github-featured / opendataloader-project--opendataloader-pdf

发布

0 发布 77 Git 标签

RSS 订阅源
  • v1.11.1 447a0963d9
    比较

    docs: fact-check and fix documentation against actual codebase

    frostbyte_neo 发布于 2026-02-26 08:35:42 +00:00

    • Remove non-existent --exclude-patterns CLI flag (ai-safety.mdx)
    • Remove non-existent convert subcommand from CLI examples (3 files)
    • Fix output_formats→format, --output-formats→-f to match actual API (tagged-pdf-rag.mdx)
    • Fix Docker image path to ghcr.io/opendataloader-project/opendataloader-pdf-cli (tagged-pdf-rag.mdx)
    • Remove non-existent GapBasedSorter.java reference (reading-order.mdx)
    • Add missing sensitive-data filter to safety table (ai-safety.mdx)
    • Update OCR status from "coming soon" to available via hybrid mode (faq.mdx)
    • Remove unverified benchmark comparison table (tagged-pdf-rag.mdx)
    • Remove LlamaIndex "Coming Soon" sections (rag-integration.mdx, tagged-pdf-rag.mdx)
    • Replace deprecated langchain import with langchain_opendataloader_pdf (tagged-pdf-rag.mdx)
    • Update Node.js requirement from 18+ to 20+ matching package.json engines (faq.mdx, quick-start-nodejs.mdx)
    • Update Maven version from 1.4.1 to 1.11.0 (quick-start-java.mdx)

    Co-Authored-By: Claude Opus 4.6 noreply@anthropic.com

    下载附件