Perl 6 language and compiler development 2.0 | Logs at irclog.perlgeek.de/perl6-dev/today | For toolchain/installation stuff see #perl6-toolchain | For MoarVM see #moarvm
Set by moderator on 14 May 2018.
00:01 MasterDuke joined 01:57 ilbot3 joined
moderator Perl 6 language and compiler development 2.0 | Logs at irclog.perlgeek.de/perl6-dev/today | For toolchain/installation stuff see #perl6-toolchain | For MoarVM see #moarvm
02:32 ufobat_ joined 03:58 commavir joined 04:03 lizmat joined 04:28 commavir joined 04:40 commavir joined 04:52 commavir joined 05:10 commavir joined 05:33 thundergnat joined
nine .tell AlexDaniel` the Inline::Perl5 failures are caused by some change in MoarVM since 2018.04. I'd call that a blocker. 06:40
yoleaux 13 May 2018 20:45Z <AlexDaniel> nine: Please update the status of R#1783 (is it fully resolved? If not, what is left?)
synopsebot R#1783 [open]: github.com/rakudo/rakudo/issues/1783 [⚠ blocker ⚠] New sporadic explosions likely related to precomp
yoleaux nine: I'll pass your message to AlexDaniel`.
06:50 Ven`` joined 06:52 robertle joined 06:56 Ven`` joined
nine .tell AlexDaniel` I've found the offending commit in MoarVM and reverted it. Inline::Perl5 is fine again! 07:03
yoleaux nine: I'll pass your message to AlexDaniel`.
07:04 Ven`` joined 07:07 |Tux| joined 07:39 Ven` joined 09:04 brrt joined
Geth nqp: c291768115 | (Elizabeth Mattijsen)++ | tools/build/MOAR_REVISION
Bump Moar for latest precomp fixes + new JITs
 09:06
¦ nqp: version bump brought these changes: github.com/MoarVM/MoarVM/compare/2...5-ga607c42
09:17 Ven`` joined
Geth rakudo: 9aa6747f8a | (Elizabeth Mattijsen)++ | tools/build/NQP_REVISION
Bump Moar to get the latest fixes / improvements
 09:30
rakudo: version bump brought these changes: github.com/perl6/nqp/compare/2018....1-gc291768
8d12d5b42d | (Elizabeth Mattijsen)++ | src/core/Array.pm6

  - uses the new nqp::splice op
  - probably faster still for larger result sets
  - fewer allocations, so more memory / GC friendly
|Tux| Rakudo version 2018.04.1-83-g9aa6747f8 - MoarVM version 2018.04.1-115-ga607c42f1
csv-ip5xs0.859 - 0.896
csv-ip5xs-208.599 - 8.628
csv-parser24.343 - 25.248
csv-test-xs-200.454 - 0.464
test8.311 - 8.752
test-t2.368 - 2.436
test-t --race0.972 - 0.978
test-t-2041.844 - 43.040
test-t-20 --race14.491 - 14.877
 09:54
tbrowder_ Hi, folks. 10:33
I’m going to try to merge my Rakudo PR #1826 tomorrow unless there are objections. 10:35
10:42 commavir joined 10:48 commavir joined 10:55 commavir joined
tbrowder_ While on the topic of Perl 6 pod tables and the Rakudo GH #1825 entered by Zoffix which resulted in PR #1826, I see that the pod table rules could be changed for the better without sacrificing backwards compatibility. For instance, the table Zoffix was trying to use could be simplified to have NO pipes (visual column separators) in any row EXCEPT the row with two columns. That currently is not the case, but the change 10:57
would be easy to make.
11:04 commavir joined 11:10 Ven`` joined, commavir joined 11:27 Ven` joined 11:55 brrt joined, commavir joined 12:07 Ven`` joined 12:23 Zoffix left 12:29 Ven`` joined 12:30 AlexDaniel joined 12:40 Ven`` joined 13:18 Ven`` joined 13:47 thundergnat joined 14:07 Ven`` joined 14:11 lizmat joined 15:02 skids joined 15:11 krunen joined 15:15 lizmat joined
16:32 lucasb joined
Geth rakudo: a5d46b4273 | (Elizabeth Mattijsen)++ | src/core/Array.pm6
Make @a.splice(offset,length,...) about 20% faster

