Geth rakudo/main: 211197cbc8 | (Will Coleda)++ | tools/create-release-announcement.raku
Fix URL in release announcement tool
 01:24
releasable6 Next release in ≈1 day and ≈15 hours. There are no known blockers. Please log your changes in the ChangeLog: github.com/rakudo/rakudo/wiki/ChangeLog-Draft 03:00
04:38 Pixi left 05:28 melezhik joined 06:38 Pixi joined 06:40 melezhik_ joined 06:48 melezhik_ left, melezhik_ joined
melezhik_ ab5tract: I believe I fixed yesterday issue with wrong job counter , please as usually pull the latest commit and upgrade your agents 06:54
07:05 melezhik_ left, melezhik_ joined 07:18 melezhik_ left, melezhik_ joined 07:24 melezhik_ left, melezhik_ joined 07:37 melezhik_ left, melezhik left, melezhik_ joined 07:42 melezhik_ left 08:03 melezhik_ joined, melezhik_ left, melezhik_ joined
melezhik_ . 08:03
08:56 melezhik_ left, melezhik_ joined 09:04 melezhik_ left, melezhik_ joined 09:09 melezhik_ left, melezhik_ joined 09:14 melezhik_ left 09:15 melezhik_ joined 09:20 melezhik_ left, melezhik_ joined 09:25 melezhik_ left 09:26 melezhik_ joined 09:34 melezhik_ left, melezhik_ joined 09:42 melezhik_ left, melezhik_ joined 09:50 melezhik_ left, melezhik_ joined 10:03 melezhik_ left 10:04 melezhik_ joined 10:09 melezhik_ left, melezhik_ joined 10:15 melezhik_ left, melezhik_ joined 10:25 melezhik_ left, melezhik_ joined 10:31 melezhik_ left, melezhik_ joined 10:43 melezhik_ left 10:44 melezhik_ joined 10:51 melezhik_ left 10:54 melezhik_ joined 10:56 melezhik_ left 11:18 melezhik joined 12:22 melezhik_ joined
melezhik_ . 12:22
12:38 melezhik_ left 12:41 melezhik_ joined 12:46 melezhik_ left, melezhik_ joined 12:53 melezhik_ left, melezhik_ joined 13:19 melezhik_ left, melezhik_ joined 13:24 melezhik_ left, melezhik_ joined 13:37 melezhik left 13:40 melezhik_ left 13:42 melezhik_ joined 14:02 librasteve_ joined 14:18 melezhik_ left 14:19 melezhik_ joined 14:27 melezhik_ left, melezhik_ joined 14:42 melezhik_ left 14:43 melezhik_ joined 14:52 melezhik_ left, melezhik_ joined 15:05 melezhik_ left, melezhik_ joined 15:40 melezhik_ left, melezhik_ joined 15:41 melezhik_ left
lizmat just realized that $¢ in RakuAST only lives in QAST land 15:43
which is probably why we have this subtle difference:
m: Q|say $¢|.EVAL
camelia Nil
lizmat m: Q|say $¢|.AST.EVAL
camelia (Mu)
lizmat so I just found that: role A { dd $?ROLE }; class B does A { } very quickly recurses on itself somehow 16:16
m: role A { say $?ROLE }; class B does A { } # but "say" doesn't
camelia (A)
lizmat I guess a DIHWIDT case 16:20
timo did you already find the spot that recurses? 16:21
lizmat gist.github.com/lizmat/b2e97ffa4e1...0504621512 # first 50 lines out of the 80K I captured :-) 16:24
m: role A { say $?ROLE.raku }; class B does A { } # it's not .raku that's doing it 16:25
camelia A
timo there's an "exit" in dd? :D 16:29
lizmat no, I used own-up
timo m: class A { method statement-list { self }; method statements { [ ] } }; my $*CU = A.new; dd; say "survived dd call"; 16:30
camelia Cannot pop from an empty Array
in block <unit> at <tmp> line 1
timo m: class A { method statement-list { self }; method statements { [ 1 ] } }; my $*CU = A.new; dd; say "survived dd call";
camelia ( no output )
timo m: class A { method statement-list { self }; method statements { [ 1 ] } }; dd; say "survived dd call";
camelia block <unit>()
survived dd call
timo it's recursing inside of sub prefix? 16:32
lizmat not sure... as it looked like a DIHWIDT, I didn't want to look much further right now 16:33
timo might have been the call to .of there 16:34
this also seems to infini-recurse: role A { say $?ROLE.of }; class B does A { } 16:35
lizmat ah, that may be it, as dd is trying to do that 16:36
timo ok, so 16:37
trying to call .of on the role forces it to be punned 16:38
that causes the body of the role to be run
that calls into dd or say again
lizmat ah.. and that causes it to be punned again, as it isn't punned yet
timo that's probably also why there's an `-e:1` frame in there, that's the role body 16:44
17:21 melezhik_ joined 17:30 melezhik_ left 17:34 melezhik_ joined 17:48 melezhik_ left 17:49 melezhik_ joined 17:53 melezhik_ left 18:22 melezhik joined
lizmat rakkable: eco-provides $*EXECUTABLE-NAME 19:03
rakkable lizmat, Running: eco-provides $*EXECUTABLE-NAME, please be patient!
lizmat, Found 4 lines in 2 files (1 distributions):
lizmat, gist.github.com/79cff38d3e97d00a29...0d278f6de6
ab5tract This transitive role thing is really weird… the role appears to be composed when the type checking arrives 21:16
er, when it arrives at type checking
R#5317
linkable6 R#5317 [open]: github.com/rakudo/rakudo/issues/5317 [roles][Will be addressed in RakuAST] role-constrained variable does not accept derived role defined inline
ab5tract That is, $value.HOW is a ClassHOW, but with the nominal archetype 21:17
[Coke] anyone want access to the Snyk issues? patrickb? 22:02
22:02 melezhik left 22:40 committable6 left, quotable6 left, releasable6 left, sourceable6 left, unicodable6 left, lizmat left 22:41 [Tux] left 22:44 lizmat_ joined, [Tux] joined
ab5tract gist.github.com/ab5tract/468ffb02c...66ce3e8682 <-- it looks like inline role's aren't composed 22:59
23:02 melezhik joined 23:24 lizmat_ left, lizmat joined

Powered by Cro and the Raku® Programming Language.

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

Thank you!