Kara Woo
@karawoo.com
software engineer | #rstats | PDX | she/her
Ironic that this screenshot also has a bug. Fixed version: github.com/karawoo/adve...
adventofcode/2025/day04.R at main · karawoo/adventofcode
Advent of Code solutions. Contribute to karawoo/adventofcode development by creating an account on GitHub.
github.com
December 5, 2025 at 7:27 PM
Ironic that this screenshot also has a bug. Fixed version: github.com/karawoo/adve...
whoops, I did some cleanup before pushing my solution and made a mistake. `neighbors[!oob(neighbors[, 1], dat) && !oob(neighbors[, 2], dat), ]` should be `neighbors[!oob(neighbors[, 1], dat) & !oob(neighbors[, 2], dat), ]`.
Updated version is here: github.com/karawoo/adve...
Updated version is here: github.com/karawoo/adve...
adventofcode/2025/day04.R at main · karawoo/adventofcode
Advent of Code solutions. Contribute to karawoo/adventofcode development by creating an account on GitHub.
github.com
December 5, 2025 at 7:25 PM
whoops, I did some cleanup before pushing my solution and made a mistake. `neighbors[!oob(neighbors[, 1], dat) && !oob(neighbors[, 2], dat), ]` should be `neighbors[!oob(neighbors[, 1], dat) & !oob(neighbors[, 2], dat), ]`.
Updated version is here: github.com/karawoo/adve...
Updated version is here: github.com/karawoo/adve...
I think day 5 has been my favorite so far. Also got my first `vector memory limit reached` error of the year. #adventofcode #rstats
![Screenshot of the following code:
dat <- readLines("input05.txt")
ranges <- dat[1:(which(dat == "") - 1)] |>
strsplit("-") |>
lapply(as.numeric)
ingredients <- as.numeric(dat[(which(dat == "") + 1):length(dat)])
## part 1
count <- 0
for (i in ingredients) {
for (j in ranges) {
if (i >= j[1] && i <= j[2]) {
count <- count + 1
break
}
}
}
count
## part 2
sorted_ranges <- ranges[order(sapply(ranges, `[[`, 1))]
highest <- 0
count <- 0
for (i in sorted_ranges) {
if (i[2] <= highest) {
next
} else if (i[1] > highest) {
count <- count + (i[2] - i[1]) + 1
} else {
count <- count + (i[2] - (highest))
}
if (i[2] > highest) {
highest <- i[2]
}
}
count](https://cdn.bsky.app/img/feed_thumbnail/plain/did:plc:o4tcgqoslmipb3j62vjhrq73/bafkreid2r4332uq6xihy3qxgcbnwtutiwlk5vrmkfocnpc2462vuf37hmu@jpeg)
December 5, 2025 at 7:18 PM
I think day 5 has been my favorite so far. Also got my first `vector memory limit reached` error of the year. #adventofcode #rstats
Day 4. Had an inscrutable bug in part 2 for the longest time; eventually scrapped it and rewrote it in almost the same way, and this time it worked 🤷♀️ #adventofcode #rstats
![Screenshot of the following code:
library("tidyverse")
file <- "input04.txt"
dat <- read_fwf(
file,
col_positions = fwf_widths(rep(1, nchar(readLines(file, n = 1))))
) |>
as.matrix()
oob <- function(x, dat) {
x == 0 | x > max(dim(dat)) # assumes square matrix
}
get_neighbors <- function(row, col, dat) {
x <- row
y <- col
possible <- list(
tl = c(row - 1, col - 1),
t = c(row - 1, col),
tr = c(row - 1, col + 1),
l = c(row, col - 1),
r = c(row, col + 1),
bl = c(row + 1, col - 1),
b = c(row + 1, col),
br = c(row + 1, col + 1)
)
neighbors <- do.call(rbind, possible)
colnames(neighbors) <- c("row", "col")
neighbors[!oob(neighbors[1, ], dat) && !oob(neighbors[2, ], dat), ]
}
is_movable <- function(neighbors, dat) {
sum(dat[as.matrix(neighbors)] == "@") < 4
}
rolls <- which(dat == "@")
rolls_xy <- arrayInd(rolls, dim(dat))
neighbors <- apply(rolls_xy, 1, function(x) get_neighbors(x[1], x[2], dat))
## part 1
neighbors |>
map_lgl(is_movable, dat) |>
sum()
## part 2
clear <- function(rolls, neighbors, dat) {
moved <- neighbors |>
map_lgl(is_movable, dat)
dat[rolls[moved]] <- "."
