dr_lineage()

Traverses dataset IDs while retaining version IDs on each returned edge. It does not infer column lineage or reconstruct a single historical DAG. Run results expose only recorded inputs for that exact execution, without querying the latest lake state or inferring unrecorded upstream edges.

Usage

dr_lineage(
  x,
  asset = NULL,
  direction = c("upstream", "downstream"),
  recursive = TRUE
)

Arguments

x: Connected lake, run result, measurement, measurement set, dbt result or dbt artifact directory. Measurement edges describe the input release and metric version; they have no execution run ID.
asset: Optional starting dataset ID (dbt unique ID for dbt inputs).
direction: Follow upstream inputs or downstream consumers.
recursive: Follow all reachable edges, with cycle protection.

Value

A tibble of lineage edges with endpoint IDs and versions. dbt edges have empty version fields because a manifest declares dependencies.

Examples

orders <- dr_product("orders", data.frame(id = 1:3))
totals <- dr_product("totals", orders) |>
  dr_add_recipe(dr_recipe() |> dr_step_summarise(n = dplyr::n()))
dr_lineage(dr_run(totals))

Documentation source snapshot · View source at d9cb454 ↗
Code examples are documented source examples; they are not executed in this website build.