MS-SQL: Cleaning up duplicates with Common Table Expressions
It is a common pattern to prefer non-semantic primary keys — auto-incremented integers or UUIDs — over semantic ones. It’s advisable to define unique indexes on the semantic identity fields, [...]