How to improve your R package (Automatically ✨, and not 🧠)



How can you write a good R package? Read a starter guide, the “R packages” book, “Writing R Extensions”, and then? This talk will feature ways to take your R package, and your R development skills, to the next level! We shall present tools that equip you with helpful flags and metrics, from R CMD check locally and on the cloud, to lintr::lint_package() and covr::package_coverage(). We shall also mention tools that automagically improve your package or its docs: the styler and pkgdown packages. Furthermore, we shall explain how rOpenSci Software Peer Review helps package authors receive human and humane feedback. Finally, we shall advocate for learning from others’ experience, be it by directly reading code; or by reading blogs such as the R-hub blog and developers’ forums such as R-package-devel mailing list. We don’t aim to go through a catalogue: we hope you’ll get away from this talk with one or a few lifechanging habits for your daily R development work.