Author: Dariusz DziewialtowskiDariusz Dziewialtowski Date: Sep 17, 2008 18:53
> Can you explain "row-by-rowcomparison"? Â Do you mean you inspect the data
> in both tables and match rows manually? Â Or do you pick columns to "pretend"
> they are the key? Â Otherwise, without a key (whether PK or UQ), how could
> the software even attempt to guess at matching rows on its own?
First of all - apologies for the long post; feel free to skip it - as
a short answer let me say that when compared tables have no key or
indexes (or: one of the tables has keys, and another - not), the user
can additionally specify "sorting" columns for the purpose of data
comparison. All options in SCT (SQL Server Comparison Tool) were
added to handle particular real life situations, and over the years
SCT "accumulated" such options like: "Columns to be ignored during
comparison" or "Comparison offset".
|