Clojurians Log v2

https://github.com/Lambdaphile/aoc/blob/main/2024/clojure/src/aoc/days/01/core.clj

(def all (slurp "input.txt"))
(def small (slurp "small.txt"))

(let [data (-> all
               (split-lines)
               (->>
                (map #(split % #"   "))))
       col-a (sort (mapv #(Integer/parseInt %) (map first data)))
       col-b (sort (mapv #(Integer/parseInt % ) (mapv second data)))
       diff (map #(abs (- %1 %2)) col-a col-b )
       total (reduce + diff)      
      ]
   total)

I like all and small @thomas ha, nice

Could have compressed even more, but no need. Awesome that we have a frequencies method. Part 1

(defn row-prep [rowstr]
  (let [x (s/split rowstr #"[ ]+")
        y (map #(Integer/parseInt %) x)]

    ;; Just return Y.
    y))

(def c (map row-prep (s/split-lines in)))
(def left (sort (map first c)))
(def right (sort (map last c)))
(reduce + (map #(abs (- %1 %2)) left right))

(def right-freqs (frequencies right))
(reduce + (map #(* % (get right-freqs % 0)) left))

https://github.com/RyanFleck/aoc24/blob/master/src/aoc24/day1.clj

My day 1 solutions In YAMLScript: https://github.com/ingydotnet/advent-of-code/blob/main/2024/01/part1.ys

always seems like the parsing is the tough bit!

(defn parse-input [source]
  (let [lines (str/split-lines source)]
    (reduce (fn [[left right] line]
              (let [[x y] (map parse-long
                               (str/split line #"\s+"))]
                [(conj left x) (conj right y)]))
            [[] []]
            lines)))

(def day-1-input
  (-> "resources/aoc2024/day1.txt"
      slurp
      parse-input))

(defn total-distance [[list1 list2]]
  (reduce + 0 (map difference (sort list1) (sort list2))))

(defn task1 []
  (total-distance day-1-input))

(defn similarity-score [[list1 list2]]
  (let [l2-frequencies (frequencies list2)]
    (transduce (map #(* % (get l2-frequencies % 0))) + 0 list1)))

(defn task2 []
  (similarity-score day-1-input))

https://github.com/bhauman/adv2024/blob/main/src/adv2024/day01/sol.clj

Refactored the overall AOC/YS repo layout. ReadMe has section to show code as Clojure. https://github.com/ingydotnet/advent-of-code/blob/main/ReadMe.md

I'm committing to not obsessively staying up every night this year to finish them when they drop, but at least for the easy days I will... 🙂 https://gitlab.com/maximoburrito/advent2024/-/blob/main/src/day01/main.clj?ref_type=heads

Good to see a thread! Btw, Is it okay to post code to a public repo?

;; ## Day 1
(def pairs
  (-> ( 1)
   (
    #"\n"
    #"\s+"
    parse-long)))

;; ### Part 1
(let [left-list  (sort (map first pairs))
      right-list (sort (map second pairs))
      distance   (fn [a b] (abs (- a b)))]
  (reduce + (map distance left-list right-list)))

;; ### Part 2
(let [left             (map first pairs)
      right            (map second pairs)
      similarity       (frequencies right)
      similarity-score (fn [x] (* x (similarity x 0)))]
  (reduce + (map similarity-score left)))

Day 1 is up. It's fun to be writing Clojure again! • Blog post: https://github.com/abyala/advent-2024-clojure/blob/main/docs/day01.md • Source code: https://github.com/abyala/advent-2024-clojure/blob/main/src/advent_2024_clojure/day01.clj

@abyala I’ve been trying to wrap by head around juxt and this line really helps: > Whenever we take one thing and return a vector of things from different functions, we call on juxt Thanks for sharing 😁

As usual, day 1 is a breeze : https://github.com/Maravedis/advent_code/blob/master/src/advent_of_code/2024/01.clj . I wish those dropped at a friendlier time in Europe 😞

@kjw0323 Public repo is ok for your own solution, but they ask you do not commit any input or puzzle text. Source : https://adventofcode.com/2024/about > Can I copy/redistribute part of Advent of Code? Please don't. Advent of Code is free to use, not free to copy. If you're posting a code repository somewhere, please don't include parts of Advent of Code like the puzzle text or your inputs. If you're making a website, please don't make it look like Advent of Code or name it something similar.

