1 of 1 people found this helpful
There is an idea here (https://community.tableau.com/ideas/9280) that would address this particular challenge. Please consider voting for it.
In the meantime, you could solve this by adding a union in the flow, after the split but prior to the pivot, with a dummy data set of 1 record that contained the maximum number of fields you anticipate having to pivot. That would give you explicit fields to perform the pivot and then you could filter out any records resulting from the pivot that were dummy data, plus any NULL values from the actual data that were fields in excess of what actually resulted from the split.
And yes, the split does only allow 10 resulting fields . You can work around it, but it's not as easy as it should be. So, here's another idea to vote for! https://community.tableau.com/ideas/9033
Hope that helps!
Please to see I hadn't missed anything. Ideas have been voted on :-)
For now, given the 10 column limitation, I've manually created the additional 3 calc fields to give me 10 split instances of the field. I then have explicitly added all 10 fields to the pivot and then removed all the rows with an empty value.