Day 7: Bridge Repair

Megathread guidelines

  • Keep top level comments as only solutions, if you want to say something other than a solution put it in a new post. (replies to comments can be whatever)
  • You can send code in code blocks by using three backticks, the code, and then three backticks or use something such as https://topaz.github.io/paste/ if you prefer sending it through a URL

FAQ

  • lwhjp
    link
    fedilink
    3
    edit-2
    1 month ago

    Haskell

    A surprisingly gentle one for the weekend! Avoiding string operations for concatenate got the runtime down below one second on my machine.

    import Control.Arrow
    import Control.Monad
    import Data.List
    import Data.Maybe
    
    readInput :: String -> [(Int, [Int])]
    readInput = lines >>> map (break (== ':') >>> (read *** map read . words . tail))
    
    equatable :: [Int -> Int -> Int] -> (Int, [Int]) -> Bool
    equatable ops (x, y : ys) = elem x $ foldM apply y ys
      where
        apply a y = (\op -> a `op` y) <$> ops
    
    concatenate :: Int -> Int -> Int
    concatenate x y = x * mag y + y
      where
        mag z = fromJust $ find (> z) $ iterate (* 10) 10
    
    main = do
      input <- readInput <$> readFile "input07"
      mapM_
        (print . sum . map fst . (`filter` input) . equatable)
        [ [(+), (*)],
          [(+), (*), concatenate]
        ]
    
    • @LeixB
      link
      31 month ago

      Love the fold on the list monad to apply the operations.

    • lwhjp
      link
      fedilink
      2
      edit-2
      1 month ago

      Since all operations increase the accumulator, I tried putting a guard (a <= x) in apply, but it doesn’t actually help all that much (0.65s -> 0.43s).

      • @VegOwOtenks
        link
        2
        edit-2
        1 month ago

        0.65 -> 0.43 sounds pretty strong, isn’t that a one-fourth speedup?

        Edit: I was able to achieve a 30% speed improvement using this on my solution

        • lwhjp
          link
          fedilink
          21 month ago

          It’s not insignificant, sure, but I’d prefer 10x faster :D

          Plus I’m not sure it’s worth the loss of generality and readability. It is tempting to spend hours chasing this kind of optimization though!

    • @VegOwOtenks
      link
      English
      21 month ago

      I wanted to this the way yo did, by repeatedly applying functions, but I didn’t dare to because I like to mess up and spend some minutes debugging signatures, may I ask what your IDE setup is for the LSP-Hints with Haskell?
      Setting up on my PC was a little bit of a pain because it needed matching ghc and ghcide versions, so I hadn’t bothered doing it on my Laptop yet.

      • @LeixB
        link
        31 month ago

        I use neovim with haskell-tools.nvim plugin. For ghc, haskell-language-server and others I use nix which, among other benefits makes my development environment reproducible and all haskellPackages are built on the same version so there are no missmatches.

        But, as much as I love nix, there are probably easier ways to setup your environment.

        • @VegOwOtenks
          link
          21 month ago

          I just checked and I have haskell-tools.nvim on my PC but it somehow crashes the default config of the autocompletion for me, which I am too inexperienced to debug. I’ll try it nonetheless, since I don’t have autocompletion on the laptop anyways, thank you for the suggestion!

      • lwhjp
        link
        fedilink
        21 month ago

        Ah, well, I have a bit of a weird setup. GHC is 9.8.4, built from git. I’m using HLS version 2.9.0.1 (again built from git) under Emacs with the LSP and flycheck packages. There are probably much easier ways of getting it to work :)

        • @VegOwOtenks
          link
          21 month ago

          I envy emacs for all of its modes, but I don’t think I’m relearning the little I know about vi. Thank you for the answer on the versions and building!