Perl 6 - the future is here, just unevenly distributed

IRC log for #perl6-dev, 2016-10-15

| Channels | #perl6-dev index | Today | | Search | Google Search | Plain-Text | summary

All times shown according to UTC.

Time Nick Message
02:30 travis-ci joined #perl6-dev
02:30 travis-ci Rakudo build passed. Zoffix Znet 'Merge pull request #904 from tbrowder/new-test
02:30 travis-ci https://travis-ci.org/rakudo/rakudo/builds/167705475 https://github.com/rakudo/rakudo/compare/8268ffbc1ae5...41c0750171cd
02:30 travis-ci left #perl6-dev
04:18 ggoebel joined #perl6-dev
04:29 Ven_ joined #perl6-dev
06:24 geekosaur joined #perl6-dev
06:39 p3rln00b Internet dead at home for some reason :( not a good release day start :(
06:39 yoleaux2 14 Oct 2016 21:53Z <jnthn> p3rln00b: I'd guess so, but do a spectest to be sure
06:44 Zoffix joined #perl6-dev
06:44 Zoffix \o/
06:44 Zoffix Though, I can always tether my lappy to my phone if I had to.
06:46 yoleaux2 joined #perl6-dev
06:47 yoleaux2 joined #perl6-dev
06:49 yoleaux2 joined #perl6-dev
06:51 yoleaux2 joined #perl6-dev
06:58 RabidGravy joined #perl6-dev
07:11 dalek rakudo/nom: 2c681c5 | (Zoffix Znet)++ | docs/ChangeLog:
07:11 dalek rakudo/nom: Log all changes to date
07:11 dalek rakudo/nom:
07:11 dalek rakudo/nom: Documents commits:
07:11 dalek rakudo/nom: 6283124  bfe6219  e5df7a7  5a79e22  8df0965  bf7945e  7f16cbb  b08ef96
07:11 dalek rakudo/nom: 8268ffb
07:11 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/2c681c5552
07:24 vendethiel- joined #perl6-dev
07:37 [Tux] This is Rakudo version 2016.09-199-g41c0750 built on MoarVM version 2016.09-47-g736364c
07:37 [Tux] csv-ip5xs        3.125
07:37 [Tux] test            15.766
07:37 [Tux] test-t           6.888
07:37 [Tux] csv-parser      17.313
07:40 dalek rakudo/nom: 873e60f | (Zoffix Znet)++ | src/core/Mu.pm:
07:40 dalek rakudo/nom: Remove unused Mu.dispatch:<.^> method
07:40 dalek rakudo/nom:
07:40 dalek rakudo/nom: The code is not hit during stresstests. Per discussion[^1] and
07:40 dalek rakudo/nom: confirmation[^2], this code is no longer used and can be removed.
07:40 dalek rakudo/nom: Stresstest passes.
07:40 dalek rakudo/nom:
07:40 dalek rakudo/nom: [1] https://github.com/perl6/doc/issues/970#issuecomment-253824298
07:40 dalek rakudo/nom: [2] https://irclog.perlgeek.de/perl6-dev/2016-10-14#i_13405760
07:40 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/873e60fb68
07:44 dalek roast: 53bdf26 | usev6++ | S (3 files):
07:44 dalek roast: Fudge tests for JVM; use ticket number
07:44 dalek roast: review: https://github.com/perl6/roast/commit/53bdf269b0
07:48 Zoffix psch, you around? So the closure nqp failures are due to an old bug, but what about the last two failures? Can they be ignored for the release? https://gist.github.com/zoffixznet/448d79d9cf422e64901089c0eb803e83
07:53 FROGGS joined #perl6-dev
07:55 travis-ci joined #perl6-dev
07:55 travis-ci Rakudo build errored. Zoffix Znet 'Log all changes to date
07:55 travis-ci https://travis-ci.org/rakudo/rakudo/builds/167837220 https://github.com/rakudo/rakudo/compare/41c0750171cd...2c681c55520d
07:55 travis-ci left #perl6-dev
07:56 bartolin committable6: 2016.09,HEAD say (my int $ = 2**32) * (my int $ = 2**63)
07:56 committable6 bartolin, ¦«2016.09»: 0␤¦«HEAD»: This type cannot unbox to a native integer: P6opaque, Failure␤  in block <unit> at /tmp/5gdU3BwADu line 1␤ «exit code = 1»
07:57 Zoffix I guess I have my answer. The test was addded just 2 days ago: https://github.com/perl6/nqp/commit/682f8253bde4fc5c9774d7e68940f7ae07044f59
07:57 psch yeah, pmurias is pretty awesome about nqp tests
07:57 psch anyway, the failure is about autoboxing, and is probably something we should look into
07:57 psch might relate to natives on jvm i think
07:58 Zoffix should look into... you mean before the release?
