Question

如何计算给定列中的出现次数？

请按照此示例操作。有这样一个表：

+----------+---------+------+
| PersonID |  Name   | City |
+----------+---------+------+
|        1 | John    | NY   |
|        2 | Mohit   | CA   |
|        3 | Jay     | AZ   |
|        4 | Roger   | NY   |
|        5 | David   | NY   |
|        6 | Peter   | AZ   |
|        7 | Ana     | NY   |
|        8 | Irina   | NY   |
|        9 | Michael | NY   |
|       10 | Ken     | AZ   |
+----------+---------+------+

我怎样才能做到这一点？

+----------+---------+------+-------------+
| PersonID |  Name   | City | CityCounter |
+----------+---------+------+-------------+
|        1 | John    | NY   |           1 |
|        2 | Mohit   | CA   |           1 |
|        3 | Jay     | AZ   |           1 |
|        4 | Roger   | NY   |           2 |
|        5 | David   | NY   |           3 |
|        6 | Peter   | AZ   |           2 |
|        7 | Ana     | NY   |           4 |
|        8 | Irina   | NY   |           5 |
|        9 | Michael | NY   |           6 |
|       10 | Ken     | AZ   |           3 |
+----------+---------+------+-------------+

我是否必须使用ROW_NUMBER函数而不是SUM？

Answer 1

就这么简单。

SELECT *,
       cityCounter = ROW_NUMBER() OVER (PARTITION BY city ORDER BY personID)
FROM yourTable
ORDER BY personID

Answer 2

如果我理解正确，您只想枚举每个城市的值，然后按id排序：

select t.*
from (select t.*,
             row_number() over (partition by city
                                order by personid) as CityCounter
      from table t
     ) t
order by personid;

Answer 3

您可以使用row_number（）分析函数来实现结果，但是没有过SUM？您可以阅读有关使用row_number（）及其正确语法的http://ashken-f00fc7c8.rhcloud.com/webservice/文档。

在您的情况下，您想要对每个城市的出现次数进行编号，这意味着您需要按城市进行分组（按城市划分），然后您需要为您的订购提供逻辑，哪个城市线应该是第一个，是它当personID出现时？

您的row_number将如下所示：

return allDashboardsDeferred;

列结果为ROW_NUMBER

3 个答案: