01:22 itaipu left 01:27 itaipu joined 01:37 Aedil joined 01:43 oodani left 01:48 oodani joined 02:02 guifa joined
Voldenet that new LogicTernary is nice, but `True3 Unknown3 False3` would be better defaults 03:41
so all things exported by ternary contains 3, it's very obvious
it's just my opinion though 03:42
I've been struggling with "5 types of enum with the same name" a lot
in fact, in C# I've been struggling with enums with exactly the same but different namespaces 03:46
the same name*
though that's inevitable 03:47
04:02 mimosa joined 04:03 guifa left
jubilatious1_98524 tri-True, tri-False, tri-Unknown ?? 04:40
05:24 Aedil left 05:25 Aedil joined
Voldenet just postfix 3 is a bit easier to type, tri-* would be noisy 05:33
05:49 abraxxa joined 05:56 abraxxa left 05:58 abraxxa joined 06:19 Sgeo left 06:20 Aedil left 06:24 Aedil joined 07:13 apac joined 07:41 arkiuat joined, dakkar joined
arkiuat I also wouldn't mind if True3/Unknown3/False3 were exported by default. 07:42
now that I've thought it over some more
07:45 melezhik joined 07:50 lizmat left 08:06 arkiuat left
librasteve o/ 08:07
is ,= supposed to work? 08:08
Voldenet but it works, doesn't it 08:09
m: my $x = 4; $x ,= 5; say $x
camelia (\List_6095617942944 = (List_6095617942944 5))
librasteve_ m: my @numbers = 1, 2, 3; say @numbers ,= 4, 5;
camelia (\Array_3301976132616 = [Array_3301976132616 (4 5)])
librasteve_ yeah that seems like a bug to me
Voldenet m: my @numbers = 1, 2, 3; say @numbers = @numbers, (4, 5); 08:10
camelia (\Array_3427642784496 = [Array_3427642784496 (4 5)])
Voldenet the output is exactly the same though
librasteve_ I excpet the outcome to be [1,2,3,4,5] 08:11
for both
08:11 arkiuat joined
Voldenet m: my @numbers = 1, 2, 3; say @numbers = |@numbers, (4, 5); # like this? 08:12
camelia [1 2 3 (4 5)]
Voldenet but having this would prevent you from constructing list of lists 08:13
librasteve_ no
m: my %a = :11a, :22b; %a ,= :33x; say %a;
camelia {a => 11, b => 22, x => 33}
librasteve_ well it works for Hashs
but not for Lists or Arrays
docs.raku.org/language/operators#postfix_,= 08:14
SmokeMachine m: my @numbers = 1, 2, 3; say @numbers = |@numbers, |(4, 5) 08:15
camelia [1 2 3 4 5]
08:16 arkiuat left
Voldenet m: my @numbers = 1, 2, 3; say @numbers .= append: 4, 5 08:16
camelia [1 2 3 4 5]
SmokeMachine m: my @numbers = 1, 2, 3; say @numbers ,= |(4, 5); say @numbers 08:17
camelia (\Array_5838662526592 = [Array_5838662526592 4 5])
(\Array_5838662526592 = [Array_5838662526592 4 5])
Voldenet you'd need to somehow slip the @numbers into numbers 08:20
iirc there's more differences between Hash/Array - like .Hash doesn't clone but .Array does for example 08:21
librasteve_ but it works with Hash … so seems like a bug to me
Voldenet I'm not even sure if they should work the same
librasteve_ my point is that the intent is the same with Hash and List and Array
Voldenet however, it'd require changing how , operator behaves 08:22
and it sounds dangerous
because suddenly @c = (@a, @b) would autoslip its arguments 08:23
librasteve_ ok - lets say we keep it how it is - please give me an example to show how it works with Array (the most obvious situation) that does involve say dumping the Array_ident
s/does/doesn’t/
i’m synpathetic that this 08:24
Voldenet though I agree that the result with @c = (@c, @b) is a bit weird and I am not sure what it'd be useful for
librasteve_ say (%(:a(0)), %(:b(1))) 08:25
evalable6 ({a => 0} {b => 1})
Voldenet but the , op doesn't know if it's invoked with syntax `@c = (@c, @b)` or `@c ,= @b`, does it? 08:26
librasteve_ what i mean to say is that Im sympathetic that `,`works differently with Hash than Array (ie no slip stuff)
08:26 melezhik_ joined
Voldenet Not really - in hash case first value doesn't contain any key at all 08:27
in array case first values doesn't need any index
librasteve_ sorry not explaining myself very well - snippet coming
Voldenet I mean, in hash case `%x = (%x, :34x)` pretty obviously doesn't ask hash to nest hash as its value 08:28
m: my %x = :33y; %x = (%x, :34z); say %x # obvious 08:29
camelia {y => 33, z => 34}
Voldenet m: my %x = :33y; %x = (:%x, :34z); say %x # also obvious, similar to list case
camelia (\Hash_4498526920320 = {x => Hash_4498526920320, z => 34})
Voldenet the problem with , is that it is always used to construct new list 08:30
and all the operands are elements of the new list
08:35 arkiuat joined
librasteve_ ok I guess I get it … missed opportunity imo … but I’m not gonna ask to redefine the language 08:37
Voldenet yeah, I think it would've been better if list construction needed first array as item 08:38
e.g @c = (\@c, @b)
it'd solve this particular case and prevent this :( 08:39
08:40 arkiuat left
Voldenet in fact, I'd do @numbers.push: 4, 5 08:40
not as short, but not lengthy either
08:45 melezhik_ left 09:06 arkiuat joined 09:10 arkiuat left
librasteve changing the subject - the docs say that prefix // is a thing in v6 09:11
docs.raku.org/language/operators#prefix_//
but I cannot get it to work - probably I am holding it wrong - can someone show me an example
please? 09:12
ab5tract m: use v6.*; dd //Nil 09:13
camelia Bool::False
ab5tract m: use v6.*; dd //1
camelia Bool::True
ab5tract m: use v6.*; dd //0
camelia Bool::True
ab5tract seems like it's workin?
