Performing Operations On An Arbitrary Number of Columns With Same Prefix

KNIME Analytics Platform
1

I am a new KNIME user attempting to create a workflow to translate a certain data source for a number of purposes.  One key aspect of this data source is that each instance contains a number of tag variables, with each such column named Tagged<something>.  However, the number of tags in each data set from the source varies, and I'd prefer not to have to re-hard code this each time.

Mainly, I need to 1.) recode the tag variables from 'x' and missing to 1 and 0, and from string to number, and 2.) At some point, convert wide to long with date, another variable, and each tag variables defining the new observations.

I've tried the Extract Column Header > etc. > TableRow to Variable Loop Start method for the first task, but that just appends rather replaces, leaving me with further problems to deal with. 

Thanks in advance for any help anyone can provide!

2

Are you using the loop end column append node to finish the loop?  Otherwise it sound like you are on the correct path.

Regards,

Aaron

3

I tried using column append, but in case 1 I still need to figure out how to delete the old versions, and I don't see and obvious way to filter out columns based on a regex of the name.  Is there another way of doing that I am not thinking of?

 

4

Have you tried the "Enforce Inclusion" option in the column filter?  This will limit the output to only the columns you specify. If you have an example, I can take a look at it. 

Regards,

Aaron

5