Calculated Table using SELECTEDVALUE

[ww1711]

Hi all,

This should be simple enough, but I can't get it to work... I've got Table A and Table B that share the same schema. I want to inspect a slicer (that chooses between 2 values in a separate table) and return either Table A or Table B as the definition of calculated table C.

Essentially, something like (pseudocode): 

• if slicerSelection == Value A
  • return Table A
• else
  • return Table B

I want the above to be the defintion of "Table C", so that I may assign it to a Matrix and the user can easily swap between table A and table B. I understand this kind of thing can be accomplished with bookmarks (hiding/showing based-on button clicks), but I'd rather do it with the above logic if possible.

Thank you!

[mwegener]

Hi @ww1711 ,

the calculated table would look something like this.

Table C =
UNION (
    ADDCOLUMNS ( Table A, "Source", "Table A" ),
    ADDCOLUMNS ( Table B, "Source", "Table B" )
)

Regards,

Marcus

Dortmund - Germany
If I answered your question, please mark my post as solution, this will also help others.
Please give Kudos for support.

[JarroVGIT · Accepted solution]

Hi @ww1711 ,

Unfortunately, a Calculated Table is evaluated upon refresh of the datamodel. Thus selecting a slicervalue doesn't return a new evaluated calculated table.

What might be a solution to your problem is to merge both table A and B in total with an additional column "SourceTable" that has value A for all rows that come from Table A and has value B for all rows from Table B. This then can be Table C. If you put the column "SourceTable" in the slicer, you can have filtered contexts of Table C in the matrix.

Hope this explanation makes sense, if not please let me know!

Kind regards

Djerro123

-------------------------------

If this answered your question, please mark it as the Solution. This also helps others to find what they are looking for.

Keep those thumbs up coming! 🙂

[jfprend]

Lol Power BI strikes out again

[TomMartens]

Hey @ww1711 ,

next to @JarroVGIT suggested it's also important to know, that the DAX formula can not return a table object or a column object.
This is one of the reasons (but not the most important one) why it's simply not possible to swap the axis of a visual or the row/column headers of a table or matrix visual.

So, it's necessary to have a column that contains all the possible values and create a measure that "maps" the values from the source tables in accordance with the selections made by the user.

The most simple way is to merge the tables as suggested by @JarroVGIT and "typed" by @mwegener.

Regards,

Tom