Geth rakudo: 9b3a0b6ff2 | (Tobias Leich)++ | 2 files
Support "const" marked return values in C++ functions
06:48
rakudo: db8b218f6b | (Tobias Leich)++ | lib/NativeCall.pm6
change attributes in role Native from int8 to int

This blew up in module Box2D which is about bindings to a C++ library. The changed attributes point to not allocated memory. The program crash was triggered by a method "Box2D::Body.CreateFixture" which heavily involves traits. So perhaps the error only shows up when a certain level of traits or mixins are involved.
  See irclog.perlgeek.de/moarvm/2018-04-11#i_16035620
06:58
lizmat Files=1239, Tests=76303, 310 wallclock secs (15.90 usr 5.69 sys + 2137.22 cusr 207.43 csys = 2366.24 CPU) 08:29
Geth tap-harness6/master: 5 commits pushed by (Naoum Hankache)++, (Juan Julián Merelo Guervós)++ 08:38
MasterDuke how do i make a conditional based on the git revision of $*PERL.compiler? 12:11
Zoffix it's encoded in $*PERL.compiler.version 12:21
m: say $*PERL.compiler.version 12:22
camelia v2018.03.236.gdb.8.b.218.f.6
Zoffix By "git revision" do you mean a particular commit?
MasterDuke i think i have a workaround, but i wanted something like `if $*PERL.compiler.version < X { } else { }` 12:24
Zoffix Well, yeah, it'd be if $*PERL.compiler.version < v2018.03.236.gdb.8.b.218.f.6 { ... } else {... } 12:27
And if user is on a commit earlier than db8b218f6, the first branch would be used 12:28
MasterDuke `git describe 23af1f4` gives '2018.03-122-g23af1f4aa', how do i turn that into the 'v...' string? 12:30
Zoffix m: dd Version.new: '2018.03-122-g23af1f4aa' 12:31
camelia v2018.03.122.g.23.af.1.f.4.aa
MasterDuke ah, it does support git describe output, cool
thanks 12:32
committable6: 2017.10,2018.01,2018.02,2018.03^,HEAD use nqp; for ^5_000 { (^100).race(batch=>1).map({ $_ }).List }; my int @a; nqp::getrusage(@a); if @a[4] > 0 { say @a[4] } else { EVAL q|use nqp; say nqp::atpos_i(nqp::getrusage(), 4)| }; 12:36
committable6 MasterDuke, ¦2017.10: «419844␤» ¦2018.01: «417048␤» ¦2018.02: «445604␤» ¦2018.03^: «446844␤» ¦HEAD(4302666): «394176␤»
MasterDuke so github.com/rakudo/rakudo/issues/1713 doesn't seem to be new?
lizmat sad commute& 17:56
jnthn hugs lizmat 18:21
brrt good * 18:44
jnthn o/ brrt 18:46
brrt \o jnthn
[Tux] Rakudo version 2018.03-236-gdb8b218f6 - MoarVM version 2018.03-112-gfa0fc1832
csv-ip5xs0.915 - 0.941
csv-ip5xs-208.888 - 8.897
csv-parser36.610 - 36.790
csv-test-xs-200.454 - 0.478
test9.540 - 9.653
test-t2.521 - 2.567
test-t --race1.059 - 1.063
test-t-2045.347 - 46.384
test-t-20 --race15.216 - 15.485
19:06
FROGGS m: my $foo = 42; constant bar = $foo div 2; 19:22
camelia 5===SORRY!5=== Error while compiling <tmp>
An exception occurred while evaluating a constant
at <tmp>:1
Exception details:
Cannot resolve caller infix:<div>(Any, Int); none of these signatures match:
(Int:D \a, Int:D \b)
FROGGS I find this kinda LTA... does somebody know if it is ticketed already?
jnthn What would you like it to do? 19:25
Somehow complain about use of $foo at BEGIN time when it can't possibly have been initialized? 19:26
geekosaur cheaty... could there be (perhaps as an option for debugging) BEGIN-time initialization of such to a Failure? 19:27
FROGGS jnthn: well, perhaps warn if throws an error? explain how time travel works? dunno yet 19:30
samcv hmm for some reason i'm getting different hash's on mips32 compared to my machine even though my test seems to have worked fine in isolation 20:41
argh
MasterDuke lizmat: around? 23:26