07:58 psch no, the bug is an old one, we just didn't test it before
07:59 psch i'm not sure in how far it actually does impact rakudo either
07:59 Zoffix bartolin, oh, I was meant to fix that.
07:59 Zoffix I thought I did
07:59 AlexDaniel when?
08:00 bartolin r: my int $iL = 2**62; say $iL * $iL
08:00 camelia rakudo-moar 873e60: OUTPUT«This type cannot unbox to a native integer: P6opaque, Failure␤  in block <unit> at <tmp> line 1␤␤»
08:00 camelia ..rakudo-jvm 2a1605: OUTPUT«0␤»
08:00 AlexDaniel m: say (my int $ = 2**32) * (my int $ = 2**63)
08:00 camelia rakudo-moar 873e60: OUTPUT«This type cannot unbox to a native integer: P6opaque, Failure␤  in block <unit> at <tmp> line 1␤␤»
08:00 Zoffix When what? I meant I said I'll fix it and I forgot :)
08:01 AlexDaniel ah ok :) Just wanted to note that whateverables are currently behind HEAD for one or two commits
08:02 bartolin I'm somewhat confused. that test failed with 'Numeric overflow' on rakudo-j a few days ago
08:03 dalek nqp: 298e228 | (Zoffix Znet)++ | t/qast/01-qast.t:
08:03 dalek nqp: Temporarily comment out tests failing on JVM
08:03 dalek nqp:
08:03 dalek nqp: The tests were added in the past few days, but the issues existed
08:03 dalek nqp: on JVM for a while[^1][^2]. Comment the tests out for the release,
08:03 dalek nqp: to avoid any confusion downstream, when someone notices "new" failures.
08:03 dalek nqp:
08:03 dalek nqp: This commit will be reverted after the release.
08:03 dalek nqp:
08:03 dalek nqp: [1] https://irclog.perlgeek.de/perl6-dev/2016-10-14#i_13401309
08:03 dalek nqp: [2] https://irclog.perlgeek.de/perl6-dev/2016-10-15#i_13407254
08:03 dalek nqp: review: https://github.com/perl6/nqp/commit/298e2281ec
08:03 bartolin and now it dies with the above error. but with --optimize=1 it returns 21267647932558653966460912964485513216 (no numeric overflow anymore)
08:05 Zoffix bartolin, it's meant to return 0. It was made to return a Failure recently, but later it was discussed that natives should not return Failures and returning 0 is perfectly fine. Whatever tests expect a Failure were added as part of coverage work and I'll revert them shortly. The optimize stuff is likely due to dispatch bug with nativevs
08:05 yoleaux2 08:04Z <japhb> Zoffix: Thanks!
08:06 Zoffix (there's a ticket for it)
08:07 bartolin Zoffix: Ahh, I see. Must have missed that discussion. Thanks!
08:09 bartolin the ticket is RT #129813, probably
08:12 Zoffix Aforementioned discussion: https://irclog.perlgeek.de/perl6-dev/2016-10-10#i_13372432
08:12 Zoffix Well, it starts a few lines above that; at https://irclog.perlgeek.de/perl6-dev/2016-10-10#i_13372307
08:13 bartolin Zoffix++
08:15 Zoffix .tell p3rln00b revert https://github.com/perl6/nqp/commit/298e2281ec after the release
08:15 yoleaux2 Zoffix: I'll pass your message to p3rln00b.
08:24 psch huh
08:24 psch the last failures in qast.t are actually automatic *un*boxing
08:25 psch ...i suppose i shoulda seen that, cause it says so in the test results
08:26 psch i don't think i understand the test
08:26 Zoffix__ joined #perl6-dev
08:26 japhb_ joined #perl6-dev
08:26 Ulti_ joined #perl6-dev
08:27 jnthn_ joined #perl6-dev
08:27 bartolin_ joined #perl6-dev
08:29 facetious joined #perl6-dev
08:29 Woodi_ joined #perl6-dev
08:31 dalek rakudo/nom: 8ad885f | (Zoffix Znet)++ | / (2 files):
08:31 dalek rakudo/nom: Remove overflow check on native infix:<*> and infix:<**>
08:31 dalek rakudo/nom:
08:31 dalek rakudo/nom: To preserve 1-type-of-return and inlining, we do not[^1] do overflow checks
08:31 dalek rakudo/nom: on native subs.
08:31 dalek rakudo/nom:
08:31 dalek rakudo/nom: [1] https://irclog.perlgeek.de/perl6-dev/2016-10-10#i_13372307
08:31 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/8ad885f0ab
08:32 dalek roast: c8fc433 | (Zoffix Znet)++ | S32-num/int.t:
08:32 dalek roast: Remove tests that expect overflow checks from native subs
08:32 dalek roast:
08:32 dalek roast: The tests were added recently, as part of coverage work and they
08:32 dalek roast: test behaviour that is unwanted and has been removed[^1] from rakudo.
