convert tibble to character vector for scale_fill_manual

[zoowalk]

I must be missing something obvious here.

How can I convert a tibble into a character vector so that I can use it in scales_fill_manual(values=...)
(I disregard the option to simply merge both tibbles here).

library(tidyverse)

this is what I have

df_my_colors<- tibble::tribble(
  ~unit,  ~color,
      "Ideal",  "blue",
      "Premium",   "red",
      "Good", "green"
  )

this is what I want

c("Ideal"="blue", "Premium"="red", "Good"="green")
#>   Ideal Premium    Good 
#>  "blue"   "red" "green"

this is what I tried (among others)

my_colors <- df_my_colors %>% 
  pivot_wider(id_cols="unit", names_from="unit", values_from="color")
  

that's how I want to use it.

diamonds %>% 
  filter(cut %in%df_my_colors$unit) %>% 
  ggplot()+
  geom_bar(aes(x=cut,
               fill=cut),
           stat="count")+
 #scale_fill_manual(values=c("Ideal"="blue", "Premium"="red", "Good"="green"))
 scale_fill_manual(values=my_colors)

[Matthias]

what you want to create is called a "named vector".
Basically you extract the information from the tibble, create a vector of the column "color", then you add the names from the column "unit" to the vector, using the "names" function.

my_colors = df_my_colors$color
names(my_colors) = df_my_colors$unit

gives: (ah sorry I used other colors for testing)

    Ideal     Premium        Good 
    "blue3" "firebrick"  "seagreen"

then you can use this with: scale_fill_manual(values = my_colors)
The advantage is: you can reorder the columns and the associated colours will be the same!