Schemas beside datasets (Table Schema JSON files)

Schemas beside datasets (Table Schema JSON files)

Each dataset stores its Table Schema JSON file directly beside the dataset file (for example, accounts.schema.json beside accounts.csv). This keeps schemas and data tightly coupled and avoids any dedicated schemas/ directory. A datapackage.json manifest may still be placed at the repository root to bind resources and schemas into a standardized Frictionless Data Package.

For Finnish compliance, schemas MUST declare primary and foreign keys for audit-trail references. See Finnish bookkeeping and tax-audit compliance.

← Repository-level README expectations BusDK Docs VAT area (reference data and filed summaries) →

Sources

• Frictionless Table Schema as the contract
• Data directory layout (principles)
• Finnish bookkeeping and tax-audit compliance