Package 'usethis'

Description

These helpers produce the markdown text you need in your README to include badges that report information, such as the CRAN version or test coverage, and link out to relevant external resources. To add badges automatically ensure your badge block starts with a line containing only ⁠<!-- badges: start -->⁠ and ends with a line containing only ⁠<!-- badges: end -->⁠.

Usage

use_badge(badge_name, href, src)

use_cran_badge()

use_bioc_badge()

use_lifecycle_badge(stage)

use_binder_badge(ref = git_default_branch(), urlpath = NULL)

use_posit_cloud_badge(url)

Arguments

Details

• use_badge(): a general helper used in all badge functions

• use_bioc_badge(): badge indicates BioConductor build status

• use_cran_badge(): badge indicates what version of your package is available on CRAN, powered by https://www.r-pkg.org

• use_lifecycle_badge(): badge declares the developmental stage of a package according to https://lifecycle.r-lib.org/articles/stages.html.

• use_binder_badge(): badge indicates that your repository can be launched in an executable environment on https://mybinder.org/

• use_posit_cloud_badge(): badge indicates that your repository can be launched in a Posit Cloud project

• use_rscloud_badge(): : Use use_posit_cloud_badge() instead.

See Also

Functions that configure continuous integration, such as use_github_actions(), also create badges.

Examples

## Not run: 
use_cran_badge()
use_lifecycle_badge("stable")

## End(Not run)

Description

These functions take you to various web pages associated with a project (often, an R package) and return the target URL(s) invisibly. To form these URLs we consult:

• Git remotes configured for the active project that appear to be hosted on a GitHub deployment

• DESCRIPTION file for the active project or the specified package. The DESCRIPTION file is sought first in the local package library and then on CRAN.

• Fixed templates:

  • Circle CI: ⁠https://circleci.com/gh/{OWNER}/{PACKAGE}⁠

  • CRAN landing page: ⁠https://cran.r-project.org/package={PACKAGE}⁠

  • GitHub mirror of a CRAN package: ⁠https://github.com/cran/{PACKAGE}⁠ Templated URLs aren't checked for existence, so there is no guarantee there will be content at the destination.

Usage

browse_package(package = NULL)

browse_project()

browse_github(package = NULL)

browse_github_issues(package = NULL, number = NULL)

browse_github_pulls(package = NULL, number = NULL)

browse_github_actions(package = NULL)

browse_circleci(package = NULL)

browse_cran(package = NULL)

Arguments

Details

• browse_package(): Assembles a list of URLs and lets user choose one to visit in a web browser. In a non-interactive session, returns all discovered URLs.

• browse_project(): Thin wrapper around browse_package() that always targets the active usethis project.

• browse_github(): Visits a GitHub repository associated with the project. In the case of a fork, you might be asked to specify if you're interested in the source repo or your fork.

• browse_github_issues(): Visits the GitHub Issues index or one specific issue.

• browse_github_pulls(): Visits the GitHub Pull Request index or one specific pull request.

• browse_circleci(): Visits the project's page on Circle CI.

• browse_cran(): Visits the package on CRAN, via the canonical URL.

Examples

# works on the active project
# browse_project()

browse_package("httr")
browse_github("gh")
browse_github_issues("fs")
browse_github_issues("fs", 1)
browse_github_pulls("curl")
browse_github_pulls("curl", 183)
browse_cran("MASS")

Description

Creates a new local project and Git repository from a repo on GitHub, by either cloning or fork-and-cloning. In the fork-and-clone case, create_from_github() also does additional remote and branch setup, leaving you in the perfect position to make a pull request with pr_init(), one of several functions for working with pull requests.

create_from_github() works best when your GitHub credentials are discoverable. See below for more about authentication.

Usage

create_from_github(
  repo_spec,
  destdir = NULL,
  fork = NA,
  rstudio = NULL,
  open = rlang::is_interactive(),
  protocol = git_protocol(),
  host = NULL
)

Arguments

Git/GitHub Authentication

Many usethis functions, including those documented here, potentially interact with GitHub in two different ways:

• Via the GitHub REST API. Examples: create a repo, a fork, or a pull request.

• As a conventional Git remote. Examples: clone, fetch, or push.

Therefore two types of auth can happen and your credentials must be discoverable. Which credentials do we mean?

• A GitHub personal access token (PAT) must be discoverable by the gh package, which is used for GitHub operations via the REST API. See gh_token_help() for more about getting and configuring a PAT.

• If you use the HTTPS protocol for Git remotes, your PAT is also used for Git operations, such as ⁠git push⁠. Usethis uses the gert package for this, so the PAT must be discoverable by gert. Generally gert and gh will discover and use the same PAT. This ability to "kill two birds with one stone" is why HTTPS + PAT is our recommended auth strategy for those new to Git and GitHub and PRs.

• If you use SSH remotes, your SSH keys must also be discoverable, in addition to your PAT. The public key must be added to your GitHub account.

Git/GitHub credential management is covered in a dedicated article: Managing Git(Hub) Credentials

See Also

• use_github() to go the opposite direction, i.e. create a GitHub repo from your local repo

• git_protocol() for background on protocol (HTTPS vs SSH)

• use_course() to download a snapshot of all files in a GitHub repo, without the need for any local or remote Git operations

Examples

## Not run: 
create_from_github("r-lib/usethis")

# repo_spec can be a URL
create_from_github("https://github.com/r-lib/usethis")

# a URL repo_spec also specifies the host (e.g. GitHub Enterprise instance)
create_from_github("https://github.acme.com/OWNER/REPO")

## End(Not run)

Description

These functions create an R project:

• create_package() creates an R package

• create_project() creates a non-package project, i.e. a data analysis project

Both functions can be called on an existing project; you will be asked before any existing files are changed.

Usage

create_package(
  path,
  fields = list(),
  rstudio = rstudioapi::isAvailable(),
  roxygen = TRUE,
  check_name = TRUE,
  open = rlang::is_interactive()
)

create_project(
  path,
  rstudio = rstudioapi::isAvailable(),
  open = rlang::is_interactive()
)

Arguments

Value

Path to the newly created project or package, invisibly.

See Also

create_tidy_package() is a convenience function that extends create_package() by immediately applying as many of the tidyverse development conventions as possible.

Description

• edit_r_profile() opens .Rprofile

• edit_r_environ() opens .Renviron

• edit_r_makevars() opens .R/Makevars

• edit_git_config() opens .gitconfig or .git/config

• edit_git_ignore() opens global (user-level) gitignore file and ensures its path is declared in your global Git config.

• edit_pkgdown_config opens the pkgdown YAML configuration file for the current Project.

• edit_rstudio_snippets() opens RStudio's snippet config for the given type.

• edit_rstudio_prefs() opens RStudio's preference file.

Usage

edit_r_profile(scope = c("user", "project"))

edit_r_environ(scope = c("user", "project"))

edit_r_buildignore()

edit_r_makevars(scope = c("user", "project"))

edit_rstudio_snippets(
  type = c("r", "markdown", "c_cpp", "css", "html", "java", "javascript", "python",
    "sql", "stan", "tex", "yaml")
)

edit_rstudio_prefs()

edit_git_config(scope = c("user", "project"))

edit_git_ignore(scope = c("user", "project"))

edit_pkgdown_config()

Arguments

Details

The ⁠edit_r_*()⁠ functions consult R's notion of user's home directory. The ⁠edit_git_*()⁠ functions (and usethis in general) inherit home directory behaviour from the fs package, which differs from R itself on Windows. The fs default is more conventional in terms of the location of user-level Git config files. See fs::path_home() for more details.

Files created by edit_rstudio_snippets() will mask, not supplement, the built-in default snippets. If you like the built-in snippets, copy them and include with your custom snippets.

Value

Path to the file, invisibly.

Description

Git operations that address a remote use a so-called "transport protocol". usethis supports HTTPS and SSH. The protocol dictates the Git URL format used when usethis needs to configure the first GitHub remote for a repo:

• protocol = "https" implies ⁠https://github.com/<OWNER>/<REPO>.git⁠

• protocol = "ssh" implies ⁠[email protected]:<OWNER>/<REPO>.git⁠

Two helper functions are available:

• git_protocol() reveals the protocol "in force". As of usethis v2.0.0, this defaults to "https". You can change this for the duration of the R session with use_git_protocol(). Change the default for all R sessions with code like this in your .Rprofile (easily editable via edit_r_profile()):

  options(usethis.protocol = "ssh")
  

• use_git_protocol() sets the Git protocol for the current R session

This protocol only affects the Git URL for newly configured remotes. All existing Git remote URLs are always respected, whether HTTPS or SSH.

Usage

git_protocol()

use_git_protocol(protocol)

Arguments

Value

The protocol, either "https" or "ssh"

Examples

## Not run: 
git_protocol()

use_git_protocol("ssh")
git_protocol()

use_git_protocol("https")
git_protocol()

## End(Not run)

Description

Get a situation report on your current Git/GitHub status. Useful for diagnosing problems. The default is to report all values; provide values for tool or scope to be more specific.

Usage

git_sitrep(tool = c("git", "github"), scope = c("user", "project"))

Arguments

Examples

## Not run: 
# report all
git_sitrep()

# report git for current user
git_sitrep("git", "user")

## End(Not run)

Description

Adds .Rproj.user, .Rhistory, .Rdata, .httr-oauth, .DS_Store, and .quarto to your global (a.k.a. user-level) .gitignore. This is good practice as it decreases the chance that you will accidentally leak credentials to GitHub. git_vaccinate() also tries to detect and fix the situation where you have a global gitignore file, but it's missing from your global Git config.

