Lesson 10 of 17 · Week 05 · Thu Oct 22

Functions Explained

Why write functions?

If you've copied and pasted a block of code twice, it's time for a function. A function gives a name to a piece of work, so you can reuse it, test it once, and fix it in one place.

mm_to_inches <- function(mm) {
  inches <- mm / 25.4
  return(inches)
}
mm_to_inches(254)          # 10
mm_to_inches(c(10, 100))   # vectorized for free, because / is

Anatomy: a name, the keyword function, the arguments in parentheses, and the body in braces. The function returns the value given to return(), or, if there's no return(), the value of the last line. Variables created inside a function (like inches) exist only while it runs.

Adapted from Software Carpentry's “Functions Explained” episode (CC BY 4.0, The Carpentries). Text, examples and exercises have been rewritten for this course.

Building functions from functions

inches_to_feet <- function(inches) inches / 12
mm_to_feet <- function(mm) {
  inches_to_feet(mm_to_inches(mm))
}

Small functions that do one thing are easy to check. Combining them builds bigger tools you can trust.

Defensive programming

A function should fail loudly and early when it gets input it can't handle, rather than quietly returning nonsense. stopifnot() stops with an error if any condition is FALSE. stop() lets you write your own message:

mm_to_inches <- function(mm) {
  stopifnot(is.numeric(mm))
  if (any(mm < 0)) {
    stop("lengths can't be negative")
  }
  mm / 25.4
}
mm_to_inches("12")    # Error: is.numeric(mm) is not TRUE

Functions on data frames, with defaults

country_rows <- function(dat, ctry, yr = NULL) {
  keep <- dat$country == ctry
  if (!is.null(yr)) {
    keep <- keep & dat$year %in% yr
  }
  dat[keep, ]
}
country_rows(gm, "Kenya")                 # all years: yr uses its default, NULL
country_rows(gm, "Kenya", yr = 2007)

An argument with a default (yr = NULL) can be left out. NULL is a common way to say “not given”, and is.null() checks for it. Arguments can be given by position or by name; names make calls easier to read.

Text in, text out

paste("tidy", "data")                             # "tidy data"   (sep = " " by default)
paste0("tidy", "data")                            # "tidydata"
paste(c("tidy", "data"), collapse = " ")          # squash a vector into one string

Exercises

Each exercise gives you up to 10 checked attempts. Use Run as many times as you like to experiment first, then click Check answer when you're ready. Your exercises use numbers/subsets unique to you, so answers won't match a classmate's exactly.