but then i tried to connect another “Nominal Value Row Filter” to filter out C (i need to implement a filtering sequence). In the node’s configuration window i can see again all the possible initial values A,B,C,D,E even if they were filtered out by previous node… while i expected to see only C,D,E.
This behaviour is the same also with row splitter node and also with the usage of the “Edit nominal domain” node.
Seems KNIME row filter nodes are working this way. For example if you use Row Filter node on some table and you go on results (Filtered) and tab Spec - columns: x you we see data which you have filtered…
This just become obvious when Nominal Value Row Filter node came.
So answer to your question would be you are not doing anything wrong
anyway, i see the ordinary Row Filter/splitter nodes work as expected (i can filter some values and then i can see filtered rows in the node configuration for any sequence of this kind of nodes),
The strange case i don’t understand, is when i use a sequence of nominal values row splitter/filter nodes: output table is ok, but the node’s configuration window show always (for each node) all the possible values already filtered in previous nodes… i thought it was similar to the ordinary row filter node (i expected not to see already filtered values in configuration)…