Usage

git_vaccinate()

Description

The ⁠git_default_branch*()⁠ functions put some structure around the somewhat fuzzy (but definitely real) concept of the default branch. In particular, they support new conventions around the Git default branch name, globally or in a specific project / Git repository.

Usage

git_default_branch()

git_default_branch_configure(name = "main")

git_default_branch_rediscover(current_local_default = NULL)

git_default_branch_rename(from = NULL, to = "main")

Arguments

Value

Name of the default branch.

Background on the default branch

Technically, Git has no official concept of the default branch. But in reality, almost all Git repos have an effective default branch. If there's only one branch, this is it! It is the branch that most bug fixes and features get merged in to. It is the branch you see when you first visit a repo on a site such as GitHub. On a Git remote, it is the branch that HEAD points to.

Historically, master has been the most common name for the default branch, but main is an increasingly popular choice.

git_default_branch_configure()

This configures init.defaultBranch at the global (a.k.a user) level. This setting determines the name of the branch that gets created when you make the first commit in a new Git repo. init.defaultBranch only affects the local Git repos you create in the future.

git_default_branch()

This figures out the default branch of the current Git repo, integrating information from the local repo and, if applicable, the upstream or origin remote. If there is a local vs. remote mismatch, git_default_branch() throws an error with advice to call git_default_branch_rediscover() to repair the situation.

For a remote repo, the default branch is the branch that HEAD points to.

For the local repo, if there is only one branch, that must be the default! Otherwise we try to identify the relevant local branch by looking for specific branch names, in this order:

• whatever the default branch of upstream or origin is, if applicable

• main

• master

• the value of the Git option init.defaultBranch, with the usual deal where a local value, if present, takes precedence over a global (a.k.a. user-level) value

git_default_branch_rediscover()

This consults an external authority – specifically, the remote source repo on GitHub – to learn the default branch of the current project / repo. If that doesn't match the apparent local default branch (for example, the project switched from master to main), we do the corresponding branch renaming in your local repo and, if relevant, in your fork.

See https://happygitwithr.com/common-remote-setups.html for more about GitHub remote configurations and, e.g., what we mean by the source repo. This function works for the configurations "ours", "fork", and "theirs".

git_default_branch_rename()

Note: this only works for a repo that you effectively own. In terms of GitHub, you must own the source repo personally or, if organization-owned, you must have admin permission on the source repo.

This renames the default branch in the source repo on GitHub and then calls git_default_branch_rediscover(), to make any necessary changes in the local repo and, if relevant, in your personal fork.

See https://happygitwithr.com/common-remote-setups.html for more about GitHub remote configurations and, e.g., what we mean by the source repo. This function works for the configurations "ours", "fork", and "no_github".

Regarding "no_github": Of course, this function does what you expect for a local repo with no GitHub remotes, but that is not the primary use case.

Examples

## Not run: 
git_default_branch()

## End(Not run)
## Not run: 
git_default_branch_configure()

## End(Not run)
## Not run: 
git_default_branch_rediscover()

# you can always explicitly specify the local branch that's been playing the
# role of the default
git_default_branch_rediscover("unconventional_default_branch_name")

## End(Not run)
## Not run: 
git_default_branch_rename()

# you can always explicitly specify one or both branch names
git_default_branch_rename(from = "this", to = "that")

## End(Not run)

Description

A personal access token (PAT) is needed for certain tasks usethis does via the GitHub API, such as creating a repository, a fork, or a pull request. If you use HTTPS remotes, your PAT is also used when interacting with GitHub as a conventional Git remote. These functions help you get and manage your PAT:

• gh_token_help() guides you through token troubleshooting and setup.

• create_github_token() opens a browser window to the GitHub form to generate a PAT, with suggested scopes pre-selected. It also offers advice on storing your PAT.

• gitcreds::gitcreds_set() helps you register your PAT with the Git credential manager used by your operating system. Later, other packages, such as usethis, gert, and gh can automatically retrieve that PAT and use it to work with GitHub on your behalf.

Usually, the first time the PAT is retrieved in an R session, it is cached in an environment variable, for easier reuse for the duration of that R session. After initial acquisition and storage, all of this should happen automatically in the background. GitHub is encouraging the use of PATs that expire after, e.g., 30 days, so prepare yourself to re-generate and re-store your PAT periodically.

Git/GitHub credential management is covered in a dedicated article: Managing Git(Hub) Credentials

Usage

create_github_token(
  scopes = c("repo", "user", "gist", "workflow"),
  description = "DESCRIBE THE TOKEN'S USE CASE",
  host = NULL
)

gh_token_help(host = NULL)

Arguments

Details

create_github_token() has previously gone by some other names: browse_github_token() and browse_github_pat().

Value

Nothing

See Also

gh::gh_whoami() for information on an existing token and gitcreds::gitcreds_set() and gitcreds::gitcreds_get() for a secure way to store and retrieve your PAT.

Examples

## Not run: 
create_github_token()

## End(Not run)
## Not run: 
gh_token_help()

## End(Not run)

Description

The ⁠issue_*⁠ family of functions allows you to perform common operations on GitHub issues from within R. They're designed to help you efficiently deal with large numbers of issues, particularly motivated by the challenges faced by the tidyverse team.

• issue_close_community() closes an issue, because it's not a bug report or feature request, and points the author towards Posit Community as a better place to discuss usage (https://forum.posit.co).

• issue_reprex_needed() labels the issue with the "reprex" label and gives the author some advice about what is needed.

Usage

issue_close_community(number, reprex = FALSE)

issue_reprex_needed(number)

Arguments

Saved replies

Unlike GitHub's "saved replies", these functions can:

• Be shared between people

• Perform other actions, like labelling, or closing

• Have additional arguments

• Include randomness (like friendly gifs)

Examples

## Not run: 
issue_close_community(12, reprex = TRUE)

issue_reprex_needed(241)

## End(Not run)

Description

Adds the necessary infrastructure to declare your package as licensed with one of these popular open source licenses:

Permissive:

• MIT: simple and permissive.

• Apache 2.0: MIT + provides patent protection.

Copyleft:

• GPL v2: requires sharing of improvements.

• GPL v3: requires sharing of improvements.

• AGPL v3: requires sharing of improvements.

• LGPL v2.1: requires sharing of improvements.

• LGPL v3: requires sharing of improvements.

Creative commons licenses appropriate for data packages:

• CC0: dedicated to public domain.

• CC-BY: Free to share and adapt, must give appropriate credit.

See https://choosealicense.com for more details and other options.

Alternatively, for code that you don't want to share with others, use_proprietary_license() makes it clear that all rights are reserved, and the code is not open source.

Usage

use_mit_license(copyright_holder = NULL)

use_gpl_license(version = 3, include_future = TRUE)

use_agpl_license(version = 3, include_future = TRUE)

use_lgpl_license(version = 3, include_future = TRUE)

use_apache_license(version = 2, include_future = TRUE)

use_cc0_license()

use_ccby_license()

use_proprietary_license(copyright_holder)

Arguments

Details

CRAN does not permit you to include copies of standard licenses in your package, so these functions save the license as LICENSE.md and add it to .Rbuildignore.

See Also

For more details, refer to the the license chapter in R Packages.

Description

Activates a project in usethis, R session, and (if relevant) RStudio senses. If you are in RStudio, this will open a new RStudio session. If not, it will change the working directory and active project.

Usage

proj_activate(path)

Arguments

Value

Single logical value indicating if current session is modified.

Description

proj_sitrep() reports

• current working directory

• the active usethis project

• the active RStudio Project

Call this function if things seem weird and you're not sure what's wrong or how to fix it. Usually, all three of these should coincide (or be unset) and proj_sitrep() provides suggested commands for getting back to this happy state.

Usage

proj_sitrep()

Value

A named list, with S3 class sitrep (for printing purposes), reporting current working directory, active usethis project, and active RStudio Project

See Also

Other project functions: proj_utils

Examples

proj_sitrep()

Description

Most ⁠use_*()⁠ functions act on the active project. If it is unset, usethis uses rprojroot to find the project root of the current working directory. It establishes the project root by looking for a .here file, an RStudio Project, a package DESCRIPTION, Git infrastructure, a remake.yml file, or a .projectile file. It then stores the active project for use for the remainder of the session.

In general, end user scripts should not contain direct calls to ⁠usethis::proj_*()⁠ utility functions. They are internal functions that are exported for occasional interactive use or use in packages that extend usethis. End user code should call functions in rprojroot or its simpler companion, here, to programmatically detect a project and build paths within it.

If you are puzzled why a path (usually the current working directory) does not appear to be inside project, it can be helpful to call here::dr_here() to get much more verbose feedback.

Usage

proj_get()

proj_set(path = ".", force = FALSE)

proj_path(..., ext = "")

with_project(
  path = ".",
  code,
  force = FALSE,
  setwd = TRUE,
  quiet = getOption("usethis.quiet", default = FALSE)
)

local_project(
  path = ".",
  force = FALSE,
  setwd = TRUE,
  quiet = getOption("usethis.quiet", default = FALSE),
  .local_envir = parent.frame()
)

Arguments

Functions

• proj_get(): Retrieves the active project and, if necessary, attempts to set it in the first place.

• proj_set(): Sets the active project.

• proj_path(): Builds paths within the active project returned by proj_get(). Thin wrapper around fs::path().

• with_project(): Runs code with a temporary active project and, optionally, working directory. It is an example of the ⁠with_*()⁠ functions in withr.

