Doing Data Science using R: A Guide for Social Scientists

Building

We assume that this will be built from inside an RStudio (server) session running in a docker container.

Create docker image

The docker image can be created from the accompanying Dockerfile. Run

make docker

to create this image. The resulting image will always be called janacek:latest.

This image is also available on DockerHub and can be obtained with the docker pull command:

docker pull xmjandrews/janacek:latest

Run docker container

The shell script run_docker.sh will run the container and allow the rocker based RStudio server session to accessed through the browser. Do

source docker/run_docker.sh 

The open a browser at http://localhost:8788. Log in with username "rstudio" and password "foo".

Make individual chapters

In the RStudio server session, open the Linux terminal (not R console). Do the following to access the book's home directory.

cd book

Next, source the setenv.sh, which sets up the environment for building the book.

source setenv.sh

Next, install sparklyr. (For some reason, this does not install automatically when making the docker container, despite being in the Dockerfile as a run command.)

Rscript -e "sparklyr::spark_install()"

The following script will make all chapter pdfs and copy them to a directory called build:

bash build.sh

The following script will make all chapter slides and copy them to a directory called build_slides:

bash build_slides.sh