08:32 dalek roast:
08:32 dalek roast: [1] https://github.com/rakudo/rakudo/commit/8ad885f0ab
08:32 dalek roast: review: https://github.com/perl6/roast/commit/c8fc43306b
08:45 Ven_ joined #perl6-dev
08:51 vendethiel- joined #perl6-dev
08:51 jdv79_ joined #perl6-dev
08:51 camelia joined #perl6-dev
08:51 TimToady joined #perl6-dev
08:51 gfldex joined #perl6-dev
08:51 mst joined #perl6-dev
09:00 Ven_ joined #perl6-dev
09:13 psch pff, silly jvm/jdb, can't even reenter nqp code...
09:14 psch "SIGSEGV (0xb) at pc=0x0000000000000001"
09:17 travis-ci joined #perl6-dev
09:17 travis-ci Rakudo build passed. Zoffix Znet 'Remove overflow check on native infix:<*> and infix:<**>
09:17 travis-ci https://travis-ci.org/rakudo/rakudo/builds/167844903 https://github.com/rakudo/rakudo/compare/873e60fb6849...8ad885f0ab8b
09:17 travis-ci left #perl6-dev
09:27 vendethiel joined #perl6-dev
09:27 psch okay, so apparently nqp-j feels like having an intBoxType doesn't necessarily mean that a variable that gets assigned an int needs to be boxed into the given intBoxType
09:28 psch it's perfectly okay with using BOOTInt
09:32 jdv79 joined #perl6-dev
09:33 mst_ joined #perl6-dev
09:33 TimToady_ joined #perl6-dev
09:33 gfldex_ joined #perl6-dev
09:40 Zoffix :( t/04-nativecall/16-rt125408.t keeps failing when I'm running a release test on my VM :/
09:42 Zoffix__ joined #perl6-dev
09:47 Zoffix joined #perl6-dev
09:49 Zoffix And unless it's a flapper, the issue is due to updated nqp/moarvm
09:49 * Zoffix tries building HEAD of each
09:55 camelia joined #perl6-dev
10:13 Zoffix Ah. It'd help if I updated my test repos :}
10:23 TimToady joined #perl6-dev
10:26 Zoffix There's no way for me to lose the ability to delete perl6/nqp repo without losing my org admin status, is there?
10:26 Zoffix I'm paranoid about deleting perl6/nqp instead of zoffixznet/nqp during pre-release tests :|
10:27 nine Inline-Perl5 (redesign_v6_inline *)> perl6 -e 'Distribution::Path.new($*CWD, :file("META.info".IO))'
10:27 nine No such method 'IO' for invocant of type '<anon|118534656>'
10:28 nine This could be a rakudo regression ^^^
10:28 Zoffix m: class {}.IO
10:28 camelia rakudo-moar 8ad885: OUTPUT«No such method 'IO' for invocant of type '<anon|63061472>'␤  in block <unit> at <tmp> line 1␤␤»
10:28 Zoffix star: class {}.IO
10:28 camelia star-m 2016.04: OUTPUT«Method 'IO' not found for invocant of class '<anon|57619280>'␤  in block <unit> at <tmp> line 1␤␤»
10:29 Zoffix s: Distribution::Path, 'new', \($*CWD, :file("META.info".IO))
10:29 SourceBaby Zoffix, Sauce is at https://github.com/rakudo/rakudo/blob/8ad885f/src/core/Distribution.pm#L122
10:29 nine m: Distribution::Path, 'new', \($*CWD, :file("META.info".IO))
10:29 camelia rakudo-moar 8ad885: OUTPUT«WARNINGS for <tmp>:␤Useless use of constant string "new" in sink context (lines 1, 1)␤»
10:30 Zoffix Wasn't there a change from a role to a class?
10:30 nine m: Distribution::Path.new($*CWD, :file("META.info".IO))
10:30 camelia rakudo-moar 8ad885: OUTPUT«No meta file located at /home/camelia/META.info␤  in block <unit> at <tmp> line 1␤␤»
10:30 Zoffix 'cause there's a weird quirk that a role is Cool and Cool has .IO
10:30 Zoffix m: role {}.IO
10:30 camelia rakudo-moar 8ad885: OUTPUT«No such method 'IO' for invocant of type '<anon|48705520>'␤  in block <unit> at <tmp> line 1␤␤»
10:30 Zoffix Oh dafuq
10:30 Zoffix m: role {}.new.IO
10:30 camelia rakudo-moar 8ad885: OUTPUT«No such method 'IO' for invocant of type '<anon|47906832>'␤  in block <unit> at <tmp> line 1␤␤»
10:30 Zoffix m: say role {} ~~ Cool
10:30 camelia rakudo-moar 8ad885: OUTPUT«True␤»
10:31 * Zoffix is uber confused now
10:31 Zoffix m: Cool.IO
10:31 camelia rakudo-moar 8ad885: ( no output )
10:32 nine I got the error with "This is Rakudo version 2016.09-116-g2a2f26c" on my laptop. Upgrading rakudo right now.
