opendataloader-project/opendataloader-pdf

OpenDataLoader PDF is a high-performance engine for extracting text, tables, and layout from PDFs into structured formats like JSON, HTML, and Markdown. It supports deterministic local processing with an optional AI hybrid mode to enhance complex page extraction. The system provides programmatic APIs and CLI tools across Java, Python, and Node.js, enabling batch processing, auto-tagging, and content sanitization.

This period focused on establishing a multi-module Java build structure and releasing the OpenDataLoader PDF engine with comprehensive support for Python, Node.js, and Java APIs. The work introduced a two-phase extraction pipeline, hybrid AI backend integration, and expanded output formats including HTML, Markdown, and enriched JSON, while removing legacy Markdown generation capabilities.

29 changes

This period focused on introducing hybrid processing modes that route PDF pages to external AI backends for enhanced content extraction, alongside the development of an MCP server for AI agent integration. Concurrently, the project stabilized the command-line interface through a refactored public API, comprehensive regression testing, and automated verification harnesses.

16 changes

opendataloader-project/opendataloader-pdf was measured the same way every project in this corpus was: the same rubric, at a pinned commit, with the result published in full. Point the surveyor at a repository you know and see whether you agree with it.