Interactive drawing on ggplot2 visualizations using D3

The drawit function allows the user to draw a single, continuous, one-to-one line. This is particularly useful for tasks in which the user is expected to estimate trends or sketch expected relationships in the data.

Usage

drawit(
  p,
  ...,
  width = NULL,
  height = NULL,
  shiny_message_loc = NULL,
  show_on_finish = FALSE,
  draw_start = NULL,
  smoother = 0.001,
  interpolator = NULL
)

Arguments

p: A ggplot2 object.
...: Additional arguments passed
width: Width of the output widget in pixels.
height: Height of the output widget in pixels.
shiny_message_loc: Name of the Shiny input/output binding used for messaging.
show_on_finish: If TRUE, displays a secondary geom after the user finishes drawing. Note that this will always be the second geom layer within your ggplot2 object.
draw_start: Specifies where drawing interaction begins, on the scale of your data.
smoother: A number greater than 0 that controls how much nearby x-values are binned together. Larger values group more points, creating a smoother but less detailed drawing (useful for dense data). Values closer to 0 keep points more separate, preserving detail (useful when the exact x-y relationship matters)
interpolator: A number greater than 0 that controls how may "in between" x-values are added to gaps between data. Values closer to zero indicate that the function is looking for smaller gaps, thus filling in more points. Higher values avoid adding interpolated points, making the drawing "chunkier."

Value

An r2d3 htmlwidget.

Details

For more flexible, free-form drawing, see sketchit().

Currently, drawit can support geom_scatter, geom_line, and geom_smooth, with a maximum of two geoms per plot.

A highlighted region indicates where drawing is permitted. Users draw from left to right, and the interface provides visual feedback when portions of the data range have not yet been covered.

The drawing resolution is tied to the x-values in the data. As a result, extending the x-axis far beyond the observed data without adjusting draw_start may lead to an overly granular drawing experience

Shiny Integration

When used within a shiny application, the widget returns the recorded drawing data as a structured object once the user has filled the drawable region. Each x-value in the original data is paired with a corresponding user-drawn y-value.

The returned data includes:

x: The x-values drawn by the user, preserving the original data as much as possible
y: The user-drawn y-values corresponding to each x

Because drawit() enforces a single continuous line, the output represents a fully specified function over the domain of the data.

Examples

if (FALSE) { # \dontrun{
# Basic usage
p <- ggplot(mtcars, aes(x = wt, y = mpg)) +
  geom_point(size = 2, colour = "magenta") +
  labs(x = "Weight", y = "MPG")

drawit(p)

# Pipe-friendly usage
(ggplot(mtcars, aes(x = wt, y = mpg)) +
  geom_point(size = 2, colour = "magenta") +
  labs(x = "Weight", y = "MPG")) |>
  drawit()
} # }