[R-Forge #2763] Reduce NAMED(2) RHS

Submitted by: Matt Dowle; Assigned to: Nobody; R-Forge link

Seems like an extra step at C level might be bumping to NAM(2) unnecessarily.

As raised by Frank in comments here :
http://stackoverflow.com/a/16152670/403310

Consider :

options(datatable.verbose = TRUE)
a = 1:3
b = 4:6
DT = data.table(DUMMYCOL=1:3)
.Internal(inspect(DT[,ifelse(!is.na(a),a,b)]))
Detected that j uses these columns:
@0x00000000080a5780 13 INTSXP g0c2 [NAM(2)](len=3, tl=0) 1,2,3
DT[,foo:=ifelse(!is.na(a),a,b)]
Detected that j uses these columns:
Assigning to all 3 rows
RHS for item 1 has been duplicated. Either NAMED vector or recycled list RHS.
DT
DUMMYCOL foo
1: 1 1
2: 2 2
3: 3 3

Start with the R-Forge report and its reproduced DT[,ifelse(...)] and DT[,foo:=...] examples. Inspect the C-level RHS assignment path involved by the second example, using the supplied verbose output and internal inspection to verify where NAM(2) or duplication is introduced. Done means the unnecessary duplication is removed while both examples retain their shown results.