(def real-input (slurp "src/2024/day01"))

(defn parse-input [input]
  (->> (str "[" input "]")
       edn/read-string
       (partition 2)))

; 1
(let [input-data (parse-input real-input)
      first-list (sort (map first input-data))
      second-list (sort (map second input-data))
      distances (map (fn [a b] (abs (- b a))) first-list second-list)]
  (reduce + distances))

; 2
(let [input-data (parse-input real-input)
      first-list (map first input-data)
      second-list (map second input-data)
      _ (prn (frequencies second-list)) ;; didn't think about that one
      diff-scores (map (fn [a]
                         (* a (count (filter #{a} second-list))))
                       first-list)]
  (reduce + diff-scores))

The edn/read-string with the [] concat is really funny to me for some reason.

Love read-string for this 🙂 Saw it first from Bruce Hauman in last year’s (correction: was 2022) advent of code threads: https://clojurians.slack.com/archives/C0GLTDB2T/p1670259468872829

(defn total-distance-between-lists
  "Given two sorted lists, return the sum of the absolute difference between each pair of numbers"
  [list-1 list-2]
  (reduce + (map #(Math/abs (- %1 %2)) list-1 list-2)))

(defn similarity-score
  "Given a frequency map and a number, return the similarity score"
  [frequency-map number]
  (* number (frequency-map number 0)))

(defn total-similarity-score
  "Given a sequence of numbers and a frequency map, return the total similarity score"
  [numbers frequency-map]
  (reduce (fn [total number]
            (+ total (similarity-score frequency-map number)))
          0
          numbers))

(defn- pairs->vectors
  "Given a sequence of pairs, return a pair of
  vectors containing the first and second elements of each pair"
  [pairs]
  (reduce (fn [[list-1 list-2] [a b]]
            [(conj list-1 a) (conj list-2 b)])
          [[] []]
          pairs))

(defn part-1
  [input]
  (let [pairs (utils/parse-long-pairs input)
        [list-1 list-2] (pairs->vectors pairs)
        sorted-list-1 (sort list-1)
        sorted-list-2 (sort list-2)]
    (total-distance-between-lists sorted-list-1 sorted-list-2)))

(defn part-2
  [input]
  (let [pairs (utils/parse-long-pairs input)
        [list-1 list-2] (pairs->vectors pairs)
        frequency-map (frequencies list-2)]
    (total-similarity-score list-1 frequency-map)))

(ns aoc2024.aoc01)

(def input (->> (slurp "/Users//aoc/aoc24/aoc1.txt")
                (re-seq #"\d+")
                (map parse-long)
                (partition 2)
                (apply map vector)))

(defn p1 [[l1 l2]]
  (reduce + (mapv (comp abs -) (sort l1) (sort l2))))

(defn p2 [[l1 l2]]
  (let [s (frequencies l2)]
    (transduce (map #(abs (* % (s % 0)))) + l1)))

That's the second use of transduce I've seen. I really need to look into how that works.

It’s map + reduce (in this case)

Generally transduce is used when you have a reduce operation, but first you map/filter etc the input sequence. Granted you can just do that inside the reducing function in reduce but it’s generally cleaner to use transduce . Also using transducers can be much faster.

Okay, I will play around with using that for my solution.

Thanks for the explanation!

@roklenarcic you define input and then two functions, but how do you connect the two?

