Re: Selecting unique count on non-unique index

Try Select field1, field2, field3, count(*) from table group by field1, field2, field3 This will list out your (unique) key any the number of rows with that key. Nick Leporacci, Jr. Ian Clark <ian@tubeway-army.demon.co.uk> wrote in article <34ded553.1455143745@news.demon.co.uk>... > Hi All, > > I've spotted a table in our database that has no unique key but is > nevertheless updated using a WHERE clause that identifies a > uniqueness. > > That is: > > UPDATE a_table SET a_field = a_value > WHERE a_field1 = a_value1 > AND a_field2 = a_value2 > AND a_field3 = a_value3> > The fields a_field1, a_field2, a_field3 have not been collated to form > a unique key ( which I intend to do! ). > Having found the total number of records on the table, how do I now > determine the number of records based on the uniqueness of the three > fields sited above? > > Cheers, Ian. > Tedious 'office speak' part 3: "Let's go scuba in the think tank." >