PDF Generation from Firefox via CDP

Foxbridge implements Page.printToPDF with IO stream support for Puppeteer v24+.

Usage

const pdf = await page.pdf({
  format: 'A4',
  printBackground: true,
  landscape: false,
  margin: { top: '1cm', bottom: '1cm' },
});
fs.writeFileSync('page.pdf', pdf);

Supported Parameters

All parameters are forwarded directly to Juggler’s Page.printToPDF.

IO Stream Protocol

Puppeteer v24+ uses the IO domain to read PDF data via streaming instead of reading the base64 data field directly. Foxbridge supports both approaches:

1. Page.printToPDF returns {data: "...", stream: "pdf-stream-123"}
2. Puppeteer calls IO.read({handle: "pdf-stream-123"})
3. Foxbridge returns the PDF data with eof: true
4. Puppeteer calls IO.close({handle: "pdf-stream-123"})

Stream handles are stored in memory and cleaned up on IO.close.

Page.printToPDF
  → Juggler Page.printToPDF
  ← {data: base64} stored as stream
  → returns {data, stream: handle}

IO.read({handle})
  ← {base64Encoded: true, data: ..., eof: true}

IO.close({handle})
  ← {} (stream deleted)

BiDi Backend

On the BiDi backend, Page.printToPDF maps to browsingContext.print, which returns the PDF as a base64 string. The same IO streaming mechanism is used.

See also

• CDP Domain Coverage — Current CDP coverage snapshot
• How-To Guides — Common foxbridge tasks and recipes
• Performance Metrics — Page timing data from Firefox