Documentation
¶
Overview ¶
Package materialize writes a compiled, reviewed assessment.ReportDataset's ranking back onto each cited OpportunityAssessment row's rank projection columns, and marks the dataset "final" — the last step in the compile → review → materialize workflow (prism-roadmap PRD FR14 / RMI-OMNIROADMAP-006).
This is the promised follow-up to RMI-OMNIROADMAP-001, which deliberately left opportunity_rank_calculated/final unset on OpportunityAssessment: ranking is inherently cross-opportunity and could not be derived when a single assessment was saved. Now that a full-corpus compile (compile.Compile) has produced a dataset and it has been reviewed (review.Apply, then a fresh compile), this writes the result back so a query against one assessment can find its current rank directly, without needing the whole dataset.
Index ¶
Constants ¶
const StatusFinal = "final"
StatusFinal marks a dataset as reviewed and rank-materialized.
Variables ¶
This section is empty.
Functions ¶
func Materialize ¶
func Materialize(ctx context.Context, s Store, datasetID string, dataset assessment.ReportDataset) error
Materialize writes dataset's ranking back onto each cited OpportunityAssessment row, then marks datasetID StatusFinal.
Excluded opportunities (RankedOpportunity.Excluded != "") are skipped — they have no CalculatedRank/FinalRank to write. Each assessment's rank is set independently (not wrapped in one cross-call transaction, since Store is an interface over persistence this package doesn't own the transaction boundary for); a failure partway through leaves some assessments updated and some not, but SetOpportunityRank is idempotent, so re-running Materialize after fixing the cause is always safe.
Source Files