passing tibble for scale_fill_manual

[nithinmkp]

I want to change the color of the fill items in a ggplot .
I know I can use like this
scale_fill_manual(values=c("A"="red","B"="blue")

is there a way to pass it something like this
scale _fill_manual(values=tibble(c("A,"B),c("red","blue"))

[JackDavison]

To the best of my understanding, scale_fill_manual(values = ...) wants a vector.

However, if it is easier for you to create category-colour pairs in a dataframe, its not too tricky to turn it into a named vector.

I've also given an example of a function you could write to make this easier to do in future (could go in a personal package)

library(dplyr, warn.conflicts = FALSE)
library(ggplot2)

# plot
plot_dat <- 
  tibble(x = c("A", "B"),
         y = c(2, 5))

plt <- ggplot(plot_dat, aes(x,y)) + geom_col(aes(fill = x))

# colors
color_dat <-
  tibble(cat = c("A", "B"),
         col = c("red", "blue"))

cols <- rlang::set_names(color_dat$col, color_dat$cat)

plt + scale_fill_manual(values = cols)

# write a function to make it easier to do!

scale_fill_tbl <- function(cols_tbl, cat = "cat", col = "col"){
  
  scale_fill_manual(
    values = rlang::set_names(color_dat[[col]], color_dat[[cat]])
  )
  
}

plt + scale_fill_tbl(cols_tbl = color_dat)

^{Created on 2022-10-15 with reprex v2.0.2}

[nithinmkp]

Thanks a Lot Jack!! that was neat

[andresrcs]

Just for variety's sake. You can also define color in your data itself and use scale_fill_identity()

library(dplyr)
library(ggplot2)

plot_dat <- 
    tibble(x = c("A", "B"),
           y = c(2, 5))

color_dat <-
    tibble(cat = c("A", "B"),
           col = c("red", "blue"))

plot_dat %>% 
    inner_join(color_dat, by = c(x = 'cat')) %>% 
    ggplot(aes(x, y)) +
    geom_col(aes(fill = col)) +
    scale_fill_identity(guide = "legend")

^{Created on 2022-10-18 with reprex v2.0.2}