[00:04] *** sjn left
[01:35] *** [Coke] left
[01:37] *** [Coke] joined
[05:43] *** hurufu joined
[09:48] *** hurufu left
[11:01] *** sjn joined
[13:04] <timo> random observation: env RAKUDO_RAKUAST=1 rakudo -e 'sub test(Num() $x where 1..*) { }; say &test.signature.params[0].constraint_list[0].raku;'   →   -> ;; Mu $_? = OUTER::<$_> { #`(Block|955247808) ... }

[13:04] <timo> without RAKUDO_RAKUAST: -> ;; Mu $_? = OUTER::<$_> { #`(Block|449520608) ... }

[13:04] <timo> wrong paste

[13:04] <timo> -> ;; $_ { #`(Block|1041456880) ... }

[13:05] <timo> the "use outer $_ as default" in rakuast is probably unnecessary, or perhaps it's even wrong, though I can't think of a way to not pass an argument to the constraint from our own code

[14:02] <Geth> ¦ rakudo/lizmat-9: e2913d03cb | (Elizabeth Mattijsen)++ | src/main.nqp

[14:02] <Geth> ¦ rakudo/lizmat-9: Make RAKUDO_RAKUAST=0 *not* activate the RakUAST logic

[14:02] <Geth> ¦ rakudo/lizmat-9: 

[14:02] <Geth> ¦ rakudo/lizmat-9: Spotted by timo++

[14:02] <Geth> ¦ rakudo/lizmat-9: review: https://github.com/rakudo/rakudo/commit/e2913d03cb

[14:02] <Geth> ¦ rakudo: lizmat++ created pull request #6214: Make RAKUDO_RAKUAST=0 *not* activate the RakUAST logic

[14:02] <Geth> ¦ rakudo: review: https://github.com/rakudo/rakudo/pull/6214

[14:07] <timo> put you a comment

[14:14] <timo> I was thinking without your change anything is valid in the env var and only env var missing or empty string in it turns rakuast off; with the change I expect RAKUDO_RAKUAST=hello will cause rakudo to crash in versions that have the check and not crash in earlier versions

[14:20] <timo> actually, it'd be easier if we just string-compare against "" or "0" maybe?

[14:20] <timo> instead of attempting to intify

[14:25] <timo> actually looks like i was wrong and +"abcd" just gives 0 so that's actually fine

[14:26] *** librasteve_ joined
[17:09] <[Coke]> m: say +"abcd"

[17:09] <camelia> rakudo-moar f628473b8: OUTPUT: «Cannot convert string to number: base-10 number must begin with valid digits or '.' in '<HERE>abcd' (indicated by <HERE>)␤  in block <unit> at <tmp> line 1␤␤»

[17:22] <timo> nqp: say(+"äbcd")

[17:22] <camelia> nqp-moarvm: OUTPUT: «␤We trust you have received the usual lecture from the local System␤Administrator. It usually boils down to these three things:␤␤    #1) Respect the privacy of others.␤    #2) Think before you type.␤    #3) With great power comes great responsibility…»

[17:22] <timo> uhhh i guess not

[17:25] <librasteve_> https://rakudoweekly.blog/2026/06/01/2026-22-some-ast-fruit/

[18:01] <lizmat> librasteve_++

[18:49] <Geth> ¦ rakudo/main: 8ec11ecc2f | (Elizabeth Mattijsen)++ | src/core.c/core_epilogue.rakumod

[18:49] <Geth> ¦ rakudo/main: Make sure .assuming passes on slurpy hashes

[18:49] <Geth> ¦ rakudo/main: 

[18:49] <Geth> ¦ rakudo/main: In reference to https://github.com/rakudo/rakudo/issues/6212

[18:49] <Geth> ¦ rakudo/main: 

[18:49] <Geth> ¦ rakudo/main: Whenever there was a slurpy hash, it would not pass on the slurpy

[18:49] <Geth> ¦ rakudo/main: hash argument.  That was wrong, it should, and it should actually

[18:49] <Geth> ¦ rakudo/main: slip the hash in.

[18:49] <Geth> ¦ rakudo/main: <…commit message has 5 more lines…>

[18:50] <Geth> ¦ rakudo/main: review: https://github.com/rakudo/rakudo/commit/8ec11ecc2f

[19:34] <Geth> ¦ rakudo/we_know_dyncall_plus_clang_is_broken_so_make_ci_green: 6a2787ef19 | (Timo Paulssen)++ | t/04-nativecall/02-simple-args.t

[19:34] <Geth> ¦ rakudo/we_know_dyncall_plus_clang_is_broken_so_make_ci_green: TODO both problematic nativecall tests for moar+clang+dyncall

[19:34] <Geth> ¦ rakudo/we_know_dyncall_plus_clang_is_broken_so_make_ci_green: 

[19:34] <Geth> ¦ rakudo/we_know_dyncall_plus_clang_is_broken_so_make_ci_green: the code clang generates assumes the upper bytes of an unsigned char or

[19:34] <Geth> ¦ rakudo/we_know_dyncall_plus_clang_is_broken_so_make_ci_green: unsigned short are zeroed out, but dyncall only has a function for passing

[19:34] <Geth> ¦ rakudo/we_know_dyncall_plus_clang_is_broken_so_make_ci_green: signed char and short, where negative numbers end up sign-extended, messing

[19:34] <Geth> ¦ rakudo/we_know_dyncall_plus_clang_is_broken_so_make_ci_green: clang's generated assembly for x == 0xFE up, which tries to comare the whole

[19:34] <Geth> ¦ rakudo/we_know_dyncall_plus_clang_is_broken_so_make_ci_green: 32bit register against 0xFE.

