do you want to add on the js with the thread context? 10:18
yoleaux 5 Mar 2019 22:06Z <MasterDuke> pmurias: you can use line directives to set the filename in EVALs. e.g., `EVAL "#line 1 made_up_file\nsay q|hi|;die q|so long|;say q|bye|"`
timotimo Kaiepi: does it give the type? 10:19
it can be that an undefined value is passed to the op in some place that it tries to turn into an int or so
Kaiepi lemme check
pmurias i'm adding getsockopts, getsockopt, and setsockopt ops to moar and the jvm 10:20
timotimo in moar i would know to "call MVM_dump_backtrace(tc)", but no clue how you would do something equivalent on jvm 10:21
pmurias what do you need the ctx for on the js?
Kaiepi: but :ctx should add one if you need them
Kaiepi wdym by the js? the javascript vm? 10:22
pmurias the js backend
Kaiepi i haven't written anything for the js yet
can't install it because of issues with a couple dependencies
timotimo just making sure the js code doesn't break? that's good form 10:23
pmurias Kaiepi: btw. usually if you add nqp level tests I can add ops easily
timotimo not everybody tests all backends when they change code. i am very guilty of this
Kaiepi tests for getsockopts/getsockopt/setsockopt would be fairly straightforward to add 10:24
10:24 lizmat joined
timotimo pmurias is responsible for a quite big portion of the nqp tests for all the ops we have 10:24
from when the JS port was being made 10:25
pmurias I do feel they add value besides helping the JS backend ;) 10:26
Kaiepi ok here's the error 10:27
10:27 lizmat left
pmurias but for the JS backend they help a lot to figure out what the op should do and to implement it without mixing rakudo itself into it 10:27
jnthn: I checked EVALFILE and it just uses EVAL_* so if we added a filename it could use it too 10:28
timotimo i agree they do
Kaiepi did i write this part of the setsockopt op right? 10:31
Geth rakudo: 2732a28646 | (Paweł Murias)++ | src/core/ForeignCode.pm6
Add a :$filename parameter to EVAL to override the filename

The filename is used in stack traces. Use this from &EVALFILE
pmurias jnthn, everyone else: commit that adds a :$filename argument to EVAL 2732a286469f97efc5ca6425573c1ca21613d8c8 10:39
10:55 leont joined
pmurias I'll add tests and docs if the :$filename to EVAL is acceptable 10:55
jnthn pmurias: That looks reasonable to me 11:03
yoleaux 6 Mar 2019 16:39Z <SmokeMachine> jnthn: its saying that there are some errors here...
jnthn pmurias: Though maybe wants Str()
Since IO::Path could be passed to EVALFILE 11:04
Kaiepi oh, i forgot to check whether rakudo even builds properly on the jvm without my changes 11:12
it doesn't! 11:16
yay...? 11:18
pmurias jnthn: a Str(...) call? 11:28
ahh, a Str() type 11:29
Kaiepi ok the R:I:JSON commits were causing the cannot unbox a type object error 11:30
11:32 lizmat joined
Geth rakudo: 69d88be375 | (Paweł Murias)++ | src/core/ForeignCode.pm6
Change :$filename type constraint in EVAL from Str to Str()
pmurias do we want to have a method form of EVAL on Buf? 12:29
there is a RT ticket with a fudge from 2014 for that
and a comment from Zoffix that he is against adding this 12:30
I don't think we need that method either
Geth roast: Kaiepi++ created pull request #517:
[IP6NS Grant] Add tests for IO socket .get-option/.set-option
roast: a80e043221 | (Paweł Murias)++ | S29-context/evalfile.t
Test that EVALFILE works with an IO::Path argument
roast: a98962a585 | (Paweł Murias)++ | S29-context/evalfile.t
Test that EVALFILE get the filename somewhere in the stacktrace
12:40 dhyan_nataraj left, dhyan_nataraj joined 13:04 travis-ci joined
travis-ci Rakudo build failed. Paweł Murias 'Change :$filename type constraint in EVAL from Str to Str()' 13:04
13:04 travis-ci left 13:09 dhyan_nataraj left
pmurias looks into the fail 13:15
timotimo could it be the apt-get trouble again? 13:17
pmurias timotimo: one fails is cause by a mistake in my commit 13:19
timotimo: the nativecall one I can't reproduce 13:20
Geth rakudo: 1ec4f171dd | (Paweł Murias)++ | src/core/ForeignCode.pm6
Fix bug causing incorrect warning when missing an optional argument
13:27 lucasb joined
timotimo pmurias: i am having a real hard time scrolling through the log on travis, can you paste the relevant bits somewhere 13:29
pmurias timotimo: it looks like something caused by my changes 13:36
13:50 Kaiepi left, Kaiepi joined 13:52 dhyan_nataraj joined 14:12 travis-ci joined
travis-ci Rakudo build passed. Paweł Murias 'Fix bug causing incorrect warning when missing an optional argument' 14:12
14:12 travis-ci left 14:19 Kaiepi left 14:20 Kaiepi joined
Geth rakudo/release-2019.somewhen: b07350a70c | (Aleks-Daniel Jakimenko-Aleksejev)++ | 2 files
Log all changes (+ announcement)