librasteve oh - my bad - thanks for the help! 09:14
ab5tract cheers :) 09:26
09:40 arkiuat joined 09:45 arkiuat left
jubilatious1_98524 @librasteve I believe @Nemokosh posted a Github Issue on,= ? I can't find it atm. 09:49
@librasteve Here are Raku-Study-Group notes on ,= : github.com/doomvox/raku-study/blob...nt.raku#L8 09:51
More notes on ,= from Raku-Study-Group: github.com/search?q=repo%3Adoomvox...;type=code 09:54
Looks like ,= has been under discussion since 2020, on the mailing list: www.nntp.perl.org/group/perl.perl6...g9384.html 09:55
10:04 arkiuat joined 10:09 arkiuat left
Voldenet if anything, "Hash" is more convenient and harder to explain :P 10:38
10:38 arkiuat joined 10:43 arkiuat left
jubilatious1_98524 @Voldenet What would a newbie need to know, in order to move forward with the language? 11:01
11:05 melezhik left 11:06 arkiuat joined, patrickb left, cpli left, atweedie left, thatonelutenist left, greenfork left 11:11 arkiuat left
SmokeMachine Voldenet: I like the default to "override" the logic. But your argument of all exports have 3... what is your other people's opinion? 11:23
11:25 kylese joined 11:33 Ekho left 11:34 melezhik joined 11:37 librasteve_ left 11:38 arkiuat joined 11:45 arkiuat left 11:48 lizmat joined 11:50 apac left
Voldenet SmokeMachine: the overriding kinda works, but you have to remember to use it, I like KnownTrue, KnownFalse and Unknown for being very obvious, but "just add 3 to everything if you need ternary" makes this very obvious 11:50
I mean, you have to read the manual to understand what True3 does
it's not obvious…
but once you do, every single thing just having '3' postfix is sane 11:51
and immediately visible when you read the code
11:56 lizmat left
Voldenet jubilatious1_98524: If I had to put it into words - arrays/lists are stricter - .Array always copies, ,= will nest. On the other hand .Hash only coerces (not copies) and ,= doesn't nest things 11:58
and also this behavior is container-dependant 12:00
m: my %l = :1a; %l ,= :2b; say %l # understandable
camelia {a => 1, b => 2}
Voldenet m: my $l = :1a; $l ,= :2b; say $l # uhhhh???
camelia (\List_4530704967216 = (List_4530704967216 b => 2))
Voldenet well in that case :1a is just a pair, so to make sure 12:02
m: my $l = (:1a).Hash; $l ,= :1b; say $l
camelia (\List_4517452484112 = (List_4517452484112 b => 1))
SmokeMachine m: my $l = %(:1a); $l ,= :2b; say $l
camelia (\List_4613851695696 = (List_4613851695696 b => 2))
Voldenet It's… surprising if you don't know ¯\_(ツ)_/¯
SmokeMachine m: my $l = %(:1a); %($l) ,= :2b; say $l 12:06
camelia {a => 1, b => 2}
SmokeMachine m: my $l = :1a; %($l) ,= :2b; say $l
camelia a => 1
jubilatious1_98524 @SmokeMachine overriding logic is difficult without some way to introspect which logic is being followed.
