Warning message with stat_function() when using purrr::map2

[yuw444]

I found a problem with stat_function(), it produces some warning messages although the result is as expected.

Here is the code to reproduce the bug:

library(ggplot2)
library(purrr)
data(mtcars)
temp <- function(y) {
    function(x) y / x
}

fun_list <- map(1:3 * 10^4, ~ temp(.x))

p <- ggplot(
    mtcars,
    aes(disp, hp, fill = wt)
) +
    geom_point(shape = 21, size = 3, aes(fill= factor(vs))) +
    purrr::map2(
        fun_list,
        c("4", "6", "8"),
        ~ stat_function(
            fun = .x,
            aes(color = .y)
        )
    )

p

Below is the warning message

Warning messages:
1: The following aesthetics were dropped during statistical transformation: fill
ℹ This can happen when ggplot fails to infer the correct grouping structure in the data.
ℹ Did you forget to specify a `group` aesthetic or to convert a numerical variable into a factor? 
2: The following aesthetics were dropped during statistical transformation: fill
ℹ This can happen when ggplot fails to infer the correct grouping structure in the data.
ℹ Did you forget to specify a `group` aesthetic or to convert a numerical variable into a factor? 
3: The following aesthetics were dropped during statistical transformation: fill
ℹ This can happen when ggplot fails to infer the correct grouping structure in the data.
ℹ Did you forget to specify a `group` aesthetic or to convert a numerical variable into a factor?

[teunbrand]

More minimal reprex to reproduce warning:

library(ggplot2)
#> Warning: package 'ggplot2' was built under R version 4.3.1

ggplot(mtcars, aes(disp, hp, fill = wt)) +
  geom_point() +
  stat_function(fun = ~ .x)
#> Warning: The following aesthetics were dropped during statistical transformation: fill
#> ℹ This can happen when ggplot fails to infer the correct grouping structure in
#>   the data.
#> ℹ Did you forget to specify a `group` aesthetic or to convert a numerical
#>   variable into a factor?

^{Created on 2023-10-26 with reprex v2.0.2}

In any case, I think the warnings are appropriate, as you've specified fill =wt at the top-level ggplot() call, which will result in non-constant columns within a group that the stat applies to.

[clauswilke]

I tend to agree that the warning is appropriate even if it is a bit confusing here. The code is asking stat_function() to do something that it cannot do (vary the fill color along the data points), and so this triggers a warning. It's difficult to write these warnings in such a way that they correctly describe the true underlying issue in all cases.

[teunbrand]

Yeah agreed. Because it is working as intended, I'll close this as 'not a bug'.

[yuw444]

Thanks for the insight. Seems like I should put fill=wt inside of geom_point().