Jobs & Results

Every upload creates a job. Single uploads return their result inline, but you can always re-fetch a result, check status (essential for bulk jobs), list past jobs, and download the original file.

Base path: /ocr

---

Check job status

GET /ocr/jobs/{job_id}/status

Returns the real-time processing status and progress for a job. Use this to poll bulk jobs to completion.

curl "https://netraocr.shivicx.com/api/v1/ocr/jobs/8f3c…/status" \
  -u "ak_live_xxxxxxxx:sk_live_xxxxxxxx"

Response 200 OK

{
  "job_id": "8f3c…",
  "document_id": "doc_…",
  "batch_id": "batch_…",
  "status": "processing",
  "progress": 60,
  "created_at": "2026-06-16T10:30:00Z",
  "updated_at": "2026-06-16T10:30:04Z",
  "message": "Extracting fields",
  "error": null,
  "processing_time_ms": null,
  "coins_consumed": 0,
  "fraud_coins_consumed": 0
}

status	Meaning
queued	Accepted, waiting to start
processing	Currently being processed
completed	Done — fetch the result
failed	Failed — see error

Polling

Poll every 2–5 seconds with backoff. Stop when status is completed or failed. For real-time updates without polling, contact us about streaming options.

---

Get a job result

GET /ocr/jobs/{job_id}/result

Returns the full OCR result for a completed job — the same payload returned inline by /ocr/upload.

curl "https://netraocr.shivicx.com/api/v1/ocr/jobs/8f3c…/result" \
  -u "ak_live_xxxxxxxx:sk_live_xxxxxxxx"

Result fields

Field	Type	Description
job_id	string	Job identifier
filename	string	Original filename
batch_id	string|null	Parent batch, for bulk jobs
structured_data	object|null	Typed fields from the selected template
page_count	integer|null	Pages processed
document_type	string|null	Detected type (e.g. invoice)
average_confidence	number|null	Mean extraction confidence
pages	array|null	Per-page detail
processing_time_ms	integer|null	Processing duration
created_at / completed_at	string|null	Timestamps
document_url / result_url	string|null	Convenience links
fraud_check_enabled	boolean	Whether an integrity check was requested
fraud_check_status	string|null	pending / processing / completed / failed
coins_consumed	number	Usage charged for this job

---

List jobs

GET /ocr/jobs

Lists your jobs (most recent first), with pagination and filtering.

curl "https://netraocr.shivicx.com/api/v1/ocr/jobs?limit=20" \
  -u "ak_live_xxxxxxxx:sk_live_xxxxxxxx"

Common query parameters include limit, offset, and status_filter. The response is a list of job summaries you can drill into with Get a job result.

---

Download the original document

GET /ocr/jobs/{job_id}/download

Streams back the original uploaded file for a job while it is retained. Uploaded originals, structured results, and metadata are kept for 30 days only for history and log keeping, then deleted completely.

curl "https://netraocr.shivicx.com/api/v1/ocr/jobs/8f3c…/download" \
  -u "ak_live_xxxxxxxx:sk_live_xxxxxxxx" \
  -o original.pdf

---

Working with batches

Bulk uploads group jobs into a batch. Each job carries the batch_id, so you can:

1. Submit with POST /ocr/bulk-upload.
2. Poll GET /batches/{batch_id} for folder-level progress.
3. Poll individual jobs with /ocr/jobs/{job_id}/status when you need per-file status.
4. Fetch each completed job's /result.

See OCR → Bulk processing and Batches.