• Status: Solved
  • Priority: Medium
  • Security: Public
  • Views: 1794
  • Last Modified:

Filter duplicate rows in SSIS data flow task

Hi Experts,

   I am retrieving data from oracle and inserting into sql server.
   I have 2 tables in oracle called Person and Contacts.
   Person:
      Person_ID, Forename, Surname, Initials, Title, AddressId, TelNo1, TelNo2, Email1, Email2
   Contact:
        Forename, Surname, Initials, Title, AddressId, TelNo1, TelNo2, Email1, Email2
        
      In SSIS, DataFlow, I am retrieving data from 2 tables and UNION ALL and assigning uniqueID for PersonID(sequence number)
      
      If I have a row in Person table and 3 rows in contacts table, after UNION ALL I am getting 4 rows for a person.
      All contacts will be inserted to Telecom table and Person details will be inserted to People table.
      But in People table I want only 1 row to be inserted(no duplicates).
      
      If AddressID is different for the same person, its not called as duplicate. so all rows
      are inserted to People table.
      
      How can I do this in SSIS data flow task after UNION ALL and assigning unique PersonID?
      I mean tracking duplicate records and not inserting into People table.
      
      I can't do UNIQUE INDEX with NO_DUP_VAL as the PersonID is generating after UNION ALL
      and is assigned to each contact i.e added to Telecom table.
      
      So, I want to do some logic after UNION ALL.
      
      Any suggestion?
0
rajvja
Asked:
rajvja
  • 2
  • 2
1 Solution
 
Bhavesh ShahLead AnalysistCommented:
Hi,

Can u use max(addressid)?
0
 
8080_DiverCommented:
  1. Do you have to do a UNION ALL instead of a UNION?  (DO you know the difference between those two? ;-)
  2. If you have to use the UNION ALL, then you might want to insert that data into a staging table and then select distinct rows from the staging table.
Remember, SSIS lets you have multiple steps if you need them and it looks like you may need them. ;-)
 
0
 
rajvjaAuthor Commented:
Hi,

 thanks for the replies. Even I used UNION I am getting the same records.
There is a column in Person table called PERSON_ID(it is unique per row) and I need to retrieve that column coz based on that column I am generate surrogate keys.

Any solution? I am really strucked.
0
 
8080_DiverCommented:
In that case, you will need to pull the data into a staging table and then determine how to identify the duplicates that you don't want.  Be sure to add an Identity column in the staging table so that you can distinguish between totally duplicate rows.
For instance, you could create a query that selects the MAX(PERSON_ID) based upon the rest of the columns and then use that to control which row you use to either move the data to your final destination table or simply as part of the process of selecting the data and generating the surrogate keys.
Hmmm, does the PERSON_ID already exist or are you generating it as part of this process?
0
 
rajvjaAuthor Commented:
thanks
0

Featured Post

 [eBook] Windows Nano Server

Download this FREE eBook and learn all you need to get started with Windows Nano Server, including deployment options, remote management
and troubleshooting tips and tricks

  • 2
  • 2
Tackle projects and never again get stuck behind a technical roadblock.
Join Now