00:00 yewscion left
aruniecrisps Thank you for the help! I'm currently playing around with the idea of making monad syntax in Raku and so I'm looking at the forgiven repo as a blueprint 00:42
guifa m: enum Foo <a b c>; say Foo.^mro 01:01
camelia ((Foo) (Int) (Cool) (Any) (Mu))
guifa m: enum Foo <a b c>; say Foo.HOW
camelia Perl6::Metamodel::EnumHOW.new
01:07 yewscion joined 01:10 AnaBanana joined 01:13 yewscion left
guifa hmmm 01:13
I'm trying to add a coerce target for an enum
given, say 01:22
enum Type <Question Answer Comment>; sub foo (Type() :$type) { ... }
How could I enable calling with `foo :type<Question>` ? 01:23
antononcube Can use where in the sub's definition signature? 01:24
guifa the where clause won't coerce the value 01:25
so COERCE gets called, but by default COERCE works on the value (0, 1, 2) not the identifier
antononcube I see. 01:26
01:29 AnaBanana left
guifa I mean I totally get I can just manually do stuff 01:29
but I'd rather handle it in the signature nicely if possible 01:30
guifa is going to try to do an mDNS module
it seems the problem is that the coerce HOW methods are applied after the trait 01:36
aha got it 01:45
multi sub trait_mod:<is> (::EnumType \enumerable, :$coerceable!) { enumerable.^add_multi_method('COERCE', anon sub COERCE (EnumType \SELF, Str \from) { SELF::.AT-KEY: from }) unless enumerable.HOW.^name eq 'Perl6::Metamodel::EnumHOW'} 01:47
antononcube 👍 01:48
guifa does mean I kinda want to use lc ones now 01:51
antononcube Are you the only one who is going to read this code? 01:52
🙂 BTW, since you are record telling me to use kebab-case instead of camelCase, I feel I can justly ask: What's up with snake_case? (In your code.) 01:54
01:54 Manifest0 left
guifa add_multi_method is a Raku HOW core method 02:12
so the idea is I can do Packet.new: :authoritative, :truncated, :want-recursive, :!recursive-available
or 02:13
:type<foo>
I added a new multi candidate for the boolean ones
antononcube How about trait_mod ? 02:17
02:19 kylese left, kylese joined 03:15 kylese left, kylese joined 04:11 yewscion joined 04:15 yewscion left 04:34 yewscion joined 04:44 yewscion left
guifa same 05:07
the end result is you say
enum Foo is enumerable <a b c>;
err coerceable
05:23 yewscion joined 05:27 yewscion left
greenfork [Coke]: I did exactly this :) I thought that maybe there's a better way to write in it repl without jumping between start/end of the line and classify seems great 05:37
timo: thank you for classify! I looked at it multiple times and it seemed like it doesn't do what I want. There's also categorize that I don't understand despite looking at it multiple times too 05:38
I think I should make a cheat sheet for simple data transformations like list -> hashmap etc. 05:39
In other languages there's a seemingly simpler alternative: there's a single "enumerable comprehension" function that specifies the final data structure 05:41
Ruby: [%w[black tree], %w[yellow tree], %w[blue sky]].each_with_object({}) { |el, acc| acc[el[1]] ||= []; acc[el[1]].push(el[0]) } 05:44
each_with_object({}) specifies `{}`
Elixir (simpler example): for [v, k] <- [~w[black tree], ~w[blue sky]], into: %{}, do: {k, v} 05:49
into: %{} specified `%{}`
I was looking in Raku a similar idea but couldn't find it
It is not bad but rather not obvious coming from other languages. A "general" solution seems easier than having to memorize specific functions that allow transitioning between different data structures 05:50
07:57 Aedil joined 08:46 dakkar joined 08:56 yewscion joined 08:59 yewscion left 09:23 Sgeo left 09:27 Manifest0 joined 09:33 sena_kun joined 09:52 sena_kun left 10:34 traxex joined
librasteve greenfork: I will try to mirror your ruby example in raku: 11:20
Ruby: [%w[black tree], %w[yellow tree], %w[blue sky]].each_with_object({}) { |el, acc| acc[el[1]] ||= []; acc[el[1]].push(el[0]) }
11:21 abraxxa-home joined 11:27 yewscion joined 11:29 yewscion left 11:35 yewscion joined
Raku: my %h; [<black tree>, <yellow tree>, <blue sky>].map: { %h{.[1]}.push: .[0] } 11:37
this is a slightly different take - since a raku Hash (%h) has a nice .push method that accumulates items as we need 11:39
11:40 abraxxa-home left, yewscion left
I agree that .classify is a bit of a sledgehammer to crack this nut 11:40
11:42 yewscion joined 11:46 yewscion left 12:07 zayd left 12:17 zayd joined
Geth ecosystem/main: 4484f6c967 | (Elizabeth Mattijsen)++ | META.list
Remove Plosurin, looks like abandoned

And no reaction from author		 14:38
ecosystem/main: 995e27e153 | (Elizabeth Mattijsen)++ | META.list
Remove modules by adaptiveoptics

They appear abandoned and author doesn't respond		 14:52
ecosystem/main: 7c2d297497 | (Elizabeth Mattijsen)++ | META.list
Remove ajs modules