I'm trying to port your solution to the squint playground... I guess just call p1 and p2 on input?

yes, just (p1 input)

usually I have input and test-input etc… so most of the time I have p1 and p2 functions that take an input so I can test with various inputs

ok cool, here's @roklenarcic’s solution on the squint playground: https://squint-cljs.github.io/squint/?boilerplate=https%3A%2F%2Fgist.githubusercontent.com%2Fborkdude%2Fcf94b492d948f7f418aa81ba54f428ff%2Fraw%2Fa6e9992b079e20e21d753e8c75a7353c5908b225%2Faoc_ui.cljs&repl=true&src=OzsgSGVscGVyIGZ1bmN0aW9uczoKOzsgKGZldGNoLWlucHV0IHllYXIgZGF5KSAtIGdldCBBT0MgaW5wdXQKOzsgKGFwcGVuZCBzdHIpIC0gYXBwZW5kIHN0ciB0byBET00KOzsgKHNweSB4KSAtIGxvZyB4IHRvIGNvbnNvbGUgYW5kIHJldHVybiB4Cgo7OyBSZW1lbWJlciB0byB1cGRhdGUgdGhlIHllYXIgYW5kIGRheSBpbiB0aGUgZmV0Y2gtaW5wdXQgY2FsbC4KKGRlZiBpbnB1dCAoLT4%2BIChqcy1hd2FpdCAoZmV0Y2gtaW5wdXQgMjAyNCAxKSkKICAgICAgICAgICAgIChyZS1zZXEgIyJcZCsiKQogICAgICAgICAgICAgKG1hcCBwYXJzZS1sb25nKQogICAgICAgICAgICAgKHBhcnRpdGlvbiAyKQogICAgICAgICAgICAgKGFwcGx5IG1hcCB2ZWN0b3IpKSkKCihkZWZuIHAxIFtbbDEgbDJdXQogIChyZWR1Y2UgKyAobWFwdiAoY29tcCBhYnMgLSkgKHNvcnQgbDEpIChzb3J0IGwyKSkpKQoKKGRlZm4gcDIgW1tsMSBsMl1dCiAgKGxldCBbcyAoZnJlcXVlbmNpZXMgbDIpXQogICAgKHRyYW5zZHVjZSAobWFwICMoYWJzICgqICUgKGdldCBzICUgMCkpKSkgKyBsMSkpKQoKI18odGltZSAocDEgaW5wdXQpKQojXyh0aW1lIChwMiBpbnB1dCkp

Using read-string trick I can make it shorter: https://gist.github.com/RokLenarcic/18a9a9cd134ca7f53c8897396208149e

https://github.com/jramosg/advent-of-code/blob/master/src/advent_of_code/year_2024/day_01/main.clj

@roklenarcic read-string isn't available in squint (yet) :)

Today was a fun Day1 puzzle, I enjoyed it. [Here](https://github.com/alexalemi/advent/blob/main/2024/clojure/p01.clj) is my clojure solution. I would have liked to make the whole thing a transduce but I couldn't figure out how to turn (apply map list) into a transducer immediately.

Fun warmup! https://github.com/CarnunMP/Advent-of-Code/blob/master/src/y2024/d1.clj

https://clojurians.slack.com/archives/C0GLTDB2T/p1733037898227799?thread_ts=1733029934.345259&cid=C0GLTDB2T Oops, I thought publishing my specific puzzle inputs was fine, as they're generated. Maybe it used to be fine, but they changed the policy? In any case, off to tweak my repo... 😅

https://github.com/ghaskins/adventofcode/blob/main/src/aoc/2024/day1.clj

I realized after that my use of zipmap was probably the wrong choice as it wouldnt have worked if there were repeating items in the lhs, but I got lucky. In any case, I didnt place on the leaderboard. This was my first AOC and it was fun

I tend to take advantage of day 1 being easy to explore a different language. just finished a Haskell book, so decided to put it to practice. https://github.com/FelipeCortez/advent-of-code/blob/master/2024/01.hs will switch to Clojure tomorrow!

I corrected my improper use of zipmap and simplfied

(ns aoc.2024.day1
  (:require [clojure.string :as string]))

(defn parse []
  (->> (slurp "inputs/day1.txt")
       (string/split-lines)
       (map #(map parse-long (string/split % #"\s+")))
       (reduce (fn [acc items] (mapv conj acc items)) [[] []])
       (map sort)))

(defn part1 []
  (let [[left right] (parse)]
    (->> (map (comp abs -) left right)
         (reduce + 0))))

(defn part2 []
  (let [[left right] (parse)
        f (frequencies right)]
    (->> (map #(* % (get f % 0)) left)
         (reduce + 0))))

@dingelsz3 - I'm glad you found it helpful!

