00:38 yuplushi left 01:10 dogbert11 joined 01:12 dogbert17 left 01:38 yuplushi joined 01:47 lucasb left 03:50 linkable6 joined, evalable6 joined 03:55 dogbert17 joined 03:59 dogbert11 left 05:19 leont left 05:22 dogbert11 joined 05:24 dogbert17 left 05:59 Kaiepi joined 06:09 Kaiepi left, Kaiepi joined 06:10 Kaiepi left, Kaiepi joined 06:42 dogbert11 left 06:44 dogbert11 joined
nwc10 release(s) \o/ 06:46
is it safe to do things on master again?
07:13 dogbert17 joined 07:14 dogbert12 joined 07:15 dogbert12 left, dogbert11 left 07:16 dogbert12 joined 07:17 dogbert12 left 07:18 dogbert12 joined, dogbert17 left 07:27 dogbert17 joined 07:28 dogbert11 joined 07:29 dogbert12 left, dogbert11 left 07:30 dogbert11 joined 07:33 dogbert17 left 07:38 dogbert11 left 08:06 dogbert11 joined 08:09 sena_kun joined 08:16 dogbert17 joined 08:21 dogbert11 left 08:22 dogbert11 joined 08:23 dogbert17 left 08:25 zostay left, kawaii left 08:27 dogbert17 joined, kawaii joined 08:31 dogbert11 left 08:32 zostay joined
sena_kun releasable6, status 08:39
releasable6 sena_kun, Next release in ≈27 days and ≈10 hours. 2 blockers. Changelog for this release was not started yet
sena_kun, Details: gist.github.com/56324ed9543fac094a...923a351812
MasterDuke huh. i've pulled, but git describe still shows 2020.07-* for nqp and rakudo 08:45
sena_kun MasterDuke, that's a known failure of mine. 08:46
I had some horrible issues yesterday trying to make a point, but now I have a plan which will likely work out.
>spectest Executed in 126,01 secs 08:47
Now I wonder if we _did_ have a perf regression...
MasterDuke ah, wasn't sure if it had all been resolved and i was just doing something wrong
sena_kun Executed in 119,65 secs 08:49
hmm, just floating it seems...
sena_kun bumps 08:50
Geth nqp: 9cbf0c07cb | Altai-man++ | tools/templates/MOAR_REVISION
[MoarVM Bump] Brings 4 commits

MoarVM bump brought: github.com/MoarVM/MoarVM/compare/2...g0e7417add 0e7417add Merge pull request #1338 from MoarVM/2020.08 9a5203395 Merge pull request #1336 from MoarVM/flags-split 7f854e655 Shrink `flags1` and `flags2` in struct MVMCollectable down to MVMuint8. 497748f71 Split `flags` in struct MVMCollectable into `flags1` and `flags2`.
rakudo: ea60eaecdd | Altai-man++ | tools/templates/NQP_REVISION
[NQP Bump] Brings 4 commits

NQP bump brought: github.com/perl6/nqp/compare/2020....g9cbf0c07c 9cbf0c07c [MoarVM Bump] Brings 4 commits 62f28e3e4 Merge pull request #659 from Raku/release-2020.08 90166ede7 [release] Bump VERSION to 2020.08 4aeb7aaef [release] Bump MoarVM revision to 2020.08
lizmat sena_kun: spectest does seem slower the past 2 weeks 09:32
09:33 Kaiepi left 09:34 Kaiepi joined 09:35 Kaiepi left, Kaiepi joined 09:42 Kaiepi left
Geth rakudo/suggest-capitalization-errors: df5b3d5308 | (Elizabeth Mattijsen)++ | src/core.c/Exception.pm6
Externalize suggestions and tips

The X::Method::NotFound error is hard to test for. This commit adds the "suggestions" and "tips" method, so that one can test for
  (not) expected suggestions, and whether any tips were generated.
Also only use numeric weights in suggestions: there was a mix of strings and integers, that can't be right :-)
rakudo/suggest-capitalization-errors: 3703469d80 | (Elizabeth Mattijsen)++ | t/05-messages/01-errors.t
Add test for specific suggestions
09:46 Kaiepi joined 09:54 dogbert11 joined 09:58 dogbert17 left 10:20 dogbert17 joined 10:22 dogbert11 left 10:36 leont joined 10:38 Altai-man joined 10:40 sena_kun left 10:47 JJMerelo joined
JJMerelo releasable6: status 10:47
releasable6 JJMerelo, Next release in ≈27 days and ≈8 hours. 2 blockers. Changelog for this release was not started yet
tellable6_ 2020-08-19T16:38:45Z #raku-dev <[Coke]> jjmerelo changing the last paragraph in the article for tomorrow. I don't like what i wrote.
