Ovrin validates extracted fields against the rules implied by the schema and tag grammar. This is the main safeguard that prevents bad output from flowing silently downstream.
Examples of rules
- required
- enum
- min
- max
- format-specific constraints
- cross-field rules
Why they matter
These rules are how the library turns vague extracted text into something that can be trusted in business logic or sent to a ledger or compliance workflow.