Package Management

Last updated: September 29, 2019

Package management

Packrat is the standard package management tool for R. It’s not perfect, but it’s better than nothing.

Possible alternatives include pak (a R Infrastructure package that is an alternative to install.packages() and devtools::install_github()) and renv (from RStudio).

Standard workflow

• Install the packrat package (you only need to do this once for your system): install.packages("packrat")
• Initialization to use packrat with a new project: packrat::init("/path/to/rstudio/project")
  • Or you can turn it on via the RStudio interface.
• Install package: install.packages("packagename") and then call packrat::snapshot() to save the changes to Packrat.
• Install everything specified in the latest snapshot: packrat::restore(). This is what you would do if you just cloned a git repository that uses Packrat to install all the necessary dependencies.
  • You can do this with RStudio by going to the “Packrat” dropdown in the “Packages” tab, and then choosing “Check Library Status…”
• Using Packrat with git: packrat/lib* is gitignored by default. You can also gitingore packrat/src if you don’t want to commit the source for all dependencies to your repo (other people with have to re-download everything themselves, which is probably fine).
  • I typically do ignore packrat/src to avoid bloat in the repo.

Storing installed packages in a custom library location

.libPaths(c("/path/to/folder", .libPaths()))

After you run that, install.packages() will install in that folder, and library() will try to load from that folder.

Installing a specific version of a package

Using the remotes package

remotes::install_github("user/repo@v1.1.1")

Using the devtools package

devtools::install_version()

ℹ️ This page is part of my R knowledge base, which attempts to use idiomatic practices with the tidyverse collection of packages as much as possible. If you have suggestions for ways to improve this code, please contact me.