dickoa / fontscales

Use Iconographic Fonts in 'ggplot2'

fontscales

The goal of fontscales is to is to let you use fancy iconographic fonts with ease in ggplot2.

This package relies heavily on ragg, please see this article for more information.

Installation

You can install the dev version of fontscales from Github with:

require("devtools")
install_github("tidymodels/fontscales")

StateFace

The StateFace font by ProPublica contains the 50 different U.S. states as tiny glyphs and can thus be used in data visualizations.

The data set usa_arrests represents a data set where this might be useful where the state variable contains the name of the state

library(fontscales)
usa_arrests
#> # A tibble: 50 x 5
#>    state       urban_pop murder assault  rape
#>    <chr>           <int>  <dbl>   <int> <dbl>
#>  1 Alabama            58   13.2     236  21.2
#>  2 Alaska             48   10       263  44.5
#>  3 Arizona            80    8.1     294  31  
#>  4 Arkansas           50    8.8     190  19.5
#>  5 California         91    9       276  40.6
#>  6 Colorado           78    7.9     204  38.7
#>  7 Connecticut        77    3.3     110  11.1
#>  8 Delaware           72    5.9     238  15.8
#>  9 Florida            80   15.4     335  31.9
#> 10 Georgia            60   17.4     211  25.8
#> # … with 40 more rows

Using ggplot2 you pass the state name variable to the label aesthetic and use geom_stateface() to draw the states. (This should function similary to geom_text() or geom_point()).

library(ggplot2)
ggplot(usa_arrests, aes(murder, assault, label = state, color = urban_pop)) +
  geom_stateface()

And other ggplot2 functions work as expected

library(tidyr)
usa_arrests %>%
  pivot_longer(-c(state, urban_pop)) %>%
  ggplot(aes(urban_pop, value, label = state)) +
  geom_stateface() +
  facet_wrap(~name, scales = "free_y")