tbrowder would it be feasible, or practical, to have PRs to the master branch always be run with "make spectest" on travis? 00:16
vrurg tbrowder: spectest is very slow and we're limited with 1hr timeframe. 00:18
s/very slow/takes long/ 00:19
tbrowder ok, thnx, i'm sure that's not an original idea! 00:20
vrurg Not really :)
00:20 softmoth left 00:21 softmoth joined
tbrowder ok, how about running our own test facility somehow magically through the power of git hooks? 00:22
vrurg tbrowder: if we have a server for this, I'd appreciate it very much. 00:23
tbrowder i've offered one before but no one was interested. 00:24
AlexDaniel it's not about having a machine, it's about setting it up
vrurg But the problem is somewhat more compilcated than just having a server and installing it. What platform would we run on it? Linux most obiously. But what about macOS and Win?
AlexDaniel and that too, yes :( 00:25
vrurg The only good solution is getting pro accounts on tracis and appveyor.
tbrowder ah, vms?
AlexDaniel vrurg: fwiw I don't even check the spectest on windows, tests were failing there before I started making releases
vrurg: here's a newer ticket: github.com/perl6/roast/issues/504 00:26
and there was another one somewhere before that
vrurg AlexDaniel: things were at very good shape a month or two ago. Don't know how it goes currently. I don't run them often too.
tbrowder vrug: i'll check on it
AlexDaniel vrurg: also, spectest used to work on jvm at some point, but I don't think it does anymore 00:27
again, when I started making releases it didn't work, so I never run spectest on jvm too…
vrurg JVM is broken for at least half a year or so.
AlexDaniel I do check that jvm runs -e 'say 42' though :)
no way? 00:28
vrurg tbrowder: that'd be great. We miss quite a coule of issues without complete testing.
discord6 <noelsken> there's a bug in create-jvm-runner at least
AlexDaniel builds
vrurg no way – for what? Fixing it? 00:29
AlexDaniel no way it doesn't work for half a year
last release was 2019.07, you sure it doesn't work there?
vrurg spectests? Did you try? I don't remember the last time they were working for me with JVM. 00:30
At some point some tests gets stuck.
discord6 <noelsken> I applied this patch here to install r-j pastebin.com/V9aePwGL
<noelsken> otherwise perl6-j is broken
<noelsken> spectest breaks somewhen in between but that's on the evalserver
<noelsken> it works - albeit really slowly - when running perl6-j directly 00:31
AlexDaniel vrurg: ah, well, spectests were not working for a long time
but you said “JVM is broke” :)
discord6 <noelsken> but the eval server grinding to a halt eventually speaks ill of r-j's potential for long-running programs
AlexDaniel vrurg: point is, I at least make sure that a hello world runs 00:32
there are times when rakudo on jvm is more broken :)
vrurg noelsken: could you make a PR for that, pls? I can't read the patch because it's wrapped and requres kinda char-by-char comparison.
discord6 <noelsken> vrurg: sure, just gotta clone on github and reset origin, gimme a bit 00:33
vrurg AlexDaniel: me too. But I can only do it for macOS and Linux. JVM never worked in my half-dead Windows VM. OpenBSD VM I just got installed and I don't to spend too much time on it.
noelsken: I'd like to ask you to explain what's wrong and why PR is needed. Either open an issue or add a comment on PR. Both will do for me. Thank you! 00:35
AlexDaniel: BTW, I got OpenBSD install fixed. Must admit that this make gives me a tick on my eye. How come does it manage to do prereqs out of the order they're specified with .NOTPARALLEL – it's out of my understanding! :( 00:36
AlexDaniel vrurg: usually there are many openbsd users willing to help, even if it's a bit surprising. Maybe open a ticket and someone will be able to explain what's going on? 00:38
vrurg AlexDaniel: I work around it by explicitly making install-main target depend on install-pre. I would work anywhere in either case. 00:39
I'm more interested in merging #3248 but it requires github.com/perl6/nqp/pull/586 and NQP bump will be needed. Shall I merge the latter and wait for the next bump? 00:40
AlexDaniel vrurg: yep 00:42
I can bump it for you :)
vrurg NP, I can do it on my own. Just wanna be sure I break nothing with it. 00:43
AlexDaniel vrurg: well, if you're worried about the PR itself, maybe a good idea to wait until the release?
most blockers have disappeared thanks to nine++ 00:44
00:44 softmoth left
AlexDaniel and vrurg++ and ugexe++ 00:45
so it wouldn't be that long of a wait, I think
00:45 ZzZombo_ joined
Geth rakudo: crnlskn++ created pull request #3254:
Fix a wrong reference
AlexDaniel in fact I'll probably start a release branch even sooner 00:46
discord6 <noelsken> vrurg: github.com/rakudo/rakudo/pull/3254
vrurg AlexDaniel: I'm worried about possible breaks in NQP since the last bump. The commit is only to support forced rebuild of the whole moar-nqp-rakudo chain when needed.
discord6 <noelsken> wonders if she still has her old github account somewhere
<noelsken> I coulda pushed with that >:3
<noelsken> though I absolutely wouldn't want to, because of process reason at the minimum
vrurg AlexDaniel: ok, I better wait anyway.
Geth rakudo: vrurg++ created pull request #3255:
Fix OpenBSD build
discord6 <noelsken> oh duh I actually linked it right after the bot I'm uh tired I'm going to bed 00:48
AlexDaniel irc messages are cheap :)
00:48 ZzZombo left, ZzZombo_ is now known as ZzZombo
AlexDaniel ok, these will now be permanently stored on discord and matrix too, but it's still cheap :) 00:49
and colabti of course…
and ideally whateverable should also keep a copy
Geth rakudo: 1ac9d65741 | (Nele Schwarz)++ | tools/templates/jvm/Makefile.in
Fix a wrong reference

