R devel - Aug 2010 - Surprising behavior of Negate()

Dear R-developers,

I find the current behavior of Negate() somewhat confusing. It does not
match the passed function 'f' until the returned function is called for
the first time. To see an example of what this can do see the following
(contrived) example:

  f <- function(x) is.integer(x)
  not_f <- Negate(f)

  f <- function(x) is.character(x)

  ## Both should, in my mind, return TRUE:
  not_f(1) == !is.integer(1)
  not_f(1L) == !is.integer(1L)

I propose to change Negate() in the following way:

  ## Easy 'fix':
  Negate <- function(f) {
    f <- match.fun(f)
    function(...) !f(...)
  }

This matches 'f' when Negate() is called and not the first time the
return
value is used. If the current behavior is desired, maybe a note in the
documentation could be added to clarify this peculiarity.

Cheers,
Olaf Mersmann

On 08/30/2010 02:01 AM, Olaf Mersmann wrote:
> This matches 'f' when Negate() is called and not the first time the
return
> value is used. If the current behavior is desired, maybe a note in the
> documentation could be added to clarify this peculiarity.
Looks like a classic lazy eval gotcha -- almost never intentional.

Testing your patch now.

-p

-- 
Peter Dalgaard
Center for Statistics, Copenhagen Business School
Phone: (+45)38153501
Email: pd.mes at cbs.dk  Priv: PDalgd at gmail.com