I've got to mention a few of the very odd interviews I've had:

1. Many years ago went for an interview as a Junior DBA. Got there and there were 18...

My bad! Mental note added. ^^.

In future will be careful not to tread on toes!

Thanks! Learned something.

Thanks!

Ha! Good question, I didn't know that a batch separator is not truly part of T-SQL!:blush:

Just as an update, I ended up going with EXCEPT using a subset of data each day for the comparison. It's performance is better than any of the other options...

Thanks, easy one for a slow Tuesday!:-D

Thanks for the info.

I find the CLR/external assembly implementation runs a lot faster for the algorithms I have implemented (Jaro, Levenstein etc), but having a T-SQL version is certainly handy.

Set up two small test tables with different schemas - turns out I can't use Tablediff. *hits head against wall*

Problem seems to have magically disappeared when I opened and ran each subreport.:rolleyes: