Perl 6 language and compiler development | Logs at colabti.org/irclogger/irclogger_log/perl6-dev | For toolchain/installation stuff see #perl6-toolchain | For MoarVM see #moarvm
Set by Zoffix on 27 July 2018.
AlexDaniel moritz: maybe you're interested in this: github.com/pierre-vigier/Perl6-Mat...-457023361 00:48
01:17 MasterDuke left 02:05 MasterDuke joined, MasterDuke left, MasterDuke joined 02:13 dogbert17 left
AlexDaniel OK, some relatively not great news :) 02:17
I just got another set of results from Blin, there are more blockers
I think they somewhat hidden previously because of some changes in zef
that is, in the previous run Blin ignored some modules because they failed to install, or something like that 02:18
timotimo: if you like gists: gist.github.com/AlexDaniel/13b7743...ab6fd6b8b4 02:33
timotimo .o( gist gallop ) 02:34
AlexDaniel timotimo: ctrl+f Bisected 02:36
timotimo the P6Repl::Helper problem is just testing output from an introspection for a sub that has since gotten return type annotations
so LHF :)
SmokeMachine AlexDaniel: I’m having that error on Red on a recent Rakudo version...
AlexDaniel SmokeMachine: you mean this one? Red – Fail, Bisected: 2b948fc38b9d685104dd2f3f0b60de51e991b9fe
uhhh that bisect result makes no sense 02:37
AlexDaniel reruns that locally…
SmokeMachine Hum!!! DateTime::Extended has no update for a long time... 02:39
AlexDaniel: where have you found that bisect result? 02:40
AlexDaniel SmokeMachine: in the gist above: ctrl+f Bisected
but that looks very unrelated :)
SmokeMachine (I’m on my cellphone...) 02:41
AlexDaniel SmokeMachine: anyway, I'm rerunning that for Red right now… will create a ticket soon!
SmokeMachine Thanks! 02:44
timotimo i'll go to sleep RSN 02:46
03:10 MasterDuke_ joined, MasterDuke left 03:15 MasterDuke_ left
SmokeMachine AlexDaniel: any news about the rerunning? 03:17
03:19 robertle left, robertle joined
AlexDaniel SmokeMachine: not really. It did finish, but here's the bisect log: gist.github.com/AlexDaniel/f0a2c56...1fa6c4a3eb 03:19
SmokeMachine: TL;DR on two revisions it got MoarVM panic, so the final answer is irrelevant 03:20
SmokeMachine: is it a common occurance that it panics during tests? 03:21
Blin attempts to deflap results, so apparently it succeeded 4 times in a row on 2018.12 03:22
so it's possible that we're looking at two issues, but more testing is needed 03:23
03:24 MasterDuke joined, MasterDuke left, MasterDuke joined
SmokeMachine No, that’s not common... 03:24
AlexDaniel: you say it succeeded 4 times because it have tried 4 times or it has tried more times and only succeeded on 4? 03:27
AlexDaniel SmokeMachine: it tried 4 times (actually 5) on 2018.12 and in all 5 runs tests were clean 03:28
so it classified it as a non-flapper (because it succeeds consistently on the old revision) and proceeded to bisect it 03:29
but during bisection it sometimes failed with (NoMatch) issue (what we're looking for) and sometimes failed with moarvm panic 03:30
SmokeMachine So, the problem is some change on rakudo, right?! (And probably 2 different problems, right?)
AlexDaniel which resulted in a useless answer :) 03:31
well, either it's the commit that introduced moarvm panics (which were later resolved?), or it's a completely random commit (more likely) :)
SmokeMachine: I'm running it again, so we'll get an answer eventually… well, we *have to* to get the answer before the release
03:31 AlexDani` joined
AlexDani` heh, you can follow my progress here: R#2638 … if you ask nicely, I can stream it next time on twitch :D 03:35
synopsebot R#2638 [open]: github.com/rakudo/rakudo/issues/2638 [⚠ blocker ⚠] [WIP] Blin 2019.01
03:36 AlexDani` left, AlexDani` joined, AlexDaniel left, AlexDani` is now known as AlexDaniel 03:49 robertle left, robertle joined 03:51 martin__ joined 03:55 ufobat_ left 04:01 MasterDuke left 04:02 MasterDuke joined 04:04 MasterDuke left, MasterDuke joined
AlexDaniel I'm done for today… 06:43
07:50 patrickb joined 07:51 martin__ is now known as ufobat 09:16 robertle left 09:17 robertle joined
Geth rakudo: 3488a65ed7 | (Elizabeth Mattijsen)++ | 3 files
Make sure $!daycount is nulled on subclasses of DateTime

Essentially the same fix as 929561642281da874c was for Date. Spotted by AlexDaniel++ / Blin++. Fixes R#2615 (again)		 10:07
synopsebot R#2615 [open]: github.com/rakudo/rakudo/issues/2615 [regression][⚠ blocker ⚠] Date/DateTime regression (expected Int:D but got Int (Int))
10:20 ExtraCrispy joined
Geth roast: 9d55fd0c0e | (Elizabeth Mattijsen)++ | S32-temporal/DateTime.t
Add tests for DateTime R#2615
 10:21
synopsebot R#2615 [open]: github.com/rakudo/rakudo/issues/2615 [regression][⚠ blocker ⚠] Date/DateTime regression (expected Int:D but got Int (Int))
lizmat hmmm... how do you rethrow a BOOTException ? 10:50
jnthn nqp::rethrow 10:52
10:59 dogbert2_ joined
lizmat jnthn++ 11:15
dogbert2_ releasable: next 11:19
releasable6 dogbert2_, Next release will happen when it's ready. 10 blockers. 116 out of 248 commits logged (⚠ 41 warnings)
dogbert2_, Details: gist.github.com/d846ff4ddff2877485...f4d14aa39b
dogbert2_ what 10 blockers 11:20
Geth rakudo: 89ca56affb | (Elizabeth Mattijsen)++ | src/Perl6/Grammar.nqp
Add some diagnostics when a module fails to precomp
 11:21
lizmat afk for some hours& 11:24
12:33 lucasb joined 12:34 travis-ci joined
travis-ci Rakudo build failed. Elizabeth Mattijsen 'Add some diagnostics when a module fails to precomp' 12:34
travis-ci.org/rakudo/rakudo/builds/483846064 github.com/rakudo/rakudo/compare/3...ca56affba1
12:34 travis-ci left 13:20 dogbert2_ left 13:23 patrickb left
Geth nqp/master: 16 commits pushed by (Paweł Murias)++
review: github.com/perl6/nqp/compare/babda...9984bbc15a		 14:07
14:24 travis-ci joined
travis-ci NQP build failed. Paweł Murias 'Test the utf16be encoding' 14:24
travis-ci.org/perl6/nqp/builds/483908103 github.com/perl6/nqp/compare/babda...9984bbc15a
14:24 travis-ci left 15:14 dalek left 15:15 dalek joined, ChanServ sets mode: +v dalek 15:18 robertle left, robertle joined
scovit Hello, does anybody knows if there is a straightforward command to run only the nativecall tests in the rakudo test directory 15:26
? 15:27
Geth rakudo: 8095ceada8 | (Elizabeth Mattijsen)++ | src/Perl6/Grammar.nqp
Disable CATCH block for precomp errors for now