otherwise we end up with an empty patterin in a split call later which breaks stuff
rakudo: 29f1b80301 | (Vadim Belman)++ (committed using GitHub Web editor) | tools/templates/jvm/Makefile.in
Merge pull request #3254 from crnlskn/master

Fix a wrong reference
00:53 ggoebel joined
Geth rakudo/master: 4 commits pushed by (Vadim Belman)++ 00:54
vrurg AlexDaniel: minus one blocker. Unless kaipei finds more peculiarities of OpenBSD. ;) 01:03
But then I at least will have pre-installed ready to go VM next time. It takes hours to get OpenBSD into a working shape... :( 01:05
AlexDaniel yaay 01:06
vrurg AlexDaniel: if you fork the release I'll try to put my commits into the changelog. Though it's gonna be a little scary task. ;) 01:44
AlexDaniel vrurg: btw I moved it back to the wiki: github.com/rakudo/rakudo/wiki/ChangeLog-Draft 01:49
draft releases have no edit history, unfortunately :(
vrurg Ok, good to know. 01:50
AlexDaniel I really liked how commit refs were working properly, maybe if they add edit history we can try it again…
releasable6: status
releasable6 AlexDaniel, Next release will happen when it's ready. 1 blocker. 7 out of 432 commits logged (⚠ 9 warnings)
AlexDaniel, Details: gist.github.com/92723a68d40b549f0b...3fbfa40309
AlexDaniel and this is back to normal! :)
vrurg BTW, ugexe confirms that 3157 is fixed. Shall we close it and fork the release? 01:52
AlexDaniel vrurg: dnoe 01:58
wait how come we're at 0 blockers 01:59
that was quick, I thought there's something else…
vrurg AlexDaniel: I see you're having so much fun with word distances that can't stop experimenting with typos? :D 02:00
AlexDaniel just my sloppy typing :)
vrurg is kidding, of course. :)
Happens to me all the time when thinking is faster than typing... 02:01
Anyway, if there not gonna be any more pre-release forks I'd merge NQP#586 then. 02:02
02:10 ggoebel left
AlexDaniel vrurg: I don't get it, you want it to be in the upcoming release or not? :) 02:11
vrurg AlexDaniel: nevermind, I misinterpreted your 'done' as if you've forked already.
AlexDaniel brrt: can you update me on what you're working on? It's something that needs to be fixed before the release, right? We're currently at 0 blockers otherwise 02:13
tellable6 AlexDaniel, I'll pass your message to brrt
02:17 softmoth joined 02:32 benjif joined 02:43 ZzZombo_ joined 02:46 ZzZombo left, ZzZombo_ is now known as ZzZombo 03:35 AlexDani` joined 03:42 ZzZombo_ joined 03:44 AlexDani` is now known as AlexDaniel, AlexDaniel left, AlexDaniel joined, ZzZombo left 03:45 ZzZombo_ is now known as ZzZombo 03:52 ZzZombo_ joined 03:54 ggoebel joined 03:55 ZzZombo left, ZzZombo_ is now known as ZzZombo 04:51 benjif1 joined 04:56 benjif1 left 04:57 benjif1 joined 04:58 benjif left, benjif1 left 04:59 benjif joined 05:00 masak joined 05:23 AlexDaniel left 05:39 benjif1 joined 05:40 benjif left
bartolin for the record: I didn't see a major breakage for the JVM backend during the last weeks (FreeBSD, using gmake). Only a few files have problems during spectest. I don't use the Evalserver, but something like 'TEST_JOBS=1 perl t/spec/test_summary rakudo.jvm'. The Evalserver has been leaking resources for a long time -- reported as RT #132104. I'm not aware of a general memory leak of rakudo-j, though. 05:43
