Query 'nycflights13'-Like Air Travel Data for Given Years and Airports.
anyflights
The anyflights
package supplies a set of functions to generate air travel data (and data packages!) similar to nycflights13
. With a user-defined year and airport, the anyflights
function will grab data on:
flights
: all flights that departed a given airport in a given year and monthweather
: hourly meterological data for a given airport in a given year and monthairports
: airport names, FAA codes, and locationsairlines
: translation between two letter carrier (airline) codes and namesplanes
: construction information about each plane found inflights
The package also exports a set of functions to query each of the above datasets individually. Each of the functions are named get_*
followed with the names of the datasets (shown above in code
), and require a varying combination of the station
(character vector of FAA LID airport codes), year
(as an integer), and month
(integer vector) arguments. Optionally, the user can also supply a dir
argument (filepath to a folder) to save the resulting datasets.
These functions require a strong connection to run; expect several minutes of runtime, especially for the anyflights
and get_flights
functions.
The FAA usually posts all of the data for a given year during February or March of the following year. All airports with an FAA LID code are supported, to my knowledge. Data queries earlier than 2013 are significantly less thoroughly tested than those 2013 and on.
Installation
anyflights
is now on CRAN! Install with the following code:
install.packages("anyflights")
To install the developmental version, use the following code:
#install.packages("remotes")
remotes::install_github("simonpcouch/anyflights")
Basic Usage
To grab a named list of 5 dataframes giving nycflights13
-like air travel data for the Portland International Airport during June 2019, you could call:
pdxflights19 <- anyflights("PDX", 2019, 6)
To grab the original nycflights13
datasets, try:
nycflights13_ <- anyflights(c("JFK", "LGA", "EWR"), 2013)
The package also provides an as_flights_package
function to create a data-only package from the data outputted by anyflights
. To create the package from the data, just supply a name:
pdxflights19 %>%
as_flights_package("pdxflights")
and check out the data-only package in the pdxflights
directory! 🐛
If you have used anyflights
to create a data package, please link me to the repository and I might feature it here! Please do not submit the outputted package to CRAN or similar repositories as original packages.
Contributing
Contributions to anyflights
, whether in the form of bug fixes, issue reports, new code or documentation improvements are encouraged and welcome. Please see CONTRIBUTING.md if you’re interested in contributing!