ggplot geom_rect() error “object not found”

问题:

I'm trying to plot a geom_rect(). Why do I receive an Error in FUN(X[[i]], ...) : object 'Month' not found? If I run df$Month in my console the object is there:

df$Month
#> [1] 2019-01 2019-02 2019-03
#> Levels: 2019-01 2019-02 2019-03

Here's my code block:

library(tidyverse)
df <- tibble(Month = factor(c("2019-01", "2019-02", "2019-03")), 
             Value = c(4, 9, 7))

ggplot(df, aes(Month, Value, group = 1)) + 
  geom_line() + 
  theme_minimal() + 
  geom_rect(data = 
              data.frame(xmin = min(as.integer(df$Month)) - 0.5,
                         xmax = max(as.integer(df$Month)) + 0.5,
                         ymin = min(df$Value),
                         ymax = max(df$Value)),
            aes(xmin = xmin, xmax = xmax, ymin = ymin, ymax = ymax),
            alpha = 0.2, fill = "green")

#> Error in FUN(X[[i]], ...) : object 'Month' not found

回答1:

You just have an extra step of setting up a dataframe in geom_rect which coincide with data in ggplot. Simply provide your max and min values to geom_rect and it works:

ggplot(df, aes(Month, Value, group = 1)) + 
  geom_line() + 
  theme_minimal() + 
  geom_rect(aes(xmin = min(as.integer(Month)) - 0.5, 
                xmax = max(as.integer(Month)) + 0.5, 
                ymin = min(Value), 
                ymax = max(Value)),
            alpha = 0.2/nrow(df), fill = "green")

回答2:

I was able to return your desired result by calling df in geom_line() after gemo_rect(). However leaving the Month field as is returned the error: Error: Discrete value supplied to continuous scale.
I worked around this by wrapping as.integer() around Month.

ggplot() + 
  theme_minimal() + 
  geom_rect(data = 
              data.frame(xmin = min(as.integer(df$Month)) - 0.5,
                         xmax = max(as.integer(df$Month)) + 0.5,
                         ymin = min(df$Value),
                         ymax = max(df$Value)),
            aes(xmin = xmin, xmax = xmax, ymin = ymin, ymax = ymax),
            alpha = 0.2, fill = "green") + 
  geom_line(data = df, aes(as.integer(Month), Value, group = 1))

You might have to clean up your x-axis label but it achieves desired outcome!

回答3:

This works:

ggplot(df, aes(Month, Value, group = 1)) + 
  geom_line() + 
  theme_minimal() + 
  geom_rect(data = 
              data.frame(xmin = min(as.integer(df$Month)) - 0.5,
                         xmax = max(as.integer(df$Month)) + 0.5,
                         ymin = min(df$Value),
                         ymax = max(df$Value)),
            aes(x = NULL,y = NULL,xmin = xmin, xmax = xmax, ymin = ymin, ymax = ymax),
            alpha = 0.2, fill = "green")

by unmapping the inherited x/y aesthetics from the top ggplot call. It's understandable that this might be confusing, though, since the description in ?geom_rect sorta kinda implies that geom_rect isn't looking for those aesthetics at all.