Home Forums SQL Server 2012 SQL Server 2012 - T-SQL How can I group near duplicate records under a new common field? RE: How can I group near duplicate records under a new common field?

  • August 31, 2015 at 8:01 pm

    #1823537

    craig.bobchin (8/31/2015)

    Thanks Jason, I do actually have the fields in the table, I just copied the earlier version of the create table script for dimMembers.

    I'll check the code against the earlier version tomorrow morning. The spreadsheet data looks good. I've been running the original version of the script today with various thresholds and as expected getting wildly varied results.

    One thing I'm seeing is that of the 30k rows in the table almost 50% have either a -1 or a NULL in the GroupID field. I'm not sure what the deal is with that.

    No problem... I did see where my original code was getting NULLs based on your sample data. I found the issue and correct that in the latest code. I have absolutely no idea how you're getting -1 for anything. The GroupID is based on existing dimMemberIDs... So unless dimMemberID = -1 exists in your data, that shouldn't be possible...

    In any case, take the new version of the code for a spin and see if it gets you a little closer.