It's gradually coming back to me. The Haskell Complex type doesn't work particularly nicely as an integer, plus the definition of division is more like "scale", so I just went with my own type.

Then I forgot which of div and quot I should use, and kept getting nearly the right answer :/

import Data.Ix  

data CNum = CNum !Integer !Integer  

instance Show CNum where  
  show (CNum x y) = "[" ++ show x ++ "," ++ show y ++ "]"  

cadd, cmul, cdiv :: CNum -> CNum -> CNum  
(CNum x1 y1) `cadd` (CNum x2 y2) = CNum (x1 + x2) (y1 + y2)  
(CNum x1 y1) `cmul` (CNum x2 y2) = CNum (x1 * x2 - y1 * y2) (x1 * y2 + y1 * x2)  
(CNum x1 y1) `cdiv` (CNum x2 y2) = CNum (x1 `quot` x2) (y1 `quot` y2)  

part1 a = iterate op (CNum 0 0) !! 3  
  where  
    op x = ((x `cmul` x) `cdiv` CNum 10 10) `cadd` a  

countEngraved = length . filter engrave  
  where  
    engrave p =  
      let rs = take 100 $ tail $ iterate (op p) (CNum 0 0)  
       in all (\(CNum x y) -> abs x <= 1000000 && abs y <= 1000000) rs  
    op p r = ((r `cmul` r) `cdiv` CNum 100000 100000) `cadd` p  

part2 a =  
  countEngraved  
    . map (\(y, x) -> a `cadd` CNum (x * 10) (y * 10))  
    $ range ((0, 0), (100, 100))  

part3 a =  
  countEngraved  
    . map (\(y, x) -> a `cadd` CNum x y)  
    $ range ((0, 0), (1000, 1000))  

main = do  
  print $ part1 $ CNum 164 56  
  print $ part2 $ CNum (-21723) 67997  
  print $ part3 $ CNum (-21723) 67997  

Ooh, challenges! Here we go!

I haven't really written any Haskell since last year's AoC, and boy am I rusty.

import Control.Monad  
import Data.List  
import Data.List.Split  
import Data.Vector qualified as V  

readInput s =  
  let [names, _, moves] = splitOn "," <$> lines s  
   in (names, map readMove moves)  
  where  
    readMove (d : s) =  
      let n = read s :: Int  
       in case d of  
            'L' -> -n  
            'R' -> n  

addWith f = (f .) . (+)  

part1 names moves =  
  names !! foldl' (addWith $ clamp (length names)) 0 moves  
  where  
    clamp n x  
      | x < 0 = 0  
      | x >= n = n - 1  
      | otherwise = x  

part2 names moves =  names !! (sum moves `mod` length names)  

part3 names moves =  
  V.head  
    . foldl' exchange (V.fromList names)  
    $ map (`mod` length names) moves  
  where  
    exchange v k = v V.// [(0, v V.! k), (k, V.head v)]  

main =  
  forM_  
    [ ("everybody_codes_e2025_q01_p1.txt", part1),  
      ("everybody_codes_e2025_q01_p2.txt", part2),  
      ("everybody_codes_e2025_q01_p3.txt", part3)  
    ]  
    $ \(input, solve) ->  
      readFile input >>= putStrLn . uncurry solve . readInput  

I'm very curious about F# since I've never used it or seen it anywhere before but I'm afraid I'm too tired to read it right now. Thank you for posting, I hope I'll remember to come back tomorrow.

I coded this along with my girlfriend who's learning python, but not motivated to share her solution. The program reads from stdin, because I usually invoke it like so: runhaskell Main.hs < input or runhaskell Main.hs < example. I think this is quite handy because I don't have to change the source code to check the example input again.

I struggled with Part 3, where I suddenly forgot I could've simply used mod, which I ended up doing anyway. I immediately recognized that Part 3 needs Mutable Arrays if I care to avoid Index hell, which is not what I wanted to with Haskell but oh well.