10:35 nine Oh. It's actually fixed in current nom :)
10:35 nine So there was a regression but also a fix :)
10:36 nine Sorry for the noise
10:39 lizmat Zoffix: still time to get in a JVM fix ?
10:40 lizmat see last mail of bartolin
10:40 lizmat j: class { has int $!n = 0; method x ($!n) {} }.new.x: 42
10:40 camelia rakudo-jvm 2a1605: OUTPUT«org.perl6.nqp.sixmodel.reprs.P6OpaqueBaseInstance$BadReferenceRuntimeException: Cannot access a native attribute as a reference attribute␤  in method x at <tmp> line 1␤  in block <unit> at <tmp> line 1␤␤»
10:43 Zoffix lizmat, if someone knows how to fix it, sure. I'm still waiting for MoarVM release
10:43 lizmat well, it's a bandaid fix really, to make sure List.tail works on JVM again
10:47 dalek rakudo/nom: 666128f | lizmat++ | src/core/Rakudo/Internals.pm:
10:47 dalek rakudo/nom: Make List.tail work on JVM for now
10:47 dalek rakudo/nom:
10:47 dalek rakudo/nom: We cannot presently bind native attributes in a signature yet.
10:47 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/666128fc6e
11:04 stmuk hmmm wondering whether to delay Rakudo * until FreeBSD/OpenBSD & Linux (HLE CPUs) works?
11:04 timotimo stmuk: any chance a bios update could potentially fix the problem on all those machines?
11:05 stmuk well maybe microcode update could fix Linux
11:07 timotimo allegedly not
11:07 timotimo but a friend told me a bios update would turn the feature off during boot
11:08 stmuk I don't think HLE is the root cause since I've seen the bug (or similar) on all BSD systems with old processors
11:08 stmuk eg. qemu virtual CPU & FreeBSD
11:08 timotimo does the documentation say unlocking a mutex that's not locked at that moment is illegal?
11:10 stmuk on what platform?
11:12 stmuk bbl
11:18 timotimo i guess i'll have a quick look if libuv gives us the ability to try to lock something only if it isn't currently locked, without blockin
11:19 timotimo then we can just do that immediately before unlocking the mutex and we won't end up unlocking an unlocked mutex
11:19 timotimo how's that sound?s
11:19 nine racy?
11:19 timotimo yeah, there's uv_mutex_trylock
11:20 timotimo it'd be racy if we looked if the lock is unlocked and then lock it
11:20 timotimo because between finding out it isn't locked and unlocking it, some other thread may have locked it
11:22 nine But if we check if it's locked and we find it to be locked, so we think it's good and just before we unlock it, someone else beats us there?
11:22 timotimo wrong. if it's locked, we know we're owning the lock
11:23 timotimo of course it'd be better if we could figure out why we end up with an unlocked mutex that we're trying to unlock again in the first place
11:23 nine Then it's fine :)
11:23 nine That's the real question, yes
11:24 timotimo even though stmuk has been complaining about this for a long time, and some other users have recently started hitting this bug, nobody wanted to do anything about it yet
11:27 timotimo actually ... i think i'll even guard this shitty hack with an environment variable
11:28 * timotimo suddenly learns about secure_getenv
11:30 timotimo hm, seems like you can compile libuv with debugging and get it to error when you do a destroy, a lock, or an unlock where you're not supposed to
11:30 timotimo so maybe we can debug this on a non-b0rked linux system that way?
11:31 timotimo ... how do we actually build libuv at all?
11:43 timotimo ugh, i'm having a really hard time finding the ticket for this
11:43 timotimo stmuk: what's the RT again?
11:45 Ven_ joined #perl6-dev
11:47 nine m: role Foo {}; my $bar = 1; $bar does Foo unless $bar.does(Foo);
11:47 camelia rakudo-moar 666128: OUTPUT«WARNINGS for <tmp>:␤Useless use of $bar in sink context (line 1)␤»
11:47 nine Ha! a new useless useless warning ^^^!
11:47 nine TimToady: ^^^
11:49 nine At least Perl 6 makes it easier to edit a postfix unless into an unless statement :)
11:49 AlexDaniel what does it mean? 「$bar does Foo」
11:50 nine mixes the role into the object
11:50 AlexDaniel m: role Foo {}; my $bar = 1; $bar does Foo unless $bar.does(Foo); say $bar.WHAT
11:50 camelia rakudo-moar 666128: OUTPUT«WARNINGS for <tmp>:␤Useless use of $bar in sink context (line 1)␤(Int+{Foo})␤»
11:50 AlexDaniel indeed
11:52 timotimo pff. on linux, libuv already sets the mutex type to ERRORCHECK, which means a re-unlock is already supposed to cause an abort() immediately
11:52 timotimo (the abort is added by libuv)
11:54 lizmat jnthn_: is there a compelling reason why Str.match returns a List instead of a Seq ?
