[R] Trying to build up functions with its names by means of lapply

[Bert Gunter]

Thanks, Bill. Clearly something I missed. I appreciate your filling in
that crack.

Another equivalent way to do it?

f2 <- function(c,nm = "gamma",...)
{
  probFunc <- paste0(c,nm)
  more <- list(...)
  function(x)do.call(probFunc,c(x,more))
}

This avoids the explicit use of get() and force(), I believe, but are
there problems here I'm missing?


-- Cheers,
Bert

On Wed, Jun 5, 2013 at 1:46 PM, William Dunlap <wdunlap at tibco.com> wrote:
> You will want to force the evaluation of the ... arguments, just as you are already
> forcing the evaluation of 'c', before returning your function.  E.g., compare the
> following two:
>
> # your 'faux', renamed and enhanced a bit
> f0 <- function (c, nm = "gamma", ...)
> {
>     probFunc <- getFunction(paste0(c, nm))
>     function(x) probFunc(x, ...)
> }
> # the above but adding a call to force the evaluation of the ... argument before returning the function
> f1 <- function (c, nm = "gamma", ...)
> {
>     probFunc <- getFunction(paste0(c, nm))
>     force(list(...))
>     function(x) probFunc(x, ...)
> }
> # Now use mapply to make a list of functions with various parameters
> z0 <-  mapply(c=c("p","d"), nm="norm", mean=c(1,-1), FUN=f0)
> z1 <-  mapply(c=c("p","d"), nm="norm", mean=c(1,-1), FUN=f1)
> z0[[1]](0.1) # expect same result as pnorm(0.1, mean=1)
> # [1] 0.8643339 # bad
> z1[[1]](0.1) # expect same result as pnorm(0.1, mean=1)
> # [1] 0.1840601 # good
> # for reference:
> pnorm(0.1, mean=1)
> # [1] 0.1840601
> pnorm(0.1, mean=-1)
> # [1] 0.8643339
>
> You don't have to say 'force(list(...))' to force their evaluation.  You could use
> 'junk <- list(...)' or 'junk <- c(...)', but using the force function should indicate
> that you are only doing this to force the evaluation of the argument at this point,
> not because you intend to use the result of list(...) or c(...).
>
> Bill Dunlap
> Spotfire, TIBCO Software
> wdunlap tibco.com
>
>
>> -----Original Message-----
>> From: r-help-bounces at r-project.org [mailto:r-help-bounces at r-project.org] On Behalf
>> Of Julio Sergio
>> Sent: Wednesday, June 05, 2013 1:26 PM
>> To: r-help at stat.math.ethz.ch
>> Subject: Re: [R] Trying to build up functions with its names by means of lapply
>>
>> Bert Gunter <gunter.berton <at> gene.com> writes:
>>
>> >
>> > faux <- function(c, nm = "gamma",...){
>> >      f <- get(paste0(c,nm))
>> >       function(x)f(x,...)
>> >     }
>> >
>> > This could be called with:
>> >
>> > > xgam <- lapply(c("p","d"), faux, shape=k, scale=theta)
>> > > xgam[[1]](1000)
>> > [1] 0.8710477
>> > > xgam[[2]](1000)
>> > [1] 0.001265311
>> >
>>
>> Excellent, Bert, and thanks a lot for your contribution! In fact, I was
>> planning to write a separate functions producer for each distribution. This
>> really saves me a lot of work!
>>
>>
>>   -Sergio
>>
>> ______________________________________________
>> 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.
>
> ______________________________________________
> 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.



-- 

Bert Gunter
Genentech Nonclinical Biostatistics

Internal Contact Info:
Phone: 467-7374
Website:
http://pharmadevelopment.roche.com/index/pdb/pdb-functional-groups/pdb-biostatistics/pdb-ncb-home.htm