They will be migrated to Raku Community modules		 14:58
antononcube @lizmat I wil put in my Raku Advent post draft within 5 hours. I assume that is a few hours before the automatic publishing. 15:07
lizmat antononcube thanks! that's a relief to hear 15:08
antononcube The post is 90% written -- I have to make coherent image links. (Always takes longer than expected.)
lizmat don't I know it :-) 15:09
antononcube @lizmat Would be fine if the image are in "dark mode" ? 15:13
jdv antononcube: did you get your dist fixed?
lizmat antononcube yes
antononcube @jdv Hmm.. I think it is just a library variable renaming. I think I did, but I have to experiment with the new version. 15:14
jdv no, its a change in how the repl works 15:17
antononcube Ok, I have misunderstood then. I can look into this more closely after 6-7 hours. 15:18
15:23 Sgeo joined
tadzik hmm. I know the best time to do this was like last year, but I reckon I should move my stuff to community modules :) 15:36
(or drop it entirely...)
but I think at least some of it is useful 15:37
lizmat tadzik: if you could transfer the repos that you think are worth preserving, I'll make sure they get a nice new release as community moduoe
*module
tadzik starts with Acme::Meow 15:38
15:44 GreaseMonkey left
Geth ecosystem/main: 5d177b246c | (Tadeusz Sośnierz)++ (committed using GitHub Web editor) | META.list
Remove my modules

They've been moved to raku-community-modules and/or are obsolete.		 15:44
tadzik I think I did this right
lizmat yeah, looks about ok :-) 15:45
15:53 yewscion joined 15:57 greaser|q joined 16:46 xinming left 17:11 MyNetAz left 17:26 MyNetAz joined 17:45 dakkar left
librasteve greenfork: if you are a purist and do not want to declare a lexical as your acc, maybe you would prefer this... 17:48
m: `[<black tree>, <yellow tree>, <blue sky>] andthen -> @_, %_={} { @_.map: { %_{.[1]}.push: .[0] }; %_ } andthen .say; 17:49
evalable6 (exit code 1) ===SORRY!=== Error while compilin…
librasteve, Full output: gist.github.com/a0f9e565b8498f9bfd...a7c19055d7
Raku eval Exit code: 1 ===SORRY!=== Error while compiling /home/glot/main.raku Bogus statement at /home/glot/main.raku:1 ------> <BOL>⏏``[<black tree>, <yellow tree>, <blue sk expecting any of: prefix statement list term
librasteve m: [<black tree>, <yellow tree>, <blue sky>] andthen -> @_, %_={} { @_.map: { %_{.[1]}.push: .[0] }; %_ } andthen .say;
evalable6 {sky => [blue], tree => [black yellow]}
Raku eval {sky => [blue], tree => [black yellow]}
18:09 yewscion left 18:37 yewscion joined 18:39 sena_kun joined 18:45 yewscion left, yewscion joined 18:46 zenmov left 19:07 Sampersand joined
antononcube @lizmat My Day 12 draft is completed. 19:17
lizmat cool!
antononcube I did not schedule it. (I am not sure do I have the rights to do it.) 19:18
lizmat ok, will do!
thanks!
antononcube 👍
19:25 PipStuart joined 19:32 traxex left, swaggboi left 19:36 traxex joined
Geth advent/main: 29b5639542 | (Elizabeth Mattijsen)++ (committed using GitHub Web editor) | raku-advent-2024/authors.md
day 12 scheduled
 19:38
antononcube 🎉 19:41
19:41 swaggboi joined
lizmat antononcube did some copy-editing as well, hope you don't mind 19:42
so you may want to double check if I didn't do an oopsie 19:43
:-)
antononcube Thanks! I was just writing that I have to proofread the post again.
I have an LLM prompt "CopyEdit", but using it usually takes me longer time, that "simple" proofreading/ 19:45
19:46 Sampersand left 19:47 yewscion left 19:50 Sampersand joined
Geth ecosystem/main: 7d7711bce6 | (Elizabeth Mattijsen)++ | META.list
Remove jnthn's modules

They are being migrated to Raku community modules		 19:53
19:57 yewscion joined 20:07 Sampersand left
greenfork @librasteve interesting! My point about avoiding a lexical variable is to type it in repl more easily. So although there's a purist in me, I for the love of God would not remember such incantation :) 20:07
Raku: my %h; [<black tree>, <yellow tree>, <blue sky>].map: { %h{.[1]}.push: .[0] } 20:09
That is very close. Interesting use of dots in .[1] 20:10
20:13 greaser|q left, greaser|q joined, greaser|q is now known as GreaseMonkey, yewscion left 20:14 swaggboi left 20:17 yewscion joined 20:21 kjp left 20:27 Sampersand joined 20:41 Aedil left 20:58 Sampersand left
librasteve yeah - the andthen is a bit of an acquired taste ... another way to write that is as an sub declaration and then to call it on your data: 21:07
my &fn = -> @_, %_={} { @_.map: { %_{.[1]}.push: .[0] }; %_ } 21:08
say fn( [<black tree>, <yellow tree>, <blue sky>] ); 21:09
another way to write the function is: 21:10
my &fn = -> @el, %acc={} { @el.map: { %acc{.[1]}.push: .[0] }; %acc };
which is familiar from the ruby hopefully...
21:22 swaggboi joined 22:38 Sampersand joined 23:14 sena_kun left 23:17 Sampersand left 23:21 yewscion left 23:37 AlexDaniel joined 23:46 unicodable6 joined 23:55 AlexDaniel left, AlexDaniel 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!