drop down row column header and put new one with a filter

[tarek_b21]

I am currently working on a project that I would like to drop an existing full row of column header to row 1 and in row 0 to insert a new row of column headers which each has a filter I could use from. For example for Formatkategorie it could be ^(16|20|21|46|48|65)$.

[ScottF]

HI @tarek_b21 and welcome to the forum.

I’m not sure exactly what you’re trying to do - perhaps you could provide an example input and output dataset? The Row to Column Names node is one way to manipulate headers, but it sounds like you want some filtering functionality as well…?

[tarek_b21]

hello,

Thank you for the help but it did not work. The target what I am trying to achieve is adding the data header as shown in the first screenshot to the knime workflow in the second workflow without damages rest of the data below.

[mlauber71]

@tarek_b21 maybe indeed you could provide us with an example. Or you could take a look at these examples on how to combine several nodes to handle Excel headers and cells.

Excel Import: Numbers recognized as string, which leads to an automatic roundup of all numbers

What you can do is adjust the settings. You could tell KNIME to start reading data from line 4 and that the header is in line 3. [image] This would give you this: [image] If the structure of your tables does change you might have to be creative. You could first import the sheet ‘as is’ then identify the header line. Turn that information into a Flow Variable and use this line in informing the real import node where to start looking for header and data.

Excel Reader Merged Cells

Well this can be done. Question is if indeed it would be easier to just rename the columns … [image] [image]

[marzukim]

as mentioned by mlauber71, if you can provide examples, it is much easier to suggest the best approach. the workflow here is just for giving an ideas to start.
KNIME_header.knwf (101.4 KB)

mm