dr_metric()

Define a reusable calculation, such as a total or cancellation rate, for a data product. Choose its grouping columns, unit and date behavior, then calculate it with dr_measure().

Usage

dr_metric(
  id,
  product,
  expr = NULL,
  compute = NULL,
  dimensions = character(),
  time_column = NULL,
  time_behavior = if (is.null(time_column)) "flow" else "stock",
  unit = "",
  owner = "",
  description = "",
  version = "1.0.0",
  approved = FALSE,
  na_policy = "reject",
  empty_policy = "error",
  code_version = NULL,
  input_columns = NULL
)

Arguments

id, version: Identity and version.
product: Input asset id.
expr: Tidy evaluation summary expression, e.g. sum(reserve).
compute: Alternative function(data, dimensions, params) for complex metrics.
dimensions: Permitted grouping columns.
time_column: Column representing the business date.
time_behavior: stock requires exactly one selected date; flow may span dates. Defaults to stock when time_column is supplied, otherwise flow.
unit: Optional unit.
owner, description: Optional business metadata.
approved: Whether business review is complete.
na_policy: Policy description; must be consistent with the expression.
empty_policy: Policy description.
code_version: Version of metric code and dependencies. Required for approved definitions; exploratory definitions may omit it.
input_columns: Optional explicit input columns for dynamic expressions or custom functions. Known expression columns are always checked as well.

Value

Metric specification. Reports execute it directly without an LLM.

Examples

metric <- dr_metric(
  "orders.total", "orders", expr = sum(amount), time_behavior = "flow",
  unit = "EUR", owner = "Analytics", description = "Total order value",
  approved = TRUE, code_version = "v1"
)
metric

Documentation source snapshot · View source at 513e09e ↗
Code examples are documented source examples; they are not executed in this website build.