number flow

not sure if this has been asked but.. could you elaborate on "There may be duplicate diameters in the file."

does that mean i could get a file containing 1, 3, 2, 4, 2, 5 with more than one of the same number?

also, is the flow constant? or is it possible that numbers will be missing? ie: 1, 3, 5, 4 (no #2) ??

numbers may occour more than once and in complete disorder.
numbers are between 1 and 1000. and you get 50 to 500 numbers in one stack.

well.. there goes my algorithm..
flush();