Geth rakudo/main: 5a742811f3 | (Elizabeth Mattijsen)++ | src/core.c/Junction.rakumod
Fix thinko in one(...).defined

one(1,Any).defined was returning False, while one(Any,1) returned True. They both should return True, and they now do.
This will need additional roast tests, as the test were only checking the second way, not the first.		 00:12
lizmat [Coke] ^^
also sleep&
02:40 lue is now known as ShimmerFairy 05:29 lue joined 05:31 ShimmerFairy left 05:46 lue is now known as ShimmerFairy 08:55 finanalyst left 09:31 melezhik joined 09:35 librasteve_ joined 11:05 finanalyst joined 11:39 finanalyst left 11:40 melezhik left 11:45 librasteve_ left
Geth roast: 7f2d7508ad | (Elizabeth Mattijsen)++ | S03-junctions/autothreading.t
Add more one() tests

Also make the junction-type.defined testing algorithmic, so that any additional tests can be easily added		 11:52
13:10 arkiuat joined
arkiuat lizmat++ thanks! 13:10
13:11 arkiuat left
lizmat yw, and thanks for reporting 13:11
fwiw, that bug must have been there since at least Feb 2020 13:12
I guess not a lot of people use one()
13:13 arkiuat joined
arkiuat at least not for definedness tests. That is a rather particular use-case, I guess 13:13
I guess this means I can go ahead and close that issue I opened then? 13:14
lizmat nah, I'll do that didn't even realize there was an issue just went by the backscroll late last night 13:17
arkiuat thanks! 13:18
15:11 arkiuat left 15:15 arkiuat joined 15:20 arkiuat left 15:21 arkiuat joined
[Coke] returns from the cold outdoors 15:52
timo it really is cold isn't it 15:53
i made the mistake of letting the outdoors come indoors for a little too long
[Coke] was -3C this morning when I got up
(and then I was outside for about 90m on and off) 15:54
releasable6: next 15:59
releasable6 [Coke], Next release in ā‰ˆ3 hours. There are no known blockers. 43 out of 46 commits logged
[Coke], Details: gist.github.com/7ebf1896b26db29648...abfcd71906
[Coke] releasable6: next 16:01
releasable6 [Coke], Next release in ā‰ˆ2 hours. There are no known blockers. 43 out of 46 commits logged
[Coke], Details: gist.github.com/0d95f62394e517ffd9...64bb654bc6
[Coke] releasable6: next 16:02
releasable6 [Coke], Next release in ā‰ˆ2 hours. There are no known blockers. 46 out of 46 commits logged
Geth rakudo/release-2025.12: c89353ced0 | (Will Coleda)++ | 3 files
Update changelog + announcement

