> What about NAof <- function(x) as(NA, class(x))?
>
That's pretty close, but it doesn't fail nicely for non-atomics:
> as(NA, class(Sys.Date()))
Error in as(NA, class(Sys.Date())) :
no method or default for coercing "logical" to "Date"
vs.
> as.Date(NA)
[1] NA
But I'm probably being too fussy (especially since I didn't ask for
that in my original post ;). I'm currently using this:
NAof <- function(x) {
length(x) <- 1
is.na(x) <- TRUE
x
}
It's rather lacking in elegance :(
Hadley
--
Assistant Professor / Dobelman Family Junior Chair
Department of Statistics / Rice University
http://had.co.nz/
______________________________________________
R-devel@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-devel
