# devtools — Tools to Make R Package Development Easier

> devtools is an R package that streamlines the entire R package development workflow. It provides functions for loading, testing, documenting, checking, and installing packages from local directories, GitHub, and other sources.

## Install

Save in your project root:

# devtools — Tools to Make R Package Development Easier

## Quick Use
```r
install.packages("devtools")
library(devtools)
load_all()       # Load package source for interactive development
test()           # Run testthat tests
document()       # Generate Rd docs from roxygen2 comments
check()          # Run R CMD check
install()        # Install the package locally
```

## Introduction
devtools is an R package by Hadley Wickham and the r-lib team that removes the friction from R package development. Before devtools, building and testing packages required manual command-line invocations and configuration. devtools wraps these steps into simple R functions that work from an interactive R session or RStudio, making the edit-load-test cycle fast and natural.

## What devtools Does
- Loads package source code into the current session with `load_all()` for rapid iteration without reinstalling
- Runs testthat test suites with `test()` and reports results inline
- Generates documentation from roxygen2 comments with `document()`
- Runs the full R CMD check suite with `check()` to catch issues before CRAN submission
- Installs packages from GitHub, GitLab, Bitbucket, and URLs with `install_github()` and friends

## Architecture Overview
devtools acts as a thin wrapper around a suite of focused packages: pkgload (for `load_all()`), pkgbuild (for building tarballs), rcmdcheck (for `check()`), remotes (for installing from remote sources), roxygen2 (for documentation), and testthat (for testing). This modular design means each component can be updated independently. devtools re-exports the key functions from these packages so developers only need to load one package.

## Self-Hosting & Configuration
- Install from CRAN: `install.packages("devtools")`
- Create a new package skeleton with `usethis::create_package("mypackage")`
- Add tests with `usethis::use_testthat()` and write test files in `tests/testthat/`
- Configure CI with `usethis::use_github_action_check_standard()` for automated R CMD check
- Set `DESCRIPTION` fields (Imports, Suggests, License) and use `usethis::use_package()` to add dependencies

## Key Features
- `load_all()` simulates package installation by sourcing all R files and loading compiled code, enabling a fast development loop
- `install_github("owner/repo")` installs packages directly from GitHub without manual download
- `check()` catches common issues (missing docs, broken examples, test failures) before CRAN submission
- Tight integration with usethis for scaffolding package infrastructure (tests, CI, badges, NEWS, vignettes)
- Works in both interactive R sessions and RStudio with dedicated build pane support

## Comparison with Similar Tools
- **usethis (R)** — complementary package for one-time setup tasks (creating files, configuring CI); devtools covers the iterative development cycle
- **pak (R)** — faster package installer; devtools focuses on development workflow, not just installation
- **remotes (R)** — handles remote installation only; devtools provides the full develop-test-check-install workflow
- **setuptools / pip (Python)** — Python packaging tools; devtools is the R equivalent but more opinionated and integrated
- **npm / cargo** — language-specific package managers; devtools is unique in combining development workflow with package management

## FAQ
**Q: Do I need devtools to develop R packages?**
A: Not strictly, but it makes the process much simpler. Without it, you would use R CMD commands from the terminal.

**Q: What is the difference between devtools and usethis?**
A: usethis handles one-time setup (create files, configure settings). devtools handles repeated actions (load, test, check, install). They work together.

**Q: Can I install packages from private GitHub repos?**
A: Yes. Set a GITHUB_PAT environment variable with a personal access token and use `install_github("owner/private-repo")`.

**Q: Does devtools work outside of RStudio?**
A: Yes. All devtools functions work in any R console, terminal, or IDE. RStudio provides additional UI integration but is not required.

## Sources
- https://github.com/r-lib/devtools
- https://devtools.r-lib.org/

---
Source: https://tokrepo.com/en/workflows/asset-95b956f4
Author: AI Open Source