Update Knowledge Graph with data from dataset 'Wereldmuseum Thesaurus', based on IRIs in the Knowledge Graph #645
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: Update Knowledge Graph with data from dataset 'Wereldmuseum Thesaurus', based on IRIs in the Knowledge Graph | |
| # "At minute 0 past hour 6 and 18" | |
| on: | |
| schedule: | |
| - cron: "0 6,18 * * *" | |
| # Run a single workflow at a time | |
| concurrency: | |
| group: knowledge-graph | |
| jobs: | |
| run: | |
| runs-on: ubuntu-latest | |
| timeout-minutes: 120 | |
| steps: | |
| - name: Checkout data repository | |
| uses: actions/checkout@v4 | |
| with: | |
| path: ./data | |
| sparse-checkout: wereldmuseum-thesaurus | |
| - name: Checkout code repository | |
| uses: actions/checkout@v4 | |
| with: | |
| path: ./code | |
| repository: colonial-heritage/integration-layer | |
| - name: Install Node | |
| uses: actions/setup-node@v4 | |
| with: | |
| node-version: "20" | |
| - name: Install app | |
| run: | | |
| cd ./code | |
| npm install --no-progress | |
| npx turbo run build --filter=@colonial-collections/graph-create | |
| - name: Create graph | |
| run: | | |
| cd ./data | |
| mkdir -p "$RUNNER_TEMP" | |
| test -f ./wereldmuseum-thesaurus/data/data.tar.zst && tar --zstd -xf ./wereldmuseum-thesaurus/data/data.tar.zst -C "$RUNNER_TEMP" | |
| ../code/apps/graph-create/dist/cli.js create \ | |
| --resource-dir ./wereldmuseum-thesaurus/resources \ | |
| --data-file "$RUNNER_TEMP/data.sqlite" \ | |
| --iterate-endpoint-url "${{ vars.SPARQL_ENDPOINT_URL_KG }}" \ | |
| --iterate-query-file ./wereldmuseum-thesaurus/queries/iterate.rq \ | |
| --generate-endpoint-url "https://api.colonialcollections.nl/datasets/nmvw/thesaurus/sparql" \ | |
| --generate-query-file ./wereldmuseum-thesaurus/queries/generate.rq \ | |
| --generate-batch-size 5000000 \ | |
| --triplydb-instance-url "${{ vars.TRIPLYDB_INSTANCE_URL }}" \ | |
| --triplydb-api-token "${{ secrets.TRIPLYDB_API_TOKEN }}" \ | |
| --triplydb-account "${{ vars.TRIPLYDB_ACCOUNT }}" \ | |
| --triplydb-dataset "${{ vars.TRIPLYDB_DATASET_KG }}" \ | |
| --triplydb-service "kg" \ | |
| --graph-name "https://data.colonialcollections.nl/wereldmuseum-thesaurus" \ | |
| --temp-dir "$RUNNER_TEMP" | |
| - name: Save changes | |
| run: | | |
| cd ./data | |
| mkdir -p ./wereldmuseum-thesaurus/data | |
| test -f "$RUNNER_TEMP/data.sqlite" && tar --zstd -cf ./wereldmuseum-thesaurus/data/data.tar.zst -C "$RUNNER_TEMP" data.sqlite | |
| git config --global user.name "github-actions[bot]" | |
| git config --global user.email "github-actions[bot]@users.noreply.github.com" | |
| git add . | |
| git commit --quiet -a -m "Save changes" || true | |
| git push --force -u origin |