PostgreSQL: Combine Count and DISTINCT ON

Question

Given this table

| id | name  | created_at                 |
| 1  | test  | 2015-02-24 11:13:28.605968 |
| 2  | other | 2015-02-24 13:04:56.968004 |
| 3  | test  | 2015-02-24 11:14:24.670765 |
| 4  | test  | 2015-02-24 11:15:05.293904 |

And this query which returns only the rows id 2 and id 4.

SELECT DISTINCT ON (documents.name) documents.*
FROM "documents"  
ORDER BY documents.name, documents.created_at DESC

How can i return the number of rows affected? Something like

SELECT COUNT(DISTINCT ON (documents.name) documents.*) FROM "documents"

Answer 1

You can use an outer query:

SELECT COUNT(1)
FROM (
    SELECT DISTINCT ON (name) *
    FROM documents
    ORDER BY name, created_at DESC
    ) alias