yfedoseev/pdf_oxide v0.3.16

v0.3.16 | Advanced Visual Table Detection and Automated Python Stubs

on GitHub

Features

• Smart Hybrid Table Extraction (#206) — Introduced a robust, zero-config visual detection engine that handles both bordered and borderless tables.
  • Localized Grid Detection: Uses Union-Find clustering to group vector paths into discrete table regions, enabling multiple tables per page.
  • Visual Line Analysis: Detects cell boundaries from actual drawing primitives (lines and rectangles), significantly improving accuracy for untagged PDFs.
  • Visual Spans: Identifies colspans and rowspans by analyzing the absence of internal grid lines and text-overflow signals.
  • Visual Headers: Heuristically identifies hierarchical (multi-row) header rows.
• Professional ASCII Tables: Added high-quality ASCII table formatting for plain text output, featuring automatic multiline text wrapping and balanced column alignment.
• Auto-generated Python type stubs (#220) — Integrated automated .pyi stub generation using mypy's stubgen in the CI pipeline, ensuring Python IDEs always have up-to-date type information for the Rust bindings.
• Python PdfDocument path-like and context manager (#223) — PdfDocument now accepts pathlib.Path (or any path-like object) and supports the context manager protocol (with PdfDocument(path) as doc:), ensuring scoped usage and automatic resource cleanup.
• Enabled by Default: Table extraction is now active by default in all Markdown, HTML, and Plain Text conversions.
• Robust Geometry: Updated Rect primitive to handle negative dimensions and coordinate normalization natively.

Bug Fixes

• Fixed segfault in nested Form XObject text extraction (#228) — Resolved aliased &mut references during recursive XObject processing using interior mutability (RefCell/Cell).
• Fixed Python Coordinate Scaling: Corrected erase_region coordinate mapping in Python bindings to use the standard [x1, y1, x2, y2] format.
• Improved ASCII Table Wrapping: Reworked text wrapping to be UTF-8 safe, preventing panics on multi-byte characters.
• Refined Rendering API: Restored backward compatibility for the render_page method.

🏆 Community Contributors

🥇 @hoesler — Huge thanks for PR #228! Your fix for the nested XObject aliasing UB is a critical stability improvement that eliminates segfaults in complex PDFs. By correctly employing interior mutability, you've made the core extraction engine significantly more robust and spec-compliant. Outstanding work! 🚀

🥈 @monchin — Thank you for the fantastic initiative on automated stub generation (#220) and the ergonomic improvements for Python (#223)! We've integrated these into the v0.3.16 release, providing consistent, IDE-friendly type hints and modern path-like/context manager support. Outstanding contributions! 🚀

Installation

Rust (crates.io)

cargo add pdf_oxide

Python (PyPI)

pip install pdf_oxide

JavaScript/WASM (npm)

npm install pdf-oxide-wasm

CLI (Homebrew)

brew install yfedoseev/tap/pdf-oxide

CLI (Scoop — Windows)

scoop bucket add pdf-oxide https://github.com/yfedoseev/scoop-pdf-oxide
scoop install pdf-oxide

CLI (Shell installer)

curl -fsSL https://raw.githubusercontent.com/yfedoseev/pdf_oxide/main/install.sh | sh

CLI (cargo-binstall)

cargo binstall pdf_oxide_cli

MCP Server (for AI assistants)

cargo install pdf_oxide_mcp

Pre-built Binaries
Download archives for Linux, macOS, and Windows from the assets below. Each archive includes both pdf-oxide (CLI) and pdf-oxide-mcp (MCP server).

Platform Support

Changelog

See CHANGELOG.md for full details.