yfedoseev/pdf_oxide v0.3.15

v0.3.15 | Header & Footer Management, Multi-Column Stability, and Font Fixes

on GitHub

Features

• PDF Header/Footer Management API (#207) — Added a dedicated API for managing page artifacts across Rust, Python, and WASM.
  • Add: Ability to insert custom headers and footers with styling and placeholders via PageTemplate.
  • Remove: Heuristic detection engine to automatically identify and strip repeating artifacts. Includes modular methods: remove_headers(), remove_footers(), and remove_artifacts(). Prioritizes ISO 32000 spec-compliant /Artifact tags when available.
  • Edit: Ability to mask or erase existing content on a per-page basis via erase_header(), erase_footer(), and erase_artifacts().
• Page Templates — Introduced PageTemplate, Artifact, and ArtifactStyle classes for reusable page design. Supports dynamic placeholders like {page}, {pages}, {title}, and {author}.
• Scoped Extraction Filtering — Updated all extraction methods to respect erase_regions, enabling clean text extraction by excluding identified headers and footers.
• Python PdfDocument.from_bytes() — Open PDFs directly from in-memory bytes without requiring a file path. (Contributed by @hoesler in #216)
• Future-Proofed Rust API — Implemented Default trait for key extraction structs (TextSpan, TextChar, TextContent) to protect users from future field additions.

Bug Fixes

• Fixed Multi-Column Reading Order (#211) — Refactored extract_words() and extract_text_lines() to use XY-Cut partitioning. This prevents text from adjacent columns from being interleaved and standardizes top-to-bottom extraction. (Reported by @ankursri494)
• Resolved Font Identity Collisions (#213) — Improved font identity hashing to include ToUnicode and DescendantFonts references. Fixes garbled text extraction in documents where multiple fonts share the same name but use different character mappings. (Reported by @productdevbook)
• Fixed Lines table strategy false positives (#215) — extract_tables() with horizontal_strategy="lines" now builds the grid purely from vector path geometry and returns empty when no lines are found, preventing spurious tables on plain-text pages. (Contributed by @hoesler)
• Optimized CMap Parsing — Standardized 2-byte consumption for Identity-H fonts and improved robust decoding for Turkish and other extended character sets.

🏆 Community Contributors

🥇 @hoesler — Huge thanks for PR #216 and #215! Your contribution of from_bytes() for Python unlocks new serverless and in-memory workflows for the entire community. Additionally, your fix for the Lines table strategy significantly improves the precision of our table extraction engine. Outstanding work! 🚀

🥈 @ankursri494 (Ankur Srivastava) — Thank you for identifying the multi-column reading order issue (#211). Your detailed report and sample document were the catalyst for our new XY-Cut partitioning engine, which makes PDFOxide's reading order detection among the best in the ecosystem! 🎯

🥉 @productdevbook — Thanks for reporting the complex font identity collision issue (#213). This report led to a deep dive into PDF font internals and a significantly more robust font hashing system that fixes garbled text for thousands of professional documents! 🔍✨

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.