@carnunmp it's from the very beginning. https://www.reddit.com/r/adventofcode/comments/7lesj5/comment/drlt9am/

I have a feeling that the last time I used frequencies was in AOC23 :-)

(def input [...])

(def lists
  [(take-nth 2 input)
   (take-nth 2 (rest input))])

(defn part1 []
  (let [l1 (sort (first lists))
        l2 (sort (second lists))]
    (reduce + (map #(abs (- %1 %2)) l1 l2))))

(defn part2 []
  (let [l1 (first lists)
        l2f (frequencies (second lists))]
    (reduce #(+ %1 (* (get l2f %2 0) %2)) 0 l1)))

https://github.com/benjamin-asdf/advent-of-code/blob/master/src/Y2024/day1.clj

take-nth I just learned about this

similar to others https://github.com/tschady/advent-of-code/blob/main/src/aoc/2024/d01.clj

People say Clojure isn't a language where you have one obvious way to do something, but looking at all the solutions they're all almost identical. Great minds think alike 😅 Here's mine https://codeberg.org/aciep/advent-of-code-2024-clojure/src/branch/master/src/aoc/d1.clj

(require '[ :as io]
         '[net.cgrand.xforms :as x])

(def xf (comp (mapcat (fn [ln] (re-seq #"\d+" ln)))
              (map parse-long)
              (x/multiplex {:xs (take-nth 2) :ys (comp (drop 1) (take-nth 2))})
              (x/by-key (comp (x/sort) (x/into [])))))

(let [{:keys [xs ys]} (into {} xf (io/lines-in *in*))]
  (println "Part A:" (reduce + (map (comp abs -) xs ys)))
  (println "Part B:" (let [frqs (frequencies ys)]
                       (transduce (map (fn [x] (* (get frqs x 0) x))) + xs))))

We could maybe make a new one? Lots of people are obviously not participating.

Good idea! Let's make 2024 edition

(preparing one, stay tuned)

current status: failed... I've created new github account for that purposes but: You are marked as spam, and therefore cannot authorize a third party application...

no worries - alternative solution works

Ok! Here is the new code for this year: 4534281-b9354839

what is the channel name?

nm

Is the code something that can be shared?

what is the join code?

In the topic of the channel 217019-4a55b8eb

it says that private leaderboard is full. 😢

I got a "that private leaderboard is full" as well....bummer!

Here is the code for that: https://github.com/genmeblog/advent-of-code/blob/master/src/common.clj#L4-L36

And the following to update the store from data/inputs and data/examples folders

(do
    (add-all-data)
    (add-all-data "examples"))

Don't forget to purge github content with the history of input commits: https://docs.github.com/en/authentication/keeping-your-account-and-data-secure/removing-sensitive-data-from-a-repository

Part 1 looks ok for me. One comment for part 2: you can get rid of map -> filter -> map chain by setting a default value in get

Thanks! Missed that 🙂

Also (get freq % _default_) can be called as (freq % _default_)

yeah, maps are functions of their keys. I tried to do it but my lsp was complaining for some reason

A very useful tool when parsing numbers, or anything that just repeats, is re-seq , instead of having to split:

(re-seq #"\d+" input) ; => ["1" "2" "3" "14" "12"])

Rudolf flies at 14km/h, for 10 seconds, before having to rest for 133 seconds.

Oh, that's neat. Thanks!

On a private leaderboard page, you can click [Ordering] to have this snippet appear:

Ty!

It's someone who used an LLM. They actually apologized and deleted the link toward their repo. They were unaware it was discouraged to use AI as a solver and not just a crutch.

How do you see the stats of other leaderboard rows? I can see mine but not others

i.e. how did you see that #1 took 4 seconds @chase-lambert?

https://adventofcode.com/2024/leaderboard/day/1

If you go to the global leaderboard there are links to the individual days at the top

@chase-lambert ty!

and thats crazy

slowest person was 1:24

I've never done it for time but I check out the times out of curiosity every now and then

i think i finished #1 in 14m, heh. Not even close