例子
考虑具有以下记录的表 COMPANY
testdb# select * from COMPANY;
id | name | age | address | salary
----+-------+-----+-----------+--------
1 | Paul | 32 | California| 20000
2 | Allen | 25 | Texas | 15000
3 | Teddy | 23 | Norway | 20000
4 | Mark | 25 | Rich-Mond | 65000
5 | David | 27 | Texas | 85000
6 | Kim | 22 | South-Hall| 45000
7 | James | 24 | Houston | 10000
(7 rows)
以下是一个示例,它将按 SALARY 升序对结果进行排序 -
testdb=# SELECT * FROM COMPANY ORDER BY AGE ASC;
这将产生以下结果 -
id | name | age | address | salary
----+-------+-----+------------+--------
6 | Kim | 22 | South-Hall | 45000
3 | Teddy | 23 | Norway | 20000
7 | James | 24 | Houston | 10000
8 | Paul | 24 | Houston | 20000
4 | Mark | 25 | Rich-Mond | 65000
2 | Allen | 25 | Texas | 15000
5 | David | 27 | Texas | 85000
1 | Paul | 32 | California | 20000
9 | James | 44 | Norway | 5000
10 | James | 45 | Texas | 5000
(10 rows)
以下是一个示例,它将按 NAME 和 SALARY 升序对结果进行排序 -
testdb=# SELECT * FROM COMPANY ORDER BY NAME, SALARY ASC;
这将产生以下结果 -
id | name | age | address | salary
----+-------+-----+--------------+--------
2 | Allen | 25 | Texas | 15000
5 | David | 27 | Texas | 85000
10 | James | 45 | Texas | 5000
9 | James | 44 | Norway | 5000
7 | James | 24 | Houston | 10000
6 | Kim | 22 | South-Hall | 45000
4 | Mark | 25 | Rich-Mond | 65000
1 | Paul | 32 | California | 20000
8 | Paul | 24 | Houston | 20000
3 | Teddy | 23 | Norway | 20000
(10 rows)
以下是一个示例,它将按 NAME 降序对结果进行排序 -
testdb=# SELECT * FROM COMPANY ORDER BY NAME DESC;
这将产生以下结果 -
id | name | age | address | salary
----+-------+-----+------------+--------
3 | Teddy | 23 | Norway | 20000
1 | Paul | 32 | California | 20000
8 | Paul | 24 | Houston | 20000
4 | Mark | 25 | Rich-Mond | 65000
6 | Kim | 22 | South-Hall | 45000
7 | James | 24 | Houston | 10000
9 | James | 44 | Norway | 5000
10 | James | 45 | Texas | 5000
5 | David | 27 | Texas | 85000
2 | Allen | 25 | Texas | 15000
(10 rows)