color_loon: Create a palette with loon's color mapping

Description

Used to map nominal data to colors. By default these colors are chosen so that the categories can be well differentiated visually (e.g. to highlight the different groups)

Usage

color_loon()

Arguments

Value

A function that takes a vector with values and maps them to a vector of 6 digit hexadecimal encoded color representation (strings). Note that loon uses internally 12 digit hexadecimal encoded color values. If all the values that get passed to the function are valid color names in Tcl then those colors get returned hexencoded. Otherwise, if there is one or more elements that is not a valid color name it uses the loons default color mapping algorithm.

Details

This is the function that loon uses by default to map values to colors. Loon's mapping algorithm is as follows:

if all values already represent valid Tk colors (see tkcolors) then those colors are taken
if the number of distinct values is less than the number of values in loon's color mapping list then they get mapped according to the color list, see l_setColorList and l_getColorList.
if there are more distinct values than there are colors in loon's color mapping list then loon's own color mapping algorithm is used. See loon_palette and the details section in the documentation of l_setColorList.

For other mappings see the col_numeric and col_factor functions from the scales package.

Examples

Run this code


pal <- color_loon()
pal(letters[1:4])
pal(c('a','a','b','c'))
pal(c('green', 'yellow'))

# show color choices for different n's
if (requireNamespace("grid", quietly = TRUE)) {
  grid::grid.newpage()
  grid::pushViewport(grid::plotViewport())
  grid::grid.rect()
  n <- c(2,4,8,16, 21)
  # beyond this, colors are generated algorithmically
  # generating a warning
  grid::pushViewport(grid::dataViewport(xscale=c(0, max(n)+1),
                     yscale=c(0, length(n)+1)))
  grid::grid.yaxis(at=c(1:length(n)), label=paste("n =", n))
  for (i in rev(seq_along(n))) {
   cols <- pal(1:n[i])
   grid::grid.points(x = 1:n[i], y = rep(i, n[i]),
                     default.units = "native", pch=15,
                     gp=grid::gpar(col=cols))
  }
  grid::grid.text("note the first i colors are shared for each n",
                  y = grid::unit(1,"npc") + grid::unit(1, "line"))
}