[ARROW-13118] [R] Improve handling of R scalars in some nse_funcs

• Type: Improvement
• Status: Open
• Priority: Major
• Resolution: Unresolved
• Component/s: R
• External issue URL: https://github.com/apache/arrow/issues/18716
• Assignee: Unassigned
• Reporter: Ian Cook
• Created: 18/Jun/21 15:14
• Updated: 11/Jan/23 08:31

Description

Some of the functions in nse_funcs do not behave properly when passed R scalar input in expressions in dplyr verbs. Some examples:

Table$create(x = 1) %>% mutate(as.character(42))
Table$create(x = 1) %>% mutate(is.character(("foo")))
Table$create(x = 1) %>% mutate(nchar("foo"))
Table$create(x = 1) %>% mutate(is.infinite(Inf))

This could be resolved by using build_expr() instead of Expression$create(), but build_expr() is somewhat heavy. The only part of it we really need to make this work is this:

    args <- lapply(args, function(x) {
      if (!inherits(x, "Expression")) {
        x <- Expression$scalar(x)
      }
      x
    })

If build_expr() is too heavy, we could make a function called wrap_r_scalar, like this:

wrap_r_scalar <- function(x) {
  if (!inherits(x "Expression")) {
    assert_that(
      length(x) == 1,
      msg = "Literal vectors of length != 1 not supported"
    )
    Expression$scalar(x)
  } else {
    x
  }
}

and use it as needed in various of the nse_funcs functions.

Issue Links

• relates to: ARROW-13037 [R] Incorrect param when creating Expression crashes R