Split a packet - NDI Platform API

Split finds logical documents inside one file. PDFs become page-range segments. Spreadsheets become one segment per worksheet, then each sheet is classified against your classes.

Python
TypeScript
curl

from ndi_sdk import NdiClient
from ndi_sdk.models.document_ops import SplitCategory

with NdiClient() as client:
    packet = client.documents.create_upload("packet.pdf")
    job = client.jobs.wait(
        client.documents.split(
            packet,
            classes=[
                SplitCategory(id="invoice", label="Invoice", description="A supplier invoice"),
                SplitCategory(id="receipt", label="Receipt", description="A payment receipt"),
            ],
            wait_seconds=30,
        ).job_id
    )
    for segment in job.result.segments:
        print(segment.start_page, segment.end_page, segment.split_class)

import { NdiClient } from "ndi-sdk";

const client = new NdiClient();
const packet = await client.documents.createUpload("packet.pdf");
const job = await client.jobs.wait(
  (
    await client.documents.split(packet, {
      classes: [
        { id: "invoice", label: "Invoice", description: "A supplier invoice" },
        { id: "receipt", label: "Receipt", description: "A payment receipt" },
      ],
      wait_seconds: 30,
    })
  ).job_id,
);
console.log(job.result);

UPLOAD_ID=$(curl -s -X POST "$NDI_BASE_URL/v1/uploads" \
  -H "X-API-Key: $NDI_API_KEY" \
  -F "file=@packet.pdf" | jq -r '.upload_id')

curl -s -X POST "$NDI_BASE_URL/v1/split?wait_seconds=60" \
  -H "X-API-Key: $NDI_API_KEY" \
  -H "Content-Type: application/json" \
  -d "{\"source\":{\"type\":\"upload\",\"upload_id\":\"$UPLOAD_ID\"},\"classes\":[{\"id\":\"invoice\",\"label\":\"Invoice\",\"description\":\"A supplier invoice\"},{\"id\":\"receipt\",\"label\":\"Receipt\",\"description\":\"A payment receipt\"}]}"

Options

Option	Default	Meaning
`classes`	you must set this	Document classes to cut on. Spreadsheets classify each worksheet against these classes
`unknown_policy`	`include`	`include` keeps unmatched pages as unclassified; `force` assigns the best class; `error` fails the job
`overlap_policy`	`exclusive`	`exclusive` keeps segments disjoint; `shared_boundary_page` includes a boundary page in both adjacent segments
`split_rules`	none	Optional free-text guidance for how to cut the packet
`output.include_content`	`false`	When true, each segment includes its extracted markdown in `content`
`output.materialize_files`	`false`	When true, each document segment includes a downloadable artifact. Spreadsheet segments always materialize as single-sheet `.xlsx`
`page_ranges`	all pages	One-based inclusive ranges. Not supported for spreadsheet sources

Result

result.segments is the ordered cut list.

Field	Meaning
`start_page` / `end_page`	Inclusive page range of this cut. For workbooks, both equal `sheet_index + 1` — one segment per worksheet
`status`	Whether this cut was assigned a class. `matched` has `class`. `unclassified` was kept because `unknown_policy` is `include`
`class`	JSON name of the winning request class for a `matched` segment. Null when unclassified. The Python SDK attribute is `split_class`
`content`	Segment markdown when `output.include_content` is true. Null otherwise
`artifacts`	Downloadable files when `output.materialize_files` ran (spreadsheet segments always materialize as a single-sheet `.xlsx`)

A parse_result source is supported. After split, extract or ground each segment (or its materialized file) separately.