On 11/13/2012 11:19 AM, Duncan Murdoch wrote:
On 13/11/2012 1:33 PM, Jamie Olson wrote:
I was surprised to notice that statements like:
h = function(...){list(...)}(x=4)
do not throw syntax errors. R claims that 'h' is now a function, but I
can't seem to call it.
> h = function(x){list(x)}(4)
> is(h)
[1] "function" "OptionalFunction" "PossibleMethod"
> h()
Error in list(x) : 'x' is missing
> h(4)
Error in h(4) : attempt to apply non-function
>
What's going on?
The body of your function is
{list(x)}(4)
The problem is,
{list(x)}
does not return a function, so you can't call it with the argument 4.
Another way to see this is
> body(h)
{
list(x)
}(4)
> eval(body(h))
Error in eval(expr, envir, enclos) : attempt to apply non-function
So the body is syntactically valid, but can not be evaluated because, as
Duncan said, {list(x)} is not a function (and that fact can not be
determined until it is being executed).
If you had
h <- function(x) { function(y) y }(4)
it would return 4 every time, because the anonymous function does that.
If what you were doing is to create an anonymous function and
immediately call it, you can do that as
h <- (function(x){list(x)})(4)
in which case
> is(h)
[1] "list" "vector"
> h
[[1]]
[1] 4
I'm not sure which behavior you were expecting.
Duncan Murdoch
--
Brian S. Diggs, PhD
Senior Research Associate, Department of Surgery
Oregon Health & Science University
______________________________________________
[email protected] mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
