Question

I’ve got a table on a MSSQL-Server with a couple of fields:

fieldA,fieldB,fieldC,DateTimeField

now I’m doing a select like this:

SELECT fieldA,fieldB,fieldC, max(DateTimeField),min(DateTimeField),
(count(*) - count(DateTimeField)) as DateTimeFieldIsNull
FROM table 
GROUP BY fieldA, fieldB, fieldC

which is working quite well.

Now I want to count all DateTimeFields that are in the past, or in another column, that are in the future.

Is there a perfomant way? (maybe even without JOINING) ?

Thnx, Harry

Answer 1

Just add a COUNT/CASE construct.
You get NULL from the implied ELSE which COUNT ignores

COUNT(CASE WHEN DateTimeField > GETDATE() THEN 1 END) AS FutureDates,
COUNT(CASE WHEN DateTimeField < GETDATE() THEN 1 END) AS PastDates,