2020-08-19T16:39:02Z #raku-dev <[Coke]> jjmerelo thanks for getting it *into* wp and setting up the publish
releasable6 JJMerelo, Details: gist.github.com/eb7b749624fa7ffb3b...8b8668e13a
JJMerelo .tell [Coke] no problem. Thanks.
tellable6_ JJMerelo, I'll pass your message to [Coke]
10:48 AlexDaniel joined, AlexDaniel left, AlexDaniel joined
AlexDaniel Altai-man: github.com/rakudo/rakudo/issues/38...-678758587 10:49
Altai-man: that being said, I don't see any new commits?… hmm 10:51
Altai-man: actually, I see some. The ones by vrurg are not in the release branch 10:58
but then I don't see how that would make a difference… 10:59
11:23 JJMerelo left 12:07 dogbert11 joined, dogbert17 left 12:09 dogbert17 joined 12:13 dogbert12 joined, dogbert11 left 12:15 dogbert11 joined 12:17 dogbert17 left, dogbert12 left
dogbert11 are there still troubles with the release? 12:31
lizmat dogbert11: Altai-man is running blin now 12:43
when that's ok, a release will follow
is my current understanding
12:54 dogbert17 joined 12:56 dogbert11 left
Altai-man I'm back from doing my own life stuff. Blin is ok, I'll start packaging ASAP after getting some food. 13:19
>Blin Executed in 122,39 mins
13:57 dogbert11 joined 13:58 dogbert17 left
Geth rakudo/release-2020.08.1: 308673a684 | Altai-man++ | 3 files
2020.08.1 release announcement + changelog
nqp/release-2020.08.1: 9 commits pushed by (Christian Bartolomäus)++, (Elizabeth Mattijsen)++, (Nicholas Clark)++, (Jonathan Worthington)++, Altai-man++ 13:59
AlexDaniel Altai-man: take it easy 14:06
Altai-man: release-2020.08.1 branch is one commit ahead, so releasing from master means that the commit won't be in the release 14:07
14:08 dogbert11 left
AlexDaniel just don't rush and it'll be fine :) 14:08
Altai-man I rebased branches for nqp and rakudo, updated announcement to include a couple of commits which has slipped in 2020.08 (mystery again?) but were not mentioned. 14:09
TEST_JOBS=24 VERSION=2020.08.1 VERSION_MOAR=2020.08 BRANCH_ROAST=master BRANCH_RAKUDO=release-2020.08.1 BRANCH_NQP=release-2020.08.1
14:09 dogbert11 joined
Altai-man Now this looks sane? 14:09
AlexDaniel yeah that looks correct to me, if roast master doesn't have new tests 14:10
Altai-man: there isn't really a mystery to me. From what it looks like the akefile was working with an old commit 14:12
ah… hold on… nevermind
yes, the Seq commits by vrurg were in master but were not in the release branch 14:13
Altai-man Yes.
The question is how something completely different was tagged as release, not actual release branch. Not even tagged, but packed an all that. 14:14
14:39 sena_kun joined 14:40 Altai-man left
Geth nqp/release-2020.08.1: 7d3a885d79 | Altai-man++ | tools/templates/MOAR_REVISION
[release] Bump MoarVM revision to 2020.08
nqp/release-2020.08.1: 49ad331635 | Altai-man++ | VERSION
[release] Bump VERSION to 2020.08.1
rakudo/release-2020.08.1: 41d8c33e45 | Altai-man++ | tools/templates/NQP_REVISION
[release] Bump NQP revision to 2020.08.1
rakudo/release-2020.08.1: b86c05cb0f | Altai-man++ | VERSION
[release] Bump VERSION to 2020.08.1
14:54 dogbert17 joined
sena_kun Ok, so seems like it is done. 14:55
Anyone wanting to check it on another machine? 14:56
[Tux] Rakudo version 2020.07-85-gea60eaecd - MoarVM version 2020.08-4-g0e7417add
csv-ip5xs0.817 - 0.843
csv-ip5xs-208.019 - 8.683
csv-parser27.589 - 27.621
csv-test-xs-200.400 - 0.410
test7.995 - 8.065
test-t1.880 - 1.941
test-t --race0.850 - 0.875
test-t-2032.363 - 33.152
test-t-20 --race8.654 - 8.773
Geth nqp: Altai-man++ created pull request #660:
Release 2020.08.1
rakudo: Altai-man++ created pull request #3870:
Release 2020.08.1
sena_kun Now tag 2020.08.1 seems right to me. 14:58
14:58 dogbert11 left
AlexDaniel sena_kun: one thing I'd like to note is that when you push the tag like that there's no way back really 15:02
sena_kun: there is no real difference between master and other branches, so just because a branch is not merged into master yet doesn't mean anything 15:03
sena_kun: so right now there's a published tag that points at a non-master commit, meaning 2020.08.1 is now released 15:04
sena_kun Yes.
