You may want to split your [Project Code] in your Datasource 2
and do blend with a Datasource 1 (as a Primary) twice -- for each split.
Please find the attached as a demo.
Hope it could help a bit.
Blend_YF.twbx 34.7 KB
sorry to revive an old thread.
I have the samme issue but with the exception that i do not strictly know how many splits my [Project Code] results in from time to time. The database containing counterpart to [Project Code] can change from day to day with users entering more comma separated values in the same field.
Is there a solution where i do not need to know upfront how many values there are in the field?
As i understand this solution i would need to to the blend once per split.
Thanks in advance,
No prob, there would be an easy way since --
especially with Join Calculations in version 10.2+.
Please find the attached.
Join_v10.2_YF.twbx 49.5 KB
So basically you join everything on everything with 1=1 and then filter using the 'contains' statement to get the 'real' matches?
Seems like exactly what i need. I do not have crazy amounts of data so it seems doable.
Thanks a lot!
I would mark your answer as a solution, but i do not think i can when it wasn't my question to start with - sorry
Yep, you got it right.
Even if there would be lots of data (up to a point :-),
or the data would be in different places (connections),
Tableau Data Engine could be doing cross-join and filter
efficiently (with a minimal impact on performance).
Besides, the string comparison (in filter) would add to the cost.
PS I've marked the question as Assumed Answered.