Limits with NA cause errors when there is no data

[paleolimbot]

The following should create a blank plot without erroring, but does not after #3426. This should probably be fixed in the same PR as #3448.

library(ggplot2)
ggplot(data.frame(x = numeric(0)), aes(x)) + 
  geom_point() +
  scale_x_continuous(limits = c(0, NA))
#> Warning in rep(yes, length.out = len): 'x' is NULL so the result will be
#> NULL
#> Error in ans[ypos] <- rep(yes, length.out = len)[ypos]: replacement has length zero

The offending code is here (where self$limits is not NULL and self$range$range is NULL):

ggplot2/R/scale-.r

Lines 439 to 452 in 10fa001

	get_limits = function(self) {
	if (self$is_empty()) {
	return(c(0, 1))
	}
	if (is.null(self$limits)) {
	self$range$range
	} else if (is.function(self$limits)) {
	# if limits is a function, it expects to work in data space
	self$trans$transform(self$limits(self$trans$inverse(self$range$range)))
	} else {
	ifelse(is.na(self$limits), self$range$range, self$limits)
	}
	},

I think that Scale$get_limits() needs another case for limits containing NA (or Scale$is_emtpy() should return TRUE for these limits). @thomasp85 any opinion on either of these solutions?

[lock]

This old issue has been automatically locked. If you believe you have found a related problem, please file a new issue (with reprex) and link to this issue. https://reprex.tidyverse.org/

[lock]

This old issue has been automatically locked. If you believe you have found a related problem, please file a new issue (with reprex) and link to this issue. https://reprex.tidyverse.org/