Yes, the rowid_object is expected to by NULL in general cases in Staging Tables. The rowid_object will only be populated in-case of "Load By Rowid_object"
Rowid_object will be automatically generated based on a sequence.
In case of load by rowID concept, we should find appropriate rowid_object value and update to landing table right? Otherwise record can not match to BO later during load process if incoming record has a match in BO/XREF?
Without load by rowid, can we do update to XREF based on Primary key only?
1) When you don't have rowID and have the Pkey + source system :
a) if Pkey+source system exists its an update
b) if pkey+source system does not exists it will create a new BO and rowId for that record.
2) When you have rowID+Pkey+ source, then it will insert the pkey under that rowID that is being passed.
Hope that helps.
Thank you very much.
Can you tell me how do i populate rowid_object for source? because this has to be passwed to stage table during stage process and then it can be used for LBR.
rowid_object should be a valid value right? So verification need to be done on the BO to see if you can choose LBR ..
1 of 1 people found this helpful
Load by rowid is useful when you want to update a particular record irrespective of the source system.
You can use external tool like ETL to identify the rowid object of particular record depending on uniqueness of that record and populate same into landing table.
If rowid is invalid the record will fail during the load process.