------------------------------------------------------------------------------------------------------------------------
CHART

select
                Count                         on 0
    NON EMPTY { Switzerland, France, Spain }  on 1
    NON EMPTY { 2020, 2021, 2022           }  on 2
              Product                         on 3     => always keep those tuples
                                                                 ([Px][0].NE = true, [Py][1].NE = true, ...)


cell evaluation:
                                                                                            Count        Count.FV    ...
    ( Switzerland ) ( 2020 )    10  => Switzerland[0].NE = true   2020[0].NE = true       row.0 = 10      10.0
    ( Switzerland ) ( 2021 )     -                                                        row.1 =  -
    ( Switzerland ) ( 2022 )     -                                                        row.2 =  -

    ( France      ) ( 2020 )     -                                                        row.3 =  -
    ( France      ) ( 2021 )     -                                                        row.4 =  -
    ( France      ) ( 2022 )     -                                                        row.5 =  -

    ( Spain       ) ( 2020 )    20  => Spain[2].NE = true         2020[0].NE = true       row.6 = 20      20.0
    ( Spain       ) ( 2021 )     -                                                        row.7 =  -
    ( Spain       ) ( 2022 )    30  => Spain[2].NE = true         2022[2].NE = true       row.8 = 30      30.0

to avoid having large empty list for every cells-columns, let's keep an index (row-index 2 row-pos) :

                         row-index 2 row-pos     Count  Count.FV ...
                                 6-0               20     20.0
                                 0-1               10     10.0
                                 8-2               30     30.0



from that:

    axis.0 : { Switzerland[0], Spain[2] }  - order is relevant: sorted using their initial [Si] !
    axis.1 : { 2020[0], 2022[2] }          - order is relevant: sorted using their initial [Si] !

    axis.0 length before non-empty :3      - required to compute the row before non-empty
    axis.1 length before non-empty :2      - required to compute the row before non-empty

redo the cell-iteration:

    and generate the members columns
    and generate the cells columns => need to retrieve initial row-index from initial tuple index!

                                                              Count
    ( Switzerland[0] ) ( 2020[0] )      => ( 0   +0 ) = 0 => 10  row.0
    ( Switzerland[0] ) ( 2022[2] )      => ( 0   +2 ) = 2 =>  -  row.1
    ( Spain      [2] ) ( 2020[0] )      => ( 2*3 +0 ) = 6 => 20  row.2
    ( Spain      [2] ) ( 2022[2] )      => ( 2*3 +2 ) = 8 => 30  row.3

--




_
