Variables of a data.frame beginning by a dot disappear in within()

Question

It is legal to call a variable by a name that begins with a dot (like .identifier). However, within() function does not preserve them. Did I miss something? Or is it a bug?

A <- data.frame(.has.a.dot=1:10,has.no.dot=letters[1:10])
within(A, new.variable<-rnorm(10,.has.a.dot))

gives :

   has.no.dot new.variable
1           a     1.300361
2           b     3.014026
3           c     2.354260
4           d     4.261637
5           e     5.159326
6           f     7.178712
7           g     6.438039
8           h     8.253819
9           i     9.463351
10          j     8.828403

Answer 1

This seems to be because of standard in the as.list method for class environment. The standard argument is all.names = FALSE. From ?as.list:

all.names a logical indicating whether to copy all values or (default) only those whose names do not begin with a dot.

You can change the within.data.frame method to the following:

within.data.frame <- function (data, expr, ...) 
{
  parent <- parent.frame()
  e <- evalq(environment(), data, parent)
  eval(substitute(expr), e)
  # l <- as.list(e) # removed this line
  l <- as.list(e, all.names=TRUE) # added this line
  l <- l[!sapply(l, is.null)]
  nD <- length(del <- setdiff(names(data), (nl <- names(l))))
  data[nl] <- l
  if (nD) 
    data[del] <- if (nD == 1) 
      NULL
  else vector("list", nD)
  data
}

Then you get your expected behavior:

within(A, new <- .has.a.dot)
##    .has.a.dot has.no.dot new
## 1           1          a   1
## 2           2          b   2
## 3           3          c   3
## 4           4          d   4
## 5           5          e   5
## 6           6          f   6
## 7           7          g   7
## 8           8          h   8
## 9           9          i   9
## 10         10          j  10