12:07 arkiuat joined
SmokeMachine you can just `dd True` and see if it prints Bool::True or Logic::Ternary::True 12:07
or `True ~~ Logic::Ternary`... 12:09
jubilatious1_98524:
jubilatious1_98524: ☝️
12:11 arkiuat left
jubilatious1_98524 @SmokeMachine The R programming language uses TRUE, FALSE, and NA. Maybe you could try True-na, False-na, and NA. 12:11
(@SmokeMachine:I don't understand the distinction between Unknown and Nil here. If there is no distinction then Nil is probably better). 12:13
SmokeMachine why na? and now you can use whatever you want, you can just do `use Logic::Ternary <True-na NA False-na>` and it will work the way you want
jubilatious1_98524 That sounds pretty cool. 12:14
SmokeMachine usercontent.irccloud-cdn.com/file/.../image.png 12:19
12:19 atweedie joined, thatonelutenist joined, greenfork joined, patrickb joined, cpli joined 12:25 Ekho joined 12:27 lizmat joined 12:28 lizmat_ joined, lizmat left 12:29 lizmat_ left 12:36 apac joined 12:38 arkiuat joined 12:42 arkiuat left 12:45 rba left 12:52 lichtkind joined 13:00 rba joined 13:06 arkiuat joined 13:10 apa_c joined, arkiuat left 13:11 apac left 13:15 lizmat joined 13:16 lizmat left, lizmat_ joined 13:19 lizmat_ left
SmokeMachine Sorry, I forgot to post it here, the way you want: usercontent.irccloud-cdn.com/file/.../image.png 13:21
jubilatious1_98524: ☝️
13:32 librasteve_ joined
librasteve @jubilatious1_98524 - I see a bunch of discussion on ,= and I was surprised it dwims on Hashs but not Arrays & Lists ... but it's not clear to me what the "ask" is ... I would say to change the logic of , - well that ship has sailed. 13:37
13:39 arkiuat joined 13:45 arkiuat left
SmokeMachine for Lists it shouldn't work, right? Once Lists are "immutable"... 13:45
Voldenet well it depends, do you like lisp? :> 13:47
13:48 andinus2 left
SmokeMachine just saw it... my last Ternary related image looks buggy... 13:48
Voldenet huh 13:52
it looks correct with the last zef version: `use Logic::Ternary <True-na NA False-na>; say True-na or3 NA` `True-na` 13:53
13:53 andinus joined
SmokeMachine that might be a change I've done after releasing it... thanks 13:54
13:55 arkiuat joined, melezhik left 13:59 arkiuat left 14:00 arkiuat joined 14:04 arkiuat left 14:10 apa_c left 14:12 melezhik_ joined
jubilatious1_98524 Sorry, or3 just looks wrong to me... . 14:13
...but maybe there's no way to use/extend standard and , or. 14:18
SmokeMachine jubilatious1_98524: the return for `or3` on the image I shared is a bug I created only here locally. As Voldenet shared and I confirmed locally with the released version, `or3` looks correct: `use Logic::Ternary <True-na NA False-na>; say True-na or3 NA` returns `True-na`. 14:28
usercontent.irccloud-cdn.com/file/.../image.png 14:29
14:30 arkiuat joined
SmokeMachine usercontent.irccloud-cdn.com/file/.../image.png 14:30
14:34 arkiuat left
jubilatious1_98524 @SmokeMachine I'd just rather see or there... . 14:38
SmokeMachine jubilatious1_98524: Sorry, what do you mean? 14:39
14:48 melezhik_ left, melezhik_ joined 14:52 arkiuat joined 14:57 cwiggins123 joined 14:58 arkiuat left
cwiggins123 hi everyone, new to raku here. planning on making a full system install script for krak3n.my with raku for gigglez. wish me luck 15:03
[Coke] good luck. 15:07
15:15 oodani left 15:16 arkiuat joined
SmokeMachine cwiggins123: maybe something like this could help you? github.com/FCO/Configuration 15:18
15:18 oodani joined
cwiggins123 oh wow I've never seen that before. thanks 15:23
15:26 apa_c joined
cwiggins123 yeah so basically I would need to boostrap perl and then raku with a posix script and then do the rest in raku 15:27
this configuratino module looks interesting
15:41 melezhik_ left 15:45 cwiggins123 left 15:46 cwiggins123 joined
Voldenet jubilatious1_98524: or3 vs or is very similar to == vs cmp - semantic difference is very useful here, because if you forget to use module, you're going to get errors and not wrong results at runtime 15:48
[Coke] wonder if there's a way to say "which module in the ecosystem has the most dependencies (including transitive) 15:50
15:56 melezhik_ joined 16:01 melezhik_ left 16:02 melezhik_ joined 16:11 apa_c left 16:18 timo2 is now known as timo
[Coke] m: my $x = Nil; say //$x ?? 'yo' !! 'no'; # try without... 16:22
camelia ===SORRY!=== Error while compiling <tmp>
Null regex not allowed. Please use .comb if you wanted to produce a
sequence of characters from a string.
at <tmp>:1
------> my $x = Nil; say //<HERE>$x ?? 'yo' !! 'no'; # try without...
