I am not sure how to built the workflow that will:
Read Excel file (OK, Excel Reader node here).
Check the content (all columns and all rows) and search for diacritic characters. Let’s simplify the set of those characters to ‘a’, ‘b’ and ‘c’ (I’ve tried String Manipulation Multi Column node here, however not sure what would be the correct expression).
Add duplicated column if the original column contains at least one diacritic character ‘a’, ‘b’ or ‘c’ (also not sure how to do that).
In appropriate duplicated column add figure ‘1’ if row contains at least one diacritic character ‘a’, ‘b’ or ‘c’.
Any suggestions will be more than welcome.
The issue is solved now. The solution is to apply Column Filter node and filter only String columns before applying String Manipulation Multi Column node.
Hi,
You should not need to use the Column Filter node if you select only the relevant columns in the String Manipulation (Multi Column) node. It has a column selection at the top and I think by default all columns are selected.
Kind regards,
Alexander
I’m getting the following error ‘The method indexOf(java.lang.CharSequence, java.lang.CharSequence) in the type org.knime.base.node.preproc.stringmanipulation.manipulator.IndexOfManipulator is not applicable for the arguments (java.lang.Double, java.lang.String)’ if working on all columns.
I’m using indexOf function in String Manipulation Multi Column node, thus I need to work on string columns only.
Hi Kaz,
Ah, I thought you might know the columns and could just select them manually. But if the names change, then you are absolutely right, the Column Filter is the way to go! Let me see if we have a ticket open to add the type-based filtering to the String Manipulation node. I think it would make sense if it used the same dialog component as the Column Filter.
Kind regards,
Alexander
Edit: I have created a ticket for this. The internal ticket number is AP-20360.
