geom_polygon with hex fill color from data-CodePudding

I have a tibble with the following structure:

library(tidyverse)
dmd <- tibble(sample = rep(c(1,2,3), each = 2),
              time   = rep(c(1,2),3),
              score  = c(0.69,0.95,1,0.90,1,0.76)) %>%
  mutate(x = map(base::strsplit(paste(sample,
                                      sample   0.5,
                                      sample,
                                      sample - 0.5,
                                      sep = ","),
                                split = ","),
                 as.numeric)) %>%
  mutate(y = map(base::strsplit(paste(time   0.5,
                                      time,
                                      time - 0.5,
                                      time,
                                      sep = ","),
                                split = ","),
                 as.numeric)) %>%
  unnest(c(x,y)) %>%
  group_by(sample,time) %>%
  mutate(group_id = cur_group_id()) %>%
  ungroup

# A tibble: 24 × 6
   sample  time score     x     y group_id
    <dbl> <dbl> <dbl> <dbl> <dbl>    <int>
 1      1     1  0.69   1     1.5        1
 2      1     1  0.69   1.5   1          1
 3      1     1  0.69   1     0.5        1
 4      1     1  0.69   0.5   1          1
 5      1     2  0.95   1     2.5        2
 6      1     2  0.95   1.5   2          2
 7      1     2  0.95   1     1.5        2
 8      1     2  0.95   0.5   2          2
 9      2     1  1      2     1.5        3
10      2     1  1      2.5   1          3
# … with 14 more rows

I then use grDevices::rgb to generate a grayscale hex value dcolor based on score. The if() statement doesn't do much in this example, but it's important that I retain this element for the functionalization of my code.

minsc <- min(dmd$score)
thrsh <- 0.75
if (minsc < thrsh) {
  floor <- minsc
} else {
  floor <- thrsh
}

dmd <- dmd %>% mutate(temp = (1 - score) / (1 - floor)) %>%
  mutate(dcolor = grDevices::rgb(temp,temp,temp)) %>%
  dplyr::select(-temp)

# A tibble: 24 × 7
   sample  time score     x     y group_id dcolor 
    <dbl> <dbl> <dbl> <dbl> <dbl>    <int> <chr>  
 1      1     1  0.69   1     1.5        1 #FFFFFF
 2      1     1  0.69   1.5   1          1 #FFFFFF
 3      1     1  0.69   1     0.5        1 #FFFFFF
 4      1     1  0.69   0.5   1          1 #FFFFFF
 5      1     2  0.95   1     2.5        2 #292929
 6      1     2  0.95   1.5   2          2 #292929
 7      1     2  0.95   1     1.5        2 #292929
 8      1     2  0.95   0.5   2          2 #292929
 9      2     1  1      2     1.5        3 #000000
10      2     1  1      2.5   1          3 #000000
# … with 14 more rows

If I plot dmd with geom_polygon, dcolor is interpreted as a factor.

ggplot(dmd)   
  geom_polygon(mapping = aes(x = x,
                             y = y,
                             group = group_id,
                             fill = dcolor),
               color = "black")  
  theme_void()

I would like to force ggplot to use the hex values given in dcolor as the fill color for each diamond. Using scale_fill_manual gets me closer, but the colors are still incorrect (#000000 should be black, #FFFFFF should be white).

pal <- dmd %>% dplyr::select(group_id, dcolor) %>%
  unique %>% dplyr::select(dcolor) %>% as_vector

ggplot(dmd)   
  geom_polygon(mapping = aes(x = x,
                             y = y,
                             group = group_id,
                             fill = dcolor),
               color = "black")  
  scale_fill_manual(values = palette(pal))  
  theme_void()

CodePudding user response：

You could achieve your desired result using scale_fill_identity:

library(tidyverse)

ggplot(dmd)   
  geom_polygon(mapping = aes(x = x,
                             y = y,
                             group = group_id,
                             fill = dcolor),
               color = "black")  
  theme_void()  
  scale_fill_identity(guide = guide_legend())