schema-bq.yml

# This workflow is used to push changed HTAN.model.csv 
# to BigQuery table located at `htan-dcc.metadata.data-model`

name: Add updated schema to BigQuery

on:
  push:
    branches: main
    paths: 'HTAN.model.csv'
  workflow_dispatch: 

jobs:
  add-to-bq:
    name: Add schema to BQ
    runs-on: ubuntu-latest

    permissions:
      contents: 'read'
      id-token: 'write'

    steps:
      - uses: actions/checkout@v2
      - uses: actions/setup-python@v4 
        with:
          python-version: '3.9'

      - id: 'auth'
        name: 'Authenticate to Google Cloud'
        uses: 'google-github-actions/auth@v1'
        with:
          workload_identity_provider: ${{ secrets.GOOGLE_WIF }}
          service_account: 'bq-schema@htan-dcc.iam.gserviceaccount.com'

      - name: 'Set up Cloud SDK'
        uses: 'google-github-actions/setup-gcloud@v1'
        with:
          version: '>= 363.0.0'

      - name: 'Use gcloud CLI'
        run: 'gcloud info'

      - name: Install Google Cloud BigQuery Client
        shell: bash
        run: |
          pip3 install google-cloud-bigquery

      - name: Load .csv to BQ with defined schema
        shell: bash
        run: |
          bq load --source_format=CSV \
          --replace=true \
          --skip_leading_rows=1 \
          htan-dcc:metadata.data-model \
          HTAN.model.csv \
          ./.github/workflows/BQ.table.schema.json