dr_measure()

Usage

\method{collect}{dr_measurement_set}(x, ...)

\method{print}{dr_measurement_set}(x, ...)

dr_measure(
  x,
  metric = NULL,
  by = character(),
  at = NULL,
  release = NULL,
  filters = list(),
  params = list(),
  record = NULL,
  metrics = NULL,
  period = c("each", "aggregate")
)

Arguments

x

Connected lake or successful dr_run_result. In-memory dataraft.core::dr_run() results support exploratory measurements with the same definitions. They cannot be recorded or saved in issued reports, even for approved metrics.

...

Reserved for future extensions.

metric

Single metric definition.

by

Grouping columns selected from the metric's permitted dimensions. Omitting by calculates an overall total, with a reminder when dimensions are available. Use by = character() to request an overall total explicitly.

at

Business date, or a vector for flow metrics.

release

Optional explicit product release id for a connected lake. For a publication result it must be omitted or match that exact release.

filters

Named list of exact-match filters on permitted dimensions.

params

Parameters passed to custom compute functions.

record

Record definition and lineage. Defaults to TRUE on a writable lake and FALSE on a read-only lake or publication result. Use a connected writable lake for record = TRUE. Unrecorded results still carry their complete metric definition and pinned release in the manifest. Exploratory metrics always default to record = FALSE and cannot be saved in reports.

metrics

Nonempty list of metric definitions for a measurement set. Optional unique names label metrics in the collected table. Supply either metric or metrics.

period

For metrics, "each" calculates each selected date separately; "aggregate" calculates across selected dates. Stock metrics still require one date. With at = NULL, dates are not split automatically.

Value

Examples

\dontshow{if (requireNamespace("dataraft.lake", quietly = TRUE) && requireNamespace("duckdb", quietly = TRUE)) withAutoprint(\{ # examplesIf}
root <- tempfile("dataraft-example-")
config <- dataraft.lake::dr_lake_config(
  dataraft.lake::dr_registry_duckdb(file.path(root, "lake.db")),
  dataraft.lake::dr_storage_local(file.path(root, "data")),
  landing = file.path(root, "landing"), backend = "duckdb"
)
lake <- dataraft.lake::dr_connect_lake(config)
path <- file.path(root, "orders.csv")
utils::write.csv(data.frame(order_id = 1:2, amount = c(25, 75)), path,
  row.names = FALSE)
source <- dataraft.core::dr_source_file("orders.file", path, reader = utils::read.csv)
contract <- dataraft.core::dr_contract(
  "orders", version = "1.0.0", columns =
  c(order_id = "integer", amount = "numeric"), key = "order_id"
)
release <- dataraft.core::dr_product("orders", contract = contract, code_version = "v1") |>
  dataraft.core::dr_add_source(source) |> dataraft.core::dr_publish(to = lake)
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"
)
release |> dr_measure(metric)
dataraft.lake::dr_disconnect_lake(lake)
unlink(root, recursive = TRUE)
\dontshow{\}) # examplesIf}