11:55 lizmat (when returning more than one Match)
11:55 bartolin_ timotimo: it's issue 410 for MoarVM, isn't it?
11:55 timotimo ah
11:55 timotimo that's why i couldn't find it on rt
11:59 timotimo stmuk: can you try out a patch for me on the systems that explode?
11:59 Ven_ joined #perl6-dev
12:02 timotimo because i just pushed a branch to moarvm that makes extra double sure that that unlock can't happen if the lock before that didn't happen
12:03 stmuk timotimo: sure!
12:05 stmuk timotimo:f2369?
12:05 timotimo 9b12a49
12:08 stmuk ah I see the branch
12:10 geekosaur joined #perl6-dev
12:18 evalable6 joined #perl6-dev
12:29 stmuk timotimo: YES! it fixes panda install on FreeBSD 11
12:29 timotimo o_O
12:29 timotimo okay, could you perhaps put a hardware watchpoint on the variable that has_user_threads reads?
12:30 timotimo because if this fixed it, that value seems to change during the body of the multi_cache_add function
12:32 timotimo well, at least i don't see a reason why this shouldn't be on master
12:32 stmuk timotimo: I can do this later but I was going out shortly
12:32 timotimo OK
12:32 timotimo i'm glad this fixed it
12:33 timotimo i hope this is the same problem linux with broken HLE has?
12:33 stmuk the stacktraces looked v similar
12:34 timotimo as long as multi_cache_add was on the stack trace, it'll potentially be the right thing
12:34 stmuk timotimo: OK I can do further testing today early Euroish evening time
12:34 timotimo thing is, it shouldn't be possible for this to fix anything
12:34 timotimo oh
12:34 timotimo yeah, it is possible i guess
12:34 timotimo if you make enough new threads that the thread id wraps
12:35 stmuk bbfn
12:36 timotimo does zef or panda create about 2**32 or 2**64 threads to install stuff?
12:38 lizmat timotimo: pretty sure it doesn'
12:38 lizmat t
12:39 timotimo yeah, i'm pretty sure, too
12:39 timotimo it'd probably take a couple of days to create that many threads
12:54 geekosaur joined #perl6-dev
13:09 lizmat afk&
13:16 timotimo m: Thread.new for ^(2**32)
13:16 camelia rakudo-moar 666128: OUTPUT«Required named parameter 'code' not passed␤  in block <unit> at <tmp> line 1␤␤»
13:16 timotimo m: Thread.new(:code({ sleep 10 })) for ^(2**32)
13:16 camelia rakudo-moar 666128: OUTPUT«Memory allocation failed; could not allocate 131072 bytes␤»
13:16 timotimo that was fast
13:17 timotimo yeah, that kind of code gets you OOM-killed rather quickly :)
13:25 evalable6 joined #perl6-dev
14:42 committable6 joined #perl6-dev
15:26 dalek roast: a08298d | (Harrison Chienjo)++ | S29-os/system.t:
15:26 dalek roast: Tests for Proc.command and Proc.signal (#169)
15:26 dalek roast:
15:26 dalek roast: * Add test for Proc.command.
15:26 dalek roast:
15:26 dalek roast: * Fix missing last parameter.
15:26 dalek roast:
15:26 dalek roast: * Add test for proc.command returned by shell().
15:26 dalek roast:
15:26 dalek roast: * Add .signal test for Proc.
15:26 dalek roast:
15:26 dalek roast: * Use correct name.
15:26 dalek roast:
15:26 dalek roast: * Fix wrong name in test description.
16:19 TimToady nine: thanks, put it into my uu file :)
16:48 mst_ joined #perl6-dev
16:51 Zoffix Release! Release! Release! Release!
16:51 * Zoffix is impatient :D
16:54 mst bloody canadians
16:54 moritz so, who does the release?
16:54 Zoffix NeuralAnomaly does.
16:55 Zoffix But we're all waiting for jnthn++ to cut MoarVM release :)
17:05 lucasb_ joined #perl6-dev
17:11 mst WE DEMAND MOAR
17:11 mst Zoffix: NeuralAnomaly is yet another of your botswarm?
17:11 * lucasb_ is waiting too :)
17:11 Zoffix mst, yup
17:12 geekosaur zoffix is a bit botty >.>
17:14 moritz taking over the world with IRC and release bots. I, for one, welcome our now electronic overlords!
