Also switch to forder on subsetted ordering in i

[MichaelChirico]

If we want to get the top five entries of a column, I guess the canonical approach is:

set.seed(45L)
DT = data.table(A=sample(3, 10, TRUE), 
         B=sample(letters[1:3], 10, TRUE), C=sample(10))

DT[order(C), C[1:5]]

But I've found in more general scenarios (for example, subsetting on the top five of C then performing some operations on other columns) the following approach to be useful :

DT[order(C)[1:5], B]

However, this approach kills forder detection:

DT[order(C)[1:5], B, verbose = TRUE]

I guess it's easy to detect this and do forder and then subset.

[jangorecki]

data.table/R/data.table.R

Lines 430 to 441 in 0bf0e33

	if (is.call(isub) && isub[[1L]] == as.name("order") && getOption("datatable.optimize") >= 1) { # optimize here so that we can switch it off if needed
	if (verbose) cat("order optimisation is on, i changed from 'order(...)' to 'forder(DT, ...)'.\n")
	isub = as.list(isub)
	isub = as.call(c(list(quote(forder), quote(x)), isub[-1L]))
	}
	if (is.null(isub)) return( null.data.table() )
	if (is.call(isub) && isub[[1L]] == quote(forder)) {
	order_env = new.env(parent=parent.frame()) # until 'forder' is exported
	assign("forder", forder, order_env)
	assign("x", x, order_env)
	i = eval(isub, order_env, parent.frame()) # for optimisation of 'order' to 'forder'
	# that forder returns empty integer() is taken care of internally within forder