Treemap layer (sugar) — mark

Creates a treemap from a hierarchy table (id/parent columns, leaf sizes in a value column) using plotit's self-contained squarified tiling. Equivalent to the pipeline as_graph(hierarchy) |> layout_treemap() |> mark_rect(data = ~leaves); the laid-out tables (nodes/edges/leaves) are stored on @graph for further tuning.

Usage

mark_treemap(
  plot,
  data = NULL,
  node_color = ._MARK_STYLE$primary,
  show_labels = TRUE,
  ...
)

Arguments

plot: A plotit object whose data is a hierarchy table with id, parent, and leaf-level value columns (build via as_graph() on the same shape). A global encode(fill = ...) maps tile fill against any hierarchy column.
data: Optional hierarchy table for this layer.
node_color: Default tile fill when no fill aesthetic is mapped (default ._MARK_STYLE$primary = "#4E79A7").
show_labels: If TRUE (default), draw leaf ids at tile centres. Labels render white over the unmapped brand-blue fill; when a fill is mapped they fall back to near-black – chain mark_text(data = ~leaves, colour = ...) for full control.
...: Unused; tiling fine-tuning lives on layout_treemap() in the explicit pipeline form.

Value

Modified plotit object; @graph holds nodes/edges/leaves.

Details

Fully self-contained: no treemapify dependency, deterministic Bruls squarify layout. Tiles receive the unified white hairline separators and coordinate axes are blanked (the diagram is coordinate-free). A mapped fill column ships with the curated token palette – friendly qualitative for categories, viridis sequential for continuous values (chain scale_fill() to replace it).

References

AntV G2: Treemap (graphlib)

Examples

h <- data.frame(
  id     = c("root", "A", "B", "a1", "a2", "b1"),
  parent = c(NA, "root", "root", "A", "A", "B"),
  value  = c(NA, NA, NA, 30, 20, 50)
)
h |>
  plotit(encode(fill = id)) |>
  mark_treemap()