ggplot2: Can I send arguments in a formatter function?

tidyverse
,
1

I want to create a diagram with numbers formatted as percent on the y-axis. The data has only 3 decimals precision so I want numbers formatted as percent with 1 digits preceision. 0.034 should be formatted as 3.4%. But when I use scales::percent as a formatter function in ggplot2 the number is formatted with 2 decimals precision (0.034 = 3.40%).

The scales::percent function has an accuracy argument. Is there any way I can send an argument from ggplot to the scales::percent function?

library(ggplot2)
library(scales)

set.seed(12)

#Create a dataset with 3-digits precision:
df <- data_frame(xvar = LETTERS[1:5],yvar = round(runif(5, min = 0, max = 0.080), 3))
df$xvar <- as.factor(df$xvar)

print(df)

# A tibble: 5 x 2
# xvar   yvar
# <fct> <dbl>
#     1 A     0.006
# 2 B     0.065
# 3 C     0.075
# 4 D     0.022
# 5 E     0.014

# The diagram shows percent with 2 decimals, but it should only be 1
ggplot(df, aes(xvar, yvar)) +
    geom_bar(stat = "identity") +
    scale_y_continuous(breaks = pretty_breaks(8), labels = percent)

# If a try to send an accuracy argument to scales:percent from ggplot it generates an error

ggplot(df, aes(xvar, yvar)) +
    geom_bar(stat = "identity") +
    scale_y_continuous(breaks = pretty_breaks(8), labels = percent(accuracy = 0.1))

# Error in number(x = x, accuracy = accuracy, scale = scale, prefix = prefix,  : 
#                     argument "x" is missing, with no default
2

This is the use case for the scales *_format() variants (in this case, percent_format()).

library(ggplot2)
library(scales)

set.seed(12)

#Create a dataset with 3-digits precision:
df <- dplyr::data_frame(xvar = LETTERS[1:5],yvar = round(runif(5, min = 0, max = 0.080), 3))
df$xvar <- as.factor(df$xvar)

# The diagram shows percent with 2 decimals, but it should only be 1
ggplot(df, aes(xvar, yvar)) +
  geom_bar(stat = "identity") +
  scale_y_continuous(breaks = pretty_breaks(8), labels = percent)


ggplot(df, aes(xvar, yvar)) +
  geom_bar(stat = "identity") +
  scale_y_continuous(breaks = pretty_breaks(8), labels = percent_format(accuracy = 0.1))

Created on 2018-10-24 by the reprex package (v0.2.1.9000)

2 Likes