How to vectorize this simple problem?

Let R(n) be a vector, f(r) and g(r) are functions. Let A and B are complementer sets of indices. For example A = R(R>0); B = R(R<=0); I would like to act f on the subvector R(A), and g on the subvector R(B), but get back the whole vector in the original order, so for the output vector V(n) = f(R(n)) if n is element of A, and V(n) = g(R(n)) if n is element of B. f and g are computationally complex functions, so I dont want to evaluate f and g on the whole input vector R, just on necessary elements, but using a vectorized method, and somehow get back automatically the vector in the original order, not to vaste time of index shuffling. Is it possible? Which is the best method?

Answer 1

[Matt J]

 V=nan(size(R));
 V(A)=f(R(A));
 V(B)=g(R(B));