Question

I’ve been trying to do is for ages but can’t seem to do it, it’s the follwing:

Select count of values in column x where columnfoo = y

Answer 1

You can use CASE expressions within aggregate functions to only aggregate based on a condition:

SELECT COUNT(CASE WHEN columnfoo = 'y' THEN 1 END) AS y_count
FROM   tbl

If columnfoo = y, then factor the row into the count, otherwise, don’t add to the count.

This avoids having to filter out records in your WHERE clause in case you also want to work with records where columnfoo is different y.

Example: You can count based on different conditions at the same time:

SELECT COUNT(CASE WHEN columnfoo = 'y' THEN 1 END) AS y_count,
       COUNT(CASE WHEN columnfoo = 'x' THEN 1 END) AS x_count,
       COUNT(CASE WHEN columnfoo = 'z' THEN 1 END) AS z_count
--     etc..
FROM   tbl