Remove all AI processing from the import-docx endpoint. Instead:
- Convert the .docx to HTML with mammoth.convertToHtml (preserves
bold, headings, lists, etc.) — the result goes straight into the
item body and renders in the lesson viewer
- Extract raw text in parallel to pick the document title from the
first non-empty line
- No external services, no AI, works on the deployed server
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
nmemmertandClaude Sonnet 4.6
19f922c4ba