Day 4

Advent of Code: Worked Solutions

Setup

Import libraries:

library(tidyverse)

Read text input from file into a matrix:

input <- read_lines("../input/day04.txt") |>
  str_split("") |> 
  enframe(name = NULL) |> 
  unnest_wider(value, names_sep = "") |> 
  as.matrix() |> 
  unname()

Part 1

Define a function to shift a matrix up, down, left, or right:

shift <- function(mtx, dir = c("up", "down", "left", "right"), ...) {
  switch(dir,
    "up"    = mtx |> lead(...),
    "down"  = mtx |> lag(...),
    "left"  = mtx |> t() |> lead(...) |> t(),
    "right" = mtx |> t() |> lag(...)  |> t()
  )
}

Define a function which, for each cell in the matrix, counts how many of its 8 immediate neighbors match a given value:

count_nbrs <- function(mtx, value) {
  list(
    mtx |> shift("up",    default = "."),
    mtx |> shift("down",  default = "."),
    mtx |> shift("left",  default = "."),
    mtx |> shift("right", default = "."),
    mtx |> shift("up",    default = ".") |> shift("left",  default = "."),
    mtx |> shift("up",    default = ".") |> shift("right", default = "."),
    mtx |> shift("down",  default = ".") |> shift("left",  default = "."),
    mtx |> shift("down",  default = ".") |> shift("right", default = ".")
  ) |> 
    map(\(x) x == value) |> 
    reduce(`+`)
}

Count how many rolls in the input have less than 4 neighboring rolls:

sum(input == '@' & count_nbrs(input, '@') < 4)

Part 2

Iteratively remove rolls until no more removals are possible:

cur <- input

repeat {
  prv <- cur
  cur[cur == '@' & count_nbrs(cur, '@') < 4] <- "."
  if (all(prv == cur))
    break
}

fnl <- cur

Count how many rolls were removed:

sum(input == "@") - sum(fnl == "@")