Feature request: Add possibility for custom country categories

[rempsyc]

Follow-up on #6 and #10.

Here is my use case: My data set has MANY countries and it is not useful to plot all countries when most of them after the top 8 have very small percentages. So I create a category called "Other" to group all other countries together, which I also want to represent on the plot. But then I get the error stated above. I wish there was a way to plot custom categories or just NA values (but that it also appears in the plot legend).

packageVersion("ggflags")
#> [1] '0.0.4'

country_4legend <- c("us", "gb", "nl", "de", "ca", "be", "au", "ru", "other")

df <- data.frame(x = rep(1:10, each = 10),
                 y = rep(1:10, 10),
                 country = c(country_4legend[-9], 
                             rep(country_4legend, each = 10), "other", "other"))

# Custom category doesn't work
ggplot2::ggplot(df, ggplot2::aes(x, y, country = country)) +
  ggflags::geom_flag(size = 8.5)
#> Error in (function (classes, fdef, mtable) : unable to find an inherited method for function 'grobify' for signature '"NULL"'

# Missing values work
country_4legend <- c("us", "gb", "nl", "de", "ca", "be", "au", "ru", NA)

df <- data.frame(x = rep(1:10, each = 10),
                 y = rep(1:10, 10),
                 country = c(rep(country_4legend, each = 11), NA))

ggplot2::ggplot(df, ggplot2::aes(x, y, country = country)) +
  ggflags::geom_flag(size = 8.5)
#> Warning: Removed 12 rows containing missing values or values outside the scale range
#> (`geom_flag()`).

# But not the legend
ggplot2::ggplot(df, ggplot2::aes(x, y, country = country)) +
  ggflags::geom_flag(size = 8.5) +
  ggflags::scale_country(breaks = country_4legend)
#> Warning: Removed 12 rows containing missing values or values outside the scale range
#> (`geom_flag()`).
#> Error in (function (classes, fdef, mtable) : unable to find an inherited method for function 'grobify' for signature '"NULL"'

country_4legend2 <- c(country_4legend[-9], "NA")
ggplot2::ggplot(df, ggplot2::aes(x, y, country = country)) +
  ggflags::geom_flag(size = 8.5) +
  ggflags::scale_country(breaks = country_4legend2)
#> Warning: Removed 12 rows containing missing values or values outside the scale range
#> (`geom_flag()`).

^{Created on 2024-05-31 with reprex v2.1.0}

[rempsyc]

Actually, the feature already exists! It is just not documented but labels can be fed directly to scale_country().

library(ggflags)

d <- data.frame(
  x = rnorm(10), y = rnorm(10),
  country = sample(c("AR", "FR"), 10, TRUE),
  stringsAsFactors = FALSE
)
p <- ggplot2::ggplot(d, ggplot2::aes(x = x, y = y, country = tolower(country))) +
  geom_flag() +
  scale_country(labels = c("AR", "FR"))
p

^{Created on 2024-05-31 with reprex v2.1.0}