Amy

I forgot that “weekdays” for a US website means something different for me here in UTC+9.

This was surprisingly fiddly, but I think I managed to do it reasonably neatly.

import Control.Arrow  
import Data.Foldable  
import Data.List (sortBy)  
import Data.List.Split  
import Data.Maybe  
import Data.Ord  

data Fishbone  
  = Fishbone (Maybe Int) Int (Maybe Int) Fishbone  
  | Empty  
  deriving (Eq)  

instance Ord Fishbone where  
  compare = comparing numbers  

readInput :: String -> [(Int, Fishbone)]  
readInput = map readSword . lines  
  where  
    readSword = (read *** build) . break (== ':')  
    build = foldl' insert Empty . map read . splitOn "," . tail  

insert bone x =  
  case bone of  
    (Fishbone l c r next)  
      | isNothing l && x < c -> Fishbone (Just x) c r next  
      | isNothing r && x > c -> Fishbone l c (Just x) next  
      | otherwise -> Fishbone l c r $ insert next x  
    Empty -> Fishbone Nothing x Nothing Empty  

spine (Fishbone _ c _ next) = c : spine next  
spine Empty = []  

numbers :: Fishbone -> [Int]  
numbers (Fishbone l c r next) =  
  (read $ concatMap show $ catMaybes [l, Just c, r])  
    : numbers next  
numbers Empty = []  

quality :: Fishbone -> Int  
quality = read . concatMap show . spine  

part1, part2, part3 :: [(Int, Fishbone)] -> Int  
part1 = quality . snd . head  
part2 = uncurry (-) . (maximum &&& minimum) . map (quality . snd)  
part3 = sum . zipWith (*) [1 ..] . map fst . sortBy (flip compareSwords)  
  where  
    compareSwords =  
      comparing (quality . snd)  
        <> comparing snd  
        <> comparing fst  

main =  
  forM_  
    [ ("everybody_codes_e2025_q05_p1.txt", part1),  
      ("everybody_codes_e2025_q05_p2.txt", part2),  
      ("everybody_codes_e2025_q05_p3.txt", part3)  
    ]  
    $ \(input, solve) -> readFile input >>= print . solve . readInput  

I liked this one!

import Control.Arrow  
import Control.Monad  
import Data.List  
import Data.Ratio  

simpleTrain = uncurry (%) . (head &&& last) . map read  

compoundTrain input =  
  let a = read $ head input  
      z = read $ last input  
      gs =  
        map  
          ( uncurry (%)  
              . (read *** read . tail)  
              . break (== '|')  
          )  
          $ (tail . init) input  
   in foldl' (/) (a % z) gs  

part1, part2, part3 :: [String] -> Integer  
part1 = floor . (2025 *) . simpleTrain  
part2 = ceiling . (10000000000000 /) . simpleTrain  
part3 = floor . (100 *) . compoundTrain  

main =  
  forM_  
    [ ("everybody_codes_e2025_q04_p1.txt", part1),  
      ("everybody_codes_e2025_q04_p2.txt", part2),  
      ("everybody_codes_e2025_q04_p3.txt", part3)  
    ]  
    $ \(input, solve) -> readFile input >>= print . solve . lines  

I thought this was going to be the knapsack problem, but no.

import Control.Monad  
import Data.List.Split  
import qualified Data.Set as Set  
import qualified Data.Multiset as MSet  

part1, part2, part3 :: [Int] -> Int  
part1 = sum . Set.fromList  
part2 = sum . Set.take 20 . Set.fromList  
part3 = maximum . MSet.toCountMap . MSet.fromList  

main =  
  forM_  
    [ ("everybody_codes_e2025_q03_p1.txt", part1),  
      ("everybody_codes_e2025_q03_p2.txt", part2),  
      ("everybody_codes_e2025_q03_p3.txt", part3)  
    ]  
    $ \(input, solve) ->  
      readFile input >>= print . solve . map read . splitOn ","  

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  

Do you mean TeX (not GNU though)?

The related package METAFONT converges on e, as well.