6 Replies Latest reply: May 8, 2012 8:49 AM by Lucky RSS

How to avoid aggregations in Bar Chart


Hello, this is a sample of my problem

This is a simple table showing an ID, a customer and a value

MWSnap 20120508120048.gif


with a simple Bar chart using the expression sum(Value) i receive this obvious result:


MWSnap 20120508120229.gif

but i would like yahoo to be drawed 2 times, like in the table, each bar showing the value 5000

I don't want the ID field to compare in the chart


how can i obtain this result?


Attached the sample project

Thankyou so much!


  • How to avoid aggregations in Bar Chart

    If you could add the ID as one of the expression IT might help you get it divided.


    Hope it helps you!

    Thanks/ Subin

    • How to avoid aggregations in Bar Chart

      No, i cannot use the ID in the chart

      • How to avoid aggregations in Bar Chart
        Erich Shino

        HI, Lucky.

        I think the chart might be confusing for users but if it makes sense in this case, you can use dual

        Remove your original dimension and include a calculated one.

        The syntax will be:

        = dual ( Customer, id)


        The expression will remain the same


        hope this helps,



      • Re: How to avoid aggregations in Bar Chart
        Dennis Hoogenboom



        So you want the value of the ID by id shown per customer but you dont want to use the ID?

        Then somehow you have to make the customer unique to seperate them.

        (because now you ask the sum of the values per costumer, so Yahoo will be summed)


        If you don't want to use the ID you could add something else to make it unique.

        Take a look at this and let me know if it works for you:




        Load          *,

                                      '1'  as counter;

        LOAD * INLINE [

            id, Customer, Value

            1548, google, 10000

            1578, yahoo, 5000

            1648, msn, 1000

            1487, yahoo, 5000

            487, altavista, 500



        Join (TestTable)

        Load Customer,

                   Sum(counter) as UQ

        Resident TestTable

        Group by Customer;


        Join (TestTable)

        Load id,

                            IF(UQ =1, Customer, Customer& ' '& RowNo()) as Customer2

        Resident TestTable


        Drop field UQ, counter from TestTable;


        This will create a new Customer field (named it Customer2) which adds the rownumber to the customer when there is more then 1 ID for this customer.


        Maybe not the perfect solution but maybe you can work from here.