scale-shape-binned
Binned shape scale: cuts a continuous variable into n-breaks bins, each bin gets one shape from palette.
The scale stays continuous: the trained domain is numeric and the resolver snaps each row to the bin its value falls into. The legend renders one glyph per bin at the midpoint.
Usage
scale-shape-binned(
n-breaks: 4,
breaks: auto,
palette: auto,
name: none,
limits: none,
oob: "drop",
labels: auto,
)Parameters
| Parameter | Default | Description |
|---|---|---|
n-breaks |
4 |
Number of bins to partition the continuous domain into. Ignored when breaks is set. |
breaks |
auto |
Array of bin edges, or auto to derive equal-width bins from n-breaks. Edges define the bin boundaries; n-breaks is ignored when set. |
palette |
auto |
Array of shape keywords, one per bin, or auto for the library default. |
name |
none |
Legend title. Overrides any name set via labels when both are present. |
limits |
none |
Continuous (lo, hi) pair pinning the domain, or none. |
oob |
"drop" |
Out-of-range policy: "drop" (default) removes rows whose value falls outside limits; "squish" clamps continuous values to the nearest endpoint. |
labels |
auto |
Array of legend labels aligned with bin midpoints, or auto. |
Returns
Scale object consumed by plot.
Examples
Bin a continuous score column into four shape buckets.
#let d = range(0, 12).map(i => (x: i, y: i, w: i + 1))
#plot(
data: d,
mapping: aes(x: "x", y: "y", shape: "w"),
layers: (geom-point(size: 3pt),),
scales: (scale-shape-binned(n-breaks: 4),),
width: 10cm,
height: 6cm,
)
