00:07 hungrydonkey joined, Kaiepi left, Kaiepi joined 00:58 pamplemousse joined
Geth nqp: 9cd9d8f6a4 | (Aleks-Daniel Jakimenko-Aleksejev)++ | tools/templates/MOAR_REVISION
[MoarVM Bump] Brings 13 commits

MoarVM bump brought: github.com/MoarVM/MoarVM/compare/2...g3d8ff61e2 3d8ff61e2 Merge pull request #1280 from MasterDuke17/use_fsa_for_cbq c59514234 Use the FSA for ConcBlockingQueue c0fe97a68 Fix variable might be clobbered by ‘longjmp’ or ‘vfork’ warnings in interp.c 5863b0235 Silence a compiler warning in add_nodes_for_typed_argument ... (9 more lines)
rakudo: 5ee3d9240d | (Aleks-Daniel Jakimenko-Aleksejev)++ | tools/templates/NQP_REVISION
[NQP Bump] Brings 2 commits

NQP bump brought: github.com/perl6/nqp/compare/2020....g9cd9d8f6a 9cd9d8f6a [MoarVM Bump] Brings 13 commits e4db2f51f Some minor optimizations to NQP matching
01:41 pamplemousse left 01:42 pamplemousse joined 01:47 pamplemousse left 02:10 pamplemousse joined
vrurg .seen altai-man 02:16
tellable6 vrurg, I saw altai-man 2020-04-29T16:33:15Z in #raku: <Altai-man_> Chicken-egg problem. Who will adapt the software if everyone says "Oh, so it is not adapted yet, we'll wait till someone else adapts that".
vrurg .tell Altai-man I think we could try unrolling the next-dispatcher stuff to unblock releases. 02:18
tellable6 vrurg, I'll pass your message to Altai-man_
03:03 maggotbrain joined 03:11 pamplemousse left 03:46 hungrydonkey left, hungrydonkey joined
guifa2 Interesting. I can't find anywhere how <{ ... }> works in regex. Per the grammar, assert is defined as '<' ~ '>' <assertion>, but there is no assertion that includes braces and a codeblock. 04:43
I mean, it's easy enough for me to implement in binex, by just adding '{' ~ '}' <codeblock> as an assertion 04:44
06:39 hungrydonkey left, hungryd99 joined
MasterDuke guifa2: don't know in the grammar, but the implementation is in INTERPOLATE and INTERPOLATE_ASSERTION i believe. could try to find out how they get called 06:42
06:42 [Tux] left 06:46 hungryd99 left 06:49 [Tux] joined 06:54 hungrydonkey joined 07:01 hungrydonkey left 07:02 hungrydonkey joined 07:30 hungrydonkey left 07:43 hungrydonkey joined 08:00 Altai-man_ joined
Kaiepi so... turns out there is a way to handle binding and listening sockets separately for the jvm for that bind before connect pr! 08:16
it's just very, very annoying to write
bah, nvm, i doubt this would be very robust 08:55
08:57 sena_kun joined, AlexDaniel left 08:58 AlexDaniel joined, AlexDaniel left, AlexDaniel joined, Altai-man_ left 09:04 hungrydonkey left 09:34 tyilanmenyn joined 09:42 Kaiepi left 09:43 Kaiepi joined 09:59 hungrydonkey joined 10:20 MasterDuke left, hungrydonkey left 10:25 tyilanmenyn is now known as tyil
lizmat Files=1306, Tests=111242, 215 wallclock secs (29.32 usr 8.45 sys + 3020.98 cusr 279.25 csys = 3338.00 CPU) 10:32
10:46 Ven`` joined 10:56 Altai-man_ joined 10:57 Ven`` left 10:58 sena_kun left
Kaiepi m: say $?LANG.dump 11:15
camelia - desigilname: LANG
- longname: LANG
- name: LANG
- identifier: LANG
- morename: isa NQPArray
- colonpair: isa NQPArray
- sigil: $
- twigil: ?
- sym: ?
Kaiepi m: say $*LANG.dump
camelia Dynamic variable $*LANG not found
in block <unit> at <tmp> line 1
Kaiepi m: BEGIN say $*LANG.dump
camelia - statement: isa NQPArray
Kaiepi are $?LANG and $*LANG supposed to be the same?
$?LANG is always the ast for the literal $?LANG variable afaik 11:16
s/ast/qast/ 11:17
er, the match i mean 11:23
11:38 Ven`` joined
lizmat Kaiepi: that part of speculatiion is rather handwavy 11:54
11:55 MasterDuke joined
Kaiepi wdym? 11:56
lizmat about $?LANG and $*LANG supposedly being the same always 12:04
Kaiepi they're not 12:06
i'm just not sure if $?LANG is supposed to be the match for the $?LANG variable itself like it is in the cases i've tested or not 12:10
12:27 hungrydonkey joined
Geth rakudo: 0aa8b2a8fd | (Elizabeth Mattijsen)++ | src/core.c/Supply-factories.pm6
Use a smarter semaphore in Supply.reduce

Since the produced values can never be an nqp::null, use nqp::null as a semaphore to indicate the first value has not yet been seen. Removes the need for a boolean semaphore.
12:32 hungryd47 joined, hungrydonkey left
Geth rakudo: 9cabea01d4 | (Elizabeth Mattijsen)++ | src/core.c/Supply-factories.pm6
Use a smarter semaphore in Supply.produce

