scale_color_manual is not working for 2 line plot with break in y-axis (ggplot2)-CodePudding

I've been trying for a while now and also doing a lot of research, but I just can't get it to add a simple legend for my two lines.

I have two lines in my chart and I just want to add a legend for the two lines.

scale_color_manual did not work. I suspect it is because I am using scale_y_break. In other plots (without scale_y_break) scale_color_manual works without problems.

Here is my code:

day <- c(1:5)
altimeter <- c(8.291, 8.872, 7.212, 8.1, 5.92)
slope_kilometers <- c(30.23, 34.8, 29.34, 32.98, 21.23)

df2 <- data.frame(day, altimeter, slope_kilometers)

library(ggbreak)
altimeter_color <- "steelblue"
slope_kilometers_color <- "darkred"

ggplot(df2, aes(x = day))  
  #Altimeter data
  geom_line(aes(y = altimeter), 
            linetype = 2, 
            linewidth = 1, 
            color = altimeter_color)   
  geom_point(y = altimeter, size = 3, color = altimeter_color)  
  #Slope kilometers data
  geom_line(aes(y = slope_kilometers), 
            linetype = 2, 
            linewidth = 1, 
            color = slope_kilometers_color)  
  geom_point(y = slope_kilometers, size = 3, color = slope_kilometers_color)  
  
  #Y-Axis
  scale_y_break( c(9, 20), scales = 1.5)  
  
  #Label
  labs(x = "Tage", 
       y = "[km]")  
  
  #Legend
  scale_color_manual(values = c(altimeter_color, slope_kilometers_color))  

  #Title
  ggtitle("Höhenmeter und Pistenkilometer meines 5-tägigen Skiurlaubs")

I tried different versions of scale_color_manual, labs, aes(fill="")

CodePudding user response：

Update: I tweaked the former plot (removed):

One way to achieve what you want is:

First bring data in long format then put color inside aesthetics:

Rule of thumb: What is in aesthetics will have a legend:

library(tidyverse)
library(ggbreak)

df2 %>% 
  pivot_longer(-day) %>% 
  ggplot(aes(x = day))  
  #Altimeter data
  geom_line(data = . %>% filter(name == "altimeter"), aes(y = altimeter, color = name),  
            linetype = 2, linewidth = 1 )  
  geom_point(data = . %>% filter(name == "altimeter"), aes(y = altimeter, color = name), size = 3)  
  #Slope kilometers data
  geom_line(data = . %>% filter(name == "slope_kilometers"), aes(y = slope_kilometers, color = name), 
            linetype = 2, linewidth = 1)  
  geom_point(data = . %>% filter(name == "slope_kilometers"), aes(y = slope_kilometers, color = name), size = 3)  
  #Y-Axis
  scale_y_break( c(9, 20), scales = 1.5)  
  #Label
  labs(x = "Tage",  y = "[km]", color = "")  
  #Legend
  scale_color_manual(values = c(altimeter_color, slope_kilometers_color))  
  #Title
  ggtitle("Höhenmeter und Pistenkilometer meines 5-tägigen Skiurlaubs")  
  theme(legend.position = "bottom")