Hi,
I was wondering whether you might consider adding a function to forcats that sets up factor levels according to a numerically-aware string-sorting method like gtools::mixedsort.
There's a number of use cases: eg, when you end up with genomes ordered like "chr1", "chr10", "chr11" ... "chr2" rather than "chr1", "chr2", ..., in a plot if you don't hack your factors correctly
eg,
fct_alphameric/fct_alnum/fct_mixedsort <- function(x, ...) {
factor(x, levels = gtools::mixedsort(unique(x), ...))
}