OkraPDF icon

OkraPDF

Extract tables and text from PDFs using OkraPDF

Actions5

Document → Extract (Async)

AI-generated

Summary

Submit a PDF for asynchronous extraction using OkraPDF.

Inputs

  • inputType (required) — Whether to fetch the PDF from a URL or use binary data from a previous node. Options: URL or Binary Data. Default: URL.
  • pdfUrl (required) — URL of the PDF to extract. Visible when inputType is URL.
  • binaryPropertyName (required) — Name of the binary property containing the PDF. Visible when inputType is Binary Data.
  • options — Optional settings for the extraction request. Sub‑properties: webhookUrl (string) – URL to receive a webhook when extraction completes (async only); maxPages (number) – maximum pages to process (0 = unlimited).

Output shape

JSON array where each item contains the response from the OkraPDF async extraction endpoint. Typically includes a jobId and initial status (e.g., queued).

Because this is the async operation, the node does not wait for extraction to finish. Use the returned jobId with the Job resource operations (Get Status, Get Results) to poll for completion and retrieve results.

Examples

Example 1: Extract tables from a PDF hosted online.

Set resource=Document, operation=Extract (Async), inputType=URL, pdfUrl='https://example.com/file.pdf', options.webhookUrl='https://myhook.com/ocr', options.maxPages=10.

Example 2: Extract from binary data produced by another node.

Set resource=Document, operation=Extract (Async), inputType=Binary Data, binaryPropertyName='data' (or the name of the incoming binary field).

Links

Discussion