Skip to content

Concepts

Background pages — the why behind the how.

  • Re-run cascade — the shared-table semantics that scope selective re-runs.
  • DAG wiring — how GetCDMTableId(@table) lookups and StoreSequence(…) writers become edges in the pipeline DAG.
  • CSV / PostgreSQL parity — why the CSV target produces the same data as the PostgreSQL target, what engine version you need, and the three asymmetries that remain (vocab loading, eras, cascade cleanup).