• local_project(): Sets an active project and, optionally, working directory until the current execution environment goes out of scope, e.g. the end of the current function or test. It is an example of the ⁠local_*()⁠ functions in withr.

See Also

Other project functions: proj_sitrep()

Examples

## Not run: 
## see the active project
proj_get()

## manually set the active project
proj_set("path/to/target/project")

## build a path within the active project (both produce same result)
proj_path("R/foo.R")
proj_path("R", "foo", ext = "R")

## build a path within SOME OTHER project
with_project("path/to/some/other/project", proj_path("blah.R"))

## convince yourself that with_project() temporarily changes the project
with_project("path/to/some/other/project", print(proj_sitrep()))

## End(Not run)

Description

The ⁠pr_*⁠ family of functions is designed to make working with GitHub pull requests (PRs) as painless as possible for both contributors and package maintainers.

To use the ⁠pr_*⁠ functions, your project must be a Git repo and have one of these GitHub remote configurations:

• "ours": You can push to the GitHub remote configured as origin and it's not a fork.

• "fork": You can push to the GitHub remote configured as origin, it's a fork, and its parent is configured as upstream. origin points to your personal copy and upstream points to the source repo.

"Ours" and "fork" are two of several GitHub remote configurations examined in Common remote setups in Happy Git and GitHub for the useR.

The Pull Request Helpers article walks through the process of making a pull request with the ⁠pr_*⁠ functions.

The ⁠pr_*⁠ functions also use your Git/GitHub credentials to carry out various remote operations; see below for more about auth. The ⁠pr_*⁠ functions also proactively check for agreement re: the default branch in your local repo and the source repo. See git_default_branch() for more.

Usage

pr_init(branch)

pr_resume(branch = NULL)

pr_fetch(number = NULL, target = c("source", "primary"))

pr_push()

pr_pull()

pr_merge_main()

pr_view(number = NULL, target = c("source", "primary"))

pr_pause()

pr_finish(number = NULL, target = c("source", "primary"))

pr_forget()

Arguments

Git/GitHub Authentication

Many usethis functions, including those documented here, potentially interact with GitHub in two different ways:

• Via the GitHub REST API. Examples: create a repo, a fork, or a pull request.

• As a conventional Git remote. Examples: clone, fetch, or push.

Therefore two types of auth can happen and your credentials must be discoverable. Which credentials do we mean?

• A GitHub personal access token (PAT) must be discoverable by the gh package, which is used for GitHub operations via the REST API. See gh_token_help() for more about getting and configuring a PAT.

• If you use the HTTPS protocol for Git remotes, your PAT is also used for Git operations, such as ⁠git push⁠. Usethis uses the gert package for this, so the PAT must be discoverable by gert. Generally gert and gh will discover and use the same PAT. This ability to "kill two birds with one stone" is why HTTPS + PAT is our recommended auth strategy for those new to Git and GitHub and PRs.

• If you use SSH remotes, your SSH keys must also be discoverable, in addition to your PAT. The public key must be added to your GitHub account.

Git/GitHub credential management is covered in a dedicated article: Managing Git(Hub) Credentials

For contributors

To contribute to a package, first use create_from_github("OWNER/REPO"). This forks the source repository and checks out a local copy.

Next use pr_init() to create a branch for your PR. It is best practice to never make commits to the default branch branch of a fork (usually named main or master), because you do not own it. A pull request should always come from a feature branch. It will be much easier to pull upstream changes from the fork parent if you only allow yourself to work in feature branches. It is also much easier for a maintainer to explore and extend your PR if you create a feature branch.

Work locally, in your branch, making changes to files, and committing your work. Once you're ready to create the PR, run pr_push() to push your local branch to GitHub, and open a webpage that lets you initiate the PR (or draft PR).

To learn more about the process of making a pull request, read the Pull Request Helpers vignette.

If you are lucky, your PR will be perfect, and the maintainer will accept it. You can then run pr_finish() to delete your PR branch. In most cases, however, the maintainer will ask you to make some changes. Make the changes, then run pr_push() to update your PR.

It's also possible that the maintainer will contribute some code to your PR: to get those changes back onto your computer, run pr_pull(). It can also happen that other changes have occurred in the package since you first created your PR. You might need to merge the default branch (usually named main or master) into your PR branch. Do that by running pr_merge_main(): this makes sure that your PR is compatible with the primary repo's main line of development. Both pr_pull() and pr_merge_main() can result in merge conflicts, so be prepared to resolve before continuing.

For maintainers

To download a PR locally so that you can experiment with it, run pr_fetch() and select the PR or, if you already know its number, call ⁠pr_fetch(<pr_number>)⁠. If you make changes, run pr_push() to push them back to GitHub. After you have merged the PR, run pr_finish() to delete the local branch and remove the remote associated with the contributor's fork.

Overview of all the functions

• pr_init(): As a contributor, start work on a new PR by ensuring that your local repo is up-to-date, then creating and checking out a new branch. Nothing is pushed to or created on GitHub until you call pr_push().

• pr_fetch(): As a maintainer, review or contribute changes to an existing PR by creating a local branch that tracks the remote PR. pr_fetch() does as little work as possible, so you can also use it to resume work on an PR that already has a local branch (where it will also ensure your local branch is up-to-date). If called with no arguments, up to 9 open PRs are offered for interactive selection.

• pr_resume(): Resume work on a PR by switching to an existing local branch and pulling any changes from its upstream tracking branch, if it has one. If called with no arguments, up to 9 local branches are offered for interactive selection, with a preference for branches connected to PRs and for branches with recent activity.

• pr_push(): The first time it's called, a PR branch is pushed to GitHub and you're taken to a webpage where a new PR (or draft PR) can be created. This also sets up the local branch to track its remote counterpart. Subsequent calls to pr_push() make sure the local branch has all the remote changes and, if so, pushes local changes, thereby updating the PR.

• pr_pull(): Pulls changes from the local branch's remote tracking branch. If a maintainer has extended your PR, this is how you bring those changes back into your local work.

• pr_merge_main(): Pulls changes from the default branch of the source repo into the current local branch. This can be used when the local branch is the default branch or when it's a PR branch.

• pr_pause(): Makes sure you're up-to-date with any remote changes in the PR. Then switches back to the default branch and pulls from the source repo. Use pr_resume() with name of branch or use pr_fetch() to resume using PR number.

• pr_view(): Visits the PR associated with the current branch in the browser (default) or the specific PR identified by number. (FYI browse_github_pulls() is a handy way to visit the list of all PRs for the current project.)

• pr_forget(): Does local clean up when the current branch is an actual or notional PR that you want to abandon. Maybe you initiated it yourself, via pr_init(), or you used pr_fetch() to explore a PR from GitHub. Only does local operations: does not update or delete any remote branches, nor does it close any PRs. Alerts the user to any uncommitted or unpushed work that is at risk of being lost. If user chooses to proceed, switches back to the default branch, pulls changes from source repo, and deletes local PR branch. Any associated Git remote is deleted, if the "forgotten" PR was the only branch using it.

• pr_finish(): Does post-PR clean up, but does NOT actually merge or close a PR (maintainer should do this in the browser). If number is not given, infers the PR from the upstream tracking branch of the current branch. If number is given, it does not matter whether the PR exists locally. If PR exists locally, alerts the user to uncommitted or unpushed changes, then switches back to the default branch, pulls changes from source repo, and deletes local PR branch. If the PR came from an external fork, any associated Git remote is deleted, provided it's not in use by any other local branches. If the PR has been merged and user has permission, deletes the remote branch (this is the only remote operation that pr_finish() potentially does).

Examples

## Not run: 
pr_fetch(123)

## End(Not run)

Description

• Moves ⁠R/{old}.R⁠ to ⁠R/{new}.R⁠

• Moves ⁠src/{old}.*⁠ to ⁠src/{new}.*⁠

• Moves ⁠tests/testthat/test-{old}.R⁠ to ⁠tests/testthat/test-{new}.R⁠

• Moves ⁠tests/testthat/test-{old}-*.*⁠ to ⁠tests/testthat/test-{new}-*.*⁠ and updates paths in the test file.

• Removes context() calls from the test file, which are unnecessary (and discouraged) as of testthat v2.1.0.

This is a potentially dangerous operation, so you must be using Git in order to use this function.

Usage

rename_files(old, new)

Arguments

Description

All functions open your .Rprofile and give you the code you need to paste in.

• use_devtools(): makes devtools available in interactive sessions.

• use_usethis(): makes usethis available in interactive sessions.

• use_reprex(): makes reprex available in interactive sessions.

• use_conflicted(): makes conflicted available in interactive sessions.

• use_partial_warnings(): warns on partial matches.

Usage

use_conflicted()

use_reprex()

use_usethis()

use_devtools()

use_partial_warnings()

Description

Execute a bit of code without usethis's normal messaging.

Usage

ui_silence(code)

Arguments

Value

Whatever code returns.

Examples

# compare the messaging you see from this:
browse_github("usethis")
# vs. this:
ui_silence(
  browse_github("usethis")
)

Description

This function helps you add a minimal RStudio Addin binding to inst/rstudio/addins.dcf.

Usage

use_addin(addin = "new_addin", open = rlang::is_interactive())

Arguments

Description

use_author() adds a person to the Authors@R field of the DESCRIPTION file, creating that field if necessary. It will not modify, e.g., the role(s) or email of an existing author (judged using their "Given Family" name). For that we recommend editing DESCRIPTION directly. Or, for programmatic use, consider calling the more specialized functions available in the desc package directly.

use_author() also surfaces two other situations you might want to address:

• Explicit use of the fields Author or Maintainer. We recommend switching to the more modern Authors@R field instead, because it offers richer metadata for various downstream uses. (Note that Authors@R is eventually processed to create Author and Maintainer fields, but only when the tar.gz is built from package source.)

