NA in character vectors

Hi all,

While R-1.5.0 is still unfrozen, I'd like to try again to generate interest in

LETTERS[c(NA,2)] in S is c("","B"), but in R is c("NA","B").

We had an interesting discussion then, and I learned (from Duncan Murdoch and
Thomas Lumley) that R does have an internal code for missing char values
(R_NaString), but it gets easily confused with the string "NA".  Check this:

  R> z <- c(LETTERS[c(2,NA)], "NA", paste("NA"))
  R> is.na(z)
     [1] FALSE  TRUE  TRUE FALSE
  R> z[3]==z[4]
     [1] TRUE
  R> z=="NA"
     [1] FALSE  TRUE  TRUE  TRUE

Thomas Lumley <tlumley@u.washington.edu> suggested that this weird behavior
essentially arises because the parser converts "NA" to R_NaString, and so...

It looks like we basically need to
 1) stop the parser generating R_NaString from \"NA\"
 2) Change PRINTNAME(R_NaString) to avoid ambiguity

While I still like the simple S-Plus model (""), Thomas's suggestions (with
PRINTNAME(R_NaString) = "<NA>" for example) would be OK too.  Thanks for
listening (again)!

-- David Brahm (brahm@alum.mit.edu)
"I write this letter not to nag nor to whine, but to prod." - Lisa Simpson
-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-devel mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-devel-request@stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._