[R-pkg-devel] Using function with same name as argument as default argument

[John Fox]

Dear Jan,

Here's yet another solution that takes advantage of partial matching:

 > foo <- function(x, schema. = schema(x)) {
+   if (is.null(schema.)) stop("schema missing")
+   x
+ }

 > schema <- function(x) attr(x, "schema")

 > y <- 1
 > z <- 2
 > attr(y, "schema") <- "bar"

 > foo(y)
[1] 1
attr(,"schema")
[1] "bar"

 > foo(z)
Error in foo(z) : schema missing

 > foo(z, schema="baz")
[1] 2


I hope this helps,
  John

On 2022-08-08 9:10 a.m., Jan van der Laan wrote:
> Not sure if this belongs on r-help or r-package-devel; decided for the 
> latter as the question is mainly relevant when writing code to be used 
> by others.
> 
> The issue I run into is that I want to write a function similar to:
> 
> 
> foo <- function(x, schema = schema(x)) {
>    if (is.null(schema)) stop("schema missing")
>    # ...
> }
> 
> with 'schema()' something like (strongly simplified):
> 
> schema <- function(x) attr(x, "schema")
> 
> 
> However using both the argument schema and calling the schema() function 
> in one of the default arguments is not allowed ans results in the 
> following somewhat cryptic error message:
> 
> Error in foo(1:3) :
>    promise already under evaluation: recursive default argument 
> reference or earlier problems?
> 
> I am looking for a clean solution to this. I can rename the argument or 
> function, but calling it something other than schema feels impractical 
> as both refer to the same thing (the schema of x). The best solution I 
> have come up with until now is to define a second function to be used in 
> default function arguments:
> 
> schema_ <- schema
> 
> foo <- function(x, schema = schema_(x)) {
>    if (is.null(schema)) stop("schema missing")
>    # ...
> }
> 
> I guess another solution would be:
> 
> foo <- function(x, schema) {
>    if (missing(schema)) schema <- schema(x)
> }
> 
> But then it is not clear for the user from the interface that the 
> 'schema' argument can be omitted.
> 
> I am hoping some of you have other suggestions.
> 
> And, is this something that could be 'fixed' in the R-parser?
> 
> 
> Thanks,
> Jan
> 
> ______________________________________________
> R-package-devel using r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-package-devel
-- 
John Fox, Professor Emeritus
McMaster University
Hamilton, Ontario, Canada
web: https://socialsciences.mcmaster.ca/jfox/