deployment.md

Apache File Directory — Deployment

Three ways to put the page somewhere. All of them serve the same static files.

On your own web server

1. Download the latest .zip from GitHub Releases.
2. Extract it.
3. Copy index.html and the icons/ folder to the directory you want listed.
4. Edit index.html to describe the files that are actually there — see Usage.
5. Upload.

The icons/ folder is easy to forget. Without it the page renders with broken image placeholders in every row, because the icons are referenced by relative path.

With Docker

An image is published to GitHub Packages:

docker pull ghcr.io/willtheorangeguy/apache-file-directory:main
docker run -d -p 8000:80 ghcr.io/willtheorangeguy/apache-file-directory:main

Then open http://localhost:8000/.

The image name is lowercase. GHCR requires it, so Apache-File-Directory will not pull even though that is the repository name.

The container is Nginx serving the static files, so it is a convenient way to preview the page rather than a production requirement — any web server will do.

Building the image yourself

docker build -t afd .
docker run -d -p 8000:80 afd

Useful when you have edited index.html and want to see your version rather than the published one.

GitHub Pages

The docs.yml workflow deploys on every push to main, which is how https://williamvdg.me/Apache-File-Directory/ stays current.

Forking and enabling Pages gives you the same for your own copy — a hosted directory listing with no server to run.

Automation

Workflow	Trigger	Does
docs.yml	push to main	Deploys the app and docs
docker-publish.yml	push to main	Builds and pushes to GHCR
gitleaks.yml	pushes and PRs	Scans for committed secrets

Dependabot updates the Actions and the Docker base image daily.

There is nothing to configure

No environment variables, no database, no build step. Deployment is copying files. That is the whole point of the design — see Architecture.