• Presence of the fake author placed by create_package() and use_description(). This happens when usethis has to create a DESCRIPTION file and the user hasn't given any author information via the fields argument or the global option "usethis.description". The placeholder looks something like ⁠First Last <[email protected]> [aut, cre]⁠ and use_author() offers to remove it in interactive sessions.

Usage

use_author(given = NULL, family = NULL, ..., role = "ctb")

Arguments

Examples

## Not run: 
use_author(
  given = "Lucy",
  family = "van Pelt",
  role = c("aut", "cre"),
  email = "[email protected]",
  comment = c(ORCID = "LUCY-ORCID-ID")
)

use_author("Charlie", "Brown")

## End(Not run)

Description

R can save and reload the user's workspace between sessions via an .RData file in the current directory. However, long-term reproducibility is enhanced when you turn this feature off and clear R's memory at every restart. Starting with a blank slate provides timely feedback that encourages the development of scripts that are complete and self-contained. More detail can be found in the blog post Project-oriented workflow.

Usage

use_blank_slate(scope = c("user", "project"))

Arguments

Description

.Rbuildignore has a regular expression on each line, but it's usually easier to work with specific file names. By default, use_build_ignore() will (crudely) turn a filename into a regular expression that will only match that path. Repeated entries will be silently removed.

use_build_ignore() is designed to ignore individual files. If you want to ignore all files with a given extension, consider providing an "as-is" regular expression, using escape = FALSE; see examples.

Usage

use_build_ignore(files, escape = TRUE)

Arguments

Examples

## Not run: 
# ignore all Excel files
use_build_ignore("[.]xlsx$", escape = FALSE)

## End(Not run)

Description

Use this if you want to encourage users of your package to cite an article or book.

Usage

use_citation()

Description

Adds a CODE_OF_CONDUCT.md file to the active project and lists in .Rbuildignore, in the case of a package. The goal of a code of conduct is to foster an environment of inclusiveness, and to explicitly discourage inappropriate behaviour. The template comes from https://www.contributor-covenant.org, version 2.1: https://www.contributor-covenant.org/version/2/1/code_of_conduct/.

Usage

use_code_of_conduct(contact, path = NULL)

Arguments

Details

If your package is going to CRAN, the link to the CoC in your README must be an absolute link to a rendered website as CODE_OF_CONDUCT.md is not included in the package sent to CRAN. use_code_of_conduct() will automatically generate this link if (1) you use pkgdown and (2) have set the url field in ⁠_pkgdown.yml⁠; otherwise it will link to a copy of the CoC on https://www.contributor-covenant.org.

Description

Adds test coverage reporting to a package, using either Codecov (⁠https://codecov.io⁠) or Coveralls (⁠https://coveralls.io⁠).

Usage

use_coverage(type = c("codecov", "coveralls"), repo_spec = NULL)

use_covr_ignore(files)

Arguments

Description

Adds infrastructure needed to use the cpp11 package, a header-only R package that helps R package developers handle R objects with C++ code:

• Creates ⁠src/⁠

• Adds cpp11 to DESCRIPTION

• Creates src/code.cpp, an initial placeholder .cpp file

Usage

use_cpp11()

Description

Creates cran-comments.md, a template for your communications with CRAN when submitting a package. The goal is to clearly communicate the steps you have taken to check your package on a wide range of operating systems. If you are submitting an update to a package that is used by other packages, you also need to summarize the results of your reverse dependency checks.

Usage

use_cran_comments(open = rlang::is_interactive())

Arguments

Description

use_data() makes it easy to save package data in the correct format. I recommend you save scripts that generate package data in data-raw: use use_data_raw() to set it up. You also need to document exported datasets.

Usage

use_data(
  ...,
  internal = FALSE,
  overwrite = FALSE,
  compress = "bzip2",
  version = 3,
  ascii = FALSE
)

use_data_raw(name = "DATASET", open = rlang::is_interactive())

Arguments

See Also

The data chapter of R Packages.

Examples

## Not run: 
x <- 1:10
y <- 1:100

use_data(x, y) # For external use
use_data(x, y, internal = TRUE) # For internal use

## End(Not run)
## Not run: 
use_data_raw("daisy")

## End(Not run)

Description

use_data_table() imports the data.table() function from the data.table package, as well as several important symbols: ⁠:=⁠, .SD, .BY, .N, .I, .GRP, .NGRP, .EACHI. This is a minimal setup and you can learn much more in the "Importing data.table" vignette: ⁠https://rdatatable.gitlab.io/data.table/articles/datatable-importing.html⁠. In addition to importing these functions, use_data_table() also blocks the usage of data.table in the Depends field of the DESCRIPTION file; data.table should be used as an imported or suggested package only. See this discussion.

Usage

use_data_table()

Description

use_description() creates a DESCRIPTION file. Although mostly associated with R packages, a DESCRIPTION file can also be used to declare dependencies for a non-package project. Within such a project, devtools::install_deps() can then be used to install all the required packages. Note that, by default, use_decription() checks for a CRAN-compliant package name. You can turn this off with check_name = FALSE.

usethis consults the following sources, in this order, to set DESCRIPTION fields:

• fields argument of create_package() or use_description()

• getOption("usethis.description")

• Defaults built into usethis

The fields discovered via options or the usethis package can be viewed with use_description_defaults().

If you create a lot of packages, consider storing personalized defaults as a named list in an option named "usethis.description". Here's an example of code to include in .Rprofile, which can be opened via edit_r_profile():

options(
  usethis.description = list(
    "Authors@R" = utils::person(
      "Jane", "Doe",
      email = "[email protected]",
      role = c("aut", "cre"),
      comment = c(ORCID = "YOUR-ORCID-ID")
    ),
    Language =  "es",
    License = "MIT + file LICENSE"
  )
)

Prior to usethis v2.0.0, getOption("devtools.desc") was consulted for backwards compatibility, but now only the "usethis.description" option is supported.

Usage

use_description(fields = list(), check_name = TRUE, roxygen = TRUE)

use_description_defaults(package = NULL, roxygen = TRUE, fields = list())

Arguments

See Also

The description chapter of R Packages

Examples

## Not run: 
use_description()

use_description(fields = list(Language = "es"))

use_description_defaults()

## End(Not run)

Description

use_directory() creates a directory (if it does not already exist) in the project's top-level directory. This function powers many of the other use_ functions such as use_data() and use_vignette().

Usage

use_directory(path, ignore = FALSE)

Arguments

Examples

## Not run: 
use_directory("inst")

## End(Not run)

Description

use_git() initialises a Git repository and adds important files to .gitignore. If user consents, it also makes an initial commit.

Usage

use_git(message = "Initial commit")

Arguments

See Also

Other git helpers: use_git_config(), use_git_hook(), use_git_ignore()

Examples

## Not run: 
use_git()

## End(Not run)

Description

Sets Git options, for either the user or the project ("global" or "local", in Git terminology). Wraps gert::git_config_set() and gert::git_config_global_set(). To inspect Git config, see gert::git_config().

Usage

use_git_config(scope = c("user", "project"), ...)

Arguments

Value

Invisibly, the previous values of the modified components, as a named list.

See Also

Other git helpers: use_git(), use_git_hook(), use_git_ignore()

Examples

## Not run: 
# set the user's global user.name and user.email
use_git_config(user.name = "Jane", user.email = "[email protected]")

# set the user.name and user.email locally, i.e. for current repo/project
use_git_config(
  scope = "project",
  user.name = "Jane",
  user.email = "[email protected]"
)

## End(Not run)

Description

Sets up a git hook using the specified script. Creates a hook directory if needed, and sets correct permissions on hook.

Usage

use_git_hook(hook, script)

Arguments

See Also

Other git helpers: use_git(), use_git_config(), use_git_ignore()

Description

Tell Git to ignore files

Usage

use_git_ignore(ignores, directory = ".")

Arguments

See Also

Other git helpers: use_git(), use_git_config(), use_git_hook()

Description

Two helpers are available:

• use_git_remote() sets the remote associated with name to url.

• git_remotes() reports the configured remotes, similar to ⁠git remote -v⁠.

Usage

use_git_remote(name = "origin", url, overwrite = FALSE)

git_remotes()

Arguments

Value

Named list of Git remotes.

Examples

## Not run: 
# see current remotes
git_remotes()

# add new remote named 'foo', a la `git remote add <name> <url>`
use_git_remote(name = "foo", url = "https://github.com/<OWNER>/<REPO>.git")

# remove existing 'foo' remote, a la `git remote remove <name>`
use_git_remote(name = "foo", url = NULL, overwrite = TRUE)

# change URL of remote 'foo', a la `git remote set-url <name> <newurl>`
use_git_remote(
  name = "foo",
  url = "https://github.com/<OWNER>/<REPO>.git",
  overwrite = TRUE
)

# Scenario: Fix remotes when you cloned someone's repo, but you should
# have fork-and-cloned (in order to make a pull request).

