I want to generate a plot which keeps all the colors in the legend, even for a subset of the data. When i create a reproducible example it works, and i am going out of my mind trying to work out what i am missing. Below is my working reproducible example and a dput of my code that won’t work. Appreciate any help correcting this.

#First my working case:

library("RColorBrewer")
library("ggplot2")

#Create the color values for the data

gear_fill = brewer.pal(n = 3, "Dark2")
names(gear_fill) <- sort(unique(mtcars$gear))

#Plot the data
mtcars %>% 
  filter(gear %in% c(4, 5)) %>% 
  ggplot(aes(x = gear, 
             y = mpg, 
             fill = factor(gear, sort(unique(mtcars$gear))))) +
  geom_col(show.legend = TRUE) + 
  scale_fill_manual(name = "Gear", 
                    drop = FALSE, 
                    values = gear_fill)

Correctly shows the color for 3 gears even though I filtered it out:

#Now my problem

#Create data from dput - data anonymized

 data1 <- structure(list(project.code = c("External Event", "External Event", 
"External Event", "External Event", "External Event", "External Event", 
"External Event", "External Event", "Internship", "Training/Course", 
"Training/Course", "Training/Course"), university.people = c(" ", 
"h", "m", "n", "o", "r", "s", "t", "r", "m", "s", "t"), university.partner = c("uni1", 
"uni1", "uni1", "uni1", "uni1", "uni1", "uni1", "uni1", "uni1", 
"uni1", "uni1", "uni1"), n = c(2L, 2L, 1L, 1L, 1L, 2L, 2L, 2L, 
2L, 3L, 2L, 3L)), class = c("grouped_df", "tbl_df", "tbl", "data.frame"
), row.names = c(NA, -12L), groups = structure(list(project.code = c("External Event", 
"External Event", "External Event", "External Event", "External Event", 
"External Event", "External Event", "External Event", "Internship", 
"Training/Course", "Training/Course", "Training/Course"), university.people = c(" ", 
"h", "m", "n", "o", "r", "s", "t", "r", "m", "s", "t"), .rows = structure(list(
    1L, 2L, 3L, 4L, 5L, 6L, 7L, 8L, 9L, 10L, 11L, 12L), ptype = integer(0), class = c("vctrs_list_of", 
"vctrs_vctr", "list"))), class = c("tbl_df", "tbl", "data.frame"
), row.names = c(NA, -12L), .drop = TRUE))

#Get the factor levels - this is a small recreation of my full dataset
data2 <- data.frame(project.code = c("Internship",
                                     "Research collaboration",
                                     "External Event", 
                                     NA, 
                                     "Training/Course", 
                                     "Internal Event", 
                                     "Staff Development"))

#Create my color values
project_code_colors <- brewer.pal(6, "Dark2")
names(project_code_colors) <- sort(unique(data2$project.code))

#Attempt to visualize
data1 %>%
  ggplot(aes(y = university.people, x = n, 
             fill = factor(project.code, levels = sort(unique(data2$project.code))))) +
  geom_col(col = "black") + 
  theme_bw() + 
  labs(x = element_blank(), y = element_blank(),
       fill = "Partnership type:", ) +
  scale_fill_manual(drop = FALSE, 
                    values = c(project_code_colors, na.value = "grey"))

Shows all the levels by has empty color values for cases that are not in the data:

If someone can spot the error, i would welcome it.