This repo holds some dummy data, which exists purely for workflow tests with build systems.
The contents of the
poems directory might have various things done with
Pushes to this repo are automatically built at Docker Hub into the image
We generate a Docker image. This is silly for data this small.
Normally you use a Docker image to hold "something executable", to treat a "userland contents" bundle like a program. In this case, there's nothing to execute. Thus usually you'd use some kind of tarball/zipfile to hold this data.
But for much larger data-sets, being able to use Docker image caching caching pays off, as you just pull when there's a difference. You can build repeatedly and get proper caching across builds which otherwise don't share content.
There are security model trade-offs involved, around private/sensitive data. Think carefully before going this route. But if you're going to, then here's some trivial sample data, consisting of a very few poems, which can help you test the workflows.
Ideally, you want to use a
VOLUME directive in your Dockerfile, and be able
to satisfy the source of that Volume directly from a Docker image. Instead,
you'll probably need some kind of workaround.