[R] Evaluation of global variables in function definitions

[William Dunlap]

fFactory <- function(args) {
    a <- complicated_a_computation(args)
    b <- complicated_b_computation(args)
    function(x) a*x + b*exp(x)
}
f1 <- fFactory(args1)
f2 <- fFactory(args2)

f1(x)
f2(x) # will get different result than f1(x)

f1 and f2 will look the same but produce different results because
environment(f1)
contains different values of a and b than environment(f2).

Bill Dunlap
TIBCO Software
wdunlap tibco.com


On Wed, Oct 8, 2014 at 2:53 AM, H. Dieter Wilhelm
<dieter at duenenhof-wilhelm.de> wrote:
> Duncan Murdoch <murdoch.duncan at gmail.com> writes:
>
>> On 07/10/2014 2:45 AM, H. Dieter Wilhelm wrote:
>>> Hello (),
>>>
>>> I'd like to do the following
>>>
>>> a <- 3
>>> f1 <- function (x){
>>>   a*x^2
>>> }
>>>
>>> a <- 5
>>> f2 <- function (x){
>>>   a*x^2
>>> }
>>>
>>> plotting f1, f2, ...
>>>
>>> but f1 and f2 are the same, how can I evaluated the variables in the
>>> function definition?
>>>
>>> Thank you
>>>        Dieter
>> See the "open.account" example in section 10.7, "Scope", of the
>> Introduction to R manual.
>
> Sorry but above section teaches my how to change global variables in
> functions.  But I would like to overcome the "lazy evaluation" of global
> variables in functions.  Because I can't put expensive computations into
> each function.
>
> a <- complicated_a_computation(args1)
> b <- complicated_b_computation(args1)
> c <- ...
> d
> ...
>
> f1 <- function(x) a*x+b*exp(x)+c/x+...
>
> a <- complicated_a_computation(args2)
> b <- complicated_b_computation(args2)
> c <- ...
> d
> ...
>
> f2 <- function(x) a*x+b*exp(x)+c/x +...
>
> Yes in principle I could rename the global variables to a1, a2, b1, b2,
> ...  But I'd like to learn the principles of R.  Is above somehow
> possible with the use of the substitute() command?
>
> Thanks
> --
> Best wishes
> H. Dieter Wilhelm
> Darmstadt, Germany
>
> ______________________________________________
> R-help at r-project.org 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.