{-# OPTIONS_GHC -Wall #-}
{-# LANGUAGE PatternSynonyms #-}
module Main (main) where

import qualified Data.Text as Text
import qualified Data.Text.IO as TextIO
import Control.Monad ((<$!>), forM_)
import Data.Text (Text, pattern (:<))
import qualified Data.List as List
import qualified Data.Array.MArray as MutableArray
import Control.Monad.ST (runST, ST)
import Data.Array.ST (STArray)

commaSepLine :: IO [Text.Text]
commaSepLine = Text.split (== ',') <$!> TextIO.getLine

readInstruction :: Text -> Int
readInstruction ('R' :< n) = read . Text.unpack $ n
readInstruction ('L' :< n) = negate . read . Text.unpack $ n
readInstruction _ = undefined

myName :: (Foldable t, Ord b, Enum b, Num b) => b -> t b -> b
myName maxPosition = List.foldl' (\ pos offset -> min (pred maxPosition) . max 0 $ pos + offset) 0

parentName1 :: [Int] -> Int
parentName1 = List.sum

newSTArray :: [e] -> ST s (STArray s Int e)
newSTArray xs = MutableArray.newListArray (0, length xs - 1) xs

swap :: (MutableArray.MArray a e m, MutableArray.Ix i) => a i e -> i -> i -> m ()
swap array i0 i1 = do
  e0 <- MutableArray.readArray array i0
  e1 <- MutableArray.readArray array i1
  MutableArray.writeArray array i0 e1
  MutableArray.writeArray array i1 e0

parentName2 :: [Text] -> [Int] -> Text
parentName2 nameList instructions = runST $ do
  names <- newSTArray nameList
  arrayLength <- succ . snd <$> MutableArray.getBounds names
  forM_ instructions $ \ offset -> do
    let arrayOffset = offset `mod` arrayLength
    swap names 0 arrayOffset
  MutableArray.readArray names 0

main :: IO ()
main = do
  names <- commaSepLine
  _ <- TextIO.getLine
  instructions <- fmap readInstruction <$> commaSepLine

  let namesLength = length names
  print $ names !! myName namesLength instructions
  print . (names !!) . (`mod` namesLength) $ parentName1 instructions
  print $ parentName2 names instructions

edit: I did the wrong quest

I wasn't very happy with myself on this one. Part 1 wound up being simple enough, but part 2 took me for a loop. My disappointment was that I asked an AI for a hint, but instead it coded up a solution revolved around checking cycle dependency.

I tried to at least redo it in a functional way, since I'm writing this in FSharp, but I ran into a bunch of hiccups around indexes and index tracking. That always costs me so much time. This took me forever and I'm near the bottom of the rankings haha.

Overall, I wasn't terribly happy with how I did on this one. Oh, I wound up switching all of the types to BigInteger because I was lost as to what was happening. It turned out to be indexes and orderings, but I thought that maybe something was getting truncated.

    /// Map from (prev, curr) pair to position  
    type PositionMap = Map<BigInteger * BigInteger, BigInteger * BigInteger>  
    let rec findCycle (pairToNextPair: PositionMap) startPair currentPair acc =  
        if currentPair = startPair && List.length acc > 0 then  
            Some (List.rev acc)  
        else  
            match Map.tryFind currentPair pairToNextPair with  
            | None -> None  
            | Some nextPair -> 
                findCycle pairToNextPair startPair nextPair (snd currentPair :: acc)  

    let rec eni2' score (n:BigInteger) (exp:BigInteger) (m:BigInteger) (pairMap: PositionMap) (scores:BigInteger list) iter =  
        if iter > exp then scores |> List.rev |> List.skip (max 0 (List.length scores - 5)) |> toBigInt  
        else  
            let newScore = (score * n) % m  
            let key = (score, newScore)  
            
            match Map.tryFind key pairMap with  
            | Some _ -> 
                match findCycle pairMap key key [] with  
                | Some cycle ->  
                    let remaining = int64 (exp - iter)  
                    let cycleValues = cycle  
                    let cycleLength = List.length cycleValues |> int64  
                    let scoresLength = List.length scores |> int64  
                    let totalLength = scoresLength + 1L + remaining // scores + newScore + remaining  
                    
                    let needCount = min 5L totalLength  
                    let startPos = max 0L (totalLength - needCount)  
                    
                    let scoresReversed = List.rev scores  
                    
                    let final5 = 
                        [startPos..totalLength - 1L]  
                        |> List.map (fun pos ->  
                            if pos < scoresLength then  
                                // Position is in scores (scores is reversed, so oldest is at end)  
                                scoresReversed.[int pos]  
                            elif pos = scoresLength then  
                                // Position is newScore  
                                newScore  
                            else  
                                let cycleOffset = pos - scoresLength  
                                let cyclePos = cycleOffset % cycleLength  
                                cycleValues.[int cyclePos]  
                        )  
                    
