Position scales — scale_x_continuous • vellumplot

Declare a position scale to override the trained default. scale_x_continuous() / scale_y_continuous() handle numeric (and date/time) axes; scale_x_discrete() / scale_y_discrete() handle categorical (band) axes and let you set the level order via limits.

Usage

scale_x_continuous(
  plot,
  limits = NULL,
  trans = "identity",
  breaks = NULL,
  labels = NULL,
  name = NULL,
  expand = NULL,
  sec.axis = NULL
)

scale_y_continuous(
  plot,
  limits = NULL,
  trans = "identity",
  breaks = NULL,
  labels = NULL,
  name = NULL,
  expand = NULL,
  sec.axis = NULL
)

scale_x_discrete(plot, limits = NULL, name = NULL)

scale_y_discrete(plot, limits = NULL, name = NULL)

Arguments

plot: A PlotSpec.
limits: For continuous scales a numeric length-2 domain c(min, max); for discrete scales a character vector of levels (sets order / subset). NULL trains from the data.
trans: Transformation: "identity" (default), "log10", "sqrt", "symlog" (symmetric log — linear through zero, logarithmic in the tails, so signed data and zero/negatives read on one axis), "reverse", or a scales::transform_log10()-style transform object.
breaks, labels: Explicit break positions (data units) and their labels, or NULL to compute them.
name: Axis title, or NULL to derive from the encoding.
expand: Axis padding beyond the data. NULL (default) keeps the usual 5% breathing room; a numeric c(mult, add) sets a proportion of the range plus a constant (either may be 0), applied to both ends. Both act in the scale's transformed space, so on a trans = "log10" axis add is in log units (matching how the axis is drawn). Use expand = c(0, 0) to clamp the axis exactly to the data (e.g. a bar chart's baseline), or expand = c(0.1, 0) for a wider margin.
sec.axis: A secondary axis from sec_axis() / dup_axis(), drawn on the opposite edge, or NULL for none. Continuous Cartesian plots only (see sec_axis() for the current limitations).

Value

The modified PlotSpec.

Examples

vplot(mtcars) |> mark_point(x = wt, y = mpg) |> scale_x_continuous(limits = c(0, 6))

vplot(mtcars) |> mark_point(x = wt, y = mpg) |> scale_y_continuous(trans = "sqrt")