r - How to build a Docker image with the right R and package versions and open Rstudio with them afterwards?

Question

I just finished writing a manuscript for publication in a scientific journal. I want my research to be reproducible, so besides sharing my original and compiled R markdown notebooks on Github, I would like to save the environment in which I was analysing my data (including the data, notebooks and specific R and package versions) in a Docker container. Moreover, I would like anyone trying to reproduce my work to be able to execute this code in an interactive Rstudio session.

I was able to create a Dockerfile with the correct environment. Here is a toy example:

FROM rocker/r-ver:3.5.1

RUN mkdir /home/working_directory
RUN mkdir /home/working_directory/bin

RUN R -e 'options(repos = \
  list(CRAN = "http://mran.revolutionanalytics.com/snapshot/2019-01-01")); \
  install.packages("ggplot2")'

COPY current/0[1-8]-*.Rmd /home/working_directory/
COPY current/bin/utils.R /home/working_directory/bin/

RUN R

However, this doesn't allow the users to read the Rmd notebooks and execute the code line by line. A workaround is to run the image rocker/rstudio, and install the packages from there, but I would like to be able to do so from a single docker build call. Unfortunately, I wasn't able to do so.

Cheers!

标签： rdocker

Answer 1

Rocker RStudio图像应该有你需要的东西 - RStudio 位于 R 之上。如果你特别想要 R 3.5.1，你可以在这里找到源代码：

• https://github.com/rocker-org/rocker-versioned/blob/master/rstudio/3.5.1/Dockerfile

有关如何使用它的更多详细信息，请阅读https://hub.docker.com/r/rocker/rstudio。

仅使用具有特定 R 版本的预构建图像

# with R version 3.5.1:
docker pull rocker/rstudio:3.5.1

以交互方式研究图像：

docker run --rm -it rocker/rstudio:3.5.1 bash

现在我们可以运行例如：

# Check R version
Rscript -e "sessionInfo()[['R.version']][['version.string']]"
# [1] "R version 3.5.1 (2018-07-02)"

# Check RStudio Server version
rstudio-server version
# 1.1.463

# Exit when done
exit