# Store origin = main repo's URL, e.g., "[email protected]:<OWNER>/<REPO>.git"
upstream_url <- git_remotes()[["origin"]]

# IN THE BROWSER: fork the main GitHub repo and get your fork's remote URL
my_url <- "[email protected]:<ME>/<REPO>.git"

# Rotate the remotes
use_git_remote(name = "origin", url = my_url)
use_git_remote(name = "upstream", url = upstream_url)
git_remotes()

# Scenario: Add upstream remote to a repo that you fork-and-cloned, so you
# can pull upstream changes.
# Note: If you fork-and-clone via `usethis::create_from_github()`, this is
# done automatically!

# Get URL of main GitHub repo, probably in the browser
upstream_url <- "[email protected]:<OWNER>/<REPO>.git"
use_git_remote(name = "upstream", url = upstream_url)

## End(Not run)

Description

use_github() takes a local project and:

• Checks that the initial state is good to go:

  • Project is already a Git repo

  • Current branch is the default branch, e.g. main or master

  • No uncommitted changes

  • No pre-existing origin remote

• Creates an associated repo on GitHub

• Adds that GitHub repo to your local repo as the origin remote

• Makes an initial push to GitHub

• Calls use_github_links(), if the project is an R package

• Configures origin/DEFAULT to be the upstream branch of the local DEFAULT branch, e.g. main or master

See below for the authentication setup that is necessary for all of this to work.

Usage

use_github(
  organisation = NULL,
  private = FALSE,
  visibility = c("public", "private", "internal"),
  protocol = git_protocol(),
  host = NULL
)

Arguments

Git/GitHub Authentication

Many usethis functions, including those documented here, potentially interact with GitHub in two different ways:

• Via the GitHub REST API. Examples: create a repo, a fork, or a pull request.

• As a conventional Git remote. Examples: clone, fetch, or push.

Therefore two types of auth can happen and your credentials must be discoverable. Which credentials do we mean?

• A GitHub personal access token (PAT) must be discoverable by the gh package, which is used for GitHub operations via the REST API. See gh_token_help() for more about getting and configuring a PAT.

• If you use the HTTPS protocol for Git remotes, your PAT is also used for Git operations, such as ⁠git push⁠. Usethis uses the gert package for this, so the PAT must be discoverable by gert. Generally gert and gh will discover and use the same PAT. This ability to "kill two birds with one stone" is why HTTPS + PAT is our recommended auth strategy for those new to Git and GitHub and PRs.

• If you use SSH remotes, your SSH keys must also be discoverable, in addition to your PAT. The public key must be added to your GitHub account.

Git/GitHub credential management is covered in a dedicated article: Managing Git(Hub) Credentials

Examples

## Not run: 
pkgpath <- file.path(tempdir(), "testpkg")
create_package(pkgpath)

## now, working inside "testpkg", initialize git repository
use_git()

## create github repository and configure as git remote
use_github()

## End(Not run)

Description

Sets up continuous integration (CI) for an R package that is developed on GitHub using GitHub Actions. CI can be used to trigger various operations for each push or pull request, e.g. running ⁠R CMD check⁠ or building and deploying a pkgdown site.

Workflows

There are four particularly important workflows that are used by many packages:

• check-standard: Run ⁠R CMD check⁠ using R-latest on Linux, Mac, and Windows, and using R-devel and R-oldrel on Linux. This is a good baseline if you plan on submitting your package to CRAN.

• test-coverage: Compute test coverage and report to https://about.codecov.io by calling covr::codecov().

• pkgdown: Automatically build and publish a pkgdown website. But we recommend instead calling use_pkgdown_github_pages() which performs other important set up.

• pr-commands: Enables the use of two R-specific commands in pull request issue comments: ⁠/document⁠ to run roxygen2::roxygenise() and ⁠/style⁠ to run styler::style_pkg(). Both will update the PR with any changes once they're done.

If you call use_github_action() without arguments, you'll be prompted to pick from one of these. Otherwise you can see a complete list of possibilities provided by r-lib at https://github.com/r-lib/actions/tree/v2/examples, or you can supply your own url to use any other workflow.

Usage

use_github_action(
  name = NULL,
  ref = NULL,
  url = NULL,
  save_as = NULL,
  readme = NULL,
  ignore = TRUE,
  open = FALSE,
  badge = NULL
)

Arguments

Examples

## Not run: 
use_github_action()

use_github_action_check_standard()

use_github_action("pkgdown")

## End(Not run)

Description

Gets the content of a file from GitHub, from any repo the user can read, and writes it into the active project. This function wraps an endpoint of the GitHub API which supports specifying a target reference (i.e. branch, tag, or commit) and which follows symlinks.

Usage

use_github_file(
  repo_spec,
  path = NULL,
  save_as = NULL,
  ref = NULL,
  ignore = FALSE,
  open = FALSE,
  overwrite = FALSE,
  host = NULL
)

Arguments

Value

A logical indicator of whether a file was written, invisibly.

Examples

## Not run: 
use_github_file(
  "https://github.com/r-lib/actions/blob/v2/examples/check-standard.yaml"
)

use_github_file(
  "r-lib/actions",
  path = "examples/check-standard.yaml",
  ref = "v2",
  save_as = ".github/workflows/R-CMD-check.yaml"
)

## End(Not run)

Description

use_github_labels() can create new labels, update colours and descriptions, and optionally delete GitHub's default labels (if delete_default = TRUE). It will never delete labels that have associated issues.

use_tidy_github_labels() calls use_github_labels() with tidyverse conventions powered by tidy_labels(), tidy_labels_rename(), tidy_label_colours() and tidy_label_descriptions().

tidyverse label usage

Labels are used as part of the issue-triage process, designed to minimise the time spent re-reading issues. The absence of a label indicates that an issue is new, and has yet to be triaged.

There are four mutually exclusive labels that indicate the overall "type" of issue:

• bug: an unexpected problem or unintended behavior.

• documentation: requires changes to the docs.

• feature: feature requests and enhancement.

• upkeep: general package maintenance work that makes future development easier.

Then there are five labels that are needed in most repositories:

• ⁠breaking change⁠: issue/PR will requires a breaking change so should be not be included in patch releases.

• reprex indicates that an issue does not have a minimal reproducible example, and that a reply has been sent requesting one from the user.

• ⁠good first issue⁠ indicates a good issue for first-time contributors.

• ⁠help wanted⁠ indicates that a maintainer wants help on an issue.

• wip indicates that someone is working on it or has promised to.

Finally most larger repos will accumulate their own labels for specific areas of functionality. For example, usethis has labels like "description", "paths", "readme", because time has shown these to be common sources of problems. These labels are helpful for grouping issues so that you can tackle related problems at the same time.

Repo-specific issues should have a grey background (⁠#eeeeee⁠) and an emoji. This keeps the issue page visually harmonious while still giving enough variation to easily distinguish different types of label.

Usage

use_github_labels(
  labels = character(),
  rename = character(),
  colours = character(),
  descriptions = character(),
  delete_default = FALSE
)

use_tidy_github_labels()

tidy_labels()

tidy_labels_rename()

tidy_label_colours()

tidy_label_descriptions()

Arguments

Examples

## Not run: 
# typical use in, e.g., a new tidyverse project
use_github_labels(delete_default = TRUE)

# create labels without changing colours/descriptions
use_github_labels(
  labels = c("foofy", "foofier", "foofiest"),
  colours = NULL,
  descriptions = NULL
)

# change descriptions without changing names/colours
use_github_labels(
  labels = NULL,
  colours = NULL,
  descriptions = c("foofiest" = "the foofiest issue you ever saw")
)

## End(Not run)

Description

Populates the URL and BugReports fields of a GitHub-using R package with appropriate links. The GitHub repo to link to is determined from the current project's GitHub remotes:

• If we are not working with a fork, this function expects origin to be a GitHub remote and the links target that repo.

• If we are working in a fork, this function expects to find two GitHub remotes: origin (the fork) and upstream (the fork's parent) remote. In an interactive session, the user can confirm which repo to use for the links. In a noninteractive session, links are formed using upstream.

Usage

use_github_links(overwrite = FALSE)

Arguments

Examples

## Not run: 
use_github_links()

## End(Not run)

Description

Activates or reconfigures a GitHub Pages site for a project hosted on GitHub. This function anticipates two specific usage modes:

• Publish from the root directory of a gh-pages branch, which is assumed to be only (or at least primarily) a remote branch. Typically the gh-pages branch is managed by an automatic "build and deploy" job, such as the one configured by use_github_action("pkgdown").

• Publish from the "/docs" directory of a "regular" branch, probably the repo's default branch. The user is assumed to have a plan for how they will manage the content below "/docs".

Usage

use_github_pages(branch = "gh-pages", path = "/", cname = NA)

Arguments

Value

Site metadata returned by the GitHub API, invisibly

See Also

• use_pkgdown_github_pages() combines use_github_pages() with other functions to fully configure a pkgdown site

• https://docs.github.com/en/pages

• https://docs.github.com/en/rest/pages

Examples

## Not run: 
use_github_pages()
use_github_pages(branch = git_default_branch(), path = "/docs")

## End(Not run)

Description

Pushes the current branch (if safe) then publishes a GitHub release for the latest CRAN submission.

If you use devtools::submit_cran() to submit to CRAN, information about the submitted state is captured in a CRAN-SUBMISSION file. use_github_release() uses this info to populate the GitHub release notes and, after success, deletes the file. In the absence of such a file, we assume that current state (SHA of HEAD, package version, NEWS) is the submitted state.

Usage

use_github_release(publish = TRUE)

Arguments

Description

These functions are not actively used by the tidyverse team, and may not currently work. Use at your own risk.

Sets up third-party continuous integration (CI) services for an R package on GitLab or CircleCI. These functions:

• Add service-specific configuration files and add them to .Rbuildignore.

