16  Function documentation

Second edition

You are reading the work-in-progress second edition of R Packages. This chapter is undergoing heavy restructuring and may be confusing or incomplete.

16.1 Introduction

In this chapter, you’ll learn about function documentation, as accessed by ? or help(). Base R provides a standard way of documenting a package where each function is documnted in a topic, a .Rd file in the man/ directory. These files use a custom syntax, loosely based on LaTeX, that are rendered to HTML, plain text, or pdf, as needed, for viewing.

In this book, we are not going to use .Rd files directly. Instead, we’ll use the roxygen2 package to generate them from specially formatted comments are written above each function. There are a few advantages to using roxygen2:

  • Code and documentation are intermingled so that when you modify your code, it’s easy to remember to also update your documentation.

  • You can use markdown, rather having to learn a new text formatting syntax.

  • There’s a lot of .Rd boilerplate that’s automated away.

  • roxygen2 provides a number of tools for sharing content across documentation topics and even between topics and vignettes.

In this chapter we’ll focus on documenting functions, but the same ideas apply to documenting datasets, classes and generics, and packages. You can learn more about those important topics in vignette("rd-other", package = "roxygen2").

16.2 roxygen2 basics

To get started, we’ll work through the basic roxygen2 workflow and discuss the overall structure of roxygen2 comments which are organised into blocks and tags.

16.2.1 The documentation workflow

You begin the documentation workflow by adding roxygen comments above your function. roxygen comments are comments that start with ', e.g.:

#' Add together two numbers
#' 
#' @param x A number.
#' @param y A number.
#' @return A numeric vector.
#' @examples
#' add(1, 1)
#' add(10, 1)
add <- function(x, y) {
  x + y
}

You then press Ctrl/Cmd + Shift + D or type devtools::document() to run roxygen2::roxygenise(). This example generates a man/add.Rd file that looks like this:

% Generated by roxygen2: do not edit by hand
% Please edit documentation in R/add.R
\name{add}
\alias{add}
\title{Add together two numbers}
\usage{
add(x, y)
}
\arguments{
\item{x}{A number.}

\item{y}{A number.}
}
\value{
The sum of \code{x} and \code{y}.
}
\description{
Add together two numbers
}
\examples{
add(1, 1)
add(10, 1)
}

If you’ve used LaTeX before, this should look vaguely familiar since the .Rd format is loosely based on LaTeX. If you’re interested you can read more about it in R Extensions, but generally you’ll never need to look at it except to check it in to git.

How does this file correspond to the documentation you see in R? When you run ?add, help("add"), or example("add"), R looks for an .Rd file containing \alias{add}. It then parses the file, converts it into HTML, and displays it. Here’s what the result looks like in RStudio:

This process looks inside installed packages, so to see the development documentation, devtools overrides the usual help functions so they know look in the source package. To activate the override, you need to run devtools::load_all() once. So if the development documentation doesn’t appear, you may need to load your package first.

To summarize, there are four steps in the basic roxygen2 workflow:

  1. Add roxygen2 comments to your .R files.

  2. Press Ctrl/Cmd + Shift + D or type devtools::document() to convert roxygen2 comments to .Rd files.

  3. Preview documentation with ?.

  4. Rinse and repeat until the documentation looks the way you want.

16.2.2 roxygen2 comments, blocks, and tags

Now that you understand the basic workflow, lets talk a little more about the syntax. roxygen2 comments start with #' and all the roxygen2 comments preceding a function are collectively call a block. Blocks are broken up by tags, which look like @tagName tagValue, and the content of a tag extends from the end of the tag name to the start of the next tag1. A block can contain text before the first tag which is called the introduction. By default, each block will generate a single documentation topic, i.e. a .Rd file2 in the man/ directory .

Throughout this chapter I’m going to show you roxygen2 comments from real tidyverse packages, focusing on stringr, since the functions there tend to be fairly straightforward leading to documentation that’s easy to excerpt. Here’s a simple first example: the documentation for stringr::str_unique().

