RANKX - rank with only one column in Table visual

[Anonymous]

Hi, I'm struggling with making rank with only one column if there are multiple columns in the Table visual. 

So basically I want to make rank with Key1 only when there is Key2 in the Table.

I tried a lot of rank measures. Basic rank meausre is RANKX(ALL('Table'[Key1]),CALCULATE(SUM('Table'[Value]))).I tried rankx with HASONEVALUE, ALLEXCEPT, ALLSELECTED, ALLEXCEPT but none of them helped me.

As you can see in the pictures below, if there is only Key1, rank works fine. But when Key2 is added, it gives a rank within the group of Key1 and Key2. 

Can anyone helps me? Any helps are welcomed. Thanks!

p.s. I tried to upload my PowerBI file but it's not supported. This is the table I used.

Table = DATATABLE(

    "Value", INTEGER,

    "Key1", STRING,

    "Key2", STRING,

    "Key3", STRING,

    {

        {100, "A", "a", "1a"},

        {300, "B", "b", "1b"},

        {150, "C", "a", "1c"},

        {120, "D", "b", "1d"}

    }

)

[Ritaf1983 · Accepted solution]

Hi @Anonymous 
Try this measure :

Rank Key1 = RANKX(

    ALL('Table 2'[Key 1]),

    CALCULATE(SUM('Table 2'[value]), ALLEXCEPT('Table 2', 'Table 2'[Key 1]))

)

Link to a sample file 

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