GDS-R: a GDS containerised environment for RStudio

This project extends the fantastic rocker-geospatial to include further libraries required by the activities of the Geographic Data Science Lab.

Containers are available at Docker Hub

Install

To run gdsr you will need Docker on your machine. Once ready, you can install the container by typing:

docker pull darribas/gdsr

Or if you want a Python distribution as well:

docker pull darribas/gdsrpy

Make sure to have a stable and good internet connection as the images are about 8G and 13G, respectively.

Use

This image builds directly on the Rocker-geospatial project, and the only difference is that it comes loaded with additional libraries. Hence, usage is similar to that of any Rocker image. Please check their documentation for guidance.

For version 2.0, you can boot up the server by:

docker run --rm -ti -e PASSWORD=<your password> -v /folder/to/mount:/home/rstudio/work -p 8787:8787 gdsr

This will mount a folder from the host on the work folder on the container side, which you can see and access when you login.

Build [Optional]

NOTE - This is not required or recommended unless you know what you're doing!

If, instead of installing a built image, you prefer to build your own locally, you will need a copy of this repository. Once that is available in your machine, navigate into it:

cd /path/to/repo/folder/

And you can build the image with:

docker build -t gdsr .