An environment to play with docker and docker compose. Simply run the repository in the GitHub codespaces.

This project is part of a blog post and provides a playground to learn.

You can play with docker itself, like running in terminal:

# Display help information for Docker (lists available commands and usage details)
docker --help

# List all Docker images currently available on your local system
docker images

# Display help information specifically for the 'docker images' command,
# including options and usage details for listing images
docker images --help

# Pull the v0.2 version of the Docker image 'polkas/rdevdash-app' from a registry (e.g., Docker Hub)
docker pull polkas/rdevdash-app:v0.2

# List Docker images again to confirm that the new image has been downloaded to your system
docker images

# Run a container from the 'polkas/rdevdash-app:v0.2' image, mapping port 3838 on your host machine to port 3838 in the container.
# This allows you to access services running inside the container via the host's IP on port 3838.
docker run -p 3838:3838 polkas/rdevdash-app:v0.2

# List all currently running containers to verify that the container has started
docker ps

# Stop all containers:
#   'docker ps -a -q' fetches the IDs of all containers (running or stopped),
#   and 'docker stop' stops them. This is a quick way to stop all running containers.
docker stop $(docker ps -a -q)

# Build a new Docker image using the Dockerfile located in the directory 'examples/py-shiny-app/'
# Tag the image as 'py-shiny-app'. If the build is successful (&&), list all Docker images.
docker build examples/py-shiny-app/ -t py-shiny-app && docker images

# Run a container from the newly built 'py-shiny-app' image, mapping port 3839 on the host to port 3839 in the container.
docker run -p 3839:3839 py-shiny-app

or you can reuse already existing images for Data Science.
Example of images from Jupyter.
Here is the list of jupyter images with descriptions

# https://quay.io/repository/jupyter/datascience-notebook
# You can and should use specific tag (version)
# docker run -p 8888:8888 quay.io/jupyter/datascience-notebook:2024-12-31
docker run -p 8888:8888 quay.io/jupyter/datascience-notebook

# https://quay.io/repository/jupyter/scipy-notebook
# You can and should use specific tag (version)
# docker run -p 8888:8888 quay.io/jupyter/scipy-notebook:2024-12-31
docker run -p 8888:8888 quay.io/jupyter/scipy-notebook

More challenging is to try to understand and work with Dockerfiles in examples.

• Path: examples/py-shiny-app
• Description: A simple Python Shiny app with a slider input and a plot output.
• Command to run: docker compose up --build py-shiny-app

Look for 3839 in the PORTS vscode tab

Press Control C in the terminal to close

• Path: examples/r-shiny-app
• Description: A simple R Shiny app with a slider input and a plot output.
• Command to run: docker compose up --build r-shiny-app

Look for 3838 in the PORTS vscode tab

Press Control C in the terminal to close

The current setup runs Shiny and Py Shiny apps on the server side. The repository does not include any implementation for running Shiny apps using WebAssembly or Shiny Live on the client side.

• Path: examples/jupyter
• Description: An environment to run Jupyter.
• Command to run: docker compose up --build jupyter

Look for 8888 in the PORTS vscode tab, look for a token printed in the bottom of the terminal.

Press Control C in the terminal to close

• Path: examples/rstudio
• Description: An environment to run RStudio.
• Command to run: docker compose up --build rstudio

Look for 8787 in the PORTS vscode tab, Login: rstudio and Password: rstudio

Press Control C in the terminal to close

The user-pw-list.txt file is used to manage user accounts for RStudio. It contains a list of users and their passwords in the following format:

username:password

Example:

rstudio:rstudio
user1:password1
user2:password2

To add more users, simply follow the format and add new lines to the file.

• Path: docker-compose.yml
• Description: A compose file to run all apps, including Shiny apps, JupyterLab, RStudio, at the same time.
• Command to run: docker compose up ; It can take more than 15+ minutes

This project includes a GitHub Actions CI workflow to automate the build and run process for all examples. The workflow is located in the .github/workflows/ci.yml file. It sets up Docker, builds and runs each example, and verifies that they work correctly. The workflow run the all apps simultaneously by docker-compose.yml.