A Comically Brief Intro to devtools

R packages are simply directories with a particular structure.

Bare bones: R/, DESCRIPTION

Recommended: man/, NAMESPACE

Optional: README, NEWS, CITATIONS, tests/, demo/, src/, etc.

You can pretty easily create an initial package yourself, but devtools makes developing a package much easier

devtools (Hadley): Base for creating and building R packages.

roxygen2 (Yihui): Parses comments to create man pages (.Rd)

testthat (Hadley): Unit testing for package functions

Two key links for using all of these things:

https://github.com/hadley/devtools

https://github.com/hadley/devtools/wiki

1. Install devtools, roxygen2, testthat

2. create(PATH)

3. Put your R files into R/

4. document(PATH)

#' Square a number
#'
#' \code{square_me} is a function that squares a number.
#'
#' @param x A number
#' @return x^2
#' @export
#' @examples
#' x <- 5
#' square_me(x)

square_me <- function(x) {
    x^2
}

http://www.nytimes.com/interactive/2012/11/02/us/politics/paths-to-the-white-house.html

http://bost.ocks.org/mike/miserables/

• Reusable visualizations
• No javascript/HTML knowledge required
• One line of R code in the familiar way
• Easy sharing and embedding of your figure

http://healthvis.org