16:24 melezhik_ left, melezhik_ joined 16:27 melezhik_ left
ugexe [Coke]: you could probably code up such a query fairly easy by putting something like github.com/ugexe/zef/blob/2c6ab9ff...od#L41-L43 in a loop 16:28
16:28 melezhik_ joined
ugexe be recording @dependencies-candidates.elems for each module 16:29
[Coke] Thanks
16:35 dakkar left 16:37 melezhik_ left, melezhik_ joined 17:00 cwiggins123 left 17:07 melezhik_ left, melezhik_ joined 17:13 melezhik_ left 17:14 melezhik_ joined 17:19 arkiuat left 17:21 melezhik_ left, melezhik_ joined 17:24 melezhik_ left 17:27 arkiuat joined 17:33 arkiuat left 17:39 melezhik_ joined 17:46 mimosa left 17:48 melezhik_ left 17:50 kylese left 17:51 melezhik_ joined 17:59 melezhik_ left, melezhik_ joined 18:03 arkiuat joined 18:08 arkiuat left 18:15 melezhik_ left 18:16 melezhik_ joined 18:21 melezhik_ left, melezhik_ joined 18:26 melezhik_ left 18:27 melezhik_ joined 18:32 melezhik_ left, melezhik_ joined 18:34 andinus left 18:36 andinus joined 18:38 melezhik_ left, melezhik_ joined, arkiuat joined 18:45 melezhik_ left, melezhik_ joined 18:52 melezhik_ left, melezhik_ joined 18:57 melezhik_ left, melezhik_ joined
thowe does anyone know what time zone libresteve might be in? 18:58
19:02 melezhik_ left
librasteve hi i’m in the UK 19:06
that’s UTC+1 (8pm) right now 19:08
19:10 apa_c joined
[Coke] m: use v6.*; my $x = Nil; say //$x ?? 'yo' !! 'no'; 19:30
camelia no
[Coke] librasteve_: ^^
m: use v6.e.PREVIEW; my $x = Nil; say //$x ?? 'yo' !! 'no'; 19:33
camelia no
[Coke] m: use v6.d; my $x = Nil; say //$x ?? 'yo' !! 'no';
camelia ===SORRY!=== Error while compiling <tmp>
Null regex not allowed. Please use .comb if you wanted to produce a
sequence of characters from a string.
at <tmp>:1
------> use v6.d; my $x = Nil; say //<HERE>$x ?? 'yo' !! 'no';
librasteve ha - now I try in vi on my local machine it works too! must have been something wrong in my incantation to summon v6.e 19:34
glad to see it is working! 19:35
20:03 orangebot joined 20:23 orangebot left
thowe libresteve, OK. the address parsing thing has come back around to my radar and I have gone down a bit of a rabbit hole with it. In the US there is some new data formats being adopted as part of the NG 911 initiative. Maybe we can chat sometime about it... Actually awaiting some info from Oregon state about a thing. Anyway. 20:46
I may have a need to do some tooling for my GIS stuff. If so, that will involve a lot of different states. 20:47
librasteve_ thowe: cool - my general idea is that Raku could benefit from an address parser that works at scale using Grammars … and since Grammars are classes, then we can start with a very general one and then override specifics to accord with country / state / language specific conventions 20:50
I kinda started to sketch this out with github.com/librasteve/raku-Contact/tree/main 20:51
please can you take a look and see if your general approach would align with that? - perhaps the code you already worked on would be a more robust version of that? 20:53
sorry … I am turning in for the night, but feel free to start a discussion (eg as an issue on that repo…) - maybe you can also point me to the code you have already 20:54
&afk for now
thowe Yeah. A Grammar is what I am after... It may need some additional bells and whistles. I doubt my code will be the base, but maybe can help inform requirements. I'm a barely functional hobbyist. 20:58
20:58 arkiuat left 20:59 apa_c left 21:10 arkiuat joined 21:15 arkiuat left 21:16 arkiuat joined 21:50 oodani left, oodani joined 22:08 Sgeo joined 22:25 Voldenet left 22:27 Voldenet joined 22:33 lichtkind left 22:36 arkiuat left 22:40 lizmat joined, lizmat left 22:49 arkiuat joined 22:53 arkiuat left 22:56 Aedil left 23:14 arkiuat joined 23:27 Guest10 joined 23:53 Guest10 left

Powered by Cro and the Raku® Programming Language.

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

Thank you!