You implying it is time to merge quickly or just clarifying?
AlexDaniel it's not wrong per se, but I just want to make it clear that this workflow means that people can't look at the PR and suggest fixes because it's too late
for future releases you can push your commits without pushing the tags 15:05
sena_kun But 2020.08.1 is saved, right? Or you noticed something after tag and now we're going for 2020.08.2?
AlexDaniel no issues so far :) 15:06
sena_kun As for noticing things before tagging - yes, this is a good idea.
AlexDaniel OK there is one issue 15:07
sena_kun Hm?
AlexDaniel let me double check
false alarm :) 15:08
sena_kun Phew.
We can continue then. 15:09
I do agree that more eyes looking for the release is better in every possible regard.
AlexDaniel hoold on… 15:10
ah again, false alarm xD
sena_kun However it really boils down to people making things. We are kind of scarce on time resources people have and options are 1)enforce rules like "Two people do release or nobody does"; 2)Do it the usual way, which is error-prone.
AlexDaniel sorry x)
sena_kun Are you mocking me?
AlexDaniel I get confused with all the VERSION files sometimes
nwc10 "continue then" means what? The Perl 5 guide ends with this step: github.com/Perl/perl5/blob/blead/P....pod#relax
is there an equivalent? 15:11
also, that document is long and irritating because it turned out to be hard to automate many parts
or, more accurately, that's what's left *after* things were automated whereever easy.
sena_kun To be honest, I had to pick up moarvm releases somewhere along the way, because samcv has not so much free time (neither many of moarvm folks we have) and if we want releases be on time, we can't ping everyday "Can you please release, we're ready?". 15:12
Otherwise rakudo releases would either stuck or delayed, or you can pack them yourself.
AlexDaniel nwc10: see github.com/rakudo/rakudo/blob/mast...e-managers
nwc10: steps 20-22 not really automated 15:13
nwc10: step 19 is new so I can't comment about it
nwc10 but the aim is to get to 22, and not backtrack? :-)
AlexDaniel nwc10: step 1 is release prep, there are tools but no automation yet 15:14
nwc10 and, thanks for doing this. I'm glad that you're doing it, and I think you're doing a much better job of it than I would (and I really don't reliably have the time)
lizmat sena_kun: if I do a pull now, should I get a 2020.08 MoarVM ?
AlexDaniel step 2 is about leap seconds. I never had it happen to me when I was the release manager, and since then it still didn't happen, so no automation for that
lizmat: it's not in master yet 15:15
lizmat ok
sena_kun I was going to wait for CI, buuuuuuut...
Maybe I'll do.
AlexDaniel btw I always wanted to change it to “Celebrate with an inappropriate amount of fun”, that's more like I felt when I was the release manager
sena_kun Now need to publish moarvm release... 15:17
And maybe twitter announcement in an hour or so.
AlexDaniel nwc10: it feels like many steps can still be automated for perl5 15:18
tagging and stuff
but yeah, it's messy and difficult sometimes, and the more you automate it, the less of it you actually see if things go wrong
sena_kun github.com/rakudo/rakudo/releases/tag/2020.08.1 15:20
Once rest of CI is green I'm merging into master.
Thinking about it, thanks to this release I tried to run blin on my own pc and knowing it is times faster here makes future blin checks easier. 15:22
AlexDaniel hmm
sena_kun In docker.
AlexDaniel even nicer
sena_kun Just in case you were `hmm` thinking I did it just so.
AlexDaniel no I'm just a bit surprised that whatever machine you have is actually faster for this kind of work 15:23
sena_kun I posted above it took 2 hours.
AlexDaniel holy shit
sena_kun 122 minutes, but network here is shit and IO is always the slowest.
AlexDaniel how? What are the specs? 15:24
sena_kun Ryzen goes brrrr, ahem, 3900X, 970 EVO, 64 gb of mem.
AlexDaniel nice 15:25
sena_kun So 24 threads on ~4000 mhz.
