name: Manual and Auto Sync to Hugging Face

on:
  workflow_dispatch:

jobs:
  sync-dataset:
    runs-on: ubuntu-latest
    steps:
      - name: Checkout Repository
        uses: actions/checkout@v4
        with:
          fetch-depth: 1 

      - name: Push Clean History to Hugging Face
        env:
          HF_TOKEN: ${{ secrets.HF_TOKEN }}
        run: |
          # 1. Delete the Git history so old large files are completely forgotten
          rm -rf .git
          
          # 2. Clean up any random .exe files just in case they were generated during a build
          find . -name "*.exe" -type f -delete || true
          
          # 3. Initialize a brand new, empty Git repository
          git init
          git checkout -b main
          
          # 4. Set up bot credentials
          git config user.name "github-actions[bot]"
          git config user.email "github-actions[bot]@users.noreply.github.com"
          
          # 5. Initialize Git LFS and track binary files so Hugging Face accepts them
          git lfs install
          git lfs track "*.pdf"
          git lfs track "*.docx"
          git add .gitattributes
          
          # 6. Add only the CURRENT files and make a single fresh commit
          git add .
          git commit -m "Auto-sync latest raw files to Hugging Face"
          
          # 7. Force push this single commit to Hugging Face, overwriting everything
          git push --force https://notamitgamer:$HF_TOKEN@huggingface.co/datasets/notamitgamer/usercontent main:main