Batch Process
document_ai.batch_process
ActionGoogle Cloudconnection required
Calls the Document AI API POST /v1/.../processors/{id}:batchProcess.
Start an async batch job from gs:// input (file or prefix) to a gs:// prefix.
Connection
- Required
- Yes — the node will not save without one
- Auth type
- OAuth 2.0
- Providers
- Google Document AI
- Secret shape
- {"client_id", "client_secret", "refresh_token"}
- Config it needs
- project_id, location, processor_id
- Resolved by
- config.connection_id, else node.connection_id
Configuration
6 fields, 2 required. A dotted path is a nested object and name[] an array of objects — the editor renders the same tree.
| Field | Type | Required | Default | Constraints | Example |
|---|---|---|---|---|---|
project_id GCP project id (overrides connection). | string | No | — | — | — |
location Processor location: us or eu (overrides connection). | string | No | — | — | — |
processor_id Processor id (overrides connection). | string | No | — | — | — |
gcs_input_uri gs://bucket/file.pdf, or gs://bucket/folder/ to process a prefix. | string | Yes | — | matches gs://.+ | — |
gcs_output_uri gs://bucket/output-prefix/ where JSON results are written. | string | Yes | — | matches gs://.+ | — |
mime_type MIME type of a single input file. | string | No | application/pdf | — | — |
Example configuration
Every value below is the contract’s own example or default — nothing invented.
{
"mime_type": "application/pdf"
}Output
One shape.
operation
| Field | Type | Example |
|---|---|---|
ok | boolean | — |
operation Long-running operation name. | string | — |
done | boolean | — |
{
"done": false,
"ok": false,
"operation": ""
}Notes
- The node does not poll the operation; results land in gcs_output_uri.
Related Google Document AI nodes
Solutions with Google Document AI
See Batch Process run in a workflow.
In a technical demo we add the Google Document AI connection, drop the node on the canvas and let Workflow Health pre-flight it before the first real run.