name: 01.2. Backup-based Curation

on:
  workflow_dispatch:
    inputs:
      backup_file:
        description: 'Path to the backup file (JSON or MD) inside the repository'
        required: true
        default: 'data/backup_posts.json'
      historical_mode:
        description: 'Activate Historical Mode (Ignores 30-day limit)'
        required: false
        default: 'true'
        type: boolean

permissions:
  contents: write
  pull-requests: write

jobs:
  backup-curation-process:
    runs-on: ubuntu-latest
    steps:
      - name: Repository Synchronization
        uses: actions/checkout@v7

      - name: Python 3.11 Environment Provisioning
        uses: actions/setup-python@v6
        with:
          python-version: '3.11'
          cache: 'pip'

      - name: Installation of Dependencies
        run: |
          python -m pip install --upgrade pip
          pip install --no-cache-dir pydantic PyGithub aiohttp beautifulsoup4 httpx fake-useragent pytz python-dotenv twikit>=2.1.2 playwright playwright-stealth tenacity
          # Playwright is required because Gemini evaluator or curator might use it indirectly, 
          # even if backup extraction doesn't strictly need it.
          playwright install chromium --with-deps

      - name: Agentic Curation Execution (Backup Mode)
        env:
          GEMINI_API_KEY_1: ${{ secrets.GEMINI_API_KEY_1 }}
          GEMINI_API_KEY_2: ${{ secrets.GEMINI_API_KEY_2 }}
          GH_TOKEN: ${{ secrets.GITHUB_TOKEN }}
          BACKUP_FILE: ${{ github.event.inputs.backup_file }}
          HISTORICAL_MODE: ${{ github.event.inputs.historical_mode }}
          PYTHONPATH: .
          PYTHONUNBUFFERED: "1"
        run: |
          if [ ! -f "$BACKUP_FILE" ]; then
            echo "❌ ERROR: File '$BACKUP_FILE' not found in repository."
            echo "Please upload the file to that path and try again."
            exit 1
          fi
          python src/main.py