Module day14

Disk Defragmentation

This problem is a blend of the hashing from Day 10 and the connected clique finding from Day 12 and reuses the same flood fill approach to count groups.

Structs

Exclusive 🔒

Regular data structures need to be protected by a mutex.

Shared

Atomics can be safely shared between threads.

Functions

dfs 🔒

Flood fill that explores the connected squares in the grid.

fill_row 🔒

Compute the knot hash for a row and expand into an array of bytes.

knot_hash 🔒

Slightly tweaked version of the code from Day 10 that always performs 64 rounds.

parse

Parallelize the hashing as each row is independent.

part1

part2

worker 🔒

Each worker thread chooses the next available index then computes the hash and patches the final vec with the result.