Back to the workflow

    How It Works

    Import

    Upload Avro, XML Schema, JSON Schema, Excel, Erwin XML, or connect via JDBC. Each ingest writes a SchemaIngested entry to the ledger — ULID‑keyed, hash‑chained, and stamped with the source fingerprint — so nothing reaches the MetaModel without a signed provenance row.

    Import is the first place the ledger discipline shows up. Before a single mapping is drafted, the source artifact has already been hashed, stamped, and chained. A reviewer who arrives six quarters later can ask 'what was the source bytes that produced this MetaModel state?' and get a SHA-256 answer in a single ledger lookup.

    Each ingest writes a `SchemaIngested` entry carrying the source fingerprint (sha256 of the raw bytes), the importer module identifier and version (e.g. `AvroImporter@x.y.z`), the actor's role at time of action, and the assigned schema ULID. The full list of supported source formats — and what each adapter preserves — is documented under Schema Formats.

    Ingest is intentionally a Contributor-class action under the seven-role RBAC model. Reader cannot ingest, Approver does not ingest as part of their approval duty, and Administrator cannot ingest in a way that would let them silently insert a schema into a downstream production mapping. The role split is enforced by the data structure: every `SchemaIngested` entry carries the actor role, and the role-at-time-of-action is what later releases will be checked against.

    Custom formats inherit the same discipline through the documented Custom Import adapter interface. There is no second-class ingest path — a niche Erwin extension, a proprietary mainframe COBOL copybook, and an Avro `.avsc` all produce `SchemaIngested` entries with the same fields and the same chain participation.

    Re-ingesting an updated source writes `SchemaIngested` for the new bytes plus `SchemaEvolutionEvaluated` against the prior published version. Additive changes auto-classify as minor; renames, deletions, and type-narrowing classify as major; both require Approver acknowledgment before reaching Published. This is what closes the schema-evolution gap that under SOX §404 ITGCs and 21 CFR Part 11 §11.10(c) is otherwise the most common audit finding.

    Import sets up everything downstream. The schema ULID assigned at ingest is what every subsequent `MappingDrafted`, `LineageEdgeDerived`, `ImpactAnalysisRun`, and `ArtifactExported` entry will reference. A clean import — hashed, stamped, role-attributed — is the precondition for a clean Map and a clean Govern phase.

    Audit ledger entries written by this capability

    Each action in this capability appends one of the following entry types to the ULID‑keyed, SHA‑256 hash‑chained audit ledger. A reviewer can replay the chain to reconstruct exactly what happened, when, and under which role.

    SchemaIngestedSchemaParsedSchemaVersionPublishedSchemaEvolutionEvaluated
    Read the full Audit & Trust architecture

    See import demonstrated against your data

    Pick a sample dataset or bring your own. We will walk the ledger from ingest through Approver release in a single working session.

    Request a demo

    Related across the platform

    Where this capability hands off — upstream adapters, downstream governance, and the underlying audit primitives.

    All How It Works