if (any(moved)) {
clear(rolls[!moved], neighbors[!moved], dat)
} else {
return(dat)
}
}
sum(dat == "@") - sum(clear(rolls, neighbors, dat) == "@")](https://cdn.bsky.app/img/feed_thumbnail/plain/did:plc:o4tcgqoslmipb3j62vjhrq73/bafkreidou7gwksiajgzsmjoxaucz4p3edsnabtzbh742kkdvzqlgol4yim@jpeg)
December 5, 2025 at 7:09 PM
Day 4. Had an inscrutable bug in part 2 for the longest time; eventually scrapped it and rewrote it in almost the same way, and this time it worked 🤷♀️ #adventofcode #rstats
yeah, I went through a number of solutions that were close to right but not quite
December 5, 2025 at 7:07 PM
yeah, I went through a number of solutions that were close to right but not quite
Quite pleased with my day 3. #adventofcode #rstats
![Screenshot of the following code:
library("purrr")
options(scipen = 9999)
dat <- readLines("input03.txt") |>
strsplit("")
joltage <- function(bank, n) {
if (n > 0) {
candidates <- head(bank, length(bank) - n + 1)
i <- which.max(candidates)
paste0(bank[i], joltage(bank[(i + 1):length(bank)], n = n - 1))
}
}
## part 1
map_dbl(dat, ~ as.numeric(joltage(.x, n = 2))) |>
sum()
## part 2
map_dbl(dat, ~ as.numeric(joltage(.x, n = 12))) |>
sum()](https://cdn.bsky.app/img/feed_thumbnail/plain/did:plc:o4tcgqoslmipb3j62vjhrq73/bafkreicienemuhrysulljy777fs7iscbrc2kh3fdhxwq7kza2km2bspmpm@jpeg)
December 5, 2025 at 7:06 PM
Quite pleased with my day 3. #adventofcode #rstats
Day 2: not efficient but it works. #adventofcode #rstats
![Screenshot of the following code:
library("stringr")
library("purrr")
dat <- readLines("input02.txt")
# discard any that have more than length/2 unique digits
candidate <- function(val) {
val_str <- str_split(val, "")[[1]]
length(unique(val_str)) <= (length(val_str) / 2)
}
to_check <- str_split(dat, ",")[[1]] |>
str_split("-") |>
map(~ seq(.x[1], .x[2])) |>
unlist() |>
keep(candidate)
## part 1
invalid1 <- function(x) {
x <- str_split(x, "")[[1]]
if (identical(head(x, length(x) / 2), tail(x, length(x) / 2))) {
as.numeric(paste0(x, collapse = ""))
} else {
0
}
}
map_dbl(to_check, invalid1) |>
sum()
## part 2
## divide up string into pieces of length n
split_group_size <- function(string, n) {
x <- str_split(string, "")[[1]]
groups <- ceiling(seq_along(x) / n)
split(x, groups) |>
map(paste0, collapse = "")
}
invalid2 <- function(x) {
for (i in seq_len(nchar(x)) / 2) {
pieces <- split_group_size(x, i)
if (length(unique(unlist(pieces))) == 1) {
return(sum(as.numeric(x)))
}
}
0
}
## it's slowww
map_dbl(to_check, invalid2) |>
sum()](https://cdn.bsky.app/img/feed_thumbnail/plain/did:plc:o4tcgqoslmipb3j62vjhrq73/bafkreib26fxkc2w5gh6ovxftt57onr67ozbvnlinvhghgwi5k5bjn6q2ea@jpeg)
December 5, 2025 at 6:54 PM
Day 2: not efficient but it works. #adventofcode #rstats
my gf and i make a lot of playlists for each other and recently listened to them all back to back on a cross country road trip, it was the best
November 24, 2025 at 4:37 AM
my gf and i make a lot of playlists for each other and recently listened to them all back to back on a cross country road trip, it was the best
I remembered those things which are indeed terrifying. I’d forgotten about the SONGS though, omg, each one scarier than the last.
November 17, 2025 at 7:27 PM
I remembered those things which are indeed terrifying. I’d forgotten about the SONGS though, omg, each one scarier than the last.
rewatched it wondering if i should show it to my kid and uh i think i’ll wait a while on that
November 17, 2025 at 7:17 PM
rewatched it wondering if i should show it to my kid and uh i think i’ll wait a while on that
If a chat window, 1️⃣. If a form, 2️⃣.
October 2, 2025 at 3:13 PM
If a chat window, 1️⃣. If a form, 2️⃣.
Reposted by Kara Woo
Not to mention these same politicians are seeing constant images of actual war-ravaged places and it has not compelled many of them to do shit so what's the joke exactly
September 29, 2025 at 3:32 PM
Not to mention these same politicians are seeing constant images of actual war-ravaged places and it has not compelled many of them to do shit so what's the joke exactly
Reposted by Kara Woo
Like yes we must clown this administration as it is one of the few languages they understand but let's not get too self-congratulatory when we have so many people with huge needs and a lot of leadership who still fundamentally rely on criminalization and real estate development as a solution
September 29, 2025 at 3:25 PM
Like yes we must clown this administration as it is one of the few languages they understand but let's not get too self-congratulatory when we have so many people with huge needs and a lot of leadership who still fundamentally rely on criminalization and real estate development as a solution