[00:11] <[Coke]> timo: https://github.com/rakudo/rakudo/pull/5017 is asking you some questions.

[00:11] <[Coke]> (from 2024)

[07:08] *** kurahaupo joined
[09:22] *** rakkable left
[09:22] *** rakkable joined
[09:47] *** [Coke]_ joined
[09:47] *** [Coke] left
[12:01] <Geth> ¦ roast/6.d-errata: 7e3e891578 | (Elizabeth Mattijsen)++ | S02-types/keyhash.t

[12:01] <Geth> ¦ roast/6.d-errata: Remove bogus test consuming QuantHash role

[12:01] <Geth> ¦ roast/6.d-errata: 

[12:01] <Geth> ¦ roast/6.d-errata: The QuantHash role is really an implementation detail which existence

[12:01] <Geth> ¦ roast/6.d-errata: should only be used to recognize QuantHashes, not to support any

[12:01] <Geth> ¦ roast/6.d-errata: other consuming classes or roles other than the ones provided in core.

[12:01] <Geth> ¦ roast/6.d-errata: review: https://github.com/Raku/roast/commit/7e3e891578

[12:06] <lizmat> After looking at fixing the QuantHash failure in 6.d-errata, I've come to the conclusion that the QuantHash / Setty / Baggy / Mixy roles are implementation details

[12:06] <lizmat> and only to be used as markers, pretty much like the Associative and Positional roles

[12:09] <lizmat> or make this work in the setting:

[12:10] <lizmat> m: role A { has $!a }; role B does A { }; class C does B { method b() { $!a } }; C.new.b

[12:10] <evalable6> lizmat, rakudo-moar a1f6cb4fb: OUTPUT: «»

[12:11] <lizmat> and even then, I wonder what consuming any of those roles would bring above subclassing any of the QuantHash classes: Set/SetHash/Bag/BagHash/Mix/MixHash

[14:30] *** finanalyst joined
[14:35] <finanalyst> .tell Lizmat: the most recent Rakudo changes are preventing my Rakuast::RakuDoc::Render modulle to seg fault. Wont even comoile. Not sure yet why

[14:35] <tellable6> finanalyst, I'll pass your message to lizmat

[14:55] *** finanalyst left
[15:11] *** [Coke]_ is now known as [Coke]

[18:36] * lizmat looks

[18:36] <tellable6> 2026-04-26T14:35:26Z #raku-dev <finanalyst> Lizmat: the most recent Rakudo changes are preventing my Rakuast::RakuDoc::Render modulle to seg fault. Wont even comoile. Not sure yet why

[18:43] <lizmat> .tell finanalyst the module installs fine for me

[18:43] <tellable6> lizmat, I'll pass your message to finanalyst

[18:46] <[Coke]> .ask finanalyst - "most recent" - are you using 2026.04 or something from head before that?

[18:46] <tellable6> [Coke], I'll pass your message to finanalyst

[19:06] <Geth> ¦ rakudo/main: b8f0e1f9ff | (Elizabeth Mattijsen)++ | src/core.c/Setty.rakumod

[19:06] <Geth> ¦ rakudo/main: Fix Set(Hash).EXISTS-KEY when parameterized with a coercive type

[19:06] <Geth> ¦ rakudo/main: review: https://github.com/rakudo/rakudo/commit/b8f0e1f9ff

[19:08] *** coleman left
[19:09] *** coleman joined
[19:59] <patrickb> the 2026.04 binaries are released. Except for MacOS ARM. MacStadium currently won't let me login to the server.

[19:59] <tellable6> 2026-04-25T19:40:23Z #raku-dev <[Coke]> patrickb binary releases for 2026.04 can start

[22:18] <Geth> ¦ rakudo/main: 287dfb25e9 | (Elizabeth Mattijsen)++ | src/core.c/SetHash.rakumod

[22:18] <Geth> ¦ rakudo/main: Fix Hyper for SetHashes parameterized with coercive type

[22:18] <Geth> ¦ rakudo/main: review: https://github.com/rakudo/rakudo/commit/287dfb25e9