On Mon, Jul 25, 2011 at 4:14 AM, Noah Silverman <noahsilver...@ucla.edu> wrote:
> Hi,
>
> I'm working on coding some more complex things in R and have need to break
> much of the logic into functions.
>
> I have several "global" variables that I want to change with a given
> function. (The variable has a different value after the function is called.)
>
> In other languages like C, this is simple. However, in R, if a function
> changes a variable, that change only occurs in the frame of that function.
> So, when the function returns, the old value is still there.
>
> Of course, I could just have the function return the value, but some
> functions change 5-6 variables. So, I could have a function return a list,
> and then parse that list every time, but that seems like an excessive amount
> of overhead. (Especially as some functions may be called many many times.)
>
> How have some of you handled this? Is there a "best practices" way?
>
Usually I would handle this by returning a list. It's not that much overhead.
If you have a set of functions that share variables (like a C static
variable) you can use lexical scope and <<- to make changes. The demos
in the tcltk package do this.
You can also use a scratchpad environment and pass it around. Since
environments are passed by references, changes made to that
environment inside a function will stay changed.
e<-new.env()
e$storage<-0
f<-function(i, scratchpad){
scratchpad$storage<-scratchpad$storage+i
scratchpad$storage
}
g<-function(i, scratchpad){
scratchpad$storage<-scratchpad$storage-i
scratchpad$storage
}
f(1,e)
g(2,e)
f(3,e)
g(4,e)
-thomas
--
Thomas Lumley
Professor of Biostatistics
University of Auckland
______________________________________________
R-help@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.
