Inject first argument in a call, usually data = (.)

In case the data = (.) is missing in a call to a "data-dot" function (a function that defines that its first argument data= could be missing, and it this case, it is considered to be .), this function does the injection of the first argument in the call and evaluate it.

eval_data_dot(
  call,
  arg = "data",
  value = as.symbol("."),
  env = caller_env(2L),
  abort_msg = gettextf(x = "Wrong '%s' argument.", arg),
  abort_msg2 = gettext(i = "Implicit data-dot (.) not permitted"),
  abort_frame = caller_env()
)

Arguments

call: A call object, usually a function call.
arg: The name of the argument to inject, usually 'data' or '.data'.
value: The value to inject, usually the symbol '.'.
env: The environment where the evaluation of the data-dot-injected call should be evaluated (by default, caller_env(2L), should rarely be changed).
abort_msg: The message to use in case the 'data' argument is wrong.
abort_msg2: An additional message to append to the error message in case data-dot-injection is not permitted (when .SciViews.implicit.data.dot != TRUE)
abort_frame: The environment to use for the error message, by default, the caller environment (should rarely be changed).

Value

The result from evaluating the data-dot-injected call.

Details

The call is not checked if it is a correct function call. When called from within a function, passing sys.call() as call, it should be always correct.

Examples

# Here is how you create a data-dot function
my_subset <- function(data = (.), i, j) {
  if (!is.data.frame(data))
    return(eval_data_dot(sys.call()))

  # Code of the function
  # Second argument (i here) must not be a data.frame to avoid confusion
  data[i, j]
}
dtf1 <- data.frame(x = 1:3, y = 4:6)
my_subset(dtf1, 1, 'y')
#> [1] 4
# If data is in '.', it can be omitted
.= dtf1
my_subset(1, 'y')
#> [1] 4

# This mechanism is potentially confusing. You can inactivate it anywhere:
.SciViews.implicit.data.dot <- FALSE
# This time next call is wrong
#my_subset(1, 'y')
rm(.SciViews.implicit.data.dot) # Reactivate it
my_subset(1, 'y')
#> [1] 4
# Note that, if you have not defined '.' and try to use it, you got
# an error: uncomment next line to see it
#get_temp(".")