1. For every elements in array

 2. Assign elem[0] of array in key1,key2,key3 etc to an empty array 
    (we call it: $AoA[0])
    So it'll be as many as 4 arrays (= size of array in hash)
    e.g. $AoA[0] ..$AoA[3] filled up with elem[0] of all hashes
         $AoA[0][0]="A",$AoA[1][0] ="B", $AoA[2][0]="C",$AoA[3][0]="D"

    Or if you will at this point we have @AoA = ([A],[B],[C],[D])

    Every element of arrays in first key *becomes* first element 
    of new array in @AoA

 3. Check if the elements of next array are equal
    e.g Check: elem[0]-key1 = elem[0]-key2

 4. If they are equal, store elem[0]-key2 in $AoA[0], 
    then go to next hash (elem[0]-key3)..etc

 5. If they are not equal, check if elem[0]-key1 = elem[1]-key2 ..etc 
    until elem[last]-key2 (exhaust all element of array in key2)
    If they are >1 matches, take only the *first* matching one.
    If after exhausting it, still cannot find, 
    assign '-' (dash) to $AoA[0] etc.

 6. While checking, check elem[0]-key1 ..elem[last]-key1 
    with elements of next array of key2-key4, 
    verify if elem[0]-key2 has been *used* for previous match,
    if yes then check elem[0]-key1 = elem[1]-key2, 
                                   = elem[2]-key2  ...
                                   = elem[last]-key2, 
    otherwise match it (assign elem[0]-key1 = elem[0]-key2 ).

 7. Finally we have an AoA with $AoA[1] ..$AoA[4] filled up 
    (see answers snippet below)