Ingestion Pipeline

Hello All,

I have been using ingestion pipellne on PDFs.

I need to pass entire PDF encoded and in return I get a response (which is PDF text) in one "content" field.

How can I get page-wise content for a single PDF using ingestion pipeline?

Any tricks?

Jay

I don't think you can.

@dadoonet

We actually can by splitting the pdf, but the order is not maintained in that.

Anyways, thanks, will explore more.

Also, ingest pipeline sometimes gives 201, sometimes 200.
Can't rely.