RecordArgs() is a utility function that captures and records the evaluated
forms of arguments passed to the parent function. It returns a tibble with
columns named after the arguments, containing their evaluated values only.
Value
A tibble containing the evaluated forms of the parent function’s
arguments and any additional named arguments passed via ..., with columns
named after the arguments (e.g., w, x, y, extra1). Evaluated values
are presented as scalars (e.g., 8) or list columns for complex objects
(e.g., <SpatRaster>). If ExportPath is provided, the tibble is saved to
the specified .RData file and NULL is returned invisibly.
Details
This function evaluates all arguments in the grandparent environment
(two frames up), with a fallback to the global environment if evaluation
fails. This ensures correct evaluation in iterative contexts like lapply.
It handles:
Scalars (e.g., numbers, strings) as single values.
Multi-element vectors or complex objects (e.g.,
SpatRaster) as list columns.NULLvalues are converted to the string"NULL".Failed evaluations result in
NA.Additional named arguments passed via
...in the parent function are also recorded.
The function must be called from within another function, as it relies on
sys.call(-1) to capture the parent call.
Examples
a <- 5
b <- 3
w_values <- 1:3
x_values <- c(a + b, 10, 15)
y_values <- c("ABCD", "XYZ123", "TEST")
Function1 <- function(w = 5, x, y, z = c(1, 2), ...) {
Args <- RecordArgs()
return(Args)
}
# ----------------------------------------------------
# Example 1: Simple function call with scalar and expression
Function1(x = a + b, y = 2)
#> # A tibble: 1 × 4
#> w x y z
#> <dbl> <dbl> <dbl> <list>
#> 1 5 8 2 <dbl [2]>
# ----------------------------------------------------
# Example 2: Using lapply with indexed arguments
lapply(
X = 1:3,
FUN = function(Z) {
Function1(
w = w_values[Z],
x = x_values[Z],
y = stringr::str_extract(y_values[Z], "B.+$"),
z = Z)
}) %>%
dplyr::bind_rows() %>%
print()
#> # A tibble: 3 × 4
#> w x y z
#> <int> <dbl> <chr> <int>
#> 1 1 8 BCD 1
#> 2 2 10 NA 2
#> 3 3 15 NA 3
# ----------------------------------------------------
# Example 3: Using pmap with mixed argument types
purrr::pmap(
.l = list(w = w_values, x = x_values, y = y_values),
.f = function(w, x, y) {
Function1(
w = w,
x = x,
y = stringr::str_extract(y, "B.+$"),
z = terra::rast(system.file("ex/elev.tif", package = "terra")))
}) %>%
dplyr::bind_rows() %>%
print()
#> # A tibble: 3 × 4
#> w x y z
#> <int> <dbl> <chr> <list>
#> 1 1 8 BCD <PckdSptR>
#> 2 2 10 NA <PckdSptR>
#> 3 3 15 NA <PckdSptR>
# ----------------------------------------------------
# Example 4: Using additional arguments via ...
Function1(x = a + b, y = "test", extra1 = "hello", extra2 = 42)
#> # A tibble: 1 × 6
#> w x y z extra1 extra2
#> <dbl> <dbl> <chr> <list> <chr> <dbl>
#> 1 5 8 test <dbl [2]> hello 42