17:14 Zoffix \o/
17:15 nine Seems like we really need a MoarVM release bot ;)
17:22 dalek roast: 58047e6 | moritz++ | S09-typed-arrays/native-decl.t:
17:22 dalek roast: Very basic my str @ tests
17:22 dalek roast: review: https://github.com/perl6/roast/commit/58047e68ba
17:37 dalek roast: c8889fc | moritz++ | S03-operators/autoincrement.t:
17:37 dalek roast: Test pre- and postfix autoincrement on Num
17:37 dalek roast: review: https://github.com/perl6/roast/commit/c8889fc78a
17:38 Zoffix moritz, I guess I should've mentioned that coverage report is a bit outdated :}
17:38 Zoffix Not that I mind extra tests.
17:38 moritz arglbargl
17:39 Zoffix :}
17:39 Zoffix The other one are in https://github.com/perl6/roast/commit/2a3cdb2e257e97d8c6dafefecc8f6221f87da80f
17:40 moritz and this being roast, there are least three sensible places where such tests should go
17:40 * Zoffix nods
17:41 dalek roast: c64cd86 | moritz++ | S03-operators/autoincrement.t:
17:41 dalek roast: Revert "Test pre- and postfix autoincrement on Num"
17:41 dalek roast:
17:41 dalek roast: This reverts commit c8889fc78a3b46a4721406b7d787fe1ccf8c1df1.
17:41 dalek roast:
17:41 dalek roast: Turns out those ops were already covered in commit
17:41 dalek roast: 2a3cdb2e257e97d8c6dafefecc8f6221f87da80f / file S02-types/num.t
17:41 dalek roast: review: https://github.com/perl6/roast/commit/c64cd86eda
17:57 dalek rakudo/nom: 1acf5bf | TimToady++ | src/Perl6/Actions.nqp:
17:57 dalek rakudo/nom: does is a mutator, so should want its left arg
17:57 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/1acf5bf88d
17:58 TimToady nine: you can go ahead and flip your 'unless' back if you like, though seems to me 'does' should be smart enough not to reapply a role, which oughta be safe since roles are immutable
18:00 TimToady and reapplying a role ought to be idempotent
18:06 nine TimToady: thanks! That means, I'm now gonna have to find out why exactly I need that unless. I'm sure I've tried it without at first.
18:08 Zoffix m: role Foo {}; my $x = 42; quietly $x does Foo for ^10; dd $x
18:08 camelia rakudo-moar 666128: OUTPUT«Int+{Foo}+{Foo}+{Foo}+{Foo}+{Foo}+{Foo}+{Foo}+{Foo}+{Foo}+{Foo} $x = 42␤»
18:09 Zoffix s: &infix:<does>
18:09 SourceBaby Zoffix, Sauce is at https://github.com/rakudo/rakudo/blob/666128f/src/core/operators.pm#L15
18:48 timotimo .comb(0) is apparently the same as .comb(1)
18:48 timotimo do we want that?
18:49 Zoffix m: 'foo'.comb(-1000000).say
18:49 camelia rakudo-moar 1acf5b: OUTPUT«(f o o)␤»
18:49 Zoffix IMO throwing would be good.
18:57 timotimo m: "hello".comb(NaN)
18:57 camelia rakudo-moar 1acf5b: OUTPUT«Cannot resolve caller comb(Str: Num); none of these signatures match:␤    (Cool $: *%_)␤    (Cool $: Regex $matcher, $limit = Inf, *%_)␤    (Cool $: Str $matcher, $limit = Inf, *%_)␤    (Str:D $: *%_)␤    (Str:D $: Int:D $size, $limit = *, *%_)…»
19:00 Zoffix Well, with it being 9PM in Prague, I don't think there'll be a release today.... So I'm free to go out, get booze, and get drunk :P
19:00 mst oh, good, you're back in the fold again
19:00 mst I was worried there for a minute
19:00 timotimo that's okay
19:00 Zoffix lol
19:01 Zoffix mst, I didn't drink since last Friday.
19:01 Zoffix It's very un-Zoffix-like
19:01 lucasb_ stay healthy Zoffix++
19:01 mst Zoffix: I'm so sorry
19:01 Zoffix lol
19:02 mst ugh. at some point I should try and write an irssi filter that strips mangled ++s like that
19:03 lucasb_ .comb returns nonsense values when you pass non-positive and a limit
19:03 lucasb_ m: say 'abcdef'.comb(0, 2).perl
19:03 camelia rakudo-moar 1acf5b: OUTPUT«("b", "c").Seq␤»
19:03 lucasb_ m: say 'abcdef'.comb(-1, 2).perl
19:03 camelia rakudo-moar 1acf5b: OUTPUT«("c", "d").Seq␤»
19:03 timotimo but is that really nonsense?
