andybotting / r

Using R with Jupyter / RStudio on Binder

Specifying an R environment with a runtime.txt file

Jupyter+R:

RStudio:

RShiny:

Binder supports using R and RStudio, with libraries pinned to a specific snapshot on packagemanager.rstudio.com.

Requirements and suggestions

You need to have a runtime.txt file that is formatted like:

r-<r-version>-<YYYY>-<MM>-<DD>

where <r-version> is a version of R (like 4.1, 4.0, etc) you want to use, and <YYYY>-<MM>-<DD> is the date for a snapshot from packagemanager.rstudio.com that will be used for installing your R packages.

Try using a date newer than 2022-01-01, as you'll get faster package installs thanks to binary packages from rstudio.packagemanager.com!

To install R libraries, add install.package("<package-name>") calls to install.R. If you want to pin to a specific version of the library, you can also do devtools::install_version("<package-name>", "<version>").

For some R packages, you might need to install system packages via apt - you can do so by writing out a list of apt package names in apt.txt. You can find the list of such packages in the page for your package at packagemanager.rstudio.com. Make sure to select "Ubuntu 22.04 (Jammy)" in the dropdown on the top right.

Both RStudio and IRKernel are installed by default, so you can use either the Jupyter notebook interface or the RStudio interface.

This repository also contains an example of a Shiny app.