github-pages-workflow.yml

name: Exportera till HTML och publicera till GitHub Pages

on:
  workflow_dispatch:  # Tillåter manuell körning
    inputs:
      html_source:
        description: 'HTML-källan: generera från JSON eller ladda från R2'
        required: false
        type: choice
        options:
          - generate
          - download
        default: 'generate'
      source_ref:
        description: 'Git ref att bygga från'
        required: false
        default: 'main'
      filter:
        description: 'Filtrera filer efter år (YYYY) eller specifik beteckning (YYYY:NNN). Kommaseparerad lista.'
        required: false
        type: string
  workflow_call:  # Tillåter anrop från andra workflows
    inputs:
      html_source:
        required: false
        type: string
        default: 'generate'
      source_ref:
        required: false
        type: string
        default: 'main'
      filter:
        required: false
        type: string

# Behörigheter för GitHub Pages deployment
permissions:
  contents: read
  pages: write
  id-token: write

# Tillåt endast en deployment åt gången
concurrency:
  group: "pages"
  cancel-in-progress: false

jobs:
  build:
    runs-on: ubuntu-latest
    environment: Test

    steps:
    - name: Checkout repository
      uses: actions/checkout@v4
      with:
        ref: ${{ inputs.source_ref || 'main' }}

    - name: Download pre-built HTML from Cloudflare R2
      if: inputs.html_source == 'download'
      run: |
        echo "📥 Laddar ner färdiga HTML-filer från Cloudflare R2..."

        # Configure AWS CLI for R2
        aws configure set aws_access_key_id ${{ secrets.CLOUDFLARE_R2_ACCESS_KEY_ID }}
        aws configure set aws_secret_access_key ${{ secrets.CLOUDFLARE_R2_SECRET_ACCESS_KEY }}
        aws configure set region us-east-1
        aws configure set output json

        # Create output directory
        mkdir -p _site

        # Download all HTML files from R2 HTMLEXPORT bucket
        echo "Syncing from bucket: ${{ vars.CLOUDFLARE_R2_HTMLEXPORT_BUCKET_NAME }}"
        aws s3 sync s3://${{ vars.CLOUDFLARE_R2_HTMLEXPORT_BUCKET_NAME }}/ _site/ \
          --endpoint-url https://${{ secrets.CLOUDFLARE_R2_ACCOUNT_ID }}.r2.cloudflarestorage.com \
          --delete

        # Verify download
        if [ ! -d "_site" ] || [ -z "$(ls -A _site)" ]; then
          echo "::error::Failed to download HTML files from R2"
          exit 1
        fi

        echo "✅ Downloaded HTML files from R2 HTMLEXPORT bucket"
        echo "Total files: $(find _site -type f | wc -l)"
        echo "Total size: $(du -sh _site | cut -f1)"
      env:
        AWS_DEFAULT_REGION: us-east-1

    - name: Set up Python
      if: inputs.html_source == 'generate' || inputs.html_source == ''
      uses: actions/setup-python@v4
      with:
        python-version: '3.11'

    - name: Install dependencies
      if: inputs.html_source == 'generate' || inputs.html_source == ''
      run: |
        python -m pip install --upgrade pip
        pip install -r requirements.txt

    - name: Get JSON source files (from git or R2)
      if: inputs.html_source == 'generate' || inputs.html_source == ''
      run: |
        # Try to use JSON files from git first
        if [ -d "data/sfs_json" ] && [ -n "$(ls -A data/sfs_json 2>/dev/null)" ]; then
          echo "✅ Found $(find data/sfs_json -name '*.json' | wc -l) JSON files in git"
          echo "Using JSON files from git checkout"
        else
          echo "⚠️  No JSON files in git, downloading from Cloudflare R2..."

          # Configure AWS CLI for R2
          aws configure set aws_access_key_id ${{ secrets.CLOUDFLARE_R2_ACCESS_KEY_ID }}
          aws configure set aws_secret_access_key ${{ secrets.CLOUDFLARE_R2_SECRET_ACCESS_KEY }}
          aws configure set region us-east-1
          aws configure set output json

          # Download all JSON files from R2
          mkdir -p data/sfs_json
          aws s3 sync s3://${{ vars.CLOUDFLARE_R2_RAWDATA_BUCKET_NAME }}/ data/sfs_json/ \
            --endpoint-url https://${{ secrets.CLOUDFLARE_R2_ACCOUNT_ID }}.r2.cloudflarestorage.com \
            --exclude "*" \
            --include "*.json"

          # Verify download
          if [ ! -d "data/sfs_json" ] || [ -z "$(ls -A data/sfs_json)" ]; then
            echo "::error::Failed to download JSON files from R2"
            exit 1
          fi
          echo "✅ Downloaded $(find data/sfs_json -name '*.json' | wc -l) JSON files from R2"
        fi
      env:
        AWS_DEFAULT_REGION: us-east-1

    - name: Generate HTML export
      if: inputs.html_source == 'generate' || inputs.html_source == ''
      run: |
        # Skapa output-katalog för GitHub Pages
        mkdir -p _site

        if [ -n "${{ inputs.filter }}" ]; then
          python sfs_processor.py --input data/sfs_json --output _site --formats html --filter "${{ inputs.filter }}"
        else
          python sfs_processor.py --input data/sfs_json --output _site --formats html
        fi
      env:
        PYTHONPATH: ${{ github.workspace }}

    - name: Generate index pages for HTML export
      if: inputs.html_source == 'generate' || inputs.html_source == ''
      run: |
        python exporters/html/populate_index_pages.py --input data/sfs_json --output _site/index.html --limit 30
        python exporters/html/populate_index_pages.py --input data/sfs_json --output _site/latest.html --limit 10
      env:
        PYTHONPATH: ${{ github.workspace }}

    - name: Add .nojekyll file
      run: |
        # Förhindra Jekyll-processing som kan störa ELI-URL:er
        touch _site/.nojekyll

    - name: Create deployment summary
      run: |
        echo "HTML export completed at $(date)" > _site/last-update.txt
        echo "Published to GitHub Pages" >> _site/last-update.txt

        if [ "${{ inputs.html_source }}" = "download" ]; then
          echo "Source: Pre-built HTML from Cloudflare R2" >> _site/last-update.txt
          echo "Bucket: ${{ vars.CLOUDFLARE_R2_HTMLEXPORT_BUCKET_NAME }}" >> _site/last-update.txt
        else
          echo "Source: Generated from JSON" >> _site/last-update.txt
          echo "Index pages: index.html (30 senaste), latest.html (10 senaste)" >> _site/last-update.txt
        fi

    - name: Setup Pages
      uses: actions/configure-pages@v5

    - name: Upload artifact
      uses: actions/upload-pages-artifact@v3
      with:
        path: '_site'

  deploy:
    environment:
      name: github-pages
      url: ${{ steps.deployment.outputs.page_url }}
    runs-on: ubuntu-latest
    needs: build
    steps:
    - name: Deploy to GitHub Pages
      id: deployment
      uses: actions/deploy-pages@v4