• Activate a service or give the user a detailed prompt.

• Provide the markdown to insert a badge into README.

Usage

use_gitlab_ci()

use_circleci(browse = rlang::is_interactive(), image = "rocker/verse:latest")

use_circleci_badge(repo_spec = NULL)

Arguments

use_gitlab_ci()

Adds a basic .gitlab-ci.yml to the top-level directory of a package. This is a configuration file for the GitLab CI/CD continuous integration service.

use_circleci()

Adds a basic .circleci/config.yml to the top-level directory of a package. This is a configuration file for the CircleCI continuous integration service.

use_circleci_badge()

Only adds the Circle CI badge. Use for a project where Circle CI is already configured.

Description

use_import_from() imports a function from another package by adding the roxygen2 ⁠@importFrom⁠ tag to the package-level documentation (which can be created with use_package_doc()). Importing a function from another package allows you to refer to it without a namespace (e.g., fun() instead of package::fun()).

use_import_from() also re-documents the NAMESPACE, and re-load the current package. This ensures that fun is immediately available in your development session.

Usage

use_import_from(package, fun, load = is_interactive())

Arguments

Value

Invisibly, TRUE if the package document has changed, FALSE if not.

Examples

## Not run: 
use_import_from("glue", "glue")

## End(Not run)

Description

use_jenkins() adds a basic Jenkinsfile for R packages to the project root directory. The Jenkinsfile stages take advantage of calls to make, and so calling this function will also run use_make() if a Makefile does not already exist at the project root.

Usage

use_jenkins()

See Also

The documentation on Jenkins Pipelines.

use_make()

Description

This helper:

• Adds lifecycle as a dependency.

• Imports lifecycle::deprecated() for use in function arguments.

• Copies the lifecycle badges into man/figures.

• Reminds you how to use the badge syntax.

Learn more at https://lifecycle.r-lib.org/articles/communicate.html

Usage

use_lifecycle()

See Also

use_lifecycle_badge() to signal the lifecycle stage of your package as whole

Description

This function helps you use a logo in your package:

• Enforces a specific size

• Stores logo image file at man/figures/logo.png

• Produces the markdown text you need in README to include the logo

Usage

use_logo(img, geometry = "240x278", retina = TRUE)

Arguments

Examples

## Not run: 
use_logo("usethis.png")

## End(Not run)

Description

use_make() adds a basic Makefile to the project root directory.

Usage

use_make()

See Also

The documentation for GNU Make.

Description

If roxygen is TRUE generates an empty NAMESPACE that exports nothing; you'll need to explicitly export functions with ⁠@export⁠. If roxygen is FALSE, generates a default NAMESPACE that exports all functions except those that start with ..

Usage

use_namespace(roxygen = TRUE)

Arguments

See Also

The namespace chapter of R Packages.

Description

This creates a basic NEWS.md in the root directory.

Usage

use_news_md(open = rlang::is_interactive())

Arguments

See Also

The other markdown files section of R Packages.

Description

use_package() adds a CRAN package dependency to DESCRIPTION and offers a little advice about how to best use it. use_dev_package() adds a dependency on an in-development package, adding the dev repo to Remotes so it will be automatically installed from the correct location. There is no helper to remove a dependency: to do that, simply remove that package from your DESCRIPTION file.

use_package() exists to support a couple of common maneuvers:

• Add a dependency to Imports or Suggests or LinkingTo.

• Add a minimum version to a dependency.

• Specify the minimum supported version for R.

use_package() probably works for slightly more exotic modifications, but at some point, you should edit DESCRIPTION yourself by hand. There is no intention to account for all possible edge cases.

Usage

use_package(package, type = "Imports", min_version = NULL)

use_dev_package(package, type = "Imports", remote = NULL)

Arguments

See Also

The dependencies section of R Packages.

Examples

## Not run: 
use_package("ggplot2")
use_package("dplyr", "suggests")
use_dev_package("glue")

# Depend on R version 4.1
use_package("R", type = "Depends", min_version = "4.1")

## End(Not run)

Description

Adds a dummy .R file that will cause roxygen2 to generate basic package-level documentation. If your package is named "foo", this will make help available to the user via ?foo or package?foo. Once you call devtools::document(), roxygen2 will flesh out the .Rd file using data from the DESCRIPTION. That ensures you don't need to repeat (and remember to update!) the same information in multiple places. This .R file is also a good place for roxygen directives that apply to the whole package (vs. a specific function), such as global namespace tags like ⁠@importFrom⁠.

Usage

use_package_doc(open = rlang::is_interactive())

Arguments

See Also

The documentation chapter of R Packages

Description

Does setup necessary to use magrittr's pipe operator, ⁠%>%⁠ in your package. This function requires the use of roxygen2.

• Adds magrittr to "Imports" in DESCRIPTION.

• Imports the pipe operator specifically, which is necessary for internal use.

• Exports the pipe operator, if export = TRUE, which is necessary to make ⁠%>%⁠ available to the users of your package.

Usage

use_pipe(export = TRUE)

Arguments

Examples

## Not run: 
use_pipe()

## End(Not run)

Description

pkgdown makes it easy to turn your package into a beautiful website. usethis provides two functions to help you use pkgdown:

• use_pkgdown(): creates a pkgdown config file and adds relevant files or directories to .Rbuildignore and .gitignore.

• use_pkgdown_github_pages(): implements the GitHub setup needed to automatically publish your pkgdown site to GitHub pages:

  • (first, it calls use_pkgdown())

  • use_github_pages() prepares to publish the pkgdown site from the gh-pages branch

  • use_github_action("pkgdown") configures a GitHub Action to automatically build the pkgdown site and deploy it via GitHub Pages

  • The pkgdown site's URL is added to the pkgdown configuration file, to the URL field of DESCRIPTION, and to the GitHub repo.

  • Packages owned by certain GitHub organizations (tidyverse, r-lib, and tidymodels) get some special treatment, in terms of anticipating the (eventual) site URL and the use of a pkgdown template.

Usage

use_pkgdown(config_file = "_pkgdown.yml", destdir = "docs")

use_pkgdown_github_pages()

Arguments

See Also

https://pkgdown.r-lib.org/articles/pkgdown.html#configuration

Description

This pair of functions makes it easy to create paired R and test files, using the convention that the tests for R/foofy.R should live in tests/testthat/test-foofy.R. You can use them to create new files from scratch by supplying name, or if you use RStudio, you can call to create (or navigate to) the companion file based on the currently open file. This also works when a test snapshot file is active, i.e. if you're looking at ⁠tests/testthat/_snaps/foofy.md⁠, use_r() or use_test() take you to R/foofy.R or tests/testthat/test-foofy.R, respectively.

Usage

use_r(name = NULL, open = rlang::is_interactive())

use_test(name = NULL, open = rlang::is_interactive())

Arguments

Renaming files in an existing package

Here are some tips on aligning file names across ⁠R/⁠ and ⁠tests/testthat/⁠ in an existing package that did not necessarily follow this convention before.

This script generates a data frame of ⁠R/⁠ and test files that can help you identify missed opportunities for pairing:

library(fs)
library(tidyverse)

bind_rows(
  tibble(
    type = "R",
    path = dir_ls("R/", regexp = "\\.[Rr]$"),
    name = as.character(path_ext_remove(path_file(path))),
  ),
  tibble(
    type = "test",
    path = dir_ls("tests/testthat/", regexp = "/test[^/]+\\.[Rr]$"),
    name = as.character(path_ext_remove(str_remove(path_file(path), "^test[-_]"))),
  )
) %>%
  pivot_wider(names_from = type, values_from = path) %>%
  print(n = Inf)

The rename_files() function can also be helpful.

See Also

• The testing and R code chapters of R Packages.

• use_test_helper() to create a testthat helper file.

Examples

## Not run: 
# create a new .R file below R/
use_r("coolstuff")

# if `R/coolstuff.R` is active in a supported IDE, you can now do:
use_test()

# if `tests/testthat/test-coolstuff.R` is active in a supported IDE, you can
# return to `R/coolstuff.R` with:
use_r()

## End(Not run)

Description

Adds infrastructure commonly needed when using compiled code:

• Creates ⁠src/⁠

• Adds required packages to DESCRIPTION

• May create an initial placeholder .c or .cpp file

• Creates Makevars and Makevars.win files (use_rcpp_armadillo() only)

Usage

use_rcpp(name = NULL)

use_rcpp_armadillo(name = NULL)

use_rcpp_eigen(name = NULL)

use_c(name = NULL)

Arguments

Description

Creates skeleton README files with possible stubs for

• a high-level description of the project/package and its goals

• R code to install from GitHub, if GitHub usage detected

• a basic example

Use Rmd if you want a rich intermingling of code and output. Use md for a basic README. README.Rmd will be automatically added to .Rbuildignore. The resulting README is populated with default YAML frontmatter and R fenced code blocks (md) or chunks (Rmd).

If you use Rmd, you'll still need to render it regularly, to keep README.md up-to-date. devtools::build_readme() is handy for this. You could also use GitHub Actions to re-render README.Rmd every time you push. An example workflow can be found in the ⁠examples/⁠ directory here: https://github.com/r-lib/actions/.

If the current project is a Git repo, then use_readme_rmd() automatically configures a pre-commit hook that helps keep README.Rmd and README.md, synchronized. The hook creates friction if you try to commit when README.Rmd has been edited more recently than README.md. If this hook causes more problems than it solves for you, it is implemented in .git/hooks/pre-commit, which you can modify or even delete.

