exportacao-multiplos-formatos

Traverse the ProseMirror JSON node tree and map nodes to docx package objects: paragraphs become Paragraph, headings become Heading with levels, and bold/italic marks become styled TextRun. Footnotes map to FootnoteReference and Footnote elements.

Render the document as styled HTML on an internal page, then use Puppeteer's page.pdf method with A4 format and printBackground enabled. Include a header with the manuscript title and a footer with page numbers.

mammoth.js extracts simplified HTML from .docx files, preserving basic styles. The extracted HTML can then be parsed with DOMParser and a custom serializer to produce valid ProseMirror JSON.

Rendering novels of 300+ pages synchronously exceeds typical 30-second request timeouts. Queue the export as a background Bull job, return 202 Accepted with a job ID, and notify the user via WebSocket when the download is ready.

Split the document at level-1 heading nodes to generate the ePub navigation structure with epub-gen. Include Dublin Core metadata such as title, author, language, and ISBN if registered in the project.

Do not use it for LGPD/GDPR user data compliance exports or GUF graph portability in JSON-LD format, which are handled by a dedicated compliance skill. This pipeline covers only manuscript format conversion.