[Coke] ... ok, apparently I was exhausted, release starts tomorrow, late morning, as per usual. :) 00:22
japhb timo: github.com/japhb/Terminal-Capabilities/issues/1 should be fixed now. :-) 01:32
03:18 kurahaupo joined
Geth rakudo: ugexe++ created pull request #6148:
RakuAST: fix generic Definedness/Coercion in parametric role bodies
04:15
05:24 finanalyst joined 05:26 kurahaupo left 06:02 kurahaupo joined
patrickb japhb++ 06:07
06:41 kurahaupo left, kurahaupo joined 06:53 kurahaupo left, kurahaupo joined 07:11 finanalyst left 07:49 kurahaupo left 08:35 _ramix_ joined 08:37 _ramix_ left 09:11 kurahaupo joined 09:43 kurahaupo left 10:28 finanalyst joined 10:38 finanalyst left 11:07 finanalyst joined 11:12 finanalyst left, finanalyst joined 11:43 finanalyst left 11:54 MasterDuke joined 12:04 ShimmerFairy left, rnddim joined 12:11 MasterDuke left 12:31 rnddim left, rnddim joined
Geth rakudo/main: b776b44ba7 | (Elizabeth Mattijsen)++ | src/core.c/RakuAST/Fixups.rakumod
RakuAST: don't complain about illegal E<> codes in rakudoc

Instead, render them as identity, and leave it to any renderer to deal with it whether they are legal or not.
Also be less verbose in the code.
Fixes #6145
13:34
lizmat [Coke]: ^^ would be nice if that could make it to the 2026.04 release :-)
[Coke] That is a very late addition. :) 13:55
No worries. 13:56
14:42 finanalyst joined 15:07 finanalyst left 15:39 finanalyst joined 15:42 rnddim is now known as ShimmerFairy
[Coke] annnnnd, starting. just missed the morning time here, sorry 16:06
releasable6: next
releasable6 [Coke], Next release in ≈2 hours. There are no known blockers. 5 out of 42 commits logged
[Coke], Details: gist.github.com/c435b1deb74c4a59e8...22203d15e8
[Coke] releasable6: next 16:08
releasable6 [Coke], Next release in ≈2 hours. There are no known blockers. 0 out of 42 commits logged
[Coke], Details: gist.github.com/56e97feaa646454414...89927fa584
[Coke] releasable6: next 16:14
releasable6 [Coke], Next release in ≈2 hours. There are no known blockers. 0 out of 42 commits logged
[Coke], Details: gist.github.com/92cd4019ebf4aaf9e1...3d41dcdcce
[Coke] releasable6: next
releasable6 [Coke], Next release in ≈2 hours. There are no known blockers. 0 out of 42 commits logged
[Coke], Details: gist.github.com/6f16c38034f32f74d6...a780e2c562
[Coke] bah. the changelog doesn't like having the real commit links in it.
releasable6: next 16:16
releasable6 [Coke], Next release in ≈2 hours. There are no known blockers. 0 out of 42 commits logged
[Coke], Details: gist.github.com/cfb73046ecaab86bc0...50b425f908
[Coke] ... or maybe it's just really slow to catch up today. :|
FYI, the IRC logs filter doesn't work on a too-narrow screen - the search filters popup but the submit button is off screen and no way to get to it. 16:34
also, search doesn't seem to let me restrict just to "sends by timo" 16:36
... if I search for [Coke], I see my sends. if I search for timo, I get just a list of dates in the RHS. Click on one, it resets my search criteria and shows me that day 16:37
[Coke] gives up on finding what timo said about his nqp commits and just uses lizmat's summary 16:38
16:38 finanalyst left
[Coke] github.com/rakudo/rakudo/wiki/ChangeLog-Draft done, you have a few minutes if you want to suggest any updates. 16:41
Geth rakudo/release-2026.04: 982ecc180c | (Will Coleda)++ | 3 files
Update changelog + announcement