19:03 timotimo m: say 'abcdef'.comb(-2, 2)'perl
19:03 camelia rakudo-moar 1acf5b: OUTPUT«===SORRY!=== Error while compiling <tmp>␤Two terms in a row␤at <tmp>:1␤------> say 'abcdef'.comb(-2, 2)⏏'perl␤    expecting any of:␤        infix␤        infix stopper␤        postfix␤        statement end␤        statement…»
19:03 timotimo m: say 'abcdef'.comb(-2, 2).perl
19:03 camelia rakudo-moar 1acf5b: OUTPUT«("d", "e").Seq␤»
19:03 timotimo m: say 'abcdef'.comb(-3, 2).perl
19:03 camelia rakudo-moar 1acf5b: OUTPUT«("e", "f").Seq␤»
19:03 timotimo that's so cool
19:04 timotimo not sensible to keep, though
19:04 * timotimo AFK
19:13 nine m: role Foo { has @.attrs }; my $i = 1; $i does Foo; $i.attrs.push: 1; $i does Foo; $i.attrs.push: 2; say $i.attrs
19:13 camelia rakudo-moar 1acf5b: OUTPUT«[2]␤»
19:13 nine TimToady: ^^^ not as idempotent as it ought to be
19:21 * Zoffix changed mind
19:22 Zoffix Don't wanna feel like crap tomorrow :}
19:26 nine Zoffix: that's why I rarely drink these days. Too much productive time lost. And too many awesome things to try :)
19:27 Zoffix :)
21:41 Ven_ joined #perl6-dev
21:43 Ven__ joined #perl6-dev
22:00 Zoffix "No appropriate parametric role variant available for 'Inline::Perl5::Perl5Attributes'Compilation failed in require at (eval 1) line 122."
22:00 Zoffix nine, you around? :)
22:01 Zoffix Full output. That's while trying to install Inline::Perl5 with bleed rakudo: https://gist.github.com/zoffixznet/578a8306e8b0aec2b26a23bde0bad386
22:03 lizmat Zoffix: perhaps this is related to TimToady's last commit ?
22:04 lizmat aka 1acf5bf88d8aea231509 ?
22:06 Zoffix m: role Foo {}; my $bar = 1; $bar does Foo unless $bar.does(Foo);
22:06 camelia rakudo-moar 1acf5b: ( no output )
22:06 Zoffix m: role Foo {}; my $bar = 1; $bar does Foo unless $bar.does(Foo); dd $bar
22:06 camelia rakudo-moar 1acf5b: OUTPUT«Int+{Foo} $bar = 1␤»
22:08 Zoffix gonna try to build weithout it
22:10 Zoffix Hmm... "No need for Perl5Attributes to be parametric": https://github.com/niner/Inline-Perl5/commit/f3a06bf815649f9e4f134f5be8341f4002eb36ae
22:15 * lizmat is tired and goes to bed
22:17 Zoffix Still fails with TT's change, so I'm betting my money on the above commit
22:18 Zoffix I mean without TT's commit :)
22:18 lizmat ok, well, it was related to roles
22:19 lizmat sleep&
22:19 Zoffix night
22:35 dalek joined #perl6-dev
22:35 synopsebot6 joined #perl6-dev
22:37 Zoffix Well, if I --notest install it, the stresstest seems to pass, so no rush on fixing it :)
22:37 Undercover joined #perl6-dev
22:37 SourceBaby joined #perl6-dev
22:38 timotimo all those bots run on hack, then?
22:39 Zoffix Just the above two
22:40 timotimo ok
22:40 timotimo but you create the data they use on a beefier bot?
22:40 timotimo obx*
22:40 timotimo box*
22:40 Zoffix The coverage stuff I run on my home box. One of the bots does build each rakudo commit though
22:41 timotimo cool
22:41 timotimo did you ever have a look at collect.p6c.org?
22:41 Zoffix Never heard of it
22:42 timotimo it's a monitoring thing
22:47 NeuralAnomaly joined #perl6-dev
22:48 Zoffix NeuralAnomaly, status
22:48 NeuralAnomaly Zoffix, [✘] Next release is today. Since last release, there are 47 new still-open tickets (1 unreviewed and 0 blockers) and 2 unreviewed commits. See http://perl6.fail/release/stats for details
22:48 Zoffix NeuralAnomaly, status
22:48 NeuralAnomaly Zoffix, [✔] Next release is today. Since last release, there are 47 new still-open tickets (0 unreviewed and 0 blockers) and 0 unreviewed commits. See http://perl6.fail/release/stats for details
22:49 timotimo cool :)
22:49 Zoffix NeuralAnomaly, it's time
22:49 NeuralAnomaly Zoffix, Oh boy! Really?! We're doing a realease‽‽ YEY!
22:49 Zoffix NeuralAnomaly, cut the release
22:49 NeuralAnomaly Zoffix, Will do! If you're feeling particularly naughty, you can watch me at http://perl6.fail/release/progress or go look at some cats http://icanhas.cheezburger.com/lolcats
22:49 NeuralAnomaly Zoffix, ♥♥♥♥♥♥ Prep done
22:49 AlexDaniel by the way, what's up with http://smoke.perl6.org/report ?
