Home Forums SQL Server 2008 T-SQL (SS2K8) Rank duplicates, but only rows involved in duplicates RE: Rank duplicates, but only rows involved in duplicates

  • sqldriver (10/25/2014)


    Eirikur Eiriksson (10/25/2014)


    You are on the right track there, a simple solution would be to "UPPER" the values before hashing. SHA1 is quite a lot cheaper computationally than MD5 and should be more than fit for purpose, gives a better distribution (less collisions) than binary_checksum but the latter is at least in the order of one magnitude faster, a little bit of a juggle there.

    😎

    Oh, yeah, let me start nesting functions in a test harness with code from you and Chris. That'll go well :hehe:

    Keep in mind, there's no such thing as a bad code (if it works), your efforts are as worthy as mine any given day, keep up the good work!

    😎