#' Remove duplicated strings
#'
#' `str_unique()` removes duplicated values, with optional control over
#' how duplication is measured.
#'
#' @param string A character vector to return unique entries.
#' @param ... Other options used to control matching behavior between
#'   duplicate strings. Passed on to [stringi::stri_opts_collator()].
#' @returns A character vector.
#' @seealso [unique()], [stringi::stri_unique()] which this function wraps.
#' @examples
#' str_unique(c("a", "b", "c", "b", "a"))
#'
#' # Use ... to pass additional arguments to stri_unique()
#' str_unique(c("motley", "mötley", "pinguino", "pingüino"))
#' str_unique(c("motley", "mötley", "pinguino", "pingüino"), strength = 1)
#' @export
str_unique <- function(string, ...) {
  ...
}

Here the introduction includes the title (“Remove duplicated strings”) and a basic description of what the function does. It’s followed by five tags: two @params, one @returns, one @seealso, one @examples, and one @export.

Note that I’ve wrapped each line of the block to be 80 characters wide (matching the wrapping of my code) and I’ve indented the second and subsequent lines of the long @param tag so it’s easier to scan. You can get more roxygen2 style advice in the tidyverse style guide.

The following sections will work through the most important tags. We’ll start with the introduction which provides the title, description, and details. Then we’ll cover the inputs (the function arguments), outputs (the return value), and examples. We’ll next discuss links and cross-references, then finish off with techniques for sharing documentation between topics.

16.3 Title, description, details

The introduction provides a title, description, and, optionally, details, for the function:

  • The title is taken from the first sentence. It should be written in sentence case, not end in a full stop, and be followed by a blank line. The title is shown in various function indexes and is what the user will usually see when browsing multiple functions.

  • The description is taken from the next paragraph. It’s show at the top of documentation and should briefly describe the most important features of the function.

  • Additional details are anything after the description. Details are optional, but can be any length so are useful if want to dig deep into some important aspect of the function.

The following sections describe each component in more detail, and then discuss a few useful related tags.

16.3.1 Title

When figuring out what to use as a title, I think it’s most important about the reference index. When someone is skimming the index, how will they know which functions will solve their current problem? This requires thinking about what your functions have in common (which doesn’t need to be repeated in every title) and what is unique to that function (which should be highlighted in the title).

As an example, take the titles of some of the key dplyr functions3:

  • mutate(): Create, modify, and delete columns.
  • summarise(): Summarize each group to fewer rows.
  • filter(): Subset rows using column values.
  • select(): Subset columns using their names and types.
  • arrange(): Arrange rows by column values.

Here we’ve tried to succinctly describe what the function does, making sure to describe whether it affects rows, columns, or groups. We do our best to use synonyms, instead of repeating the function name, to hopefully give folks another chance to understand the intent of the function.

At the time we wrote this chapter, I don’t think the function titles for stringr were that successful. But they provide a useful negative case study:

  • str_detect(): Detect the presence or absence of a pattern in a string.
  • str_extract(): Extract matching patterns from a string.
  • str_locate(): Locate the position of patterns in a string.
  • str_match(): Extract matched groups from a string.

There’s a lot of repetition (“pattern”, “from a string”) and the verb used for the function name is repeated in the title, so if you don’t understand the function already, the title seems unlikely to help much. Hopefully we’ll have improved those titles by the time you read this.

16.3.2 Description

The purpose of the description is to summarize the goal of the function, usually in under a paragraph. This can be challenging for simple functions, because it might feel like you’re repeating the title of the function. But it’s okay for the description to be a little duplicative; it’s often useful for the reader to see the same thing expressed in two different ways. It’s a little extra work keeping it all up to date, but the extra effort is often worth it.

#' Detect the presence/absence of a pattern
#'
#' `str_detect()` returns a logical vector `TRUE` if `pattern` is found within
#' each element of `string` or a `FALSE` if not. It's equivalent
#' `grepl(pattern, string)`.

