Build a Geoscale from a flat table of leaf regions

The general constructor. Takes a wide data.frame with one row per atom (the finest region) and one column per level, and returns a Geoscale.

Usage

geoscale_from_leaves(
  leaves,
  levels,
  key = NULL,
  weights = NULL,
  default_weight = NULL,
  members = NULL,
  geometry = NULL,
  name = "",
  desc = "",
  ...
)

Arguments

leaves

data.frame with one row per atom, one column per name in levels, and optionally numeric weight columns.

levels

Ordered character vector of level names, coarsest first. Each must be a column of leaves.

key

Name of the column holding the unique atom key. Defaults to "region" if present, otherwise the finest level (the last entry of levels), which is copied into a region column.

weights

Character vector naming the weight columns. Defaults to all numeric columns that are neither levels nor reserved names.

default_weight

The weight used when a caller does not name one. Defaults to the first entry of weights.

members

Optional named list giving the ordered code vocabulary per level. Derived from leaves when NULL (first-appearance order).

geometry

Optional sfc with one geometry per row of leaves.

name, desc

Short name and description.

...

Further named entries merged into meta (e.g. crs, source).

Value

A Geoscale.

Details

Blank strings ("") in level columns are normalised to NA, meaning "this atom has no code at this level" — partial coverage is normal in real region tables.

Examples

df <- data.frame(
  country = c("C1", "C1", "C1", "C2"),
  zone    = c("Z1", "Z1", "Z2", "Z3"),
  atom    = c("A1", "A2", "A3", "A4"),
  km2     = c(100, 200, 300, 400)
)
geoscale_from_leaves(df, levels = c("country", "zone", "atom"))
#> Geoscale: <unnamed> 
#> Levels (3, coarsest first):
#>   - country (2)
#>     - zone (3)
#>       - atom (4)
#> Atoms: 4
#> Weights: km2 (default: km2)