AlexDaniel not sure it can sustain that frequency on all cores at the same time?
sena_kun 3950X with 32 threads would crunch it faster and rippers would laugh, I guess.
It can.
AlexDaniel but yes, very very nice
AlexDaniel looks sadly at his 3700X 15:26
sena_kun At least if I trust `watch -n.1 "cat /proc/cpuinfo | grep \"^[c]pu MHz\""` output, I saw it reliably going around 4 ghz for all cores doing boinc.
And on idle all cores are at around 1800-2200. 15:27
Well, I was using 7500U until a month ago all this time, so I guess 3700X is nothing to be ashamed of. :P
AlexDaniel though mine is at 0 on idle, I only use it for our livestreams :P
Geth rakudo/Int-as-Numeric: 93e89573fb | (Elizabeth Mattijsen)++ | 2 files
Make Mu.Int work the same as Mu.Numeric

Calling `Int` on a type object, fails with:
   Invocant of method 'Int' must be an object instance of type 'Str', not
   a type object of type 'Str'. Did you forget a '.new'?
... (7 more lines)
rakudo: lizmat++ created pull request #3871:
Make Mu.Int work the same as Mu.Numeric
nqp: 7d3a885d79 | Altai-man++ | tools/templates/MOAR_REVISION
[release] Bump MoarVM revision to 2020.08
nqp: 49ad331635 | Altai-man++ | VERSION
[release] Bump VERSION to 2020.08.1
nqp: 8936ed989c | Altai-man++ (committed using GitHub Web editor) | 2 files
Merge pull request #660 from Raku/release-2020.08.1

Release 2020.08.1
sena_kun m: $*RAKU.compiler.version 15:42
camelia ( no output )
sena_kun m: $*RAKU.compiler.version.say
camelia v2020.07.85.gea.60.eaecd
sena_kun hehe
AlexDaniel e: $*RAKU.compiler.version.say 15:49
evalable6 v2020.07.85.gea.60.eaecd
sena_kun Rakudo PR is still not.
JVM takes time. :)
AlexDaniel c: release-2020.08.1 $*RAKU.compiler.version.say
committable6 AlexDaniel, ¦release-2020.08.1: «v2020.08.1␤»
AlexDaniel :P
sena_kun Fair enough. :P
AlexDaniel I actually don't know why this works. I don't remember. 15:50
I tried to get branches working for a long long time, and it didn't because there was some kind of blocker
then I just touched it and poof now it's working
AlexDaniel shrugs
sena_kun AlexDaniel, thanks a lot for your help with this case, anyway. You are really great and supportive even though you are not a release manager for the time being. 15:51
AlexDaniel sena_kun: I'm also not great and not supportive at the same time, as lizmat will probably agree. I'm like a Junction :P 15:52
sena_kun Well, as well I am appreciating how lizmat tirelessly spends time in attempts to make Rakudo better. 15:53
lizmat AlexDaniel: you *can* be a pain
AlexDaniel sorry
sena_kun I would like to try to fix a couple of easy rakudo tickets, but firstly need to play a match or something and rest a bit from all this before Comma week. 15:55
AlexDaniel take a well deserved break :) have fun 15:57
nwc10 Comma week?
sena_kun nwc10, Comma complete release next week. 15:58
nwc10 aha, that comma 15:59
sena_kun Yup.
nwc10 er, that Comma
if it's at the start of a sentance, I didn't realise that it was a proper noun.
is the coffee machine ready for this?
Geth rakudo/master: 4 commits pushed by Altai-man++ 16:02
16:10 lucasb joined
AlexDaniel m: my $x = all(); say $x.insert-funny-method-name-here 16:23
camelia all()
AlexDaniel kinda makes sense, still a bit surprising :) 16:24
they're also truthy and defined, which is probably right because identity? 16:25
and most ops on them just give you the same junction 16:26
or… something
m: say all() + any()
camelia all()
lizmat m: dd ("42" | "666").Int 16:27
camelia any(42, 666)
lizmat m: dd 4 && 6 + 2 | 3 16:28
camelia any(8, 3)
lizmat m: dd 4 & 6 + 2 | 3
camelia any(all(4, 8), 3)
AlexDaniel lizmat: precedence I think? 16:29
m: dd 4 & 8 | 3
camelia any(all(4, 8), 3)
AlexDaniel m: dd (4 & 6) + (2 | 3)
camelia all(any(6, 7), any(8, 9))
AlexDaniel and yeah they're meant to grow like this by design… 16:30
16:30 JJMerelo joined
JJMerelo Test-driving this year's annual survey docs.google.com/forms/d/e/1FAIpQLS...Q/viewform 16:37
we will release it to the wide public when any possible flaw it's ironed out, so suggestions are welcome. It's longer-ish than previous versions, but it includes also questions inserted by nige123++ for marketing purposes. 16:38
16:43 JJMerelo left
lizmat sena_kun: confirmed all clean and test clean 16:43
16:47 samcv joined 16:49 samcv left 16:50 samcv joined 16:54 JJMerelo joined
sena_kun \o/ 17:03
JJMerelo endorses lizmat++ whatever the reason 17:07
dogbert17 sena_kun: I can add that Linux Mint, both 32- and 64-bit, plus Raspberry Pi OS (running on an RPi 4) are all clean 17:10
sena_kun dogbert17, I hope `raku --version` gives something sensible now. :) 17:11
JJMerelo also sena_kun++
dogbert17 This is Rakudo version 2020.08.1-1-g92dc00c20 built on MoarVM version 2020.08 17:12
sena_kun dogbert17, this is for master or release? 17:13
dogbert17 I just built master
sena_kun Ah, ok.
