📄 HTML → PDF/DOCX Converter v18.0

v18.0 NEW: DOCX PATHWAY

✓ mode: 'docx' or 'word' → Word document

✓ Same iterative page-fitting algorithm as PDF

✓ Page count via LibreOffice → PDF conversion

✓ Orphan heading handling, coverage tracking

PDF Features (from v17.9.1):

✓ Tiling: Oversized visuals sliced into multiple pages

✓ Coverage audit: Tracks all blocks, reports missing content

✓ Emergency fallback: Falls back to page.pdf() if coverage < 95%

✓ Visual container detection, embedded videos, clickable images

✓ Unified list handling: listMode = together | split | smart

Endpoints

GET /health

Health check

POST /convert

Convert HTML → PDF or DOCX

Body: { html, mode?, listMode?, filename? }
mode: portrait (default), landscape, docx, word, pdf-paginated, pdf-legacy
listMode: smart (default), together, split

POST /debug/convert

Convert with debug session (stores intermediate artifacts)

POST /debug/blocks

Extract blocks only (Stage 1) - returns session info

GET /debug/session/:id/intermediate

Get intermediate HTML (all blocks before pagination)

GET /debug/session/:id/blocks

Get extracted blocks JSON

GET /debug/session/:id/final

Get final merged PDF or DOCX