Question

我有一张看起来像这样的表：

+-----------+--------------+---------------------+
| ENTRY_ID  | EMPLOYEE_ID  |      ACCESS_TIME    |
+-----------+--------------+---------------------+
|   001     |     1001     | 6/3/2014 8:08:00 AM |
+-----------+--------------+---------------------+
|   777     |     1201     | 8/3/2015 7:07:00 AM |
+-----------+--------------+---------------------+
|   888     |     1901     | 9/9/2015 4:04:00 PM |
+-----------+--------------+---------------------+

为了获得特定日期范围的唯一身份用户数，我有以下代码：

SELECT COUNT(UNIQUE Employee_Id) Unique_Monthly_Users FROM 
    (SELECT * FROM CORE.DATE_TEST
        WHERE ACCESS_TIME  >= TO_DATE('8/1/2015 00:00:00', 'MM/DD/YYYY HH24:MI:SS')
        AND ACCESS_TIME  < TO_DATE('9/1/2015 00:00:00', 'MM/DD/YYYY HH24:MI:SS')
    ORDER BY ACCESS_TIME ASC)

为了获得该范围的当前月份，我正在使用它：

SELECT UNIQUE TO_CHAR(ACCESS_TIME, 'Month') Month_Name FROM (
    SELECT * FROM CORE.DATE_TEST
        WHERE ACCESS_TIME  >= TO_DATE('8/1/2015 00:00:00', 'MM/DD/YYYY HH24:MI:SS')
        AND ACCESS_TIME  < TO_DATE('9/1/2015 00:00:00', 'MM/DD/YYYY HH24:MI:SS')
    ORDER BY ACCESS_TIME ASC)

这是我最后的期望输出：

+--------------+-----------+
| UNIQUE_USERS |   MONTH   |
+--------------+-----------+
|  702931      |  JANUARY  |
+--------------+-----------+
|  900000      |  FEBRUARY |
+--------------+-----------+
|  2139041022  |  MARCH    |
+--------------+-----------+

现在我尝试加入这些陈述：

SELECT COUNT(UNIQUE EMPLOYEE_ID) Unique_Emp FROM 
    ((SELECT * FROM CORE.DATE_TEST
          WHERE ACCESS_TIME  >= TO_DATE('8/1/2015 00:00:00', 'MM/DD/YYYY HH24:MI:SS')
          AND ACCESS_TIME  < TO_DATE('9/1/2015 00:00:00', 'MM/DD/YYYY HH24:MI:SS')
      ORDER BY ACCESS_TIME ASC) col1) 
INNER JOIN (SELECT UNIQUE TO_CHAR(ACCESS_TIME, 'Month') Month_Name FROM (
    (SELECT * FROM CORE.DATE_TEST
         WHERE ACCESS_TIME  >= TO_DATE('8/1/2015 00:00:00', 'MM/DD/YYYY HH24:MI:SS')
         AND ACCESS_TIME  < TO_DATE('9/1/2015 00:00:00', 'MM/DD/YYYY HH24:MI:SS')
    ORDER BY ACCESS_TIME ASC
    ) col2)
 ) ON COL1.ENTRY_ID = COL2.ENTRY_ID

......但显然这不起作用。我无法获得“id”，因为任何一个结果集都不应该有ID。两者都是存在于外部的集合。我收到以下错误：ORA-00904: "COL2"."ENTRY_ID": invalid identifier

如何正确连接这些结果集，以便获得所需的输出？

Answer 1

您可以使用两个子句执行此操作，第一个选择每个月的不同用户，第二个用于计算：

SELECT
    COUNT(*) AS UNIQUE_USERS,
    T1.MONTH
FROM
(
    SELECT DISTINCT
        USER_ID,
        TO_CHAR(ACCESS_TIME) AS MONTH
    FROM DATE_TEST
) AS T1
GROUP BY T1.MONTH

Answer 2

查找GROUP BY子句，然后尝试：

select to_char(access_time, 'Month'), count(distinct employee_id)
from core.date_test
where usage_time >= TO_DATE('8/1/2015 00:00:00', 'MM/DD/YYYY HH24:MI:SS')
and usage_time < TO_DATE('9/1/2015 00:00:00', 'MM/DD/YYYY HH24:MI:SS')
group by to_char(access_time, 'Month');

加入两个select语句并对齐它们

2 个答案: