Site Indexer #1326
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| name: Site Indexer | |
| on: | |
| schedule: | |
| - cron: '30 4 * * *' | |
| workflow_dispatch: | |
| inputs: | |
| force: | |
| description: 'Enter true to force the indexer to run.' | |
| required: false | |
| jobs: | |
| activate: | |
| runs-on: ubuntu-latest | |
| if: | | |
| (github.event_name == 'schedule' && github.repository_owner == 'asciidoctor') || | |
| github.event_name == 'workflow_dispatch' | |
| steps: | |
| - run: echo ok go | |
| build: | |
| needs: activate | |
| runs-on: ubuntu-latest | |
| container: algolia/docsearch-scraper:v1.12.0 | |
| steps: | |
| - name: Checkout | |
| uses: actions/checkout@v3 | |
| - name: Index | |
| env: | |
| APPLICATION_ID: ${{ secrets.ALGOLIA_APP_ID }} | |
| API_KEY: ${{ secrets.ALGOLIA_API_KEY }} | |
| CONFIG: ${{ github.workspace }}/docsearch/config.json | |
| run: | | |
| curl -sL -o $CONFIG https://docs.asciidoctor.org/docsearch-config.json | |
| INDEX_NAME=$(node -p "JSON.parse(require('fs').readFileSync('$CONFIG')).index_name") | |
| SITEMAP_URL=$(node -p "JSON.parse(require('fs').readFileSync('$CONFIG')).sitemap_urls[0].replace('.xml', '-ROOT.xml')") | |
| # only run indexer if index is more than 2 days out of date | |
| STALENESS_THRESHOLD=172800000 | |
| SITE_LAST_MODIFIED=$(curl -s $SITEMAP_URL | awk '/<lastmod>/{gsub(/<\/?lastmod>/,"");print;exit}' | node -p "Date.parse(require('fs').readFileSync(0).toString().trim()) - $STALENESS_THRESHOLD") | |
| SITE_LAST_INDEXED=$(curl -s -H X-Algolia-Application-Id:$APPLICATION_ID -H X-Algolia-API-Key:$API_KEY https://$APPLICATION_ID-dsn.algolia.net/1/indexes | node -p "Date.parse(JSON.parse(require('fs').readFileSync(0).toString()).items.find((it) => it.name === '$INDEX_NAME').updatedAt)") | |
| if [ "${{ github.event.inputs.force }}" = "true" ] || [ $SITE_LAST_MODIFIED -gt $SITE_LAST_INDEXED ]; then | |
| export HOME=/root | |
| export INDEX_NAME_TMP="${INDEX_NAME}_${GITHUB_RUN_ID}" | |
| cd $HOME | |
| pipenv run python -m src.index | |
| else | |
| echo 'SKIP: Index is already up to date.' | |
| fi |