Hi Praveen,
I have followed this at the time of development. But if you look into the message mapping the ID field is directly mapped to the target field from the ID of data type of first file.
My confusion is where the checking is done to match the common ID of both the files and where is the logic of rejecting the IDs not present in both the files.
After creating an interface following the link provided by you, I have checked that if there is any mismatch in IDs in both the files then its not generating the target file.
Thanks,
Apu