Skip to content

Split PDF

The "Split PDF" service task splits a PDF file into multiple separate documents. This can be used to break up multi-page PDFs into individual documents or extract specific page ranges.

Input parameters

The task expects the following fields:

{
  "pdf": {
    "referenceId": "...",
    "filename": "batch-invoice.pdf",
    "contentType": "application/pdf"
  },
  "splits": {
    "0": { "from": 1, "to": 3 },
    "1": { "from": 4, "to": 6 }
  },
  "filenamePattern": "{name}_part-{n}"
}

Explanation:

  • pdf: File reference of the PDF to be split (required).
  • splits: An object with numbered entries, each defining a page range (optional).
    • from: First page of the range (1-based). If omitted, the range starts at page 1.
    • to: Last page of the range (1-based, inclusive). If omitted, the range extends to the last page.
    • filename: Optional custom filename for this part.
  • filenamePattern: Pattern for naming the output files (optional). Available placeholders:
    • {name} – Original filename (without .pdf)
    • {n} – Sequential part number (1, 2, 3, ...)
    • The .pdf extension is added automatically if missing.

Without splits

If splits is not specified, each page is extracted as a separate PDF file.

Output

The task returns an array of file references – one per output document.

{
  "files": [
    {
      "referenceId": "...",
      "filename": "batch-invoice_part-1.pdf",
      "contentType": "application/pdf"
    },
    {
      "referenceId": "...",
      "filename": "batch-invoice_part-2.pdf",
      "contentType": "application/pdf"
    }
  ],
  "count": 2
}

Explanation:

  • files: Array of file references for all output documents.
  • count: Number of PDF files created.

JSONata examples

Extract each page individually (simplest case):

{
  "pdf": $.document
}

Split into specific page ranges:

{
  "pdf": $.document,
  "splits": {
    "0": { "from": 1, "to": 2 },
    "1": { "from": 3 }
  }
}

Notes

  • Page numbers are 1-based and inclusive: { "from": 1, "to": 3 } produces a PDF containing pages 1, 2 and 3.
  • Invalid page numbers (e.g. higher than the total page count) are automatically clamped to the valid range.
  • Overlapping splits produce documents with duplicate pages – this is intentional and can be used to create copies.

Tip

Combine this with the "Read PDF metadata" service to determine the page count in advance, then use it as a gateway condition – for example to only split PDFs with more than 10 pages.