19 Oct 2025
SmokeMachine I Amy very impressed on how easy Raku made that possible… the only problems are the same I had with Red (but on Red I was able to find a workaround). The short circuit operators not being overridable. I still plan to change that with RakuAST, but I’m not sure if it will be possible/enough. 21:05
I'm rethinking it... what would be better to be the name of a operator to receive 2 args, coerce them to Logic::Ternary and return the `and` of them? `AND` or `and3`? 22:09
arkiuat I prefer and3 of those two. 22:24
lizmat wouldn't it make more sense to put the code in a role, and mix that into the enum ? 22:26
also, I think method gist could just be self.Str ? 22:27
afk&
SmokeMachine I tried putting that in a role, but couldn’t make it work… 22:36
I prefer this way too: github.com/FCO/Logic-Ternary 23:44
20 Oct 2025
Logic::Ternary sent to Zef... :) 00:04
m_zero trand and then, course tror and trxor or if you prefer triand, trior, and trixor 00:40
(and for bonus, these are fun to say out loud! 😁 ) 00:41
SmokeMachine I released and3, or3, xor3, not3 and so3... 00:59
raku.land/zef:FCO/Logic::Ternary 01:18
Voldenet SmokeMachine: I think and3 for regular `Bool:U, Bool::True, Bool::False` values would be more obvious 03:32
m: sub infix:<and3>($x, $y) { return False if $x eqv False; return False if $y eqv False; return Bool unless $x.defined; return Bool unless $y.defined; $x and $y }; my @v = (Bool, False, True); for (@v X @v) -> ($x, $y) { say ($x,$y,$x and3 $y).map(*.gist) } 03:37
camelia ((Bool) (Bool) (Bool))
((Bool) False False)
((Bool) True (Bool))
(False (Bool) False)
(False False False)
(False True False)
(True (Bool) (Bool))
(True False False)
(True True True)
Voldenet I'm sure it can be more elegantly written maybe by even mapping Bool:U to Logic::Ternary::Unknown 03:41
current problem with Logic::Ternary is that it overrides Bool making it very non-obvious trap 03:46
m: enum Ternary ( False => -1, Unknown => 0, True => +1 ); sub thing(Bool $x) { say $x }; thing(True) # uhhhh 03:47
camelia ===SORRY!=== Error while compiling <tmp>
Calling thing(Ternary) will never work with declared signature (Bool $x)
at <tmp>:1
------> => +1 ); sub thing(Bool $x) { say $x }; <HERE>thing(True) # uhhhh
Voldenet also `Unknown` should become `Bool:U`, not False 03:49
after coercing
but really coercing should be done for internal logic, because Bool is enough 03:50
m: multi sub infix:<and3>(Bool $x, Bool $y) { given any($x, $y) { when * eqv False { False }; when not .defined { Bool }; default { $x and $y }} }; my @v = (Bool, False, True); for (@v X @v) -> ($x, $y) { say ($x,$y,$x and3 $y).map(*.gist) } 04:03
camelia ((Bool) (Bool) (Bool))
((Bool) False False)
((Bool) True (Bool))
(False (Bool) False)
(False False False)
(False True False)
(True (Bool) (Bool))
(True False False)
(True True True)
Voldenet m: multi sub infix:<and3>(Bool $x, Bool $y) { any($x, $y) eqv False ?? False !! $x and $y }; my @v = (Bool, False, True); for (@v X @v) -> ($x, $y) { say ($x,$y,$x and3 $y).map(*.gist) } 04:26
camelia ((Bool) (Bool) (Bool))
((Bool) False False)
((Bool) True (Bool))
(False (Bool) False)
(False False False)
(False True False)
(True (Bool) (Bool))
(True False False)
(True True True)
Voldenet m: multi sub infix:<and3>(Bool $x, Bool $y) { any($x, $y) !eqv False && $x and $y }; my @v = (Bool, False, True); for (@v X @v) -> ($x, $y) { say ($x,$y,$x and3 $y).map(*.gist) } 04:29
camelia ((Bool) (Bool) (Bool))
((Bool) False False)
((Bool) True (Bool))
(False (Bool) False)
(False False False)
(False True False)
(True (Bool) (Bool))
(True False False)
(True True True)
melezhik. . 06:36
SmokeMachine Voldenet: Bool:U is already mapped to Unknown (github.com/FCO/Logic-Ternary/blob/...kumod#L59) 07:11
Voldenet: yes, you will need to use Bool::* if you need to use boolean values. But I don't see it as a problem... 07:12
Voldenet but raku is dynamic language, it's not java where you can count on types 07:15
m: enum X <True>; sub Y(Bool $x) { }; sub Z($x) { Y($x) }; Z(True) # consider this 07:17
camelia Type check failed in binding to parameter '$x'; expected Bool but got X (X::True)
in sub Y at <tmp> line 1
in sub Z at <tmp> line 1
in block <unit> at <tmp> line 1
Voldenet redefining those is a bit like redefining Int or Bool 07:18
m: enum X<Bool>; say Bool::True; # this is what I mean
camelia Could not find symbol '&True' in 'X'
in block <unit> at <tmp> line 1
Voldenet Yes, you can still access it through CORE::Bool::True, but it's still messy code 07:24
so I'd still use Bool:U and actual Bool values, because they already exist 07:25
SmokeMachine Voldenet: about Unknown -> Bool:U, I plan to do that... I only forgot about it on 0.0.2... 07:37
Voldenet it'd work, but I'd say that the module shouldn't export True/False by default either way 07:40
in fact, I think the following should be valid: `use Logic::Ternary;; sub foo(Logic::Ternary() $_) { .say }; foo(Bool); foo(Bool::True); foo(Bool::False);` 07:45
so, `True` and `False` from Bool are being used, you can convert things to and from Logic::Ternary but it doesn't override True/False anyhow 07:46
melezhik Sparky ci has switched to Rakudo 2025.10 , if someone wants to test their modules against fresh Rakudo … 08:30
El_Che . 10:01
librasteve o/ 13:34
[Coke] not really a zef question but: if I do zef look META6 in powershell, and then hit ^C for some reason... the shell gets VERY confused, and keeps asking me More? More? More? - if this were bash, I'd try 'stty sane', but it looks completely messed up - only option is to open another powershell, run zef look again. 15:54
(and after the ^C, I appear to back in the original powershell, not the zef look shell)
ah, ^Z helps.
... ah, no, it's just not asking me more any more, still borked. 15:55
ugexe exit so shell(%*ENV<SHELL> // %*ENV<ComSpec> // %*ENV<COMSPEC>, :cwd($dist-path)) ?? 0 !! 1; 15:56
that is what zef does for zef look

Powered by Cro and the Raku® Programming Language.

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

Thank you!