link to RT ticket: rt.perl.org/Ticket/Display.html?id=132104 05:44
Geth roast: b13cbc50a9 | usev6++ | S16-io/watch.t
[JVM] Skip test for IO::Path.watch (NYI)
06:03 benjif1 left 06:36 sena_kun joined
lizmat Files=1276, Tests=109544, 204 wallclock secs (28.29 usr 7.81 sys + 2896.46 cusr 250.61 csys = 3183.17 CPU) 07:04
07:26 ggoebel left 07:48 [Tux] joined 08:39 sena_kun left 08:45 sena_kun joined 08:54 patrickb joined 08:58 patrickb left 09:53 Guest13443 joined 10:01 pmurias joined
|Tux| Rakudo version 2019.07.1-432-gdb849a659 - MoarVM version 2019.07.1-287-g89aa45f56
csv-ip5xs0.693 - 0.697
csv-ip5xs-206.231 - 6.397
csv-parser21.529 - 21.712
csv-test-xs-200.420 - 0.423
test6.251 - 7.154
test-t1.740 - 1.764
test-t --race0.785 - 0.785
test-t-2029.102 - 29.183
test-t-20 --race9.023 - 9.033
10:21 pmurias left 10:48 pmurias joined, brrt joined
discord6 <noelsken> good news, the jvm profiler visualvm doesn't work with java 13 so I gotta downgrade and that means the deprecation warning for -noverify probably disappears and I can put off looking into what's up with that :) 11:11
11:22 brrt left
Geth ¦ problem-solving: FCO assigned to jnthn Issue List should have `.do` method as Supply github.com/perl6/problem-solving/issues/122 11:35
¦ problem-solving: FCO assigned to jnthn Issue `Supply.map` should respect the `Slip` semantics as `List.map` does github.com/perl6/problem-solving/issues/123 11:38
12:12 ggoebel joined 12:24 AlexDaniel joined 12:25 AlexDaniel left, AlexDaniel joined
Geth nqp/truffle: 431a5b5595 | (Paweł Murias)++ | 2 files
[truffle] Implement nqp::getstderr
nqp/truffle: a000457b78 | (Paweł Murias)++ | 2 files
[truffle] Implement nqp::xor
12:58 ggoebel left 13:11 ggoebel joined 13:41 awwaiid joined
lizmat notable6: weekly 13:47
notable6 lizmat, 5 notes: gist.github.com/a3ca4046a7f98097f1...959c250c25
lizmat notable6: weekly reset 13:48
notable6 lizmat, Moved existing notes to “weekly_2019-10-21T13:48:33Z”
lizmat and the first Rakudo Weekly hits the Net: rakudoweekly.blog/2019/10/21/2019-42-answered/ 13:59
14:02 pmurias left 14:13 pmurias joined
Geth nqp/truffle: 4 commits pushed by (Paweł Murias)++ 14:28
14:40 lucasb joined 15:03 patrickb joined 15:07 pmurias left 15:42 Guest13443 left 16:34 patrickb left
discord6 <noelsken> so the Proc::Async leaks apparently don't occur when running with RAKUDO_SCHEDULER_DEBUG{,_STATUS} set 16:36
<noelsken> that doesn't seem to be the only leak that the eval server runs into though
16:42 Kaiepi left
tbrowder i would like to implement a change to pod declarator blocks so the user has a choice of preserving the original text or normalizing all text into a single string (the current behavior). should i use a pragma or env var to turn on the proposed feature? 16:51
Geth ¦ problem-solving: AlexDaniel assigned to jnthn Issue What's exactly our stance regarding other Raku implementations? github.com/perl6/problem-solving/issues/124 16:57
roast: 2b0d1968bc | (Ben Davies)++ | S32-io/signals.t
Add tests for R#3035