Deliberately not logged:
  [e66655c0][bce5c96a][6f06f9f1][eb8c9b13][2d4a8c83][5140c3c6][a1336399][31ff2eb2]
  [c5c33ba9][fbd475fb][6d8be995][0a90592a][cfaee8a0][63bc8b83]		 16:09
[Coke] doing the moarvm build have status that includes: 16:24
modified: 3rdparty/libuv (new commits)
correct way to say "just give me what is recommend for that submodule" ? 16:25
(I made no changes to that folder directly. if I go in to the folder, i get
HEAD detached at 8fb9cb9
timo that means the repo inside has something different checked out from what the outer repo has committed 16:26
does the command that "git status" outputs above the "modified: ..." section help?
[Coke] I ended up doing unix.stackexchange.com/questions/5...ew-commits 16:28
(git diff in moarvm reo, show the old commit it was using, go to 3rdparty/libuv, git checkout that commit) 16:29
Now all is blissfully silent again.
I assume there's a "git submodule" command I can use to do the same thing
timo [Coke]: is there a section that the azure pipeline thing could go, maybe "CI" or something? 16:38
in moarvm's changes
16:41 finanalyst joined
patrickb I fundamentally failed to fix up the libuv submodule after I changed the upstream repo. (I tried many different submodule commands). I ended up deleting the submodule dir *and* submodule cache dir and reinitialized from scratch. That worked. 16:42
[Coke] timo: your timing on changelog update questions is historially "just missed the window" 16:45
timo damn :) 16:46
[Coke] timo: I put it under "Core" which I have been treating as the catchall
timo the thing about submodules is that you're supposed to use the regular git commands and pretend the submodule is just a single file
so add, reset, stuff like that
[Coke] ... and it never works. :)
I never touch 3rdparty stuff. and whenever someone *else* does, my checkout gets fubar'd. 16:47
75% of the time, I just throw away my clone and get a new one once 3rdparty shows up in the status. just didn't want to do that with the release clone.
timo you have the additional constraint that in order for your commit of the outer repo to actually be usable to others, the inner repo must be not just committed but also pushed to the same thing other people have cloned there as well 16:48
but yeah i agree that it's a total pain
[Coke] ok. MoarVM release complete. slight break before picking up the nqp/rakudo releases. 16:49
back 17:00
spec test running on master... 17:17
looks like some issues on the errata branch, I believe these already mentioned post-unicode 17 merge but never fixed: 17:32
t/spec/S15-nfg/GraphemeBreakTest.t (Wstat: 768 (exited 3) Tests: 1943 Failed: 3) Failed tests: 749-751 17:33
^^ 6.c
Someone please open a ticket about that. I won't make it a release blocker, but we need a ticket to track until it's clean
some TODO passing in both 6c and 6d, but IIRC, that's fine. 17:42
Yup. same failure on 6.d for the s15-nfg test 17:45
can someone who isn't a core dev please open a ticket for that on roast? (and or just fix the errata branches?)
Geth nqp/main: 1b297e299e | (Will Coleda)++ | tools/templates/MOAR_REVISION
[release] Bump MoarVM revision to 2025.12
 17:52
nqp/main: 6a2c32b0d5 | (Will Coleda)++ | VERSION
[release] Bump VERSION to 2025.12
rakudo/release-2025.12: 3197500aa5 | (Will Coleda)++ | tools/templates/NQP_REVISION
[release] Bump NQP revision to 2025.12
rakudo/release-2025.12: 7b6eb332bd | (Will Coleda)++ | VERSION
[release] Bump VERSION to 2025.12
rakudo/release-2025.12: 8e7caeaeef | (Will Coleda)++ | docs/announce/2025.12.md
Fix next scheduled release date
 18:00
rakudo: coke++ created pull request #6049:
Release 2025.12
rakudo/main: 5 commits pushed by (Will Coleda)++ 18:01
[Coke] patrickb, el_che - should be OK to start binary releases. 18:07
Release is done. 18:11
Please squawk about any issues. I managed to avoid the one I made last time about putting the releases in the wrong dir.
Thanks to all the devs/docs/site updates everyone made this month, this was a pretty substantial release, esp. in MoarVM! 18:12
arkiuat teensy tiny one: in line 52 of docs/ChangeLog, might want to change 2025 to 2026 in the phrase "Upcoming releases in 2025 will include"
also thanks and yay!
[Coke] oops. 18:26
19:00 arkiuat left 19:03 rakkable left, rakkable joined 19:13 arkiuat joined 20:08 arkiuat left 20:33 finanalyst left
[Coke] upcoming year is calculated by a script - please open a ticket, tag it with RELENG, and mention that it should either be removed or be smarter at year end. 20:39
I would be fine if it were removed. 20:40
20:46 disbot7 joined, disbot6 left
patrickb the binary releases on rakudo.org (and thus rakubrew) for 2025.12 is done. 21:59
22:41 arkiuat joined
arkiuat [Coke], done, except that I don't think I have the ability to tag issues on rakudo. I mentioned it should be tagged RELENG in the text 22:42
well, in a comment 22:46
[Coke] arkiuat++ patrickb++ 23:59

Powered by Cro and the RakuĀ® Programming Language.

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

Thank you!