Question

I have a table like this:

column_a  column_b
  foo        1
  bar        1
  bar        2
  baz        4

I’d like to have following result, containig all rows, that have a unique field value in column_b:

  bar        2
  baz        4

foo(0) and bar(1) both have the column_b value ‘1’, so the ‘1’ is not unique.
Is there a MySQL-Expression that filters all rows that have a unique column_b?

Maybe something like that:

select * from table where isunique(column_b)

In Addition: I do not need something like DISTINCT!

edit (solved):

Thanks to YaK and jcho360, this one is the solution:

select column_a, column_b from table group by column_b having count(column_b) = 1

OR

select column_a, column_b, COUNT(1) AS total from table group by column_b having total = 1

(thx to andrewtweber)

Answer 1

Using the count function will provide a selection where the value in column_b only occurs once. Try this:

select * from table group by (column_b) having count(column_b) = 1;