Deliberately not logged:
  [7f2cd084][08bd698d][ef3975df]
16:51
[Coke] You are out of time. :) 16:52
MoarVM 2026.04 cut 17:20
Having rounded off many sharp corners (to me) in previous releases, I have reached the point where the complicated release process is... fine. Time to get someone new in the mix so they can find stuff that jdv and I have gotten used to that sucks and can be fixed. :) 17:25
github.com/Raku/roast/issues/883 is still failing 17:56
Geth roast/6.c-errata: 4eb0f75c3f | (Elizabeth Mattijsen)++ | S32-io/open.t
Make sure any deprecations don't interfere with tests
18:02
lizmat [Coke]: no, it is not :-)
[Coke] Got one new one in 6.d, one sec. 18:18
github.com/Raku/roast/issues/886 18:20
QuantHash, which is something touched this release.
lizmat aah... interesing, lemme have a look 18:24
Geth nqp/main: fae3140310 | (Will Coleda)++ | tools/templates/MOAR_REVISION
[release] Bump MoarVM revision to 2026.04
18:27
nqp/main: 7a2d33ee4a | (Will Coleda)++ | VERSION
[release] Bump VERSION to 2026.04
rakudo/release-2026.04: e32963fd17 | (Will Coleda)++ | tools/templates/NQP_REVISION
[release] Bump NQP revision to 2026.04
rakudo/release-2026.04: 3584a14680 | (Will Coleda)++ | VERSION
[release] Bump VERSION to 2026.04
lizmat [Coke]: give me 10 mins to look at te QuantHash failure
[Coke] OK. hopefully it's just a test change. :| 18:28
lizmat no, the error is legit :-(
[Coke] wish I had waited before running that last command.
lizmat m: class :: does QuantHash { }.ne 18:29
evalable6 (exit code 1) No such method 'ne' for invocant of type '<anon|1>'
in block <unit> at /tmp/0VYJjiU5pA line 1
lizmat m: class :: does QuantHash { }.new
evalable6 (exit code 1) P6opaque: no such attribute '$!elems' on type <anon|1> in a <anon|1> when trying to bind a value
in block <unit> at /tmp/9HA8iWGuVB line 1
[Coke] interesting that it only died on 6.d and not master or 6.c 18:30
lizmat ok, I see the issue... and it's goiing to need some thought 18:32
it *only* happens when creating your own class doing QuantHash
[Coke] ok. I'll take a break and figure out if I have to do anything other than "just start over". That will probably work.
lizmat which should be rare
need to go afk for a bit now...:-( 18:33
[Coke] I had just tagged and pushed, but I'm pretty sure I can just do it again.
Sure, no worries. Thanks for looking at it.
If I disappear, feel free to text me.
It also didn't trip anything in the ecosystem, so while I'd rather have a clean release... We can also do a spot check with the devs online and see if we think collectively it's a blocker. 18:40
19:17 ShimmerFairy left, ShimmerFairy joined
lizmat is back, looking at the issue again 19:21
[Coke]: I feel the changes are going to be so intrusive, that I don't trust them to be in the release 19:22
the problem comes from the $!elems attribute in QuantHashes not being defined in the QuantHash role
so a class doing the QuantHash role, would need to define it (which Setty and Baggy now do) 19:23
moving the attribute to QuantHash means that Setty and Baggy cannot use the `$!elems` syntax anymore, but would have to use the nqp::getattr(self,QuantHash,'$!elems') syntax everywhere 19:24
[Coke] ok. So let's go ahead as is, this is a low risk if we keep this one issue, and I'm sure we can fix it by .05 19:25
lizmat and that feels rather intrusive at this late date
ah yes, most definitely fixed in .05.... with a blin run
[Coke]++ 19:26
Geth rakudo: coke++ created pull request #6149:
Release 2026.04
19:33
rakudo/main: 4 commits pushed by (Will Coleda)++
lizmat whee! 19:35
[Coke]++ 19:38
[Coke] ... the reference to @needles makes github think they committed to the release. oops
not quite done, close.
timo i tend to put stuff in `backticks` which often prevents such automatic linkage of issues and users 19:39
[Coke] I did @<!-- -->needles for the version attached to the release
.tell patrickb binary releases for 2026.04 can start 19:40
tellable6 [Coke], I'll pass your message to patrickb
[Coke] .tell el_che binary releases for 2026.04 can start
tellable6 [Coke], I'll pass your message to El_Che
Geth rakudo/main: eb79837852 | (Will Coleda)++ | docs/release_guide.pod
Match variable name above
19:42
rakudo/main: f6d89e38af | (Will Coleda)++ | docs/release_guide.pod
claim next release
[Coke] OK. Done, I think. 19:46
I'll open a rakudo ticket for that quanthash issue.
Geth ¦ rakudo: coke assigned to lizmat Issue Minor QuantHash issue github.com/rakudo/rakudo/issues/6150 19:49
[Coke] tagged as a blocker for the next release.
OK, time for video games or something. Cheers. 19:50
lizmat cheers!
[Coke] I would say give it a day to see if anyone complains about the release, then we can start pulling in all of the remaining submissions. 20:13
PRs, that is
20:33 donaldh left
Geth rakudo/main: a1f6cb4fb4 | (Will Coleda)++ | docs/release_guide.pod
Note: run errata tests more often

  ... than during the release process.
23:25
[Coke] FYI I had the "edit release" window still open on the moarvm release, just hit save - this was the step that added the .tgz file 23:58