All combinations - One List - per key

[Anonymous]

Hi guys,

I need your help on this. I have a list of possible results. I want a combination of them.

To be more specific , please check the table below.

And lets take an example.

For each match, i want all the combinations (as a list) between all players (i need exact one line per player ) but just with one success for each one. 

Let me make it more clear for you. Lets check match 2 as is easier for me to populate all combo 

The combinations would be: 

So i want only one success line but for all players per game ( game A  - Players ( A,B,C,D,E)) and i need all those combination of success.

Any ideas?

Thanks in advance

[amitchandak]

@Anonymous , Not very clear

A new table like ?

crossjoin(distinct(Table[Player]), distinct(table[Match]), distinct(Table[Field]) )

[Anonymous]

ok i will  try to make it easier

1. CASE A (ACCEPTABLE VALUE 1 OR 2)
2. CASE B ( ACCEPTABLE VALUE A)
3. CASE C ( ACCEPTABLE VALUES 2,5,8)

I want all combo with one acceptable value per case. 

For the example above :

A_1, B_A, C_2

A_2,  B_A, C_2

A_1, B_A, C_5

A_2, B_A, C_5

A_1, B_A, C_8

A_2, B_A, C_8.

All the possible values per key  in my dataset are presented as a line as the screenshot below :

I have to find a way to populate all the possible combinations (per match) in one list

Better?

[v-chenwuz-msft · Accepted solution]

Hi @Anonymous ,

Is that you want a table that contains the Cartesian product of all rows?

Please consider use CROSSJOIN().Perhaps you can follow the steps below.
In the first step, get the unique value of each column, using values().
In the second part, combine these unique values. Use crossjoin()
Please modify the following MEASURE according to your dataset.

measure =
VAR _name =
    VALUES ( 'table'[name] )
VAR _player =
    VALUES ( 'table'[player] )
VAR _field =
    VALUES ( 'table'[field] )
VAR _values =
    VALUES ( 'table'[values] )
RETURN
    CROSSJOIN ( _name, _player, _field, _values )

Best Regards

Community Support Team _ chenwu zhu

If this post helps, then please consider Accept it as the solution to help the other members find it more quickly.