Hi Jon,

Interesting idea. Have you tested this against large tables? This would result in a table scan which might have a big impact on performance...