If you want to use multiple paragraphs or a bulleted list, you can use the explicit @description tag4. Here’s an example from stringr::str_like(), which mimics the LIKE operator from SQL:

#' Detect the a pattern in the same way as `SQL`'s `LIKE` operator.
#'
#' @description
#' `str_like()` follows the conventions of the SQL `LIKE` operator:
#'
#' * Must match the entire string.
#' * `_` matches a single character (like `.`).
#' * `%` matches any number of characters (like `.*`).
#' * `\%` and `\_` match literal `%` and `_`.
#' * The match is case insensitive by default.

Finally, it’s often particularly hard to write a good description if you’ve just written the function because the purpose often seems very obvious. Do your best, and then come back in a couple of months when you’ve forgotten exactly what the function does, and re-write the description to jog your memory.

16.3.3 Details

The “details” are just any additional details or explanation that you think your function needs. Most functions don’t need details, but some functions need a lot. If you have a lot of information to convey, I recommend using markdown headings to break the documentation up into sections. Here’s a example from dplyr::mutate(). We’ve elided some of the details to keep this example short, but you should still get a sense of how we used headings to break up the content in to skimmable chunks:

#' Create, modify, and delete columns
#'
#' `mutate()` adds new variables and preserves existing ones;
#' `transmute()` adds new variables and drops existing ones.
#' New variables overwrite existing variables of the same name.
#' Variables can be removed by setting their value to `NULL`.
#'
#' # Useful mutate functions
#'
#' * [`+`], [`-`], [log()], etc., for their usual mathematical meanings
#'
#' ...
#'
#' # Grouped tibbles
#'
#' Because mutating expressions are computed within groups, they may
#' yield different results on grouped tibbles. This will be the case
#' as soon as an aggregating, lagging, or ranking function is
#' involved. Compare this ungrouped mutate:
#'
#' ...

Note the use of [log()] etc — this is special roxygen2 markdown syntax that will generate a link to the documentation for the log() function. You can link to a topic in another package is [pkg::function()], and to non-function documentation with [topic] and [pkg::topic].

Also note that even though these headings come immediately after the description in the roxygen block they are shown much later (after the function arguments and return value) in the rendered documentation.

In older code, you might also see the use of @section title: which was used to create sections before roxygen2 fully supported RMarkdown. If you’ve used these in the past, you can now turn them into markdown headings.

16.4 Arguments

For most functions, the bulk of your work will go towards documenting how each argument affects the output of the function. For this purpose, you’ll use @param (short for parameter, a synonym of argument) followed by the argument name and a description of its action.

The most important job of the description is to provide a succinct summary of the allowed inputs and what the parameter does. For example, here’s str_detect():

#' @param string Input vector. Either a character vector, or something
#'  coercible to one.

And here are three of the arguments to str_flatten():

#' @param collapse String to insert between each piece. Defaults to `""`.
#' @param last Optional string use in place of final separator.
#' @param na.rm Remove missing values? If `FALSE` (the default), the result 
#'   will be `NA` if any element of `string` is `NA`.

Note that @param collapse and @param na.rm describe their default arguments. This is good practice because the function usage (which shows the default values) and the argument description are often quite far apart. The primary downside is that introducing this duplication means that you’ll need to update the docs if you change the default value; we believe this small amount of extra work is worth it to make the life of the user easier.

If an argument has a fixed set of possible parameters, you should list them. If they’re simple, you can just list them in a sentence, like in str_trim():

#' @param side Side on which to remove whitespace: `"left"`, `"right"`, or
#'   `"both"` (the default).

If they need more explanation, you might use a bulleted list, as in str_wrap():

#' @param whitespace_only A boolean.
#'   * `TRUE` (the default): wrapping will only occur at whitespace.
#'   * `FALSE`: can break on any non-word character (e.g. `/`, `-`).

The documentation for most arguments will relatively short, often one or two sentences. But you should take as much space as you need, and you’ll see some examples of multi-paragraph argument docs shortly.

16.4.1 Multiple arguments

If the behavior of multiple arguments is tightly coupled, you can document them together by separating the names with commas (with no spaces). For example, x and y are interchangeable in str_equal(), so they’re documented together:

#' @param x,y A pair of character vectors.

