Thank you @ScottF, it helped. Now i got another problem.
My workflow is separates empty and not empty files. Empty files should be ignored while not empty files need to be recorded.
For some reason my workflow is ignoring 1st file and starts processing from 2nd, do you know the potential reason?
UPD: i think it is because of the loop end. it does not trigger the “excel writer” before it shifts the iteration.
If the loop end connected to the excel writer, then it stops looping once the “empty file” is identified.
How can i close the loop and still have the workflow filter empty and not empty files?
As you have discovered, there isn’t a place to anchor the flow variable when the table is empty.
This is an example use case for the “NodePit Power nodes”(formerly the “NoOp” / “No operation” ) nodes.
You don’t need those nodes, but they do make life simple and their purpose is clearer. The alternative is is place any node that does nothing (or even does something that you can safely ignore) and then use the flow port from that node.
e.g. before installing “NoOp”, I regularly used Add Empty Rows in this scenario (and had my own component that wrapped it!).
Just configure it to Add “an additional” zero rows!
e.g. Here is what the flow could look like using the alternative options: