DataStage | Remote Column Propagation ( RPC )

I actually love this feature in DataStage. I think this is one of the excellent feature of dataStage and yet not been effective used by everyone. So I thought let me document it.

Things covered in this Blogs are

1. RPC features and highlights
2. Manipulating just few column from the list of RPC'd columns.
3. Deleting a column from the RPC list as it is not required for the target.
4. Add new columns at the bottom of the existing RPC list.

RPC highlights

1. As you all know, when we enable the check box "Runtime column propagation" there is no need to define any column definition. The column name, type, nullable characteristics and the value gets carried to the target stage. 
• Saves a lot of time in manual column defining ( using "load" table definition manually is not that great).
• Secondly when you make changes to you tables, you don't have to do anything on your DS job ( you can also argue this as a negative thing, but in most case I think it is helpful).
2. If you want to change any specific column value or assign the value to some other column, you can define only that required column in your col definition and all the other columns get automatically propagated along with this defined column.
• This feature helps to just define the column that we need to manipulate or required for data manipulation.
• Use a "transform" stage to do all the data manipulation.
3. If your target ( eg. target is "insert into a table") doesn't have a specific column, but its there on the source then you can use  a "modify" stage to remove that column.
4. Use a transformer stage to add any "new columns"

Here are some screenshots

Overall DS JOB Flow with RPC

DRS_source: Defining only the required columns and setting RPC

TRANSFORMER: To add new columns, to do data_manipulations etc.

MODIFY Stage : remove a column

MODIFY Stage: make sure you enable RPC on the modify stage output.