Took it as an opportunity to learn me some Vegalite too.

Thank you @kwando for inspiring me to do so.

Screenshot 2022-12-12 at 9.19.54 PM1200×489 47.8 KB

Basic stuff, will dive deeper at night.

Update: I could add animation to the first part. The second part graph looks ugly though. I need to play more with this stuff. This is amazing stuff!

Here’s the code for Livemd version: https://github.com/code-shoily/advent_of_code/blob/master/lib/2022/day_12.livemd

And the regular one (This one has cache based part 2 since I don’t need to worry about any graphs) advent_of_code/lib/2022/day_12.ex at main · code-shoily/advent_of_code · GitHub

I initially went with :digraph, since that is already part of the BEAM:

https://github.com/maennchen/advent-of-code/blob/b35a260de09c05404912e2c51c0ecd053f0b0476/2022/12/1_closest_path.exs#L36

I felt that that was cheating since that did 90% of the given problem. Therefore, I also implemented Dijkstra naively based on the Wikipedia description of the algorithm.

https://github.com/maennchen/advent-of-code/blob/cee1c818aca1de99b0ba8b1292869aff79ffffe8/2022/12/graph.exs#L1

:digraph was the thing I was looking for, I knew there were something like that already bundled but didn’t remember the name

Is there a way to get “all the distances” (djikstra table) from :digraph? It was quick enough to :digraph.get_short_path(g, <a>, goal) for every a though..