HAVING

When using the GROUP BY clause, it is possible to supplement your query with a HAVING clause. The HAVING clause is like a filter, but it works on aggregations of the data rather than the rows of data prior to the aggregation. Hence, it has to be included with a GROUP BY clause. It will also include the aggregation you want to check. The code would therefore look as follows:

The aggregation_condition would be where you place the aggregation and the test you want to perform. For example, my bank charges me if I have more than 20 nonregular items pass through my account in a month. In this case, the query would group by customer ID, counting the number of nonregular transactions for each calendar month. If the count were less than or equal to 20 items, then you would like this list to not include the customer in question. To clarify this, the query code would look something like the following if you were running this in August 2011: