RGB to hex

library(dplyr)
library(colorspace)
library(ggplot2)

Let’s say we have a dataframe with R, G, B values, but we want the hex. Why? Maybe we want to use scale_color_identity to plot the values in some other columns. Let’s say x and y.

set.seed(17)
rgbtib <- tidyr::expand_grid(x = 1:10, y = 1:10) %>% 
  mutate(R = sample(0:255, 100),
                         G = sample(0:255, 100),
                         B = sample(0:255, 100))
rgbtib

# A tibble: 100 × 5
       x     y     R     G     B
   <int> <int> <int> <int> <int>
 1     1     1   177   166    44
 2     1     2   231    80    48
 3     1     3   224   188   136
 4     1     4   221   168   181
 5     1     5   107    82    93
 6     1     6   238   111   179
 7     1     7   246   242   124
 8     1     8   230    57   213
 9     1     9   172   151   130
10     1    10   166    93    84
# ℹ 90 more rows

We can use {colorspace}, but it’s convoluted- have to make an RGB object first, and then convert to hex. And the RGB need to be on 0-1, not 0-255.

Writing that out doesn’t work because the colorspace RGB object can’t get stuffed in the dataframe. But this is the idea

rgbtib_writeout <- rgbtib %>% 
  # Convert to 0-1
  mutate(across(all_of(c('R', 'G', 'B')), ~./255)) %>%
  # Create the rgb object
  mutate(rgbobj = colorspace::RGB(R, G, B)) %>% 
  # Get the hex values
  mutate(hexval = colorspace::hex(rgbobj))

So, make a function. Have a maxval the user can pass (don’t assume it’s 1 or 255).

rgb2hex <- function(R, G, B, maxval = 255) {
  rgbobj <- colorspace::RGB(R/maxval, G/maxval, B/maxval)
  hexval <- colorspace::hex(rgbobj)
  return(hexval)
}

Test that

rgb2hex(177, 41, 147)

[1] "#D970C8"

apparently quarto doesn’t do the cool printing of color thing in output, just input.

"#D970C8"

[1] "#D970C8"

Now, use that in the mutate

rgbtib <- rgbtib %>% 
  mutate(hexvals = rgb2hex(R, G, B))
rgbtib

# A tibble: 100 × 6
       x     y     R     G     B hexvals
   <int> <int> <int> <int> <int> <chr>  
 1     1     1   177   166    44 #D9D373
 2     1     2   231    80    48 #F49878
 3     1     3   224   188   136 #F1DFC1
 4     1     4   221   168   181 #EFD4DB
 5     1     5   107    82    93 #AD9AA3
 6     1     6   238   111   179 #F7B0DA
 7     1     7   246   242   124 #FBF9B9
 8     1     8   230    57   213 #F482EC
 9     1     9   172   151   130 #D6CABD
10     1    10   166    93    84 #D3A39B
# ℹ 90 more rows

Plot to show it works

ggplot(rgbtib, aes(x = x, y = y, fill = hexvals)) + geom_tile() + theme(legend.position = 'none')