P2M

Fast, pure-Rust PDF to Markdown converter. No ML, no OCR, no external dependencies.

Features

• Text extraction — CMap/ToUnicode decoding, CID fonts, TrueType fallback, ligature expansion
• Layout analysis — multi-column detection (newspaper & tabular), reading-order reconstruction
• Table detection — rect-based (union-find clustering) and line-based (H/V grid intersection)
• Markdown generation — headings, lists, code blocks, bold/italic, hyperlinks, page breaks
• Tagged PDF support — structure-tree roles (H1–H6, P, L, Code, BlockQuote, Table)

Installation

cargo install p2m-cli

Or build from source:

git clone https://github.com/qntx-labs/p2m
cd p2m
cargo build --release

CLI Usage

p2m document.pdf                     # convert to stdout
p2m document.pdf -o output.md        # write to file
p2m document.pdf --pages 1,3,5       # specific pages only
p2m document.pdf --page-breaks       # insert page markers
p2m document.pdf --raw               # no metadata on stderr

Library Usage

// Simple conversion
let doc = p2m::convert("document.pdf")?;
println!("{}", doc.markdown);

// With options
let opts = p2m::Options::new().pages([1, 2, 3]);
let doc = p2m::convert_with("document.pdf", &opts)?;

// From bytes
let bytes = std::fs::read("document.pdf")?;
let doc = p2m::convert_bytes(&bytes)?;

License

Licensed under either of:

• Apache License, Version 2.0 (LICENSE-APACHE or https://www.apache.org/licenses/LICENSE-2.0)
• MIT License (LICENSE-MIT or https://opensource.org/licenses/MIT)

at your option.

Unless you explicitly state otherwise, any contribution intentionally submitted for inclusion in this project shall be dual-licensed as above, without any additional terms or conditions.

---

A QNTX open-source project.

Code is law. We write both.