In str_sub(), start and end define the range of characters to replace. But instead of supplying both, you can use just start if you pass in a two-column matrix. So it makes sense to document them together:

#' @param start,end Two integer vectors. `start` gives the position
#'   of the first character (defaults to first), `end` gives the position
#'   of the last (defaults to last character). Alternatively, pass a two-column
#'   matrix to `start`.
#'
#'   Negative values count backwards from the last character.

In str_wrap(), indent and exdent define the indentation for the first line and all subsequent lines respectively:

#' @param indent,exdent A non-negative integer giving the indent for the
#'   first line (`indent`) and all subsequent lines (`exdent`).

16.4.2 Inheriting arguments

If your package contains many closely related functions, it’s common for them to have arguments that share the same name and meaning. It would be both annoying and error prone to copy and paste the same @param documentation to every function so roxygen2 provides @inheritParams which allows you to inherit argument documentation from another package.

stringr uses @inheritParams extensively because most functions have string and pattern arguments. So str_detect() documents them in detail:

#' @param string Input vector. Either a character vector, or something
#'  coercible to one.
#' @param pattern Pattern to look for.
#'
#'   The default interpretation is a regular expression, as described
#'   `vignette("regular-expressions")`. Control options with [regex()].
#'
#'   Match a fixed string (i.e. by comparing only bytes), using
#'   [fixed()]. This is fast, but approximate. Generally,
#'   for matching human text, you'll want [coll()] which
#'   respects character matching rules for the specified locale.
#'
#'   Match character, word, line and sentence boundaries with
#'   [boundary()]. An empty pattern, "", is equivalent to
#'   `boundary("character")`.

Then the other stringr functions use @inheritParams str_detect to get a detailed documentation for string and pattern without having to duplicate that text.

@inheritParams only inherits docs for arguments that aren’t already documented, so you can document some arguments and inherit others. str_match() uses this to inherit its standard string argument but document its unusual pattern argument:

#' @inheritParams str_detect
#' @param pattern Unlike other stringr functions, `str_match()` only supports
#'   regular expressions, as described `vignette("regular-expressions")`. 
#'   The pattern should contain at least one capturing group.

You can inherit documentation from a function in another package by using the standard :: notation, i.e. @inheritParams anotherpackage::function. This does introduce one small annoyance: now the documentation for your package is no longer self-contained and the version of anotherpackage can affect the generated docs. Beware of spurious diffs caused by contributors with different installed versions.

16.5 Return value

As important as a function’s inputs is it output. Documenting the outputs is the job of the @returns5 tag. Here the goal of the docs is not to describe exactly how the values are computed (which is the job of the description and details), but to roughly describe the overall “shape” of the output, i.e. what sort of object it is, and its dimensions (if that makes sense). For example, if your function returns a vector you might describe its type and length, or if your function returns a data frame you might describe the names and types of the columns and the expected number of rows.

The return documentation for functions in stringr are straightforward because almost all functions return some type of vector with the same length as one of the inputs. For example, here’s str_like():

#' @returns A logical vector the same length as `string`.

A more complicated case is the joint documentation for str_locate() and str_locate_all()6. str_locate() returns an integer matrix, and str_locate_all() returns a list of matrices, so the text needs to describe what determines the rows and columns.

#' @return `str_locate()` returns an integer matrix with two columns and
#'   one row for each element of `string`. The first column, `start`,
#'   gives the position at the start of the match, and second column, `end`,
#'   gives the position of the end.
#'
#'   `str_locate_all()` returns a list of integer matrices as above, but
#'   the matrices have one row for each match in the corresponding element
#'   in `string`.

