If you have a dataset in which the very first column in the table is string with lots of possible values, then when you connect a Nominal Value Row Filter node it gives an error as soon as you open the dialog of the node.
This is because the first column is selected by default and it has so many values it needs a Domain Calculator node to be run beforehand. This is even if this is the column you are not planning on filtering on.
This can be problematic if you have a million rows, as then you will need to calculate the domain of that column for a million different values.
Simon.
Hey @richards99,
thanks for reporting. We are aware of this issue and will improve the error handling in the future. If you click on the red ‘X’ and close the error message you can continue to work with the node, even if you have a column with too many possible values.
Greetings and sorry for the inconvenience,
Daniel
1 Like