What is the role of the RETURN clause in a pipelined function?

Question

I wrote several pipelined functions and they work just fine without the return clause. However in the oracle's documentation there are returns in the pipelined functions:
CREATE FUNCTION StockPivot(p refcur_pkg.refcur_t) RETURN TickerTypeSet
PIPELINED IS
  out_rec TickerType := TickerType(NULL,NULL,NULL);
  in_rec p%ROWTYPE;
BEGIN
  LOOP
    FETCH p INTO in_rec;
    EXIT WHEN p%NOTFOUND;
    -- first row
    out_rec.ticker := in_rec.Ticker;
    out_rec.PriceType := 'O';
    out_rec.price := in_rec.OpenPrice;
    PIPE ROW(out_rec);
    -- second row
    out_rec.PriceType := 'C';  
    out_rec.Price := in_rec.ClosePrice;
    PIPE ROW(out_rec);
  END LOOP;
  CLOSE p;
  RETURN;
END;
/http://docs.oracle.com/cd/E11882_01/appdev.112/e10765/pipe_paral_tbl.htm#CHDJEGHC
Moreover, at least in PL/SQL developer I recieve hints that my pipelined functions return no values, though as I said the functions work smoothly. So is there a need in the return clause in pipelined functions or it is redundant?
My oracle version is 11.2.0.2.0

Answer 1

As already mentioned, in later versions of the database it doesn't appear to be necessary, however, it IS good programming practice (and necessary in non-pipelined functions) to always ensure a function has a RETURN statement in it. It could be a 'side effect' (read that as bug if you like) that it doesn't appear to be necessary, so you would be best to ensure you DO include a RETURN statement, as documented.