In other cases it can be easier to figure out what to describe by thinking about the set of functions and how they differ. For example, most dplyr functions return data frames, so just saying @return A data frame is not very useful. Instead we sat down and thought about exactly what makes each function different. We decided it makes sense to describe each function in terms of how it affects the rows, the columns, the groups, and the attributes. For example, here’s dplyr::filter():

#' @returns
#' An object of the same type as `.data`. The output has the following properties:
#'
#' * Rows are a subset of the input, but appear in the same order.
#' * Columns are not modified.
#' * The number of groups may be reduced (if `.preserve` is not `TRUE`).
#' * Data frame attributes are preserved.

@returns is also a good place to describe any important warnings or errors that the user might see here. For example readr::read_csv() mentions what happens if there are any parsing problems:

#' @returns A [tibble()]. If there are parsing problems, a warning will alert you.
#'   You can retrieve the full details by calling [problems()] on your dataset.
Submitting to CRAN

For your initial CRAN submission, all functions must document their return value. This is not required for subsequent submission, but it’s still good practice. There’s currently no way to check that you’ve documented the return value of every function (we’re working on it) which is why you’ll notice some tidyverse functions lack output documentation.

16.6 Examples

Describing what a function does great, but showing how it works is even better. That’s the role of the @examples tag, which uses executable R code to demonstrate what a function can do. Unlike other parts of the documentation where we’ve focused mainly on what you should write, here we’ll briefly give some content advice and then focus mainly on the mechanics. The mechanics of examples are complex because they must never error and they’re run in four different situations:

  • Interactively using the example() function.
  • During R CMD check on your computer, or another computer you control (e.g. GitHub action).
  • During R CMD check run by CRAN.
  • When building your pkgdown website.

After discussing what to put in your examples, we’ll talk about keeping your examples self-contained, how to display errors if needed, handling dependencies, running examples conditionally, and

16.6.1 Contents

Use examples to first show the basic operation of the function, then to highlight any particularly important properties. For example, str_detect() starts by showing a few simple variations and then highlights a property you might easily miss from reading the docs: as well as passing a vector of strings and one pattern, you can also pass one string and vector of patterns.

#' @examples
#' fruit <- c("apple", "banana", "pear", "pineapple")
#' str_detect(fruit, "a")
#' str_detect(fruit, "^a")
#' str_detect(fruit, "a$")
#' 
#' # Also vectorised over pattern
#' str_detect("aecfg", letters)

Try to stay focused on the most important features without getting into the weeds of every last edge case: if you make the examples too long, it becomes hard for the user to find the key application that they’re looking for. If you find yourself writing very long examples, it may be a sign that you should write a vignette instead.

There aren’t any formal ways to break up your examples into sections but you can use sectioning comments that use many === or --- to create a visual breakdown. Here’s an example from tidyr::chop():

#' @examples
#' # Chop ==============================================================
#' df <- tibble(x = c(1, 1, 1, 2, 2, 3), y = 1:6, z = 6:1)
#' # Note that we get one row of output for each unique combination of
#' # non-chopped variables
#' df %>% chop(c(y, z))
#' # cf nest
#' df %>% nest(data = c(y, z))
#'
#' # Unchop ============================================================
#' df <- tibble(x = 1:4, y = list(integer(), 1L, 1:2, 1:3))
#' df %>% unchop(y)
#' df %>% unchop(y, keep_empty = TRUE)
#' 
#' #' # Incompatible types -------------------------------------------------
#' # If the list-col contains types that can not be natively
#' df <- tibble(x = 1:2, y = list("1", 1:3))
#' try(df %>% unchop(y))

Strive to keep the examples focused on the specific function that you’re documenting. If you can make the point with a familiar built-in dataset, like iris, do so. If you find yourself needing to do a bunch of setup to create a dataset or object to use in the example, it may be a sign that you need to create a package dataset. See Chapter 8 for details.

16.6.2 Pack it in; pack it out

As much as possible, keep your examples self-contained. For example, this means:

  • If you modify options(), reset them at the end of the example.
  • If you create a file, create it somewhere in tempdir() and make sure to delete it at the end of the example.
  • Don’t change the working directory.
  • Don’t write to the clipboard.
  • Avoid accessing websites in examples. If the website is down, your example will fail and hence R CMD check will error.

Due to the way that examples are run during R CMD check there’s unfortunately no way to use familiar tools like withr to enforce these constraints. Instead you’ll need to do it by hand.

These constraints are often in tension with good documentation if you’re trying to document a function that somehow changes the state of the world. So if you’re finding it really hard to follow these rules, this might be another sign to switch to a vignette.

Submitting to CRAN

Many of these constraints are also mentioned in the CRAN repository policy, which you must adhere to when submitting to CRAN. Use find in page to locate “malicious or anti-social” to see the details.

Additionally, you want your examples to send the user on a short walk, not a long hike. Examples need to execute relatively quickly so users can quickly see the results, it doesn’t take ages to build your website, automated checks happen quickly, and it doesn’t take up computing resources when submitting to CRAN.

Submitting to CRAN

All examples must run in under 10 minutes.

16.6.3 Errors

What can you do if you want to include code that causes an error for the purposes of teaching. There are two basic options:

  • You can wrap the code in try() so that the error is shown, but doesn’t stop execution of the error. For example, dplyr::bind_cols() uses try() to show you what happens if you attempt to column-bind to data frames with different numbers of rows:

    #' @examples
    #' ...
    #' # Row sizes must be compatible when column-binding
    #' try(bind_cols(tibble(x = 1:3), tibble(y = 1:2)))
  • You can wrap the code \dontrun{}7 so it is never run by example(). The example above would look like this if you used \dontrun{} instead of try().

    #' # Row sizes must be compatible when column-binding
    #' \dontrun{
    #' bind_cols(tibble(x = 1:3), tibble(y = 1:2)))
    #' }

We generally recommend using try() so that the reader can see an example of the error in action.

16.6.4 Dependencies and conditional execution

An additional source of errors in examples are external dependencies: you can only use packages in examples that your package depends on (i.e. that appear in Imports or Suggests). Furthermore, example code is run in the user’s environment, not the package environment, so you’ll have to either explicitly attach the package with library() or refer to each function with ::. For example, dbplyr is a dplyr extension package so all of its examples start with library(dplyr):

#' @examples
#' library(dplyr)
#' df <- data.frame(x = 1, y = 2)
#'
#' df_sqlite <- tbl_lazy(df, con = simulate_sqlite())
#' df_sqlite %>% summarise(x = sd(x, na.rm = TRUE)) %>% show_query()

In the past, we recommended only using code from suggested packages inside an if block that used if (requireNamespace("suggested_package", quietly = TRUE)). Today, we believe that you no longer need that layer of protection because:

  • We expect that suggested packages are installed when running R CMD check8.
  • The cost of wrapping code in {} is high: you can no longer see intermediate results. The cost of a package not being installed is low: users can usually recognize the package not loaded error and can resolve it themselves.

In other cases, your example code may depend on something other than a package. For example, if your examples talk to a web API, you probably only want to run them if the user is authenticated, and want to avoid such code being run on CRAN. In this case you can use @examplesIf instead of @examples. The code in an @examplesIf block will only be executed if some condition is TRUE:

#' @examplesIf some_condition()
#' some_other_function()
#' some_more_functions()

For example, googledrive uses @examplesIf in almost every function because the examples can only work if you have an authenticated and active connection to Google Drive as judged by googledrive::drive_has_token(). For example, here’s googledrive::drive_publish():

#' @examplesIf drive_has_token()
#' # Create a file to publish
#' file <- drive_example_remote("chicken_sheet") %>%
#'   drive_cp()
#'
#' # Publish file
#' file <- drive_publish(file)
#' file$published
Submitting to CRAN

For initial CRAN submission of your package, all functions must contain some runnable examples (i.e. there must be examples and they must not all be wrapped in \dontrun{}).

16.6.5 Intermixing examples and text

An alternative to examples is to use RMarkdown’s code blocks, either ```R if you just want to show some code or ```{r} if you want the code to be run. These can be effective techniques but there are downsides to each:

  • The code in ```R blocks is never run; this means it’s easy to accidentally introduce syntax errors or forget to update it when your package changes.
  • The code in ```{r} blocks is run every time you document the package. This has the nice advantage of including the output in the documentation (unlike examples), but the code can’t take very long to run or your iterative documentation workflow will become quite painful.

