I am trying to import multiple excel files with
- list files node
- Table Row to variable loop start
- xls reader
When running the workflow, I encounter the following problem : "Input table's structure differs from reference (first iteration) column2 <empty_63> ...". The problem comes from the second column that has no name. I use a column rename node to rename it but each time Knime loops on the file it gives the column a different name (different number as suffix).
How to keep a constant column name so I can rename it?
Thank you very much, Romain
I would propose that you activate the option in the Loop End node allowing changing table specifications.
In this case the data can change between iterations, and the loop end takes care of these changes.
Does this help?
Which option should I activate in the loop end? I don't see anything about table specifications. I am using Knime 2.12.1.
Here is in attachment the table schema.
There is an error in the xls reader too: "DataSpec generated by configure does not match spec after execution". Is it caused by the "empty_73" column?
In loop end, I have checked "Ignore emty input tables" and "Allow variable column types". I still get the same error
Is there a special reason you are still using 2.12? We added quite some features with the release of KNIME 3. and one of it is the new option for loop end nodes.
There was no reason... I switched to Knime 3.2 and it works now, I ticked Allow changing table specifications as you suggested. Thank you very much.