Is there are convention or guideline on the use of the table name. Clearly, there is little point replicating the output name. (When computing pharmacophore profiles it is useful to know some ancillary information such as the number of molecules processd which I'm currently envisaging storing as part of the table name).
Thanks
What name are you referring to? Are you talking about the field in the class DataTableSpec? I don't think this field is exposed to the user in any way (there is no way for the end user to see this programmatic table name, I guess).
I see different possible solutions:
1) attach this information using DataColumnProperties (meta information attached to individual columns, you can view them in the outport view, third tab called "DataColumnProperties"
2) attach it as new column (if the property is specific to individual rows?)
3) add a view to the node that computes the properties
I would suggest to use 1) or 2) only if downstream-nodes make use of that information. Otherwise you should use solution 3); there are other examples for this, for instance the entropy scorer node or different learners that give a summarization of what they did during execution.
Regards
Bernd
Option (1) works nicely - although this also a case where the documentation could be better for developers who are relatively new to Java.
Incidentally the table name is displayed as part of the Window caption.
Interesting. I wasn't aware of the fact that the table name is displayed in the title. I expect in most cases it's simply "default"?!
Anyway. Good that you find (1) acceptable.
