PDF Office Toolkit

Problem Context

A PDF that opens locally may still fail in a business system: strict server-side parsers can reject missing %%EOF, broken startxref, or unsupported AES-256 encryption. Splitting pages, correcting skew, adding watermarks, and checking sensitive data are often fragmented across tools.

How It Works

PDF Office Toolkit groups common PDF operations into scripts/pdf_toolkit.py subcommands: split-orient corrects and splits pages, merge combines files by name, annotate adds a translucent highlight and red border, watermark and generate create CJK-friendly PDFs, crypt performs AES-256 encryption or decryption, compress targets a file size with lossless-first rewriting, check and repair validate headers, EOF, xref, strict parsing, and first-page rendering, while batch aggregates directory-level PASS/FAIL results. Write-out commands such as orient, merge, generate, watermark, crypt, and flatten rerun a strict self-check after saving, preserving output for debugging when validation fails.

Boundaries And Notes

Orientation correction relies on Tesseract; sensitive-data scanning covers only the text layer, so scanned PDFs need OCR first. Flattening rasterizes pages and removes selectable text. AES-256 encrypted files may be rejected by legacy server previewers, so confirm compatibility before upload.