  - uses the new nqp::splice op
  - probably faster still for larger result sets
  - fewer allocations, so more memory / GC friendly		 16:46
16:56 travis-ci joined
travis-ci Rakudo build failed. Elizabeth Mattijsen 'Make @a.splice(offset) about 20% faster 16:56
travis-ci.org/rakudo/rakudo/builds/379292859 github.com/rakudo/rakudo/compare/9...12d5b42d9c
16:56 travis-ci left
buggable [travis build above] ☠ All failures are due to: failed make test (1 failure). Across all jobs, only t/05-messages/10-warnings.t test file failed. 16:56
Geth rakudo: 5ce24929f5 | (Aleks-Daniel Jakimenko-Aleksejev)++ | 2 files
Re-add $#foo 5to6 error message

Accidentally thrown away in 4ac9915ef66ce054335fccc938a5fc594ec97c24. We didn't have any test for it in roast master, but there is a test in v6.c-errata.
The code for `$#` was used for both $# and $#foo feature. The former is a pre-perl5 variable, and the latter is used frequently in perl5.
I consider this particular 5to6 error message to be useful, so adding it back (but only in $#foo form).		 17:06
roast/6.c-errata: 99c07bc1e9 | (Aleks-Daniel Jakimenko-Aleksejev)++ | S32-exceptions/misc.t
Remove some tests for 4to6 and 5to6 error messages

Part of rakudo/rakudo#1356.		 17:10
synopsebot RAKUDO#1356 [open]: github.com/rakudo/rakudo/issues/1356 [RFC] Let's remove most (if not all) perl5-oriented error messages for variables
Geth roast: ca4093b22d | (Aleks-Daniel Jakimenko-Aleksejev)++ | S32-exceptions/misc2.t
Re-add a test for $#foo 5to6 error message

