Anytime I create a computed column in a view it is not available for use in the visual data lineage tool. If I move the computation into a table and then use the column from the table in a view, then it is usable in data lineage. This effectively removes view level computations from the data flow. So far I have worked around it by using non persisted computed columns in the source table, but I can see challenges when a computation depends on a more complex dataset than what a row level computation in a single table can deliver.
Is this "feature" by design? If so, why? Is there a something I am unaware of that will allow computed columns in a view to be used in a data lineage? Has anyone else faced this challenge, and what are suggested "work-arounds"?
Off the top of my head, the work-around would be to create a 'dummy' table with the same content as the View - you could a script to manage this.
There are a number of deficiencies with the way that Visual data Lineage was implemented. Have you raised a support case, so they'll consider changing the tool?