huggingface_hub
# LightRAG packages
lightrag-hku
# MinerU 2.0 packages (replaces magic-pdf) - handles PDF parsing with multiple backends
# >=3.4.1: earlier versions iterate pdftext's PageChars return value directly and
# crash with "TypeError: 'PageChars' object is not iterable" against pdftext>=0.7
mineru[core]>=3.4.1
# Progress bars for batch processing
tqdm
# Note: Optional dependencies are now defined in setup.py extras_require:
# - [image]: Pillow>=10.0.0 (for BMP, TIFF, GIF, WebP format conversion)
# - [text]: reportlab>=4.0.0 (for TXT, MD to PDF conversion)
# - [paddleocr]: paddleocr + pypdfium2 (for parser='paddleocr' - better OCR for scanned PDFs)
# - [office]: requires LibreOffice (external program, not Python package)
# - [all]: includes all optional dependencies
#
# Install with: pip install raganything[image,text] or pip install raganything[all]
# For best PDF handling: pip install raganything[paddleocr]