Since the produced values can never be an nqp::null, use nqp::null as a semaphore to indicate the first value has not yet been seen. Removes the need for a boolean semaphore.
12:41 pamplemousse joined 12:57 sena_kun joined 12:58 Altai-man_ left 13:07 hungryd47 left
Geth roast/6.d-errata: 5f50410fcf | (Elizabeth Mattijsen)++ (committed by Altai-man) | S32-list/iterator.t
Properly test for Mu.new.iterator
sena_kun releasable6, status 13:12
releasable6 sena_kun, Next release will happen when it's ready. 3 blockers. 166 out of 372 commits logged (⚠ 3 warnings)
sena_kun, Details: gist.github.com/96365d4eed1fcfe3cb...b2c49f339e
13:28 hungrydonkey joined 14:05 hungrydonkey left 14:07 hungrydonkey joined
Geth rakudo/dispatcher-revert: 5 commits pushed by Altai-man++ 14:11
AlexDaniel c: dispatcher-revert say 42 14:31
committable6 AlexDaniel, ¦dispatcher-revert: «Cannot find this revision (did you mean “segv-coercer-qast”?)»
AlexDaniel hmmmmm 14:32
c: HEAD say 42
committable6 AlexDaniel, ¦HEAD(9cabea0): «42␤»
Geth roast/dispatcher-revert: c148a44a98 | Altai-man++ | 2 files
Revert dispatcher tests
AlexDaniel c: dispatcher-revert say 42
committable6 AlexDaniel, ¦dispatcher-revert: «42␤»
vrurg It wasn't good idea to revert in roast. Mark as TODO is more correct. 14:41
AlexDaniel why? 14:42
sena_kun AlexDaniel, to not forget to bring it back once we have proper impl, maybe?
vrurg, can dispatcher reverts lead to races in case if something was not undone properly? 14:43
14:56 Altai-man_ joined 14:58 sena_kun left
Geth roast: Altai-man++ created pull request #638:
[WIP] Revert dispatcher tests
rakudo: Altai-man++ created pull request #3651:
Dispatcher revert
vrurg AlexDaniel: as it said, tests are checking for the correct behavior.
Altai-man_: races? where?
Altai-man_: In either case, we'd need to test it first. 15:05
vrurg is afk again.
Altai-man_ vrurg, e.g. I saw failures in stresstest files, e.g. github.com/Raku/roast/blob/master/...n/method.t but not reproducible.
s/e.g.// 15:07
15:22 Summertime joined
Altai-man_ AlexDaniel, I saw you mentioning blin now can (or soon will be) work with branches? 15:43
AlexDaniel Altai-man_: I'm pretty sure it works now, yeah
Altai-man_: can you try it?
Altai-man_ AlexDaniel, I am using an image created 8 days ago, but it sayd endpoint is not found, do I need a more recent version? 15:44
AlexDaniel Altai-man_: yes
Altai-man_ Ok, will build one, thanks!
AlexDaniel Altai-man_: just tested it like this: --new=dispatcher-revert 15:55
that is, without even providing a SHA
it works. Magic
Altai-man_ AlexDaniel++ 16:00
16:42 Ven`` left
[Tux] Rakudo version 2020.02.1-372-g9cabea01d - MoarVM version 2020.02.1-162-g3d8ff61e2
csv-ip5xs0.898 - 0.938
csv-ip5xs-209.781 - 9.951
csv-parser25.340 - 27.295
csv-test-xs-200.384 - 0.408
test7.984 - 8.095
test-t2.765 - 2.861
test-t --race1.010 - 1.132
test-t-2047.618 - 49.001
test-t-20 --race12.651 - 13.410
16:57 sena_kun joined 16:59 Altai-man_ left
vrurg sena_kun: did you see failures after reverting the PRs? 17:35
18:02 hungrydonkey left
guifa2 MasterDuke: Aha, found it! The P6Regex gets subclassed in NQP::Grammar and that's where it's hiding. Strange that it was done that way 18:11
sena_kun vrurg, yes. But I wasn't able to spot them again when running stresstest about 10 times in a row after that. 18:34
18:56 Altai-man_ joined 18:59 sena_kun left 19:18 Kaiepi left 19:19 Kaiepi joined 19:22 Kaiepi left
timotimo i am now running pugs, but i haven't found pugscc yet 19:29
i see the compile options for JS and such. JS says "couldn't find anything", some PIR-based ones say "invalid name: ..." 19:34
19:35 pamplemousse left, pamplemousse joined
[Coke] PIR? 19:52
timotimo stack exec -- pugs -BPIR-YAML ../test.p6 19:53
pugs: invalid name: ...
[Coke] oh, pugs.
timotimo do you have a clue how pugs could be made to output ansi-c code of your program? 19:54
19:58 Kaiepi joined
[Coke] nope. I was using it once to run the test suite on the regular, but that's it 20:03
timotimo OK, thank you anyway! 20:10
20:57 sena_kun joined 20:59 Altai-man_ left 21:32 pamplemousse left 21:34 Kaiepi left, Kaiepi joined 21:35 Kaiepi left, Kaiepi joined 22:07 reportable6 left, coverable6 left, quotable6 left, go|dfish left, reportable6 joined, coverable6 joined, quotable6 joined 22:08 go|dfish joined 22:10 Summertime left 22:13 Summertime joined 22:49 sena_kun left