nested_works_xml_upload.yml

name: Upload Works TEI Files to S3

on:
  push:
    branches:
      - 'gaddel_development'
    paths:
      - 'works/tei/**/tei/*.xml'

permissions:
  id-token: write
  contents: read

jobs:
  upload-works:
    runs-on: ubuntu-latest

    steps:
      - name: Checkout repository
        uses: actions/checkout@v4
        with:
          fetch-depth: 0

      - name: Configure AWS credentials
        uses: aws-actions/configure-aws-credentials@v2
        with:
          role-to-assume: ${{ secrets.AWS_SROPHE_ROLE }}
          aws-region: us-east-1
          role-session-name: GitHub-OIDC-data
          role-duration-seconds: 14400
          
      # Optional: for full data upload
      # - name: Identify XML files
        # run: |
        #   mkdir -p logs
        #   find ./data/works/tei -type f -path '*/tei/*.xml' > logs/xml_files.txt
        #   count=$(wc -l < logs/xml_files.txt)
        #   echo "Found $count XML file(s) to upload."
        
      - name: Identify updated XML files
        id: files
        run: |
          echo "Fetching full commit history..."
          git fetch --unshallow || echo "Already unshallowed."
      
          echo "Looking for added/modified XML files under .data/works/tei/**/tei/"
          UPDATED_FILES=$(git diff --name-only --diff-filter=AM origin/${{ github.event.before }} HEAD | grep '^\.data/works/tei/.*/tei/.*\.xml$' || true)
      
          if [ -z "$UPDATED_FILES" ]; then
            echo "No updated XML files in .data/works/tei/**/tei/"
            exit 0
          fi
      
          mkdir -p logs
          echo "$UPDATED_FILES" > xml_files.txt
      
          COUNT=$(wc -l < logs/xml_files.txt)
          echo "Found $COUNT updated XML files"
  
      - name: Upload XML files to S3
        run: |
          uploaded=0
          while IFS= read -r file; do
            filename=$(basename "$file")
            echo "Uploading $file to s3://srophe-syriaca-front-end/work/$filename"
            if aws s3 cp "$file" "s3://srophe-syriaca-front-end/work/$filename"; then
              uploaded=$((uploaded + 1))
            else
              echo "Failed to upload: $file"
            fi
          done < xml_files.txt

          echo "Total files uploaded: $uploaded"