DSXchange

I am sorry if this has been asked before but I couldn't find it in search so asking here.

I have this scenario - In a job I am creating a dataset and in that dataset I have defined 4 columns i.e. column1, column2, column3 and column4 with column1 as unique column. Now I am using this dataset in a separate job as a reference for a lookup. In the lookup I need to join on column1 and move the value of column3 in the next stage. Now in the reference link where I have this dataset stage, do I need to define all the 4 columns in the column definition or defining only column1 and column3 will do the work? I tried it and didn't get any warning. So I am assuming it works but just want to know if there is any other thing I need to keep in mind?

Thanks in advance for your help.

Welcome aboard.

You tried it. It worked. You only need to define the columns that you need.

Indeed, with Runtime Column Propagation, you can get away with defining even fewer!