Error when variable used for color is a factor and contains unused levels

[jbengler]

Reprex

library(tidyverse)
library(tidyplots)

# works
study %>% 
  tidyplot(x = treatment, y = score, color = treatment) %>% 
  add_mean_dot()

# works
study %>% 
  filter(treatment != "A") %>% 
  tidyplot(x = treatment, y = score, color = treatment) %>% 
  add_mean_dot()

# works
study %>% 
  mutate(treatment = factor(treatment)) %>% 
  tidyplot(x = treatment, y = score, color = treatment) %>% 
  add_mean_dot()

# error
study %>% 
  mutate(treatment = factor(treatment)) %>% 
  filter(treatment != "A") %>% 
  tidyplot(x = treatment, y = score, color = treatment) %>% 
  add_mean_dot()

# works
study %>% 
  mutate(treatment = factor(treatment)) %>% 
  filter(treatment != "A") %>% 
  mutate(treatment = fct_drop(treatment)) %>% 
  tidyplot(x = treatment, y = score, color = treatment) %>% 
  add_mean_dot()

[jbengler]

One option would be to routinely run fct_drop() on the color variable when starting the plot with tidyplot().