finished creating the flights dataset

.gitignore

@@ -6,3 +6,4 @@
 .quarto
 docs
 inst/doc
+kaggle.json

data-raw/flights.R

@@ -0,0 +1,70 @@
+library(jsonlite)
+library(readr)
+library(tidyverse)
+library(httr)
+library(airports)
+
+#Requires a kaggle account and API key
+
+credentials = fromJSON("kaggle.json")
+username = credentials$username
+key = credentials$key
+
+auth <- authenticate(user = username, password = key)
+
+url <- paste0("https://www.kaggle.com/api/v1/datasets/download/",
+              "patrickzel/flight-delay-and-cancellation-dataset-2019-2023"
+              , "/flights_sample_3m.csv")
+
+response <- GET(url, auth)
+
+if (response$status_code == 200) {
+  # Save the content to a temporary file
+  zip_file <- tempfile(fileext = ".zip")
+  writeBin(content(response, "raw"), zip_file)
+
+  files_in_zip <- unzip(zip_file, list = TRUE)
+
+  csv_file_name <- files_in_zip$Name[1]
+
+  # Read the CSV directly from the ZIP file
+  flights_data_raw <- read_csv(unz(zip_file, csv_file_name))
+
+} else {
+  stop("Failed to download the dataset. Status code: ", response$status_code)
+}
+
+top_100_airports <- flights_data_raw |>
+  group_by(ORIGIN) |>
+  summarise(
+    total_flights = n()
+  ) |>
+  ungroup() |>
+  arrange(desc(total_flights)) |>
+  head(100)
+
+sample_airports <- sample(top_100_airports$ORIGIN, 10)
+
+flights_by_monthyear <- flights_data_raw |>
+  filter(ORIGIN %in% sample_airports) |>
+  mutate(
+    month = month(FL_DATE),
+    year = year(FL_DATE)
+  ) |>
+  group_by(ORIGIN, month, year) |>
+  summarise(
+    total_flights = n()
+  )
+
+flights_by_month <- flights_by_monthyear |>
+  group_by(ORIGIN, month) |>
+  summarise(
+    max = max(total_flights),
+    min = min(total_flights),
+  )
+
+flights <- left_join(flights_by_month, usairports,
+  by = c("ORIGIN" = "location_id")) |>
+  select(ORIGIN, month, max, min, facility_name, arp_latitude, arp_longitude)
+
+usethis::use_data(flights, overwrite = TRUE)

data-raw/stations.R

@@ -1,5 +1,5 @@
 ## code to prepare `stations` dataset goes here
-library(tidyverse)
+library(dplyr)
 library(rnoaa)
 
 station_data <- ghcnd_stations() |>