rggplot2facet

Facet nested with mathematical symbols and text with spaces in labels (ggplot2)

I want to create a nested_facet plot but the labels contain a less or equal sign and a greater or equal sign and spaces. The following is an example:

df <- iris
df$factor <- factor(rep(c('Values 7', 'Values 123', 'Values very-high'), length.out = nrow(df)))

df <- df %>%  mutate(factor2 = case_when(factor =='Values 7' ~'Values <= 7', 
                                         factor == 'Values 123' ~ 'Values >= 123', 
                                         TRUE ~ factor))


ggplot(df, mapping = aes(x = Sepal.Length, y = Sepal.Width)) +
  geom_line() +
  facet_nested("x" + Species ~ "y-and z" +  factor2, labeller = label_parsed)

This code does not work. If I remove all spaces (like the code below) the code works fine, but does not have the correct labels (which are provided in the previous example). I have tried unicodes as well, but they are not saved correctly as a pdf.

df <- iris
df$factor <- factor(rep(c('Values 7', 'Values 123', 'very-high'), length.out = nrow(df)))

df <- df %>%  mutate(factor2 = case_when(factor =='Values 7' ~'Values <= 7', 
                                         factor == 'Values 123' ~ 'Values >= 123', 
                                         TRUE ~ factor))


ggplot(df, mapping = aes(x = Sepal.Length, y = Sepal.Width)) +
  geom_line() +
  facet_nested("x" + Species ~ "y-and" +  factor2, labeller = label_parsed)

How can I get this to work?

Solution

  • You have to use ~ in your ?plotmath strings instead of spaces:

    library(ggplot2)
library(ggh4x)
library(dplyr, warn=FALSE)

df <- iris
# Add a ~ between "Values" and "very"
df$factor <- factor(rep(c("Values 7", "Values 123", "Values~very-high"), length.out = nrow(df)))

df <- df %>% mutate(factor2 = case_when(
  factor == "Values 7" ~ "Values <= 7",
  factor == "Values 123" ~ "Values >= 123",
  TRUE ~ factor
))

ggplot(df, mapping = aes(x = Sepal.Length, y = Sepal.Width)) +
  geom_line() +
  # Add ~'s in "y and z"
  ggh4x::facet_nested("x" + Species ~ "y~and~z" + factor2,
    labeller = label_parsed
  )