Roadmap

Roadmap

pdfcpu2 evolves driven by real pipeline needs. Status of the fork-specific features:

Feature	Command	Status
Plain-text extraction (PDFBox-compatible)	extract -m text	✅ Available — Text Extraction
Embedded offline OCR (PP-OCRv6 tiny)	ocr	✅ Available — OCR
Page rendering / rasterization	render	✅ Available — Render
Content freezing (image-only pages)	solidify	✅ Available — Solidify
Glyph obfuscation (anti text extraction)	obfuscate	✅ Available — Obfuscate
Text replacement (embedded vivo Sans subsets)	replace	✅ Available — Replace

Ideas under consideration

• Garbage-collecting original fonts that lost all glyphs to replace (subset fonts are already reused and extended across runs; font garbage is currently left to optimize).
• Long-tail replace coverage: horizontal-line matching leaves rare residues (skewed baselines, split columns); PDFBox occasionally extracts fewer replacement glyphs than drawn.
• JPX (JPEG 2000) and JBIG2 image decoding in the rasterizer (currently a gray placeholder + warning).
• OCR model upgrades (PP-OCRv6 small/medium) behind the same ocr interface.

Nothing here is promised; priorities follow real pipeline needs.

Everything else

All upstream pdfcpu commands (validate, optimize, encrypt, stamp, form, …) are inherited, maintained and documented in this wiki. Feature requests that fit the project's direction are welcome via issues.