Chapter 10 Extras
10.1 License
Choose A License is a great resource for choosing a license in general.
For R packages, a safe bet is MIT
Karl Broman has some good advice on choosing a license for your R package.
10.2 continuous integration
CI is integral to a healthy package workflow.
The “continuous” part means each deployment to github/gitlab is checked.
10.3 vignettes
TLDR: if nothing else, you should have examples in your README, correct? Just make a vignette out of that.
vignettes are super important - if users don’t know how to use your package they won’t use it.
See R pkgs vignettes chapter for more.
10.4 resources
if/when you run into trouble:
- Writing an R package from scratch - Hilary Parker
- Developing R packages - Jeff Leek
- stat545 - Write your own R package - Jenny Bryan’s statistics 545 class at UBC
- R package primer - Karl Boman
- Making Your First R Package - Fong Chun Chan
- R Package Development Pictorial - Matthew Denny
- Coursera course on building R packages
- R Packages by Hadley for a full treatment of the subject.
- From time to time you may need to reference CRAN’s Writing R Extensions.
10.5 CRAN/Bioconductor
No experience with Bioconductor, though I can put you in touch with people that do.
CRAN is a slippery target. devtools::release() makes this easier, reminding you to check many things that may get your package rejected.