You can do this with a transformer. Write a flow service that takes an object as input, the transformer tests the input for null, if so, put the substitution into the output pipeline, otherwise put the input variable back to the output pipeline. Instead of direct maps, map each field through this transformer.
thanks Eric,
that’s what I was thinking about before. I was looking for a more simple solution, because if I had 50 fields in the record I would have to use 50 transformers.
I used the next method:
I created a new record with the same structure as the source one
I assigned ‘*’ to all the fields
I mapped the source record to this new one
As the result, I have the new record with the same structure and ‘*’ for the empty fields.
Question: maybe someone can tell me if there is another way to keep the same record structure if in run-time there are empty fields.
Hello Vassili, not sure if you have found the better way. Anyhow, when set a variable value, you can untick the overwrite pipeline value box. This will subsitute the variable with your subsitude sympol if the value is empty.