Matching is always done on the Base Object records and not the XREF data.
If you are doing this for IDL ..you can generate the matches with source ABC first but not merge the data. This way you will have ABC records still available to match with XYZ.
Not sure if the above helps, but you may want to provide more details around your use case.
Thank you for your quick response,
I have some screen shot may be it will help to understand,
1st screen shot have only 1 source system details were match and merge is done, we see only golden record at base object and xref have two record..
2nd screen shot we have 2nd source data coming as portal in base object. and xref has pervious and new entry.
After the 2nd source data is loaded we are using only one rule i.e exact match on Account number column, but if you see the new portal record account number is matching with xref record as it was merged among 1st source before.
1 of 1 people found this helpful
You can create Match paths on the XREF table and match columns from these paths. This will help to achieve your case.
But consider below items and design accordingly
-- If you're using Subject areas, then you cant use these XREF based match paths to create Primary or child subject areas. You need to have another set of BO based match paths for using in IDD
-- XREFs will have more data volume than BO and indexes on join columns may not be there, so match jobs may take much longer. Check the oracle AWR report during job run and make necessary DB tuning
Thank you for your time,
I will try this scenario, and we are using 10.3 HF2 so we are having provisioning tool for any C360 related development.