Introduction

This folder builds a docker container with all the pre-requisites for the FSDA Toolbox. In addition there are automated github actions to build the same docker container and store it in ghcr.io so that users can simply pull the pre-built container.

Getting Pre-Built Images

An example of how to pull a specific instance (both FSDA version and MATLAB version) is given below:

docker pull ghcr.io/uniprjrc/fsda:8.5.29-r2022a

Fo a full list of the available tags for these docker containers please see the repository packages.

Running Container

This container is built directly FROM mathworks/matlab:latest and has all the same command-line arguments as listed on its help page.

Build from Repo

To build this container clone the repository and execute

docker build -t fsda --build-arg FSDA_RELEASE=${REQUIRED_RELEASE} ./docker

You can specify previous versions of FSDA by modifying the FSDA_RELEASE=X.Y.Z build argument (see available releases).

By default this will build against the latest MATLAB docker container but you can specify a particular release by setting the build-arg MATLAB_DOCKER_TAG to something like r2022a (or any other tag supported for the docker container).

If you wish to install from your own forked repo of UniprJRC/FSDA then please set the build-arg SRC_REPO to the name of your own github repository.

GitHub Automated Build

The automated build procedure is driven when a specific tag of the form X.Y.Z is pushed to the repository. In addition you can trigger a manual build of a particular release using the GitHub CLI (or browse to the action on github). The CLI trigger action is something like

gh workflow run -F fsda-release=8.5.29 -F matlab-release=r2022a build-docker-container

This will trigger a build that results in a container called ghcr.io/uniprjrc/fsda:8.5.29-r2022a.