We have made a couple of enhancements to the aggregate query feature of Fetch XML. These updates have shipped for version 4.0 as part of UR2 and for version 3.0 as documented in KB Article 960079. So far we only supported the count aggregate function which essentially did a count(*) on the base table of the entity.
We have enhanced this functionality to add a new aggregate operation countcolumn which now performs the SQL count operation for a given column. It also support the distinct keyword which only counts unique columns. If the distinct keyword is not specified with a countcolumn aggregate, its value defaults to false.
Here is a table that summarizes the Fetch aggregate types and its equivalent SQL translation:
| CRM Fetch XML | SQL |
|
<attribute name=’address1_city’ aggregate=’count’ alias=’citycount’ /> |
SELECT COUNT(*) AS ‘citycount’ FROM Account |
|
<attribute name=’address1_city’ aggregate=’countcolumn’ alias=’citycount’ /> |
SELECT COUNT(address1_city) AS ‘citycount’ FROM Account |
|
<attribute name=’address1_city’ aggregate=’countcolumn’ alias=’citycount’ distinct=’true’ /> |
SELECT COUNT(DISTINCT address1_city) AS ‘citycount’ FROM Account |
Key: Existing Feature, New Feature
Here are a couple of examples for using these features:
Query to return a count of all Accounts which have a county specified in their address.
1: Query to return a count of all Accounts which have a county specified in their address.
2: <fetch aggregate='true'>
3: <entity name='account'>
4: <attribute name='address1_county' aggregate='countcolumn' alias='countyAccounts' />
5: </entity>
6: </fetch>
Query to return a count of states in which we have an account. This uses the distinct keyword which counts a state only once.
1: <fetch aggregate='true'>
2: <entity name='account'>
3: <attribute name='address1_stateorprovince' aggregate='countcolumn' alias='countStates' distinct='true' />
4: </entity>
5: </fetch>