The test existed before but was removed for some reason.		 17:13
japhb nine: Are you able to `zef install Inline::Python` on a completely clean Rakudo? I've been having all sorts of spurious errors. So far I've tried installing System::Query manually first, and now I'm doing System::Query and Distribution::Builder::MakeFromJSON, hoping to get a little further, but it seems like something in that module dependency chain confuses the heck out of zef. 17:25
17:32 lizmat joined
nine japhb: there's definitely something wrong with zef there 17:37
samcv good * from the usa again 18:03
timotimo hey samcv, how was your flight?
samcv long 18:04
nine japhb: pushed a fix to zef
timotimo were the seats comfortable at least?
samcv no 18:09
:)
nine Python is gonna get real concurrency - through multiple interpreters: lwn.net/SubscriberLink/754162/4cb890926a1fe5a9/
This will catapult Python straight into the 90s! 18:10
timotimo that's basically just multiplicity for python?
is that what that perl feature is called?
nine That's its name, yes. Proposal seems to be multiplicity + channels to share data between interpreters 18:11
timotimo ah, i see zero-copy data sharing in there
18:11 travis-ci joined
travis-ci Rakudo build passed. Aleks-Daniel Jakimenko-Aleksejev 'Re-add $#foo 5to6 error message 18:11
travis-ci.org/rakudo/rakudo/builds/379317329 github.com/rakudo/rakudo/compare/a...e24929f5bd
18:11 travis-ci left
timotimo we don't currently have a good way to do that with heap-managed datastructures 18:11
nine "A key characteristic of those operations is that channels transmit data derived from Python objects rather than the objects themselves. When objects are sent, their data is extracted. When the "object" is received in the other interpreter, the data is converted back into an object." 18:13
timotimo :\\
nine Doesn't sound like zero-copy to me
timotimo so basically serialization and deserialization
when that one PEP mentioned there lands, we can have multiple Inline::Python instances active in the same process, too, right? 18:16
japhb Zero copies *of the serialized data* perhaps? 18:22
nine, thank you re: zef fix. 18:23
nine Yes, we can
japhb nine: Are zef commits reported in any of our channels?
I was curious what the fix was, but didn't see it here or in perl6-toolchain. 18:24
nine github.com/ugexe/zef/commit/004eea...d97e7940d1 18:26
japhb thx 18:35
releasable6 Next release in ≈3 days and ≈23 hours. 1 blocker. Please log your changes in the ChangeLog 19:00
AlexDaniel speaking of which, releasable is happy 19:01
canary is green: travis-ci.org/perl6/rakudo-pkg-can...tification
will get toaster results in less than an hour I think
this blocker is still there though: R#1810 19:04
synopsebot R#1810 [open]: github.com/rakudo/rakudo/issues/1810 [regression][⚠ blocker ⚠] Recent regression "ctxlexpad needs an MVMContext"
19:23 reportable6 joined 19:24 greppable6 joined, undersightable6 joined, shareable6 joined
AlexDaniel there we go: toast.6lang.org/ 19:25
though I'm not sure why anybody but me may need to look at it :) 19:26
samcv AlexDaniel++ more toaster
AlexDaniel well, I run it several times every month, I just never published it 19:27
timotimo hm, Crane gots quite a few test failures; it shows up in the assixt log 19:35
AlexDaniel yeah, looking
timotimo suddenly containers where it didn't have any before? 19:36
19:46 Ven`` joined
AlexDaniel heh, I guessed it and bisectable confirmed, it's github.com/rakudo/rakudo/commit/77...5caee85a70 19:56
19:58 Ven`` joined 20:16 Ven`` joined 20:28 Ven`` joined
timotimo man, i'm glad we have an ecosystem that can catch things like this 21:23
AlexDaniel timotimo: and we were not using it before 2017.05 :) 21:31
timotimo mhm
AlexDaniel I think we'll need a special note for this release about the order of .keys in hashes 21:32
some modules flopped because of that 21:33
github.com/ugexe/Perl6-App--ecogen/pull/3 github.com/kalkin/License-Software/pull/2
Magento ( github.com/scmorrison/perl6-Magento ) module also has this issue 21:34
but I'm not doing a pull request for it today
I also haven't looked at Koos and Spit failures yet, they're weird 21:35
I'm suspecting that a lot more modules are expecting .keys in a particular module, but you'd have to run the tests many times to see any failures… 21:36
s/module/order/
btw for SQUASHathon 2018-08 I think we should go through all always-red modules and try to revive them 21:38
318 of them
well, minus a few that we will fix right now (I think?) 21:39
22:11 ufobat_ joined
lucasb committable6: 2018.04 my @a; for ^5 { push @a, my @o = Empty, $_ }; say @a 22:16
committable6 lucasb, ¦2018.04: «[[0] [1] [2] [3] [4]]␤»
lucasb m: my @a; for ^5 { push @a, my @o = Empty, $_ }; say @a 22:17
camelia [[4] [4] [4] [4] [4]]
timotimo hah, neat. missing a decont in the new version
Geth ¦ rakudo: zoffixznet self-assigned Regression exposed in Crane and other modules github.com/rakudo/rakudo/issues/1832 22:18
¦ rakudo: zoffixznet self-unassigned Regression exposed in Crane and other modules github.com/rakudo/rakudo/issues/1832 22:38
lucasb Zoffix++ for issue commenting 23:07
BTW We are in 2018 already :D 23:08
23:11 MasterDuke joined
Geth rakudo/car-grant-midrat: 73 commits pushed by 16 authors
review: github.com/rakudo/rakudo/compare/6...f400e87824		 23:15
roast/car-grant-midrat: 41 commits pushed by 10 authors
review: github.com/perl6/roast/compare/967...b3137fc26c

Powered by Cro and the Raku® Programming Language.

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

Thank you!