edgeparse

High-performance PDF-to-structured-data extraction CLI.

Convert PDF documents to Markdown, JSON, HTML, or plain text with a single command. Built on top of edgeparse-core.

Installation

cargo install edgeparse

Usage

# Convert to Markdown (default format is JSON)
edgeparse report.pdf -f markdown -o output/

# Convert to multiple formats
edgeparse report.pdf -f json,markdown,html -o output/

# Extract specific pages
edgeparse report.pdf --pages 1,3,5-7 -f markdown -o output/

# Use XY-Cut reading order (enabled by default)
edgeparse report.pdf --reading-order xycut -f markdown

# Extract with cluster-based table detection
edgeparse report.pdf --table-method cluster -f json

# Extract images externally
edgeparse report.pdf --image-output external --image-format png -f markdown

Features

• Multiple output formats — JSON, Markdown, HTML, plain text, DOCX, CSV
• Table detection — border-based and cluster detection methods
• Reading order — XY-Cut++ algorithm for correct multi-column reading order
• Image extraction — embedded base64 or external file output (PNG/JPEG)
• Content safety — filters hidden text, off-page content, watermarks
• Encrypted PDFs — password-based decryption support
• Tagged PDFs — uses PDF structure tree when available
• PII sanitization — optional personal data redaction

License

Apache-2.0