Usage

use_readme_rmd(open = rlang::is_interactive())

use_readme_md(open = rlang::is_interactive())

Arguments

See Also

The other markdown files section of R Packages.

Examples

## Not run: 
use_readme_rmd()
use_readme_md()

## End(Not run)

Description

When preparing to release a package to CRAN there are quite a few steps that need to be performed, and some of the steps can take multiple hours. This function creates a checklist in a GitHub issue to:

• Help you keep track of where you are in the process

• Feel a sense of satisfaction as you progress towards final submission

• Help watchers of your package stay informed.

The checklist contains a generic set of steps that we've found to be helpful, based on the type of release ("patch", "minor", or "major"). You're encouraged to edit the issue to customize this list to meet your needs.

Customization

• If you want to consistently add extra bullets for every release, you can include your own custom bullets by providing an (unexported) release_bullets() function that returns a character vector. (For historical reasons, release_questions() is also supported).

• If you want to check additional packages in the revdep check process, provide an (unexported) release_extra_revdeps() function that returns a character vector. This is currently only supported for Posit internal check tooling.

Usage

use_release_issue(version = NULL)

Arguments

Examples

## Not run: 
use_release_issue("2.0.0")

## End(Not run)

Description

Performs set up for checking the reverse dependencies of an R package, as implemented by the revdepcheck package:

• Creates ⁠revdep/⁠ directory and adds it to .Rbuildignore

• Populates revdep/.gitignore to prevent tracking of various revdep artefacts

• Prompts user to run the checks with revdepcheck::revdep_check()

Usage

use_revdep()

Description

Adds files and directories necessary to add a custom rmarkdown template to RStudio. It creates:

• inst/rmarkdown/templates/{{template_dir}}. Main directory.

• skeleton/skeleton.Rmd. Your template Rmd file.

• template.yml with basic information filled in.

Usage

use_rmarkdown_template(
  template_name = "Template Name",
  template_dir = NULL,
  template_description = "A description of the template",
  template_create_dir = FALSE
)

Arguments

Examples

## Not run: 
use_rmarkdown_template()

## End(Not run)

Description

If you are already using roxygen2, but not with markdown, you'll need to use roxygen2md to convert existing Rd expressions to markdown. The conversion is not perfect, so make sure to check the results.

Usage

use_roxygen_md(overwrite = FALSE)

Arguments

Description

It is likely that you want to use create_project() or create_package() instead of use_rstudio()! Both ⁠create_*()⁠ functions can add RStudio Project infrastructure to a pre-existing project or package. use_rstudio() is mostly for internal use or for those creating a usethis-like package for their organization. It does the following in the current project, often after executing proj_set(..., force = TRUE):

• Creates an .Rproj file

• Adds RStudio files to .gitignore

• Adds RStudio files to .Rbuildignore, if project is a package

Usage

use_rstudio(line_ending = c("posix", "windows"), reformat = TRUE)

Arguments

Description

This function allows you to set global RStudio preferences, achieving the same effect programmatically as clicking buttons in RStudio's Global Options. You can find a list of configurable properties at https://docs.posit.co/ide/server-pro/reference/session_user_settings.html.

Usage

use_rstudio_preferences(...)

Arguments

Value

A named list of the previous values, invisibly.

Description

Adds a unit test to automatically run a spell check on documentation and, optionally, vignettes during ⁠R CMD check⁠, using the spelling package. Also adds a WORDLIST file to the package, which is a dictionary of whitelisted words. See spelling::wordlist for details.

Usage

use_spell_check(vignettes = TRUE, lang = "en-US", error = FALSE)

Arguments

Description

A "standalone" file implements a minimum set of functionality in such a way that it can be copied into another package. use_standalone() makes it easy to get such a file into your own repo.

It always overwrites an existing standalone file of the same name, making it easy to update previously imported code.

Usage

use_standalone(repo_spec, file = NULL, ref = NULL, host = NULL)

Arguments

Supported fields

A standalone file has YAML frontmatter that provides additional information, such as where the file originates from and when it was last updated. Here is an example:

---
repo: r-lib/rlang
file: standalone-types-check.R
last-updated: 2023-03-07
license: https://unlicense.org
dependencies: standalone-obj-type.R
imports: rlang (>= 1.1.0)
---

Two of these fields are consulted by use_standalone():

• dependencies: A file or a list of files in the same repo that the standalone file depends on. These files are retrieved automatically by use_standalone().

• imports: A package or list of packages that the standalone file depends on. A minimal version may be specified in parentheses, e.g. ⁠rlang (>= 1.0.0)⁠. These dependencies are passed to use_package() to ensure they are included in the ⁠Imports:⁠ field of the DESCRIPTION file.

Note that lists are specified with standard YAML syntax, using square brackets, for example: ⁠imports: [rlang (>= 1.0.0), purrr]⁠.

Examples

## Not run: 
use_standalone("r-lib/rlang", file = "types-check")
use_standalone("r-lib/rlang", file = "types-check", ref = "standalone-dep")

## End(Not run)

Description

Creates a file from data and a template found in a package. Provides control over file name, the addition to .Rbuildignore, and opening the file for inspection.

Usage

use_template(
  template,
  save_as = template,
  data = list(),
  ignore = FALSE,
  open = FALSE,
  package = "usethis"
)

Arguments

Details

This function can be used as the engine for a templating function in other packages. The template argument is used along with the package argument to derive the path to your template file; it will be expected at fs::path_package(package = package, "templates", template). We use fs::path_package() instead of base::system.file() so that path construction works even in a development workflow, e.g., works with devtools::load_all() or pkgload::load_all(). Note this describes the behaviour of fs::path_package() in fs v1.2.7.9001 and higher.

To interpolate your data into the template, supply a list using the data argument. Internally, this function uses whisker::whisker.render() to combine your template file with your data.

Value

A logical vector indicating if file was modified.

Examples

## Not run: 
  # Note: running this will write `NEWS.md` to your working directory
  use_template(
    template = "NEWS.md",
    data = list(Package = "acme", Version = "1.2.3"),
    package = "usethis"
  )

## End(Not run)

Description

This function creates (or opens) a test helper file, typically tests/testthat/helper.R. Test helper files are executed at the beginning of every automated test run and are also executed by load_all(). A helper file is a great place to define test helper functions for use throughout your test suite, such as a custom expectation.

Usage

use_test_helper(name = NULL, open = rlang::is_interactive())

Arguments

See Also

• use_test() to create a test file.

• The testthat vignette on special files vignette("special-files", package = "testthat").

Examples

## Not run: 
use_test_helper()
use_test_helper("mocks")

## End(Not run)

Description

Creates ⁠tests/testthat/⁠, tests/testthat.R, and adds the testthat package to the Suggests field. Learn more in https://r-pkgs.org/testing-basics.html

Usage

use_testthat(edition = NULL, parallel = FALSE)

Arguments

See Also

use_test() to create individual test files

Examples

## Not run: 
use_testthat()

use_test()

use_test("something-management")

## End(Not run)

Description

Does minimum setup such that a tibble returned by your package is handled using the tibble method for generics like print() or [. Presumably you care about this if you've chosen to store and expose an object with class tbl_df. Specifically:

• Check that the active package uses roxygen2

• Add the tibble package to "Imports" in DESCRIPTION

• Prepare the roxygen directive necessary to import at least one function from tibble:

  • If possible, the directive is inserted into existing package-level documentation, i.e. the roxygen snippet created by use_package_doc()

  • Otherwise, we issue advice on where the user should add the directive

This is necessary when your package returns a stored data object that has class tbl_df, but the package code does not make direct use of functions from the tibble package. If you do nothing, the tibble namespace is not necessarily loaded and your tibble may therefore be printed and subsetted like a base data.frame.

Usage

use_tibble()

Examples

## Not run: 
use_tibble()

## End(Not run)

Description

These helpers follow tidyverse conventions which are generally a little stricter than the defaults, reflecting the need for greater rigor in commonly used packages.

Usage

use_tidy_github_actions(ref = NULL)

create_tidy_package(path, copyright_holder = NULL)

use_tidy_description()

use_tidy_dependencies()

use_tidy_contributing()

use_tidy_support()

use_tidy_issue_template()

use_tidy_coc()

use_tidy_github()

use_tidy_style(strict = TRUE)

use_tidy_logo(geometry = "240x278", retina = TRUE)

use_tidy_upkeep_issue(last_upkeep = last_upkeep_year())

Arguments

Details

• use_tidy_github_actions(): Sets up the following workflows using GitHub Actions:

  • Run ⁠R CMD check⁠ on the current release, devel, and four previous versions of R. The build matrix also ensures ⁠R CMD check⁠ is run at least once on each of the three major operating systems (Linux, macOS, and Windows).

  • Report test coverage.

  • Build and deploy a pkgdown site.

  • Provide two commands to be used in pull requests: ⁠/document⁠ to run roxygen2::roxygenise() and update the PR, and ⁠/style⁠ to run styler::style_pkg() and update the PR.

    This is how the tidyverse team checks its packages, but it is overkill for less widely used packages. Consider using the more streamlined workflows set up by use_github_actions() or use_github_action_check_standard().

• create_tidy_package(): creates a new package, immediately applies as many of the tidyverse conventions as possible, issues a few reminders, and activates the new package.

• use_tidy_dependencies(): sets up standard dependencies used by all tidyverse packages (except packages that are designed to be dependency free).

• use_tidy_description(): puts fields in standard order and alphabetises dependencies.

• use_tidy_eval(): imports a standard set of helpers to facilitate programming with the tidy eval toolkit.

