Alexander Enge 2022-12-01
- Day 1: Calorie Counting :pizza:
- Day 2: Rock Paper Scissors :scissors:
- Day 3: Rucksack Reorganization 🎒
- Day 4: Camp Cleanup 🧹
- Day 5: Supply Stacks :building_construction:
- Day 6: Tuning Trouble 📻
- Day 7: No Space Left On Device 📁
- Day 8: Treetop Tree House 🌳
Hi! 👋
This repository contains my solutions for the 2022 edition of Advent of Code.
From the Advent of Code website:
Advent of Code is an Advent calendar of small programming puzzles for a variety of skill sets and skill levels that can be solved in any programming language you like. People use them as interview prep, company training, university coursework, practice problems, a speed contest, or to challenge each other.
I’ll be using a mix of Python, Base R, and tidyverse-style R.
max_elf = 0
this_elf = 0
with open('data/day1.txt') as f:
for line in f:
if this_elf > max_elf:
max_elf = this_elf
if line == '\n':
this_elf = 0
else:
this_elf += int(line.strip())
print(max_elf)64929
lines <- readLines("data/day1.txt", warn = FALSE)
elf_indices <- cumsum(lines == "")
split(lines, elf_indices) |>
lapply(as.numeric) |>
lapply(sum, na.rm = TRUE) |>
unlist() |>
max()[1] 64929
max_elves = [0, 0, 0]
this_elf = 0
with open('data/day1.txt') as f:
for line in f:
max_elves.sort()
if this_elf > max_elves[0]:
max_elves[0] = this_elf
if line == '\n':
this_elf = 0
else:
this_elf += int(line.strip())
print(sum(max_elves))193697
lines <- readLines("data/day1.txt", warn = FALSE)
elf_indices <- cumsum(lines == "")
split(lines, elf_indices) |>
lapply(as.numeric) |>
lapply(sum, na.rm = TRUE) |>
unlist() |>
sort() |>
tail(3) |>
sum()[1] 193697
library(tidyverse)
read_table("data/day2.txt", col_names = c("other", "me")) %>%
unite(col = "round", other, me, remove = FALSE) %>%
mutate(
game_score = case_when(
round %in% c("A_Y", "B_Z", "C_X") ~ 6,
round %in% c("A_X", "B_Y", "C_Z") ~ 3,
round %in% c("A_Z", "B_X", "C_Y") ~ 0
),
play_score = recode(me, "X" = 1, "Y" = 2, "Z" = 3),
score = game_score + play_score
) %>%
pull(score) %>%
sum()[1] 14297
read_table("data/day2.txt", col_names = c("other", "outcome")) %>%
unite(col = "round", other, outcome, remove = FALSE) %>%
mutate(
me = case_when(
round %in% c("A_X", "B_Z", "C_Y") ~ "Z",
round %in% c("A_Y", "B_X", "C_Z") ~ "X",
round %in% c("A_Z", "B_Y", "C_X") ~ "Y"
),
game_score = recode(outcome, "X" = 0, "Y" = 3, "Z" = 6),
play_score = recode(me, "X" = 1, "Y" = 2, "Z" = 3),
score = game_score + play_score
) %>%
pull(score) %>%
sum()[1] 10498
from string import ascii_lowercase, ascii_uppercase
with open('data/day3.txt') as f:
lines = [line.strip() for line in f.readlines()]
halves = [(line[:len(line) // 2], line[len(line) // 2:]) for line in lines]
items = [set(half_1).intersection(half_2).pop() for half_1, half_2 in halves]
letters = ascii_lowercase + ascii_uppercase
priorities = [letters.index(item) + 1 for item in items]
print(sum(priorities))7795
group_size = 3
groups = [lines[ix:ix + group_size] for ix in range(0, len(lines), group_size)]
badges = [set.intersection(*map(set, group)).pop() for group in groups]
priorities = [letters.index(badge) + 1 for badge in badges]
print(sum(priorities))2703
def is_contained(min_1, max_1, min_2, max_2):
return (min_1 >= min_2 and max_1 <= max_2 or
min_2 >= min_1 and max_2 <= max_1)
contained = 0
with open('data/day4.txt') as f:
for line in f:
elf_1, elf_2 = line.strip().split(',')
min_1, max_1 = [int(section) for section in elf_1.split('-')]
min_2, max_2 = [int(section) for section in elf_2.split('-')]
contained += is_contained(min_1, max_1, min_2, max_2)
print(contained)475
read_csv("data/day4.txt", col_names = c("elf_1", "elf_2")) %>%
transmute(across(.fns = str_split, pattern = "-")) %>%
unnest_wider(c(elf_1, elf_2), names_sep = "_", transform = as.integer) %>%
mutate(
contained_1 = elf_1_1 >= elf_2_1 & elf_1_2 <= elf_2_2,
contained_2 = elf_2_1 >= elf_1_1 & elf_2_2 <= elf_1_2,
contained = contained_1 | contained_2
) %>%
pull(contained) %>%
sum()[1] 475
def is_overlap(min_1, max_1, min_2, max_2):
return (max_1 >= min_2 and max_2 >= min_1)
overlaps = 0
with open('data/day4.txt') as f:
for line in f:
elf_1, elf_2 = line.strip().split(',')
min_1, max_1 = [int(section) for section in elf_1.split('-')]
min_2, max_2 = [int(section) for section in elf_2.split('-')]
