feat: Publish Docker image

.github/workflows/docker.yml

@@ -0,0 +1,41 @@
+name: Publish Docker image
+
+on:
+  release:
+    types: [published]
+
+permissions:
+  contents: read
+
+jobs:
+  push_to_registry:
+    runs-on: ubuntu-latest
+    permissions:
+      packages: write
+      contents: read
+      attestations: write
+      id-token: write
+    steps:
+      - uses: actions/checkout@v4
+      - uses: docker/login-action@v3
+        with:
+          registry: ghcr.io
+          username: ${{ github.repository_owner }}
+          password: ${{ secrets.GITHUB_TOKEN }}
+      - uses: docker/metadata-action@v5
+        id: meta
+        with:
+          images: ghcr.io/${{ github.repository }}
+      - uses: docker/build-push-action@v5
+        id: push
+        with:
+          context: .
+          file: docker/Dockerfile
+          push: true
+          tags: ${{ steps.meta.outputs.tags }}
+          labels: ${{ steps.meta.output.labels }}
+      - uses: actions/attest-build-provenance@v1
+        with:
+          subject-name: ghcr.io/${{ github.repository }}
+          subject-digest: ${{ steps.push.outputs.digest }}
+          push-to-registry: true

README.md

@@ -34,12 +34,24 @@ so users will not be locked into a proprietary tool or technology.
 
 ## Usage
 
-To get started with LD Workbench, first install [NodeJS](https://nodejs.org), then run:
+To get started with LD Workbench, you can either use the NPM package or a Docker image.
+
+To use the NPM package, install [Node.js](https://nodejs.org), then run:
 
 ```sh
 npx @netwerk-digitaal-erfgoed/ld-workbench@latest --init
 ````
 
+Alternatively, to run the Docker image,
+first create a directory to store your pipeline configurations,
+then run the Docker image (mounting the `pipelines/` directory with `-v` and using `-t` for colorful console output):
+
+```sh
+mkdir pipelines
+docker run -t -v $(pwd)/pipelines:/pipelines ghcr.io/netwerk-digitaal-erfgoed/ld-workbench:latest
+```
+
+
 This creates an example LD Workbench pipeline in the `pipelines/configurations/example` directory
 and runs that pipeline right away. The output is written to `pipelines/data`.
 
@@ -61,7 +73,9 @@ You can find more examples in the [ld-workbench-configuration](https://github.co
 
 #### Iterator
 
-Each stage has a single iterator. The iterator SPARQL SELECT query must return a `$this` binding for each URI that will be passed to the generator(s).
+Each stage has a single iterator.
+The iterator selects URIs from a dataset that match certain criteria.
+The iterator SPARQL SELECT query must return a `$this` binding for each URI that will be passed to the generator(s).
 
 The query can be specified either inline:
 

docker/Dockerfile

@@ -0,0 +1,8 @@
+FROM node:lts-alpine AS build
+
+RUN npm install -g @netwerk-digitaal-erfgoed/ld-workbench@latest \
+    && npm cache clean --force
+
+COPY docker/entrypoint.sh /entrypoint.sh
+ENTRYPOINT ["/entrypoint.sh"]
+USER node

docker/entrypoint.sh

@@ -0,0 +1,7 @@
+#!/bin/sh
+
+if [ -d "/pipelines/configurations/example" ]; then
+  ld-workbench $@
+else
+  ld-workbench --init
+fi