// 0pt.dataset.schema.v1 // // A dataset 0pt produces or manages -- the thing `dataset_iceberg_mapping` // (src/types/dataset-iceberg-mapping.ts) points at, and the thing // IcebergWriteService (src/services/iceberg-write-service.ts) writes through // ST0RE's table registry on behalf of. This type intentionally stays small: // it is a reference/registry entry, not a copy of dataset content, schema, // or quality state (that boundary is documented in // docs/ICEBERG_WRITE_BOUNDARY.md -- 0pt owns dataset production, but this // specific type is only what DatasetIcebergMappingService/IcebergWriteService // need to confirm a dataset_id is real before pointing a table at it). // Union of both dataset-type vocabularies that existed in this repo's two // remotes ("structured" from one, "feature"/"derived" from the other); // nothing is dropped by the GitHub<->Gitea unification merge. export type DatasetType = "raw" | "structured" | "enriched" | "curated" | "feature" | "derived"; export type DatasetStatus = "active" | "archived" | "deprecated"; export interface Dataset { dataset_id: string; tenant_id: string; project_id: string; name: string; dataset_type: DatasetType; status: DatasetStatus; description?: string; correlation_id?: string; created_at: string; updated_at?: string; created_by?: string; } export interface CreateDatasetInput { dataset_id: string; tenant_id: string; project_id: string; name: string; dataset_type: DatasetType; description?: string; status?: DatasetStatus; correlation_id?: string; created_by?: string; }