As this is not something release build should show.
JJMerelo requests everyone take a look and/or answer the new Raku User Survey docs.google.com/forms/d/e/1FAIpQLS...Q/viewform 17:16
Please also send suggestions or errors my way.
Geth ¦ problem-solving: holli-holzer assigned to jnthn Issue .max on Hashes and Bags sorts on keys instead of values. github.com/Raku/problem-solving/issues/221 17:21
sena_kun JJMerelo, if you can edit questions now, "Opción 8" should probably be deleted. 17:25
JJMerelo sena_kun thanks.
probably nobody heard about Raku that way :-)
BTW, survey results will be shared with everyone in the community (and a bit of analysis will be done) 17:26
sena_kun JJMerelo, thanks for organizing it.
JJMerelo, also, "What are the top three values you would like to associate with the Raku language in future?" probably does not want options like "too slow" and other negative ones. ;) 17:29
JJMerelo Correct
sena_kun I guess that's the only one, right? 17:30
sena_kun JJMerelo, hm? Which should be removed? Yes, I think. Maybe it wants "fast" option as well instead. 17:31
JJMerelo OK, added high performance 17:32
17:32 dogbert11 joined 17:36 dogbert17 left 17:46 JJMerelo left 17:47 dogbert17 joined
sena_kun hopes it is anonymous 17:49
17:51 dogbert11 left 18:38 Altai-man joined 18:41 sena_kun left 19:25 dogbert11 joined 19:29 dogbert17 left 19:38 dogbert17 joined 19:40 dogbert11 left 19:43 dogbert11 joined, dogbert11 left, dogbert11 joined 19:44 dogbert17 left 19:48 dogbert17 joined, dogbert17 left 19:49 dogbert17 joined 19:50 dogbert11 left 19:56 patrickb joined 20:09 patrickb left, dogbert17 left 20:13 dogbert17 joined 20:14 dogbert11 joined 20:18 dogbert17 left
lizmat m: sub a(--> 0+0i) { } 21:09
camelia 5===SORRY!5=== Error while compiling <tmp>
Malformed return value (return constraints only allowed at the end of the signature)
at <tmp>:1
------> 3sub a(--> 07⏏5+0i) { }
lizmat m: sub a(--> 42) { }
camelia ( no output )
lizmat apparently Complex literals are not values :-) 21:10
21:22 dogbert17 joined 21:27 dogbert11 left
timotimo m: sub a(--> <0+0i>) { } 21:31
camelia ( no output )
21:36 dogbert11 joined
[Coke] m: sub a(--> <0+0i>) { } ; dd a() 21:39
camelia <0+0i> 21:40
[Coke] m: sub a(--> <0+0i>) { } ; dd a().^name
camelia "Complex"
[Coke] huh. Surprised that kept the type
21:40 dogbert17 left
gfldex m: sub a(--> <0+0i>) { }; &a.signature.say; 21:40
camelia ( --> <0+0i>) 21:41
21:41 kawaii left 21:42 lucasb left 21:44 lucasb joined 21:45 lucasb left 21:46 kawaii joined 21:48 kawaii left 21:50 kawaii joined, lucasb joined
timotimo it's direct syntax for a complex literal 21:53
whereas 0+1i relies on constant folding a little later down the line to give a Complex
22:39 sena_kun joined 22:40 Altai-man left 23:13 yuplushi left 23:21 dogbert17 joined 23:25 dogbert11 left 23:53 dogbert11 joined 23:56 dogbert17 left