overlaps += is_overlap(min_1, max_1, min_2, max_2)
print(overlaps)825
read_csv("data/day4.txt", col_names = c("elf_1", "elf_2")) %>%
transmute(across(.fns = str_split, pattern = "-")) %>%
unnest_wider(c(elf_1, elf_2), names_sep = "_", transform = as.integer) %>%
mutate(overlap = elf_1_2 >= elf_2_1 & elf_2_2 >= elf_1_1) %>%
pull(overlap) %>%
sum()[1] 825
transpose_list <- function(l) as.list(as.data.frame(t(as.data.frame(l))))
drop_empty <- function(x) x[x != " "]
input <- read.csv("data/day5.txt", header = FALSE)
input_1 <- subset(input, grepl("\\[", V1))$V1
stack_ixs <- seq(2, max(nchar(input_1)), by = 4)
stacks <- lapply(input_1, substring, first = stack_ixs, last = stack_ixs) |>
rev() |>
transpose_list() |>
lapply(drop_empty)
input_2 <- subset(input, grepl("move", V1))$V1
moves <- strsplit(input_2, split = " ")
for (move in moves) {
from = as.integer(move[4])
to = as.integer(move[6])
n = as.integer(move[2])
stacks[[to]] <- c(stacks[[to]], rev(tail(stacks[[from]], n)))
stacks[[from]] <- head(stacks[[from]], -n)
}
lapply(stacks, tail, n = 1) |>
paste(collapse = "")[1] "QNNTGTPFN"
stacks <- lapply(input_1, substring, first = stack_ixs, last = stack_ixs) |>
rev() |>
transpose_list() |>
lapply(drop_empty) # Same as before
for (move in moves) {
from = as.integer(move[4])
to = as.integer(move[6])
n = as.integer(move[2])
stacks[[to]] <- c(stacks[[to]], tail(stacks[[from]], n)) # Don't reverse
stacks[[from]] <- head(stacks[[from]], -n)
}
lapply(stacks, tail, n = 1) |>
paste(collapse = "")[1] "GGNPJBTTR"
def find_marker(input, n_letters):
for ix in range(len(input)):
letters = input[ix:ix + n_letters]
if len(letters) == len(set(letters)):
return ix + n_letters
with open('data/day6.txt') as f:
input = f.readline()
find_marker(input, n_letters=4)1802
find_marker(input, n_letters=14)3551
input <- readLines("data/day7.txt", warn = FALSE)
dirs = list(`/` = list(".dir_size" = 0))
for (cmd in input) {
if (cmd == "$ cd /") path <- '/'
else if (cmd == "$ cd ..") path <- head(path, -1)
else if (startsWith(cmd, "$ cd")) {
dir_name <- tail(strsplit(cmd, " ")[[1]], 1)
path <- c(path, dir_name)
}
else if (cmd == "$ ls") dirs[[path]] <- list(".dir_size" = 0)
else if (!startsWith(cmd, "dir")) { # Only consider files, not directories
file_size <- strsplit(cmd, " ")[[1]] |> head(1) |> as.numeric()
for (ix in seq_along(path)) { # Add file size to all parent directories
dir_size_path <- c(head(path, ix), ".dir_size")
dirs[[dir_size_path]] <- dirs[[dir_size_path]] + file_size
}
}
else next
}
str(dirs, list.len = 3) # Sanity checkList of 1
$ /:List of 5
..$ .dir_size: num 44376732
..$ lhrfs :List of 1
.. ..$ .dir_size: num 240152
..$ nwh :List of 2
.. ..$ .dir_size: num 834387
.. ..$ pmdj :List of 1
.. .. ..$ .dir_size: num 514336
.. [list output truncated]
dir_sizes <- unlist(dirs)
sum(dir_sizes[dir_sizes < 1e5])[1] 1243729
space_left <- 7e7 - dirs[[c("/", ".dir_size")]]
space_needed <- 3e7 - space_left
min(dir_sizes[dir_sizes > space_needed])[1] 4443914
import numpy as np
def is_visible(arr_1d):
"""Tests for each element in arr if it's visible from the front or back."""
return [all(arr_1d[:ix] < elem) or all(arr_1d[ix + 1:] < elem)
for ix, elem in enumerate(arr_1d)]
input = np.genfromtxt('data/day8.txt', delimiter=1)
vis = [np.apply_along_axis(is_visible, ax, input) for ax in range(input.ndim)]
np.any(vis, axis=0).sum()1854
def score_elem(arr, elem):
"""Computes the score (other visible trees) in one direction."""
if arr.size == 0:
score = 0 # Edge tree
elif all(elem > arr):
score = len(arr) # Largest tree, sees all other trees in the row
else:
score = np.where(arr >= elem)[0][0] + 1 # Sees until next equal tree
return score
input = np.genfromtxt('data/day8.txt', dtype=int, delimiter=1)
scores = np.zeros((4,) + input.shape, dtype=int)
for row_ix, row in enumerate(input):
for col_ix, elem in enumerate(row):
left = np.flip(row[:col_ix])
right = row[col_ix + 1:]
up = np.flip(input[:row_ix, col_ix])
down = input[row_ix + 1:, col_ix]
for dir_ix, arr in enumerate([left, right, up, down]):
scores[dir_ix, row_ix, col_ix] = score_elem(arr, elem)
scores = np.prod(scores, axis=0)
print(scores.max())527340
That’s how far I’ve got for the 2022 edition of Advent of Code. :facepalm: Wish me better luck next time!