Error when a marginal has a single string category

[dkyleward]

While explaining how to resolve issue #1, I realized that the following code will generate an error:

survey <- tibble(
  size = c(1, 2, 1, 1),
  autos = c("any", "any", "any", "any"),
  weight = 1
)
targets <- list()
targets$size <- tibble(
  `1` = 75,
  `2` = 25
)
targets$autos <- tibble(
  any = 100
)
result <- ipu(survey, targets)

This is because I'm using a single string "any" as the marginal category. The error message:

 Error in `contrasts<-`(`*tmp*`, value = contr.funs[1 + isOF[nn]]) : 
  contrasts can be applied only to factors with 2 or more levels 

This is an error in the stats::contrasts() function. The code:

function (x, how.many, value) 
{
  if (is.logical(x)) 
    x <- factor(x, levels = c(FALSE, TRUE))
  if (!is.factor(x)) 
    stop("contrasts apply only to factors")
  if (nlevels(x) < 2L) 
    stop("contrasts can be applied only to factors with 2 or more levels")
# ...

Oddly enough, using all ones instead of "any" works. My package relies on the mlr::createDummyFeatures() function. Perhaps I should write my own version of the function that handles this issue (and also remove a package dependency in the process).