[Rd] Surprising behavior of letters[c(NA, NA)]
murdoch.duncan at gmail.com
Fri Dec 17 16:37:21 CET 2010
On 17/12/2010 10:18 AM, Gabor Grothendieck wrote:
> On Fri, Dec 17, 2010 at 9:58 AM, Duncan Murdoch
> <murdoch.duncan at gmail.com> wrote:
> > On 17/12/2010 9:32 AM, Gabor Grothendieck wrote:
> >> Consider this:
> >> > letters[c(2, 3)]
> >>  "b" "c"
> >> > letters[c(2, NA)]
> >>  "b" NA
> >> > letters[c(NA, 3)]
> >>  NA "c"
> >> > letters[c(NA, NA)]
> >>  NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA
> >> NA NA
> >>  NA
> >> The result is a 2-vector in each case until we get to c(NA, NA) and
> >> then it unexpectedly changes from returning a 2-vector to returning a
> >> 26-vector. I think most people would have expected that the answer
> >> would be c(NA, NA).
> > This is because c(NA, NA) is a logical vector, so it gets recycled to the
> > length of letters, whereas c(NA, 3) and the others are numeric vectors, so
> > they aren't recycled, they're converted to integer indices. So the surprise
> > is due to not recognizing that NA is logical. You wouldn't expect a length
> > 1 result from letters[TRUE], would you?
> One tends not to distinguish between logical NA's and integer NA's.
> In fact R represents both of them as NA on output so this does seem
> highly error prone.
> > NA # logical
>  NA
> > NA_integer_ # integer
>  NA
I agree it's error prone, but I don't know a good solution. The ability
to distinguish them on input is a relatively recent addition (in
2.5.0). Changing the display on output would confuse a lot of people.
More information about the R-devel