I have a workflow that flows from a [rule engine node] to a [Concatenate node] and to a [excel writer node].
I notice that after the [concatenate node] there are 2 of my columns that the type become “?” the rest are in string.
This result in the [excel writer node] throwing an error “Execute Failed” Unsupoorted column type ‘?’, Please remove corresponding colunn(s) from the input table".
Any advice on how i can solve this? I note from the execution results preview table that along the workflow nodes before the [Concatenate node] the 2 columns having issues are always showing ‘string’ typem up till the [Concatenate node] where it shows ‘?’ type.
The unknown column type “?” occurs when KNIME is unable to determine the data type of a column, often following a concatenate node where the two sets of data being combined have different data types, and it is not possible for KNIME to automatically set the new column data type.
I would be very surprised if this occurred where the columns in both tables heading into the concatenate were String, since there is no way that concatenating strings from two tables should be interpreted as anything other than String.
I suspect therefore that this is not actually the case, and that you are actually attempting to concatenate String with Local DateTime, or numeric data, or data that is already an unknown type.
Given you have numbers in a column named “DateTime…” I suspect you should be checking what is happening to your data prior to the Concatenate node, as from my perspective that looks suspicious.
Now you could simply drop a Colum Auto Type Cast node after the Concatenate to have KNIME attempt to determine a “best fit” column type for the data. That may just convert both columns to String, or you could alternatively use a couple of String Manipulation nodes with string($mycolumnname$) to force the data to strings, and keep Excel Writer happy, BUT I think that this will simply be masking a fundamental problem with your data which you ought to resolve upstream.
If I am wrong about your data, please post some screen shots of the problematic columns in the data tables that are feeding the Concatenate node, as it is not possible to give more help from simp looking at a few rows of the data after it has been concatenated.
