Discussion on: Advent of Code 2019 Solution Megathread - Day 6: Universal Orbit Map

View post

Replies for: Part 1 was a no-brainer in clojure: ; Returns the number of orbits the given object is contained in (i.e. direct orbit + indirect orbits) (defn o...

I am fake upset you didn't break out the clojure zipper library for this! (it's like using a cannon for flies in this case)

Johnny • Dec 6 '19 • Edited

Oh, I will definitely look into that, thanks for the suggestion.
I only knew about (tree-seq),
I'm still a newbie in clj.

Jon Bristow • Dec 7 '19

Zipper is extremely powerful for navigating trees efficiently.

It’s also as easy to read as overly point free Haskell.

Johnny • Dec 7 '19

I vastly improved it without using zippers now.

; Counts the amount of elements in coll before element appears. Returns (count coll) if it doesn't appear at all.
(defn count-until [element coll]
  (count (take-while (partial not= element) coll)))

; Calculates the minimum amount of traversals required in an orbit map to move from object to destination
(defn traversals [orbit-map object destination]
  (let [object-centers (orbit-centers orbit-map object)
        destination-centers (orbit-centers orbit-map destination)
        first-common-center (some (set destination-centers) object-centers)]
    (+ (count-until first-common-center object-centers)
       (count-until first-common-center destination-centers))))

I just fetch the first common center of both objects and add the steps it takes to get there for both.