Its causing too many weird test errors		 15:28
lizmat scovit: make t/04-nativecall/*.t
scovit it complains that there is no t/spec/fudgeall 15:30
15:33 lucasb left
scovit I found the way 15:33
thanks lizmat
15:34 [Coke] left
scovit I get 1 failure in 00-misc.t , with rakudo compiled --has-libffi , is it normal? 15:35
it just says Non-zero exit status: 1 15:52
perl6 t/harness6 t/04-nativecall/00-mist.t
Mac OSX 15:53
lizmat scovit: don't you mean perl6 t/harness6 t/04-nativecall/00-misc.t ? 15:55
aka s/mist.t/misc.t/ ?
scovit yep
lizmat that passes for me ok 15:56
on MacOS High Sierra
scovit I'm on Sierra but I guess that is not the difference
did you try --has-libffi ?
it is not the default 15:57
lizmat ah, ok, no I didn't build with that
scovit it is a moarvm option, but you need libffi installd
lizmat where did you specify that --has-libffi ?
ah... well then I hope someone with that installed can help you further 15:58
lizmat is in the middle of some other stuff
scovit Ok, no-prob
lizmat afk again& 16:03
scovit lizmat, all, do not worry about that test: I foind out the reason it is not working here and it is a consequence of my hacking 16:29
it is the only test that tests stdout, and I added debugging there 16:30
16:41 travis-ci joined
travis-ci Rakudo build passed. Elizabeth Mattijsen 'Disable CATCH block for precomp errors for now 16:41
travis-ci.org/rakudo/rakudo/builds/483945602 github.com/rakudo/rakudo/compare/8...95ceada8bb
16:41 travis-ci left
Geth nqp: 5ce0d49f44 | (Paweł Murias)++ | src/vm/jvm/runtime/org/perl6/nqp/runtime/Ops.java
[jvm] Implement utf16be support
 16:43
16:59 travis-ci joined
travis-ci NQP build passed. Paweł Murias '[jvm] Implement utf16be support' 16:59
travis-ci.org/perl6/nqp/builds/483980025 github.com/perl6/nqp/compare/6e998...e0d49f4452
16:59 travis-ci left 17:03 lizmat left 17:28 patrickb joined 18:24 dogbert17 joined 18:25 patrickb left 19:55 robertle left, robertle joined 19:56 robertle left, robertle joined 20:03 robertle left 20:05 lizmat joined 20:10 robertle joined 20:28 dct joined 20:29 AlexDani` joined 20:31 AlexDaniel left 20:40 AlexDani` is now known as AlexDaniel
AlexDaniel o/ 20:40
yoleaux 12:32Z <brrt> AlexDaniel: how can I submit an idea?
21:19 robertle left
jnthn releasable6: status 21:20
releasable6 jnthn, Next release will happen when it's ready. 9 blockers. 157 out of 250 commits logged
jnthn, Details: gist.github.com/6d6e30ead214e1aa84...c1e841c9bc
jnthn o.O
Hopefully I'll have some time to help with those tomorrow :)
21:40 Kaiepi left 21:46 Kaiepi joined
gfldex m: my $v = Version.new("1.2.3"); 21:46
camelia ( no output )
gfldex m: my $v = Version.new("1.2.3"); say $v > Version.new(1,2,2); 21:47
camelia Default constructor for 'Version' only takes named arguments
in block <unit> at <tmp> line 1
gfldex m: my $v = Version.new("1.2.3"); say $v > Version.new("1,2,2");
camelia True
gfldex m: my $v = Version.new("1.2.3"); say $v > Version.new("1,2,2"); dd Version.new("1.2.3"), Version.new("1,2,2"); 21:48
camelia True
v1.2.3
v1.2.2
21:59 MasterDuke left
Geth rakudo: 4fd8e8731e | (Elizabeth Mattijsen)++ | src/core/Exception.pm6
Somehow bypass issue that causes LTA error of R#2640

Not sure why this makes a difference at all, but by replacing the
   when Failure
by a much more low level
   if nqp::istype(Failure)
we at least get a decent error message. I assume this is somehow related to $_ scoping optimizations that have happened.		 22:26
synopsebot R#2640 [open]: github.com/rakudo/rakudo/issues/2640 [⚠ blocker ⚠] Some sort of regression in Red module
23:12 MasterDuke joined, MasterDuke left, MasterDuke joined

Powered by Cro and the Raku® Programming Language.

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

Thank you!