1 Reply Latest reply on Oct 11, 2014 11:27 AM by philmodjunk

    Import - Update Matching Records - not working for consecutive records in source file



      Import - Update Matching Records - not working for consecutive records in source file


      I am importing a source file of contacts that contains many duplicate contacts.  Some duplicate contact records are identical and some are slightly different but with the same email address. I want to eliminate duplicates during the import process, keeping the data from the latest records and overwriting data from older records.  The source file is sorted in chronological order, oldest records first, to newest records at the bottom.  The older records may be incomplete so I want them updated with information from the newer records. 

      I do observe that when duplicate contacts that are not identical appear several records apart in the source file, the older record does in fact get updated with the info from the newer contact record.

      My problem:  But it appears that when the duplicate contact rows are consecutive (i.e. next to each other) and identical, both records get added to the target table.  

      Question:  How do I get the second record to not be added?

      Here's my script:




        • 1. Re: Import - Update Matching Records - not working for consecutive records in source file

          It should not make any difference whether the records are adjacent in the sort order or not. Two issues might affect your results, It looks like you have dealt with one by using Show All Records. (Many users fail to set up their target fields with a found set of all records.) But make sure that you are importing your records into T05_Contacts and not some other occurrence of the contacts table. The other is in how you set up the field matching and whether the values in the fields are truly identical. (The addition of an extra space character, for example, is all that it takes to keep the values from matching even though it looks like they should match.) So it looks like your next step is to review what field matching has been set up in the import records step and then to double check the actual email address values for the presence of non-visible characters preventing a match to the data.

          The third possibility is that you have encountered a new bug. If you can't find a reason for the records not to match by email address (as a test you might try setting up a self join on your T5_Contacts table to another occurrence of the same table--matching by email address to see if the imported data matches...) Possible bugs should be reported in Report an Issue.