Skip to main content
Once a document is ingested, it moves through the async parsing process. To retrieve the result, your application follows a standard Polling and Retrieval pattern.

​
Status Polling

You can check the progress of a parse job using the GET /api/v1/documents/{documentId}/status endpoint. The status response includes the current lifecycle state:
  • queued: The job is waiting to be processed.
  • processing: The engine is actively parsing the document.
  • completed: The result has been materialized and is ready for download.
  • failed: An error occurred during extraction.

​
Retrieving Results

The final, materialized artifact is retrieved via GET /api/v1/documents/{documentId}/result.

​
Retrieval Flow

  1. Submit: Documents are accepted and queued for processing via the ingestion endpoint.
  2. Poll: Your application polls the status endpoint until the state reaches completed.
  3. Fetch: Call the result endpoint to receive the final structured data.

​
Platform Boundary

To ensure security and stability, TrueParser proxies all result access. Your application interacts only with the platform’s REST endpoints and never connects directly to the underlying S3-compatible storage. This allows us to manage per-request authorization and regional routing transparently.

​
Output Formats

TrueParser delivers the final parsed result as canonical JSON. That is the primary product-facing artifact returned through the API.

​
1. JSON Document (Primary)

The standard and recommended format for most applications. TrueParser materializes the parsed data into a final, valid JSON artifact.
  • Best For: Web applications, mobile clients, and standard data processing pipelines.

​
Schema Versioning

Every result artifact includes a schema_version and document_format marker. These contracts are stable within a major version, ensuring your downstream logic remains robust even as underlying parser engines are updated.
Last modified on April 1, 2026