Question

我有这样的表结构：

  id - person_name - date

- 1 - Tester - 15-10-2014 
- 2 - Tester - 15-10-2014
- 3 - Tester - 15-09-2014
- 4 - Coder  - 15-10-2014

我需要选择人和每个人最小。 3个月的记录数。

赞：name =＆gt;测试仪; month1 =＆gt; 2; month2 =＆gt; 1 ;; name =＆gt;编码器; month1 =＆gt; 1; month2 =＆gt; 0

SELECT person_name, 
       COUNT(person_name) AS month1
FROM table
WHERE MONTH(date) = "10" GROUP BY person_name

这给了我很好的结果，比如name =＆gt;测试; month1 =＆gt; 2 ...但我还需要其他选择选择其他月份，我尝试工会等等，没有任何成功。感谢

Answer 1

如果您需要获得所有月份和所有人，则需要更复杂的查询。从您需要生成的所有行开始，然后使用left join和group by：

select n.person_name, ym.y, ym.m, count(t.person_name)
from (select distinct person_name from table) n cross join
     (select distinct year(date) as y, month(date) as m from table) ym left join
     table t
     on t.person_name = n.person_name and year(t.date) = ym.y and month(t.date) = ym.m
group by n.person_name, ym.y, ym.m;

Answer 2

如果我理解这个问题，你只需要在月份上分组，并举例说明文件管理器。沿着：

SELECT person_id
      ,MONTH(date)
      ,COUNT(person_id)
  FROM table
 WHERE MONTH(date) IN (8,9,10)
 GROUP BY person_id
         ,MONTH(date)

[EDIT] with pivot。

SELECT * 
  FROM (SELECT person_id
             , MONTH(dttm) MonthId 
          FROM tbl) SourceTable     
 PIVOT (COUNT(MonthId) 
          FOR MonthId IN ([8],[9],[10])
       ) PivotTable

如何在一个表中选择列的记录数

2 个答案: