Lessons Learned from Building R Packages

Practical tips and hard-won lessons from developing the Silhouette and blockclusterPDQ R packages — from CRAN submission to documentation.

From Research Code to CRAN Package

Turning a collection of research scripts into a polished, CRAN-ready R package is a journey. Here are some lessons I’ve learned while developing Silhouette and blockclusterPDQ.

1. Start with usethis and devtools

# Scaffold your package
usethis::create_package("myPackage")
usethis::use_testthat()
usethis::use_vignette("getting-started")

2. Write Tests Early

Don’t wait until the package is “done” — write tests as you code each function.

3. Document with roxygen2

Keep documentation close to the code. Future-you will thank present-you.

4. CRAN Submission Tips

• Run R CMD check --as-cran locally before submitting
• Pay attention to NOTEs — CRAN reviewers will
• Keep examples fast (< 5 seconds each)

More Details Coming Soon

I’ll expand each of these into detailed posts with real examples from my package development experience.

Stay tuned!