[19:34] <Geth> ¦ rakudo/we_know_dyncall_plus_clang_is_broken_so_make_ci_green: review: https://github.com/rakudo/rakudo/commit/6a2787ef19

[19:34] <Geth> ¦ rakudo: timo++ created pull request #6215: TODO both problematic nativecall tests for moar+clang+dyncall

[19:34] <Geth> ¦ rakudo: review: https://github.com/rakudo/rakudo/pull/6215

[19:34] <timo> this one's for you [Coke] 

[19:35] <[Coke]> timo: do we want skip or todo?

[19:36] <timo> dunno, i stole the skip from the other test which i assume had the exact same root cause but no "only macos" condition for the skip

[19:38] <lizmat> and now for something completely different:

[19:38] <lizmat> m: sub a(Int:D $b) { }; dd :type(.type), :modifier(.modifier) given &a.signature.params.head

[19:39] <camelia> rakudo-moar 8ec11ecc2: OUTPUT: «:modifier(":D")␤:type(Int)␤»

[19:39] <lizmat> so it looks that Parameters don't actually store Int:D as the type, but as Int and some flags

[19:41] <lizmat> this breaks this case of .assuming:

[19:41] <lizmat> m: sub a($a, Int:D $b) { }; dd &a.assuming(42).signature

[19:41] <camelia> rakudo-moar 8ec11ecc2: OUTPUT: «:(Int $b)␤»

[19:42] <lizmat> that should have been :(Int:D $b)

[19:42] <lizmat> m: sub a($a, Int:D $b) { }; dd &a.assuming(42)(Int)

[19:42] <camelia> rakudo-moar 8ec11ecc2: OUTPUT: «Parameter '$b' of routine 'a' must be an object instance of type 'Int',␤not a type object of type 'Int'. Did you forget a '.new'?␤  in sub a at <tmp> line 1␤  in block <unit> at <tmp> line 1␤␤»

[19:43] <lizmat> it will still break because of the inner signature, but still

[19:44] <[Coke]> timo++

[20:12] <timo> well, at least CI will be green now

[20:18] <[Coke]> That will be very helpful.

[20:19] <[Coke]> and now we have in the test a note about what we don't expect to work.

[20:19] <[Coke]> timo++

[21:07] <Geth> ¦ raku.org: librasteve self-assigned Dead "Install" link on Tools page https://github.com/Raku/raku.org/issues/321

[21:09] <Geth> ¦ rakudo: ugexe++ created pull request #6216: RakuAST: allow anonymous declarations across all package kinds

[21:09] <Geth> ¦ rakudo: review: https://github.com/rakudo/rakudo/pull/6216

[23:51] <Geth> ¦ setup-raku/dependabot/npm_and_yarn/vitest-4.1.0: cb90dc9e2d | dependabot[bot]++ (committed using GitHub Web editor) | 2 files

[23:51] <Geth> ¦ setup-raku/dependabot/npm_and_yarn/vitest-4.1.0: Bump vitest from 4.0.18 to 4.1.0

[23:51] <Geth> ¦ setup-raku/dependabot/npm_and_yarn/vitest-4.1.0: 

[23:51] <Geth> ¦ setup-raku/dependabot/npm_and_yarn/vitest-4.1.0: Bumps [vitest](https://github.com/vitest-dev/vitest/tree/HEAD/packages/vitest) from 4.0.18 to 4.1.0.

[23:51] <Geth> ¦ setup-raku/dependabot/npm_and_yarn/vitest-4.1.0: - [Release notes](https://github.com/vitest-dev/vitest/releases)

[23:51] <Geth> ¦ setup-raku/dependabot/npm_and_yarn/vitest-4.1.0: - [Changelog](https://github.com/vitest-dev/vitest/blob/main/docs/releases.md)

[23:51] <Geth> ¦ setup-raku/dependabot/npm_and_yarn/vitest-4.1.0: - [Commits](https://github.com/vitest-dev/vitest/commits/v4.1.0/packages/vitest)

[23:51] <Geth> ¦ setup-raku/dependabot/npm_and_yarn/vitest-4.1.0: 

[23:51] <Geth> ¦ setup-raku/dependabot/npm_and_yarn/vitest-4.1.0: <…commit message has 8 more lines…>

[23:51] <Geth> ¦ setup-raku/dependabot/npm_and_yarn/vitest-4.1.0: review: https://github.com/Raku/setup-raku/commit/cb90dc9e2d

[23:51] <Geth> ¦ setup-raku: dependabot[bot]++ created pull request #53: Bump vitest from 4.0.18 to 4.1.0

[23:51] <Geth> ¦ setup-raku: review: https://github.com/Raku/setup-raku/pull/53