Unequal breaks and label lengths error when alpha

问题:

In ggplot2 version 2.2.0, E.g.:

tmp_df <- data.frame(x = 1:3, y = 1:3, alpha = rep(0.5, 3))

#   x y alpha
# 1 1 1   0.5
# 2 2 2   0.5
# 3 3 3   0.5

ggplot(tmp_df, aes(x, y, alpha = alpha)) +
    geom_bar(stat = 'identity') +
    scale_alpha(breaks = c(0.25, 0.5, 1), labels = c('a', 'b', 'c'))

Produces the error:

Error in f(..., self = self) : Breaks and labels are different lengths

Manually removing the extra alpha values in scale_alpha fixes the problem, but surely this can be handled some how by ggplot?

回答1:

You must supply the limits for the scale because tmp_df$alpha is always the same, and ggplot does not know the 'range' of the scale.

library(ggplot2)
tmp_df <- data.frame(x = 1:3, y = 1:3, alpha = rep(0.5, 3))
tmp_df
#>   x y alpha
#> 1 1 1   0.5
#> 2 2 2   0.5
#> 3 3 3   0.5

ggplot(tmp_df, aes(x, y, alpha = alpha)) +
    geom_bar(stat = 'identity') +
    scale_alpha(breaks = c(0.25, 0.5, 1), labels = c('a', 'b', 'c'), limits = c(0, 1))

If the alpha dimension has a range itself, limits are no longer necessary, but note the in the following example the first break is ignored, as it is outside the range. limits would again be necessary if you want to include it.

tmp_df <- data.frame(x = 1:3, y = 1:3, alpha = seq(.5, 1.5, .5))
tmp_df
#>   x y alpha
#> 1 1 1   0.5
#> 2 2 2   1.0
#> 3 3 3   1.5

ggplot(tmp_df, aes(x, y, alpha = alpha)) +
    geom_bar(stat = 'identity') +
    scale_alpha(breaks = c(0.25, 0.5, 1), labels = c('a', 'b', 'c'))