16.7 Re-using documentation

roxygen2 provides a number of features that allow you to reuse documentation across topics. They are documented in vignettes("reuse", package = "roxygen2") so here we’ll focus on the three most important:

  • Documenting multiple functions in one topic.
  • Inheriting documentation from another topic.
  • Use child documents to share prose between topics, or to share between documentation topics and vignettes.

16.7.1 Multiple functions in one topic

By default, each function gets its own documentation topic, but if two functions are very closely connected you can combine the documentation for multiple functions into a single topic. For example, take str_length() and str_width() which provide two different ways of computing the size of a string. As you can see from the description, both functions are documented together, because this makes it easy to see how they differ:

#' The length/width of a string
#'
#' @description
#' `str_length()` returns the number of codepoints in a string. These are
#' the individual elements (which are often, but not always letters) that
#' can be extracted with [str_sub()].
#'
#' `str_width()` returns how much space the string will occupy when printed
#' in a fixed width font (i.e. when printed in the console).
#'
#' ...
str_length <- function(string) {
  ...
}

To merge the two topics, str_width() uses @rdname str_length to add its documentation to an existing topic:

#' @rdname str_length
str_width <- function(string) {
  ...
}

This technique is best used for functions that have not just similar arguments, but also similar return value and related examples, as discussed next.

16.7.2 Inheriting documentation

In other cases, functions in a make might share many related behaviors, but aren’t closely enough connected that you want to document them together. We’ve discussed @inheritParams above, but there are three variations that allow you to inherit other things:

  • @inherit source_function will inherit all supported components from source_function.

  • @inheritSection source_function Section title will inherit the single section with title “Section title” from source_function().

  • @inheritDotParams automatically generates parameter documentation for ... for the common case where you pass ... on to another function.

See https://roxygen2.r-lib.org/articles/reuse.html#inheriting-documentation for more details.

16.7.3 Child documents

Finally, you can use the same .Rmd or .md document in the documentation, README.Rmd, and vignettes by using RMarkdown child documents. The syntax looks like this:

#' ```{r child = "man/rmd/filename.Rmd"}
#' ```

This is not a feature we use terribly commonly in the tidyverse, but once place we do use it is in dplyr, because a number of functions use the same syntax as select() and we want to provide all the info in one place:

#' # Overview of selection features
#'
#' ```{r, child = "man/rmd/overview.Rmd"}
#' ```

Then man/rmd/overview.Rmd contains the repeated markdown:

Tidyverse selections implement a dialect of R where operators make}
it easy to select variables:

- `:` for selecting a range of consecutive variables.
- `!` for taking the complement of a set of variables.
- `&` and `|` for selecting the intersection or the union of two
  sets of variables.
- `c()` for combining selections.

...

If the Rmd file contains roxygen (Markdown-style) links to other help topics, then some care is needed. See https://roxygen2.r-lib.org/dev/articles/reuse.html#child-documents for details.


  1. Or the end of the block, if it’s the last tag.↩︎

  2. The name of the file is automatically derived from the object you’re documenting.↩︎

  3. Like all the examples, these might have changed a bit since we wrote this book, because we’re constantly striving to do better. You might compare what’s in the book to what we now use, and consider if you think if it’s an improvement.↩︎

  4. You can also use explicit @title and @details tags if needed, but we don’t generally recommend them because they add extra noise to the docs without enabling any extra functionality.↩︎

  5. For historical reasons, you can also use @return, but I think you should use @returns because it reads a little nicer.↩︎

  6. We’ll come back how to document multiple functions in one topic in Section 16.7.1.↩︎

  7. You used to be able to use \donttest{} for a similar purpose, but we no longer recommended it because CRAN sets a special flag that causes it to be executed.↩︎

  8. This is certainly true for CRAN and is true in most other automated checking scenarios.↩︎