Lightweight PDF parser with layout, tables, formulas and bounding boxes

(github.com)

42 points by beatrizalmeidaf3 hours ago

5 comments

antonly1 hour ago
I just tried this and the results look very off. I put in the MiniSat[1] Paper, and the result appears catastrophically wrong: <a href="https://imgur.com/a/i3RHwcQ" rel="nofollow">https://imgur.com/a/i3RHwcQ</a><p>[1]: <a href="http://minisat.se/downloads/MiniSat.pdf" rel="nofollow">http://minisat.se/downloads/MiniSat.pdf</a>
phenomen1 hour ago
I currently use <a href="https://github.com/firecrawl/anydoc" rel="nofollow">https://github.com/firecrawl/anydoc</a> in my PDF pipelines. In most cases it performs well. I'll test your lib to compare.
- coinfused1 hour ago
  Do you know if this one can also crop the PDF to a region of interest before converting to markdown?
JaumeGar1 hour ago
Really nice work, thanks for sharing. The table and formula extraction look great.
archeantus1 hour ago
Great work, thanks for sharing.
beatrizalmeidaf3 hours ago
I built this because extracting PDFs into Markdown/JSON often loses reading order, tables, formulas, figures, and their original locations.<p>The goal is a lightweight document extraction pipeline that preserves document structure and bounding boxes while exporting to Markdown, JSON, Excel and Word.<p>I'm also working on structure-aware semantic chunking for RAG, so retrieved chunks can retain their section, page and exact visual location in the PDF.<p>The project is open source and I'd love feedback on the architecture, extraction quality, and useful use cases.
- thatcherc2 hours ago
  This looks fantastic! The table and formula extraction features are especially interesting. My immediate question is: can this be integrated into Zotero? Most of the PDFs I read are research papers and extracting tables and formulas directly from my zotero collection would be super handy.