Wait for a chart or font to finish before capturing a PDF

Conditions to satisfy before the PDF is taken. All of them apply, in order.

Fields

What it is for

Conditions to satisfy before the PDF is taken — a CSS selector, network idle, a fixed delay. All of them apply, in that order.

The trap

networkIdle never settles on a page with a long-poll, a websocket or an analytics beacon, and the render dies at timeoutMs instead. Prefer selector: have your chart library set a data attribute when it has drawn, and wait for that. It is faster and it cannot hang.

Example

curl -X POST https://api.pdfcraft.dev/v1/render \
  -H "Authorization: Bearer $PDFCRAFT_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "html": "<h1>hello</h1>",
    "options": {
      "waitFor": {
        "selector": "#chart-ready",
        "networkIdle": false,
        "delayMs": 0
      }
    }
  }' \
  --output out.pdf

Every option

• format — Paper size. Ignored if the page CSS declares its own @page size.
• landscape — Rotate the paper to landscape orientation.
• margin — Page margins. Any side may be omitted; omitted sides default to 0.
• scale — Rendering scale factor.
• printBackground — Print background colours and images. Off by default in browsers; on by default here.
• pageRanges — Subset of pages to keep, e.g. "1-5" or "1,4,7-9". Empty means all pages.
• headerHtml — HTML for the running header. Supports Chromium print classes: date, title, url, pageNumber, totalPages. Needs a top margin to be visible.
• footerHtml — HTML for the running footer. Same classes as headerHtml; needs a bottom margin.
• waitFor — Conditions to satisfy before the PDF is taken. All of them apply, in order.
• emulateMedia — Which CSS media type the page sees.
• timeoutMs — Hard ceiling on the whole render. Exceeding it returns 408 render_timeout.

Try any of these in the playground without writing a line of code, or read the full reference.