22:49 dalek nqp: 7914a19 | (Zoffix Znet)++ | tools/build/MOAR_REVISION:
22:49 dalek nqp: bump MoarVM version to 2016.10
22:49 dalek nqp: review: https://github.com/perl6/nqp/commit/7914a19fd6
22:49 dalek nqp: 223373d | (Zoffix Znet)++ | VERSION:
22:49 dalek nqp: bump VERSION to 2016.10
22:49 dalek nqp: review: https://github.com/perl6/nqp/commit/223373d86c
22:52 timotimo oh, the release comes with a java thing now?
22:53 Zoffix What java thing?
22:53 timotimo it's building rakudo-j right now
22:53 timotimo no, nqp-j
22:53 Zoffix No, it's building nqp-j
22:53 Zoffix It does build and test nqp-j, but not rakudo-j
22:54 timotimo OK!
22:58 NeuralAnomaly Zoffix, ♥♥♥♥♥♥ nqp tests OK
23:00 dalek joined #perl6-dev
23:01 timotimo it's building nqp-j again?
23:02 Zoffix Yes. It made the tarball and it's now testing it.
23:05 mst you killed dalek
23:05 mst * Zoffix is now known as ZoffixWho
23:06 Zoffix :D
23:06 Zoffix I don't think it was me
23:06 Zoffix It was AlexDaniel
23:06 AlexDaniel yeah…
23:07 NeuralAnomaly Zoffix, ♥♥♥♥♥♥ nqp release tarball tests OK
23:08 NeuralAnomaly Zoffix, ♥♥♥♥♥♥ nqp release DONE
23:08 dalek rakudo/nom: 090c392 | (Zoffix Znet)++ | docs/announce/2016.10.md:
23:08 dalek rakudo/nom: Generate release announcement for 2016.10
23:08 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/090c392542
23:08 dalek rakudo/nom: 7c62a41 | (Zoffix Znet)++ | tools/build/NQP_REVISION:
23:08 dalek rakudo/nom: [release] bump NQP revision
23:08 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/7c62a41d9d
23:08 dalek rakudo/nom: eb3123e | (Zoffix Znet)++ | VERSION:
23:08 dalek rakudo/nom: [release] bump VERSION to 2016.10
23:08 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/eb3123e5e6
23:11 evalable6 joined #perl6-dev
23:15 NeuralAnomaly Zoffix, ♥♥♥♥♥♥ Rakudo stresstest (master) OK
23:17 NeuralAnomaly Zoffix, ♥♥♥♥♥♥ Rakudo stresstest (6.c-errata) OK
23:20 lucasb_ joined #perl6-dev
23:25 NeuralAnomaly Zoffix, ♥♥♥♥♥♥ Rakudo release DONE
23:25 NeuralAnomaly Zoffix, ♥♥♥♥♥♥ Post: upload tarballs to rakudo.org
23:25 NeuralAnomaly Zoffix, ????????????????????????????????????????????????????????????
23:25 NeuralAnomaly Zoffix, The release of **Rakudo #104 2016.10** has now been completed
23:25 NeuralAnomaly Zoffix, ????????????????????????????????????????????????????????????
23:25 * NeuralAnomaly celebrates with an appropriate amount of fun
23:25 Zoffix NeuralAnomaly, thanks
23:25 NeuralAnomaly Zoffix, any time, buddy!
23:26 p3rln00b
23:26 yoleaux2 08:15Z <Zoffix> p3rln00b: revert https://github.com/perl6/nqp/commit/298e2281ec after the release
23:28 dalek nqp: 3e14289 | (Zoffix Znet)++ | t/qast/01-qast.t:
23:28 dalek nqp: Revert "Temporarily comment out tests failing on JVM"
23:28 dalek nqp:
23:28 dalek nqp: This reverts commit 298e2281ec01c7c61e446024ca59078e3a3748bb.
23:28 dalek nqp:
23:28 dalek nqp: Unfudges failing JVM tests
23:28 dalek nqp: review: https://github.com/perl6/nqp/commit/3e142894cf
23:28 Zoffix And now I can retreat to play all three parts of Star Craft II that should finish downloading in 10-15 minutes \o/
23:30 dalek rakudo/nom: 1942243 | (Zoffix Znet)++ | docs/release_guide.pod:
23:30 dalek rakudo/nom: 2016.10 is now in the past
23:30 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/19422430e4
23:35 dalek rakudo/nom: b744de3 | (Zoffix Znet)++ | docs/ChangeLog:
23:35 dalek rakudo/nom: Prep 2016.11 section
23:35 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/b744de35b0
23:36 evalable6 joined #perl6-dev
23:37 * timotimo only played the first two parts of that
23:37 evalable6 joined #perl6-dev
23:40 evalable6 joined #perl6-dev
23:40 Zoffix I played them a few times in the past, but never the last one
23:41 * timotimo to bed

| Channels | #perl6-dev index | Today | | Search | Google Search | Plain-Text | summary