Deliberately not logged:
495ac7c0 afccc2c0 daeb7f26 5625ebc1 c47d95f2 4a152da5 3f8fea9b f976c85e b20ff803 2b948fc3 05cc8cd5 8591fd28 626a99b1 7674b666 c9e2c63b 335f32ff 08851be1 3a68cc96 d41bc6dd 09c88430 a7620cef ... (9 more lines)
rakudo/release-2019.somewhen: 905b9e32d3 | (Aleks-Daniel Jakimenko-Aleksejev)++ | docs/release_guide.pod
Actual date, and not claiming the next release
AlexDaniel timotimo: please review before I hit the big red button :)
14:53 lizmat left 14:58 pmurias left 15:01 pmurias joined
timotimo i feel honored :) 15:07
AlexDaniel: or was i just the closest living developer?:))
AlexDaniel timotimo: no, it's just that you always commit some nice fixes but *after the release* :)
15:07 lizmat joined
timotimo class specification sounds a little strange 15:09
perhaps decllaration is more commonly used?
what is "silent stuffing"?:) 15:10
15:14 lizmat left
timotimo doh i opened the file itself in tree "mode" to hopefully be able to click the commit shas 15:14
but no
AlexDaniel yeah that's annoying 15:15
timotimo looks like that parameterization in parents is mentioned again in the next section 15:17
lucasb so, on a non-existent path, IO::Path.dir was made to throw rather than return a Failure. 15:19
AlexDaniel timotimo: not seeing it, where's that?
timotimo: also feel free to fix it
lucasb should this have some symmetry with IO::Path.slurp which returns a Failure, rather than throw? 15:20
I mean, 'nonexitent'.IO.dir throws and 'nonexitent'.IO.slurp returns a failure 15:21
timotimo i've read through it once now
AlexDaniel: i'm confined to my smartphone, i would really rather have someone do it for me :) 15:22
Geth rakudo/release-2019.somewhen: f9ee5b0832 | (Aleks-Daniel Jakimenko-Aleksejev)++ | 2 files
ChangeLog tweaks

