Problem
Custom document extraction schemas must currently be created and maintained manually in the Document AI workspace. This results in duplication of configuration and increases the risk of the CDS model and the deployed Document AI schemas becoming out of sync.
Proposal
Add support for defining custom extraction schemas using CDS annotations and automatically synchronize them with Document AI.
The synchronization could be performed during application startup or deployment, ensuring that the Document AI workspace reflects the schema defined in the CDS model.
This would allow extraction schemas to be managed as code, simplify deployment, and reduce manual configuration effort.
Problem
Custom document extraction schemas must currently be created and maintained manually in the Document AI workspace. This results in duplication of configuration and increases the risk of the CDS model and the deployed Document AI schemas becoming out of sync.
Proposal
Add support for defining custom extraction schemas using CDS annotations and automatically synchronize them with Document AI.
The synchronization could be performed during application startup or deployment, ensuring that the Document AI workspace reflects the schema defined in the CDS model.
This would allow extraction schemas to be managed as code, simplify deployment, and reduce manual configuration effort.