• use_tidy_style(): styles source code according to the tidyverse style guide. This function will overwrite files! See below for usage advice.

• use_tidy_contributing(): adds standard tidyverse contributing guidelines.

• use_tidy_issue_template(): adds a standard tidyverse issue template.

• use_tidy_release_test_env(): updates the test environment section in cran-comments.md.

• use_tidy_support(): adds a standard description of support resources for the tidyverse.

• use_tidy_coc(): equivalent to use_code_of_conduct(), but puts the document in a ⁠.github/⁠ subdirectory.

• use_tidy_github(): convenience wrapper that calls use_tidy_contributing(), use_tidy_issue_template(), use_tidy_support(), use_tidy_coc().

• use_tidy_github_labels() calls use_github_labels() to implement tidyverse conventions around GitHub issue label names and colours.

• use_tidy_upkeep_issue() creates an issue containing a checklist of actions to bring your package up to current tidyverse standards. Also records the current date in the Config/usethis/last-upkeep field in DESCRIPTION.

• use_tidy_logo() calls use_logo() on the appropriate hex sticker PNG file at https://github.com/rstudio/hex-stickers.

use_tidy_style()

Uses the styler package package to style all code in a package, project, or directory, according to the tidyverse style guide.

Warning: This function will overwrite files! It is strongly suggested to only style files that are under version control or to first create a backup copy.

Invisibly returns a data frame with one row per file, that indicates whether styling caused a change.

Description

Derives a list of GitHub usernames, based on who has opened issues or pull requests. Used to populate the acknowledgment section of package release blog posts at https://www.tidyverse.org/blog/. If no arguments are given, we retrieve all contributors to the active project since its last (GitHub) release. Unexported helper functions, releases() and ref_df() can be useful interactively to get a quick look at release tag names and a data frame about refs (defaulting to releases), respectively.

Usage

use_tidy_thanks(repo_spec = NULL, from = NULL, to = NULL)

Arguments

Value

A character vector of GitHub usernames, invisibly.

Examples

## Not run: 
# active project, interval = since the last release
use_tidy_thanks()

# active project, interval = since a specific datetime
use_tidy_thanks(from = "2020-07-24T00:13:45Z")

# r-lib/usethis, interval = since a certain date
use_tidy_thanks("r-lib/usethis", from = "2020-08-01")

# r-lib/usethis, up to a specific release
use_tidy_thanks("r-lib/usethis", from = NULL, to = "v1.1.0")

# r-lib/usethis, since a specific commit, up to a specific date
use_tidy_thanks("r-lib/usethis", from = "08a560d", to = "2018-05-14")

# r-lib/usethis, but with copy/paste of a browser URL
use_tidy_thanks("https://github.com/r-lib/usethis")

## End(Not run)

Description

Creates a new tutorial below ⁠inst/tutorials/⁠. Tutorials are interactive R Markdown documents built with the learnr package. use_tutorial() does this setup:

• Adds learnr to Suggests in DESCRIPTION.

• Gitignores ⁠inst/tutorials/*.html⁠ so you don't accidentally track rendered tutorials.

• Creates a new .Rmd tutorial from a template and, optionally, opens it for editing.

• Adds new .Rmd to .Rbuildignore.

Usage

use_tutorial(name, title, open = rlang::is_interactive())

Arguments

See Also

The learnr package documentation.

Examples

## Not run: 
use_tutorial("learn-to-do-stuff", "Learn to do stuff")

## End(Not run)

Description

This opens an issue in your package repository with a checklist of tasks for regular maintenance of your package. This is a fairly opinionated list of tasks but we believe taking care of them will generally make your package better, easier to maintain, and more enjoyable for your users. Some of the tasks are meant to be performed only once (and once completed shouldn't show up in subsequent lists), and some should be reviewed periodically. The tidyverse team uses a similar function use_tidy_upkeep_issue() for our annual package Spring Cleaning.

Usage

use_upkeep_issue(year = NULL)

Arguments

Examples

## Not run: 
use_upkeep_issue()

## End(Not run)

Description

usethis supports semantic versioning, which is described in more detail in the version section of R Packages. A version number breaks down like so:

<major>.<minor>.<patch>       (released version)
<major>.<minor>.<patch>.<dev> (dev version)

use_version() increments the "Version" field in DESCRIPTION, adds a new heading to NEWS.md (if it exists), commits those changes (if package uses Git), and optionally pushes (if safe to do so). It makes the same update to a line like PKG_version = "x.y.z"; in src/version.c (if it exists).

use_dev_version() increments to a development version, e.g. from 1.0.0 to 1.0.0.9000. If the existing version is already a development version with four components, it does nothing. Thin wrapper around use_version().

Usage

use_version(which = NULL, push = FALSE)

use_dev_version(push = FALSE)

Arguments

See Also

The version section of R Packages.

Examples

## Not run: 
## for interactive selection, do this:
use_version()

## request a specific type of increment
use_version("minor")
use_dev_version()

## End(Not run)

Description

Creates a new vignette or article in ⁠vignettes/⁠. Articles are a special type of vignette that appear on pkgdown websites, but are not included in the package itself (because they are added to .Rbuildignore automatically).

Usage

use_vignette(name, title = name)

use_article(name, title = name)

Arguments

General setup

• Adds needed packages to DESCRIPTION.

• Adds inst/doc to .gitignore so built vignettes aren't tracked.

• Adds ⁠vignettes/*.html⁠ and ⁠vignettes/*.R⁠ to .gitignore so you never accidentally track rendered vignettes.

See Also

The vignettes chapter of R Packages.

Examples

## Not run: 
use_vignette("how-to-do-stuff", "How to do stuff")

## End(Not run)

Description

User-configurable options consulted by usethis, which provide a mechanism for setting default behaviors for various functions.

If the built-in defaults don't suit you, set one or more of these options. Typically, this is done in the .Rprofile startup file, which you can open for editing with edit_r_profile() - this will set the specified options for all future R sessions. Your code will look something like:

options(
  usethis.description = list(
    "Authors@R" = utils::person(
      "Jane", "Doe",
      email = "[email protected]",
      role = c("aut", "cre"),
      comment = c(ORCID = "YOUR-ORCID-ID")
    ),
    License = "MIT + file LICENSE"
  ),
  usethis.destdir = "/path/to/folder/", # for use_course(), create_from_github()
  usethis.protocol = "ssh", # Use ssh git protocol
  usethis.overwrite = TRUE # overwrite files in Git repos without confirmation
)

Options for the usethis package

• usethis.description: customize the default content of new DESCRIPTION files by setting this option to a named list. If you are a frequent package developer, it is worthwhile to pre-configure your preferred name, email, license, etc. See the example above and the article on usethis setup for more details.

• usethis.destdir: Default directory in which to place new projects downloaded by use_course() and create_from_github(). If this option is unset, the user's Desktop or similarly conspicuous place will be used.

• usethis.protocol: specifies your preferred transport protocol for Git. Either "https" (default) or "ssh":

  • usethis.protocol = "https" implies ⁠https://github.com/<OWNER>/<REPO>.git⁠

  • usethis.protocol = "ssh" implies ⁠[email protected]:<OWNER>/<REPO>.git⁠

  You can also change this for the duration of your R session with use_git_protocol().

• usethis.overwrite: If TRUE, usethis overwrites an existing file without asking for user confirmation if the file is inside a Git repo. The rationale is that the normal Git workflow makes it easy to see and selectively accept/discard any proposed changes.

• usethis.quiet: Set to TRUE to suppress user-facing messages. Default FALSE.

• usethis.allow_nested_project: Whether or not to allow you to create a project inside another project. This is rarely a good idea, so this option defaults to FALSE.

Description

Functions to download and unpack a ZIP file into a local folder of files, with very intentional default behaviour. Useful in pedagogical settings or anytime you need a large audience to download a set of files quickly and actually be able to find them. The underlying helpers are documented in use_course_details.

Usage

use_course(url, destdir = getOption("usethis.destdir"))

use_zip(
  url,
  destdir = getwd(),
  cleanup = if (rlang::is_interactive()) NA else FALSE
)

Arguments

Value

Path to the new directory holding the unpacked ZIP file, invisibly.

Functions

• use_course(): Designed with live workshops in mind. Includes intentional friction to highlight the download destination. Workflow:

  • User executes, e.g., use_course("bit.ly/xxx-yyy-zzz").

  • User is asked to notice and confirm the location of the new folder. Specify destdir or configure the "usethis.destdir" option to prevent this.

  • User is asked if they'd like to delete the ZIP file.

  • If new folder contains an .Rproj file, a new instance of RStudio is launched. Otherwise, the folder is opened in the file manager, e.g. Finder or File Explorer.

• use_zip(): More useful in day-to-day work. Downloads in current working directory, by default, and allows cleanup behaviour to be specified.

Examples

## Not run: 
# download the source of usethis from GitHub, behind a bit.ly shortlink
use_course("bit.ly/usethis-shortlink-example")
use_course("http://bit.ly/usethis-shortlink-example")

# download the source of rematch2 package from CRAN
use_course("https://cran.r-project.org/bin/windows/contrib/3.4/rematch2_2.0.1.zip")

# download the source of rematch2 package from GitHub, 4 ways
use_course("r-lib/rematch2")
use_course("https://api.github.com/repos/r-lib/rematch2/zipball/HEAD")
use_course("https://api.github.com/repos/r-lib/rematch2/zipball/main")
use_course("https://github.com/r-lib/rematch2/archive/main.zip")

## End(Not run)