Merge pull request #12 from vespa-engine/feature-jekyll-build-json-ac…

…tion

feat: Add Jekyll-based build for Vespa Search

.github/workflows/docker-publish.yml

@@ -0,0 +1,95 @@
+name: Docker
+
+# This workflow will generate container images for each of the Github Actions
+# which are using a Dockerfile.
+
+on:
+  push:
+    branches:
+      - "main"
+      - "feature-*"
+  pull_request:
+    branches:
+      - "main"
+
+jobs:
+  prepare:
+    # Finds all the Dockerfiles in the repository and sets the action names
+    # as output to reuse as matrix strategy in the build job
+    runs-on: ubuntu-latest
+
+    outputs:
+      container_names_json: ${{ steps.set-output.outputs.container_names_json }}
+
+    steps:
+      - uses: actions/checkout@v4
+
+      - name: Find Dockerfiles
+        id: set-output
+        run: |
+          # Find all Dockerfiles in the repository
+          paths_with_docker=$(find -type f -name 'Dockerfile' | sed 's|/Dockerfile||' | sed 's|./||')
+          # Transform the list of paths into a valid JSON array to be used as matrix strategy
+          action_names=$(echo $paths_with_docker | jq --raw-input --compact-output 'split(" ") |  . ')
+
+          echo "Found Dockerfiles in the repository:"
+          echo "${action_names}"
+
+          echo "container_names_json=${action_names}" >> "${GITHUB_OUTPUT}"
+
+  build:
+    runs-on: ubuntu-latest
+
+    needs:
+      - prepare
+
+    permissions:
+      contents: read
+      packages: write
+
+    strategy:
+      matrix:
+        action-name: ${{fromJson(needs.prepare.outputs.container_names_json)}}
+        docker-platform: [linux/amd64, linux/arm64]
+
+    env:
+      REGISTRY: ghcr.io
+      IMAGE_NAME: ${{ github.repository }}/${{ matrix.action-name }}
+
+    steps:
+      - uses: actions/checkout@v4
+
+      - name: Log into registry ${{ env.REGISTRY }}
+        uses: docker/login-action@v3
+        with:
+          registry: ${{ env.REGISTRY }}
+          username: ${{ github.actor }}
+          password: ${{ secrets.GITHUB_TOKEN }}
+
+      - name: Set up QEMU
+        uses: docker/setup-qemu-action@v3
+
+      - name: Set up Docker Buildx
+        uses: docker/setup-buildx-action@v3
+
+      - name: Extract Docker metadata
+        id: meta
+        uses: docker/metadata-action@v5
+        with:
+          images: ${{ env.REGISTRY }}/${{ env.IMAGE_NAME }}
+          tags: |
+            type=ref,event=pr
+            type=ref,event=branch
+            type=raw,value=latest,enable={{is_default_branch}}
+
+      - name: Build and push Docker image
+        uses: docker/build-push-action@v6
+        with:
+          context: ${{ matrix.action-name }}
+          file: ${{ matrix.action-name }}/Dockerfile
+          push: true
+          tags: ${{ steps.meta.outputs.tags }}
+          labels: ${{ steps.meta.outputs.labels }}
+          platforms: ${{ matrix.docker-platform }}
+          cache-from: type=gha
+          cache-to: type=gha,mode=max

.github/workflows/jekyll-feed-to-vespa.yml

@@ -0,0 +1,33 @@
+---
+name: (WIP) Feed Jekyll posts to Vespa
+
+on:
+  workflow_call:
+    inputs:
+      log_level:
+        description: 'Log level for Jekyll build.'
+        required: false
+        default: 'info'
+        type: string
+
+defaults:
+  run:
+    # Specify to ensure "pipefail and errexit" are set.
+    # Ref: https://docs.github.com/en/actions/writing-workflows/workflow-syntax-for-github-actions#defaultsrunshell
+    shell: bash
+
+jobs:
+  lint:
+    runs-on: ubuntu-latest
+
+    steps:
+      - uses: actions/checkout@v4
+
+      - name: Build with Jekyll
+        uses: vespa-engine/gh-actions/jekyll-build-json@main
+        with:
+          log_level: ${{ github.event.inputs.log_level }}
+
+      - name: List JSON
+        run: |
+          find _site -type f -name '*.json'

jekyll-build-json/Dockerfile

@@ -0,0 +1,30 @@
+ARG RUBY_VERSION=3.3
+FROM ruby:$RUBY_VERSION-slim
+
+RUN apt-get update \
+  && apt-get install -y \
+  build-essential \
+  git \
+  locales
+
+COPY Gemfile Gemfile
+
+ENV NOKOGIRI_USE_SYSTEM_LIBRARIES=true
+
+RUN gem env && \
+  bundle install --jobs 20 --retry 5
+
+RUN \
+  echo "en_US UTF-8" > /etc/locale.gen && \
+  locale-gen en-US.UTF-8
+
+ENV LANG=en_US.UTF-8
+ENV LANGUAGE=en_US.UTF-8
+ENV LC_ALL=en_US.UTF-8
+
+COPY entrypoint.sh /entrypoint.sh
+
+RUN mkdir -p /opt/jekyll/plugins
+COPY vespa_index_generator.rb /opt/jekyll/plugins/vespa_index_generator.rb
+
+ENTRYPOINT ["/entrypoint.sh"]

jekyll-build-json/Gemfile

@@ -0,0 +1,17 @@
+# Copyright Vespa.ai. Licensed under the terms of the Apache 2.0 license. See LICENSE in the project root.
+# frozen_string_literal: true
+
+source "https://rubygems.org"
+
+# Manage our dependency on the version of the github-pages gem here.
+gem "github-pages", "= 232"
+
+# Explicitly include this gem here.
+# It is not directly included in the github-pages gem list of dependencies,
+# even though it is included in the original GitHub Pages build infrastructure.
+gem "jekyll-include-cache", "= 0.2.1"
+gem "jekyll-octicons", "~> 14.2"
+gem "minima", "~> 2.0"
+
+# Work-around for webrick no longer included in Ruby 3.0 (https://github.com/jekyll/jekyll/issues/8523)
+gem "webrick"

jekyll-build-json/action.yml

@@ -0,0 +1,27 @@
+name: 'Build Jekyll for Vespa Search'
+description: 'A simple GH Action for producing Jekyll build artifacts compatible with Vespa JSON'
+author: 'Vespa.ai'
+inputs:
+  source:
+    description: 'Directory where the source files reside (relative to the workspace).'
+    required: false
+    default: ./
+  destination:
+    description: 'Output directory of the build. Although it can be nested inside the source, it cannot be the same as the source directory (relative to the workspace).'
+    required: false
+    default: ./_site
+  config:
+    description: 'Jekyll configuration file path relative to the workspace root.'
+    required: false
+    default: ./_config.yml
+  log_level:
+    description: 'Log level for Jekyll build.'
+    required: false
+    default: 'info'
+  token:
+    description: 'GitHub token'
+    required: true
+    default: ${{ github.token }}
+runs:
+  using: 'docker'
+  image: 'docker://ghcr.io/vespa-engine/gh-actions/jekyll-build-json'

jekyll-build-json/entrypoint.sh

@@ -0,0 +1,58 @@
+#!/bin/bash
+
+if [[ "${RUNNER_DEBUG:-}" == "1" ]]; then
+  set -x
+fi
+
+set -e
+set -o pipefail
+set -o nounset
+
+SOURCE_DIRECTORY="${GITHUB_WORKSPACE}/${INPUT_SOURCE}"
+DESTINATION_DIRECTORY="${GITHUB_WORKSPACE}/${INPUT_DESTINATION}"
+JEKYLL_CONFIG_FILE="${GITHUB_WORKSPACE}/${INPUT_CONFIG}"
+JEKYLL_ARGS=("-p" "/opt/jekyll/plugins")
+
+# Set environment variables required by supported plugins
+export JEKYLL_ENV="production"
+export JEKYLL_GITHUB_TOKEN="${INPUT_TOKEN}"
+export JEKYLL_BUILD_REVISION="${GITHUB_SHA}"
+export JEKYLL_LOG_LEVEL="${INPUT_LOG_LEVEL}"
+
+if [[ "${JEKYLL_LOG_LEVEL}" = "verbose" || "${JEKYLL_LOG_LEVEL}" = "debug" ]]; then
+  JEKYLL_ARGS+=("-V")
+fi
+
+if [[ "${RUNNER_DEBUG:-}" == "1" ]]; then
+  gem env
+  bundle env
+
+  echo "SOURCE_DIRECTORY: ${SOURCE_DIRECTORY}"
+  echo "DESTINATION_DIRECTORY: ${DESTINATION_DIRECTORY}"
+  echo "JEKYLL_CONFIG_FILE: ${JEKYLL_CONFIG_FILE}"
+  echo "GITHUB_WORKSPACE: ${GITHUB_WORKSPACE}"
+
+  echo "====== START Config file ====="
+  cat "${JEKYLL_CONFIG_FILE}"
+  echo "====== END Config file ====="
+fi
+
+{ cd "${BUNDLE_APP_CONFIG}" || { echo "::error::pages gem not found"; exit 1; }; }
+
+# Run the command, capturing the output
+build_output="$(bundle exec jekyll build ${JEKYLL_ARGS[@]} --config "${JEKYLL_CONFIG_FILE}" --source "${SOURCE_DIRECTORY}" --destination "${DESTINATION_DIRECTORY}")"
+
+# Capture the exit code
+exit_code=$?
+
+if [ $exit_code -ne 0 ]; then
+  # Remove the newlines from the build_output as annotation not support multiline
+  error=$(echo "$build_output" | tr '\n' ' ' | tr -s ' ')
+  echo "::error::$error"
+else
+  # Display the build_output directly
+  echo "$build_output"
+fi
+
+# Exit with the captured exit code
+exit $exit_code