timotimo TYVM
15:38 lizmat joined 15:54 lizmat left 16:30 pmurias left
japhb On Rakudo HEAD I get a weird test failure installing Test::Output, which I think is coming in as a dependency of Pod::To::HTML: 16:50
===> Testing: Test::Output:ver<1.001001>
No exception handler located for catch at SETTING::src/core/Exception.pm6:427 (/usr/local/google/home/gjb/perl6/rakudo-moar-1ec4f171dd343441a47a2313f115d116864a39b1/install/share/perl6/runtime/CORE.setting.moarvm:print_exception)
AlexDaniel: When were you planning to do the Rakudo release? 16:51
Geth rakudo: vrurg++ created pull request #2749:
Fix for rakudo/rakudo#2178
synopsebot RAKUDO#2178 [open]: .wrap of a method breaks callsame/nextsame/etc.
17:48 leont left
AlexDaniel japhb: right now, but I'm cutting it from the release branch, not HEAD 20:17
yoleaux 19:53Z <jmerelo> AlexDaniel: can I do the extraofficial official announcement of the next version tomorrow?
.oO( why are we referring to master HEAD as just HEAD? :) )
20:24 dhyan_nataraj left, AlexDani` joined 20:26 AlexDaniel left, AlexDani` is now known as AlexDaniel
Geth rakudo/release-2019.somewhen: 043bd6f550 | (Aleks-Daniel Jakimenko-Aleksejev)++ | 2 files
Tiny changelog tweaks
20:35 reportable6 left, evalable6 left, notable6 left, greppable6 left, bisectable6_ left, notable6 joined, ChanServ sets mode: +v notable6
jdv79 technically HEAD and $ref/head are different animals:) 20:37
20:38 greppable6 joined, ChanServ sets mode: +v greppable6 20:39 bisectable6 joined, ChanServ sets mode: +v bisectable6, reportable6 joined, ChanServ sets mode: +v reportable6, evalable6 joined 20:40 bloatable6 left, coverable6 left, undersightable6 left, statisfiable6 left 20:41 statisfiable6 joined, ChanServ sets mode: +v statisfiable6 20:43 undersightable6 joined, ChanServ sets mode: +v undersightable6, bloatable6 joined 20:44 coverable6 joined, ChanServ sets mode: +v coverable6 21:19 pmurias joined 21:47 sortiz joined
Geth nqp/release-2019.somewhen: f816dd23c2 | (Aleks-Daniel Jakimenko-Aleksejev)++ | tools/build/MOAR_REVISION
[release] Bump MoarVM revision to 2019.03
nqp/release-2019.somewhen: ed69a2a9a1 | (Aleks-Daniel Jakimenko-Aleksejev)++ | VERSION
[release] Bump VERSION to 2019.03
rakudo/release-2019.somewhen: 0f4d2074d9 | (Aleks-Daniel Jakimenko-Aleksejev)++ | tools/build/NQP_REVISION
[release] Bump NQP revision to 2019.03
rakudo/release-2019.somewhen: 47e34445f4 | (Aleks-Daniel Jakimenko-Aleksejev)++ | VERSION
[release] Bump VERSION to 2019.03
22:06 sortiz left 22:07 sortiz joined
Geth nqp: f816dd23c2 | (Aleks-Daniel Jakimenko-Aleksejev)++ | tools/build/MOAR_REVISION
[release] Bump MoarVM revision to 2019.03
nqp: version bump brought these changes:
ed69a2a9a1 | (Aleks-Daniel Jakimenko-Aleksejev)++ | VERSION
nqp: 462af6d827 | (Aleks-Daniel Jakimenko-Aleksejev)++ | 7 files
Merge branch 'master' into release-2019.somewhen
rakudo/master: 7 commits pushed by (Aleks-Daniel Jakimenko-Aleksejev)++
¦ rakudo/master: version bump brought these changes:
sortiz lexDaniel: Is the last part of commit #c4c32ab47c expected? 22:31
AlexDaniel: Is the last part of commit #c4c32ab47c expected?
AlexDaniel sortiz: thanks for looking! I think it is fine 22:32
yea it does look weird, but it will function correctly
let's try bumping… 22:33
Geth nqp: 87771ea33b | (Aleks-Daniel Jakimenko-Aleksejev)++ | tools/build/MOAR_REVISION
[MoarVM Bump] Brings 2 commits

MoarVM bump brought: bb2dc6619 Fix warnings about const from prev commit ce2152df1 Fix typo in strings documentation
¦ nqp: version bump brought these changes:
rakudo: 70d61b2144 | (Aleks-Daniel Jakimenko-Aleksejev)++ | tools/build/NQP_REVISION
[NQP Bump] Brings 7 commits

NQP bump brought: 87771ea33 [MoarVM Bump] Brings 2 commits 462af6d82 Merge branch 'master' into release-2019.somewhen ed69a2a9a [release] Bump VERSION to 2019.03 f816dd23c [release] Bump MoarVM revision to 2019.03 2d4f2a977 [js] Make nqp::getcodelocation work in the browser 89587e5f2 [js] Fix evalling js code at runtime in the browser 374e9fddb [js] Fix .wasm for source maps loading
¦ rakudo: version bump brought these changes:
AlexDaniel sortiz: it's because the release branch and master diverged a bit, with their own corresponding bumps
so during the merged I kept the one from master 22:35
22:36 sortiz left 22:54 sortiz joined
sortiz Ok, seems better now. 22:57
I was out of the loop for some time. I want to be useful :-) 22:58
23:03 sortiz left 23:37 lucasb left 23:39 pmurias left
AlexDaniel grep: IO\.dir 23:42
greppable6 AlexDaniel, 122 lines, 64 modules: