Modify stage - Runtime column propagation

[dsedi]

All,

We are using Modify stage to Keep a few columns of the input(Data from sequential file stage)
The Sequential file input stage we are using Schema file Option to define the metadata of the input file. So there is NO metadata columns definition explicitly.

In the Modify stage, we are using keep option to keep few columns, and we are forced to define the column names (metadata) explicitly on the output link of Modify stage - to the target dataset.

Now, is it possible to achieve the RCP without defining the target column names on the Output link? (As we have already mentioned the column names on KEEP statement)

Is there any other simple/better idea's ??

Thanks in advance, Edi

[balajisr]

Do you get any error? It seems to be possible.

I did a test job as
Seq File ---> Modify ----> Dataset

SeqFile: Reads metadata from Schema File having three columns
Modify: Keep First two columns

I am able to load data into dataset. Metadata is not defined explicitly in any of these stages.