Thanks, could the resultset handle information be added to the SQL docs?
As an aside, I decided to modify my function to only require one handle. I temporarily save the result set to a local DIMX array, and then iterate the temporary array to make the recursive calls. This way there shouldn't be a un-freed resultset memory issue possible.
Does a previous resultset get free'd on subsequent queries if it was not explicitly free'd via a SQLOP_FREE_RESULT call?