Fixes rakudo/rakudo#3035
roast: 19e4d36244 | (Ben Davies)++ | spectest.data
Add S32-io/signals.t to spectest.data
roast: fb73c62f75 | (Vadim Belman)++ (committed using GitHub Web editor) | 2 files
Merge pull request #584 from Kaiepi/signal

Add tests for R#3035
17:39 softmoth left 18:02 sena_kun left 18:43 brrt joined
Geth ¦ rakudo: AlexDaniel assigned to bdw Issue Regression introduced with new moarvm expr templates github.com/rakudo/rakudo/issues/3256 19:02
19:26 MasterDuke joined 19:28 cognominal joined 19:39 pmurias joined, tellable6 left 19:41 tellable6 joined 19:44 releasable6 joined
Geth ¦ rakudo: vrurg self-assigned DYNAMIC:: pseudo-package doesn't do lookups same way &DYNAMIC does. github.com/rakudo/rakudo/issues/3257 19:51
tbrowder ok, then i'll use env var "RAKUDO_POD_DECL_BLOCK_USER_FORMAT" 20:14
discord6 <noelsken> +"TEMPORARY_PROPOSAL_FOR_TESTING 20:18
<noelsken> snrk :)
20:30 softmoth joined 20:53 ggoebel left
Geth nqp: 6abc58cdf7 | (Aleks-Daniel Jakimenko-Aleksejev)++ | tools/templates/MOAR_REVISION
[MoarVM Bump] Brings 4 commits

MoarVM bump brought: github.com/MoarVM/MoarVM/compare/2...gd5f5185b2 d5f5185b2 [JIT] We should not overwrite assigned type 97e26c018 [JIT] Fix bug in takedispatcher 89aa45f56 Merge pull request #1202 from ugexe/patch-21 b0b3fc8c5 Fix obscure deadlock
rakudo: 5b1f341e2d | (Aleks-Daniel Jakimenko-Aleksejev)++ | tools/templates/NQP_REVISION
[NQP Bump] 6abc58cdf [MoarVM Bump] Brings 4 co […]

NQP bump brought: github.com/perl6/nqp/compare/2019....g6abc58cdf
22:09 squashable6 joined
Geth rakudo: daa16b9c92 | (Aleks-Daniel Jakimenko-Aleksejev)++ | src/core.c/CompUnit/Repository/FileSystem.pm6
Add support for .rakumod extension
rakudo: 2058cfbbbc | (Aleks-Daniel Jakimenko-Aleksejev)++ (committed using GitHub Web editor) | src/core.c/CompUnit/Repository/FileSystem.pm6
Merge pull request #3246 from rakudo/new-extensions

Add support for .rakumod extension
22:28 brrt left
Geth nqp: cd1442c9f2 | (Aleks-Daniel Jakimenko-Aleksejev)++ | tools/templates/MOAR_REVISION
[MoarVM Bump] 5685d40b1 [JIT] C99 fix

MoarVM bump brought: github.com/MoarVM/MoarVM/compare/2...g5685d40b1
rakudo: 8d8dcbf8d9 | (Aleks-Daniel Jakimenko-Aleksejev)++ | tools/templates/NQP_REVISION
[NQP Bump] cd1442c9f [MoarVM Bump] 5685d40b1 [ […]

NQP bump brought: github.com/perl6/nqp/compare/2019....gcd1442c9f
23:17 pmurias left