Mapping Import

Start with what you already have

Most teams already maintain source‑to‑target mappings in spreadsheets. Magellan lets you import them directly — preserving your existing work while unlocking Semantic Match suggestions, lineage derivation, and the ULID‑keyed audit trail.

Compare mapping-import steps

Each step is a discrete pass with its own input artifact, its own role gate, and its own ledger entries. Predictive mapping enters only after validation and enrichment are complete — no Semantic Match Engine proposal reaches a Reviewer queue without a typed, validated mapping behind it, and a named SME signs every released decision.

Step	Input artifact	Predictive mapping	Human-in-the-loop	Ledger entries written
Import existing spreadsheets	.xlsx · .xls · .csv (column-mapping profile)	No	Contributor — resolve deferred rows	`SchemaIngestedMappingImportedMappingImportDeferredColumnProfileApplied`
Automatic MetaModel conversion	Imported rows + registered schema ULIDs	No	Contributor — resolve unmatched paths	`MetaModelProjectedTypeNarrowingDetectedMappingImportDeferred`
Validation & enrichment	Projected MetaModel mappings	Partial	Contributor — acknowledge warnings · Reviewer — clear errors	`ValidationRunCompletedEntityAnnotatedValidationFindingRaised`
Immediate predictive enhancement	Validated mappings + adjacent target attributes	Yes	Contributor — accept/reject proposals · Approver — release	`SemanticSuggestionReturnedSemanticSuggestionAcceptedLineageEdgeDerivedImpactAnalysisRun`

Import existing spreadsheets

Input artifact

.xlsx · .xls · .csv (column-mapping profile)

Predictive mapping

Human-in-the-loop

Contributor — resolve deferred rows

Ledger entries

SchemaIngestedMappingImportedMappingImportDeferredColumnProfileApplied

Automatic MetaModel conversion

Input artifact

Imported rows + registered schema ULIDs

Predictive mapping

Human-in-the-loop

Contributor — resolve unmatched paths

Ledger entries

MetaModelProjectedTypeNarrowingDetectedMappingImportDeferred

Validation & enrichment

Input artifact

Projected MetaModel mappings

Predictive mapping

Partial

Human-in-the-loop

Contributor — acknowledge warnings · Reviewer — clear errors

Ledger entries

ValidationRunCompletedEntityAnnotatedValidationFindingRaised

Immediate predictive enhancement

Input artifact

Validated mappings + adjacent target attributes

Predictive mapping

Yes

Human-in-the-loop

Contributor — accept/reject proposals · Approver — release

Ledger entries

SemanticSuggestionReturnedSemanticSuggestionAcceptedLineageEdgeDerivedImpactAnalysisRun

Import Existing Spreadsheets

Bring your current source‑to‑target mapping spreadsheets directly into Magellan — no rework, no re‑keying. Each import atomically creates the source schema(s), the target schema(s), and a mapping (all sharing one version number) and writes MappingImported to the ULID‑keyed ledger so the spreadsheet that was the prior source of truth becomes a hash‑checkable provenance artifact.

Automatic MetaModel Conversion

Imported mappings are normalized into Magellan's governed MetaModel — gaining version control, lineage derivation, and impact analysis from the first ingest. The conversion writes MetaModelProjected entries so the projection is replayable from the ledger alone.

Validation & Enrichment

Imported mappings are validated against target schemas, type-checked, and annotated with parent-context, abbreviation, and domain-vocabulary metadata before they reach a Contributor's queue — every check writes a ValidationRunCompleted entry to the ledger.

Immediate Predictive Enhancement

Once imported, the Semantic Match Engine proposes mappings for the rest of the target schema, lineage derivation activates, and impact analysis runs — extending the value of work your team has already done. Every model call writes a SemanticSuggestionReturned entry to the ledger.