                    // final 5 comes out in reverse order  
                    final5 |> List.rev |> toBigInt  
                | None -> 
                    eni2' newScore n exp m (Map.add key ((newScore, (newScore * n) % m)) pairMap) (newScore::scores) (iter + BigInteger 1)  
            | None ->  
                let nextPair = (newScore, (newScore * n) % m)  
                eni2' newScore n exp m (Map.add key nextPair pairMap) (newScore::scores) (iter +  BigInteger 1)  

    let eni2 (n) (exp) (m) = eni2' (BigInteger 1) n exp m Map.empty [] (BigInteger 1)  

    let part2 (input: IDictionary<string, BigInteger> array)=  
        input  
        |> Array.map (fun line ->  
            //printfn $"""running line {line.AsEnumerable() |> Seq.map(fun kv -> kv.Key + "=" + kv.Value.ToString()) |> fun x -> String.Join(", ", x)}"""  
            let a = eni2 line["A"] line["X"] line["M"]  
            let b = eni2 line["B"] line["Y"] line["M"]  
            let c = eni2 line["C"] line["Z"] line["M"]  
            let ret = a + b + c  
            //printfn $"found {ret}"  
            ret  
        ) |> Seq.max  

    let part2Answer = 
        File.ReadAllLines "Inputs/Q01_P02.txt"  
        |> part1Lines  
        |> part2  

Rust

use log::debug;
use std::collections::HashSet;

use regex::Regex;

#[derive(PartialEq, Eq, Hash, Clone)]
struct Number(isize, isize);

impl Number {

  fn add(self: &Number, b: &Number) -> Number {
    Number(self.0 + b.0, self.1 + b.1)
  }

  fn mul(self: &Number, b: &Number) -> Number {
    Number(self.0 * b.0 - self.1 * b.1, self.0 * b.1 + self.1 * b.0)
  }

  fn div(self: &Number, b: &Number) -> Number {
    Number(self.0 / b.0, self.1 / b.1)
  }
}

pub fn solve_part_1(input: &str) -> String {
  let re = Regex::new(r"A=\[(\d+),(\d+)\]").unwrap();
  let (_, [x, y]) = re.captures(input).unwrap().extract();
  let a = Number(x.parse().unwrap(), y.parse().unwrap());
  let mut res = Number(0, 0);
  for _ in 0..3 {
    res = res.mul(&res);
    res = res.div(&Number(10, 10));
    res = res.add(&a);
  }
  format!("[{},{}]", res.0, res.1)
}

pub fn solve_part_2(input: &str) -> String {
  let re = Regex::new(r"A=\[([-0-9]+),([-0-9]+)\]").unwrap();
  let (_, [x, y]) = re.captures(input).unwrap().extract();
  let a = Number(x.parse().unwrap(), y.parse().unwrap());
  let mut engraved_points = 0;
  let mut pts: HashSet<_> = HashSet::new();
  for i in 0..=100 {
    for j in 0..=100 {
      let pt = Number(a.0 + 10 * i, a.1 + 10 * j);
      let mut res = Number(0, 0);
      engraved_points += 1;
      pts.insert(pt.clone());
      for _ in 0..100 {
        res = res.mul(&res);
        res = res.div(&Number(100_000, 100_000));
        res = res.add(&pt);
        if res.0.abs() > 1_000_000 || res.1.abs() > 1_000_000 {
          engraved_points -= 1;
          pts.remove(&pt);
          break;
        }
      }
    }
  }
  for i in 0..=100 {
    debug!("{}", (0..=100).map(|j| if pts.contains(&Number(a.0 + 10*i, a.1 + 10*j)) { 'X' } else {'.'}).collect::<String>());
  }
  engraved_points.to_string()
}

pub fn solve_part_3(input: &str) -> String {
  let re = Regex::new(r"A=\[([-0-9]+),([-0-9]+)\]").unwrap();
  let (_, [x, y]) = re.captures(input).unwrap().extract();
  let a = Number(x.parse().unwrap(), y.parse().unwrap());
  let mut engraved_points = 0;
  for i in 0..=1000 {
    for j in 0..=1000 {
      let pt = Number(a.0 + i, a.1 + j);
      let mut res = Number(0, 0);
      engraved_points += 1;
      for _ in 0..100 {
        res = res.mul(&res);
        res = res.div(&Number(100_000, 100_000));
        res = res.add(&pt);
        if res.0.abs() > 1_000_000 || res.1.abs() > 1_000_000 {
          engraved_points -= 1;
          break;
        }
      }
    }
  }
  engraved_points.to_string()
}

I like the idea of using this community for other similar seasonal challenges. By its nature AoC isn't going to be very active outside of December so this would be a nice way to have the community be active.

Keep Advent Of Code and Everybody.codes together => More activity in this community, more discoverability for either competitions. Only drawback => polluted feed. Which can be fixed by the tags.