Question

我有一个查询，它接受个人员工的详细信息，然后根据位置和最后的总数来总结。在第一个选择语句中我需要采取max（column9），但在联合中我需要得到总和（ column9）。我收到错误“数据类型不匹配”。以下是查询：

SELECT COLUMN1 AS LASTNAME,
         COLUMN2 AS FIRSTNAME,
         COLUMN3 AS LOCATION,
         SUM(COLUMN4) AS ACTIVITYNM1,
         SUM(COLUMN5) AS ACTIVITYNM2,
         SUM(COLUMN6) AS ACTIVITYNM3,
         SUM(COLUMN7) AS ACTIVITYNM4,
         SUM(COLUMN8) AS ACTIVITYNM5,
         MAX(COLUMN9) AS REG_HRS,
         MAX(COLUMN10) AS OT_HRS,
         MAX(COLUMN11) AS TOTAL_HRS,
         SUM(COLUMN12) AS PRODUCTIVITY_PERCENTAGE,
         COLUMN13 AS FULL_LOCATION                     
    FROM TABLE(ACTY_COL(V_STARTDATE,V_ENDDATE))
GROUP BY COLUMN1, COLUMN2, COLUMN3, COLUMN13
  UNION     
  SELECT NULL,
         NULL,
         CONCAT(SUBSTR(COLUMN3,3,3),' Total') AS LOCATION,
         SUM(COLUMN4) AS ACTIVITYNM1,
         SUM(COLUMN5) AS ACTIVITYNM2,
         SUM(COLUMN6) AS ACTIVITYNM3,
         SUM(COLUMN7) AS ACTI,
         VITYNM4,
         SUM(COLUMN8) AS ACTIVITYNM5,
         SUM(COLUMN9) AS REG_HRS,
         SUM(COLUMN10) AS OT_HRS,
         SUM(COLUMN11) AS TOTAL_HRS,
         (SUM(COLUMN12)/COUNT(DISTINCT(COLUMN1))) AS PRODUCTIVITY_PERCENTAGE,
         COLUMN13 AS FULL_LOCATION
    FROM TABLE(ACTY_COL(V_STARTDATE,V_ENDDATE))
GROUP BY COLUMN3, COLUMN13

=============================================== ========================================

嗨，非常感谢您的所有答案。我得到了解决方案。我在第二个联合中使用子查询。我在这里粘贴代码。

SELECT 
   COLUMN1 AS EMPID,
   COLUMN2 AS FIRSTNAME,
   COLUMN3 AS LASTNAME,
       COLUMN4 AS LOCATION,
       SUM(COLUMN5) AS ACTIVITYNM1,
       SUM(COLUMN6) AS ACTIVITYNM2,
       SUM(COLUMN7) AS ACTIVITYNM3,
       SUM(COLUMN8) AS ACTIVITYNM4,
       SUM(COLUMN9) AS ACTIVITYNM5,
        MIN(COLUMN10) AS EVENTDATE,
       TO_NUMBER(COLUMN11) AS REG_HRS,
       TO_NUMBER(COLUMN12) AS OT_HRS,
       TO_NUMBER(COLUMN13) AS TOTAL_HRS,
       SUM(COLUMN14) AS PRODUCTIVITY_PERCENTAGE

FROM 
   TABLE(ACTY_COL(V_STARTDATE,V_ENDDATE))
GROUP BY 
   COLUMN1,
   COLUMN2,
   COLUMN3,
   COLUMN4,
   COLUMN11,
   COLUMN12,
   COLUMN13
UNION 

SELECT
   NULL AS EMPID,
   NULL AS LASTNAME,
   NULL AS FIRSTNAME,
       SUBSTR(INNER_REC.LOCATION,2,5) AS LOCATION,
       SUM(INNER_REC.ACTIVITYNM1) AS ACTIVITYNM1,
       SUM(INNER_REC.ACTIVITYNM2) AS ACTIVITYNM2,
       SUM(INNER_REC.ACTIVITYNM3) AS ACTIVITYNM3,
       SUM(INNER_REC.ACTIVITYNM4) AS ACTIVITYNM4,
       SUM(INNER_REC.ACTIVITYNM5) AS ACTIVITYNM5,
       MIN(INNER_REC.EVENTDATE) AS EVENTDATE,
       SUM(INNER_REC.REG_HRS) AS REG_HRS,
       SUM(INNER_REC.OT_HRS) AS OT_HRS,
       SUM(INNER_REC.TOTAL_HRS) AS TOTAL_HRS,
       SUM(INNER_REC.PRODUCTIVITY_PERCENTAGE) AS PRODUCTIVITY_PERCENTAGE
  FROM
  (
SELECT 
   COLUMN1 AS EMPID,
   NULL AS LASTNAME,
   NULL AS FIRSTNAME,
       COLUMN4 AS LOCATION,
       SUM(COLUMN5) AS ACTIVITYNM1,
       SUM(COLUMN6) AS ACTIVITYNM2,
       SUM(COLUMN7) AS ACTIVITYNM3,
       SUM(COLUMN8) AS ACTIVITYNM4,
       SUM(COLUMN9) AS ACTIVITYNM5,
       MIN(COLUMN10) AS EVENTDATE,
      SUM(TO_NUMBER(COLUMN11))/(SELECT COUNT(1) FROM TABLE(ACTY_COL(V_STARTDATE,V_ENDDATE)) AB WHERE AB.COLUMN4 = 

CD.COLUMN4 AND AB.COLUMN1 = CD.COLUMN1) AS REG_HRS,
       SUM(TO_NUMBER(COLUMN12))/(SELECT COUNT(1) FROM TABLE(ACTY_COL(V_STARTDATE,V_ENDDATE)) AB WHERE AB.COLUMN4 = 

CD.COLUMN4 AND AB.COLUMN1 = CD.COLUMN1) AS OT_HRS,
       SUM(TO_NUMBER(COLUMN13))/(SELECT COUNT(1) FROM TABLE(ACTY_COL(V_STARTDATE,V_ENDDATE)) AB WHERE AB.COLUMN4 = 

CD.COLUMN4 AND AB.COLUMN1 = CD.COLUMN1) AS TOTAL_HRS,
       SUM(COLUMN14) AS PRODUCTIVITY_PERCENTAGE      
FROM 
   TABLE(ACTY_COL(V_STARTDATE,V_ENDDATE)) CD
GROUP BY 
   COLUMN4,
   COLUMN1
 ) INNER_REC
 GROUP BY
    INNER_REC.LOCATION

Answer 1

在第二个查询中尝试使用to_char（null）而不是null。另外，如果两个结果集是互斥的，或者你不关心重复，那么我建议你使用UNION ALL而不仅仅是UNION（如果不需要，不需要告诉Oracle执行删除重复的额外步骤。）

Answer 2

你能把每个人都变成同一类型吗？

编辑：如果您使用9i或更高版本：

http://download.oracle.com/docs/cd/B19306_01/server.102/b14200/functions016.htm

自从我使用Oracle以来已经有一段时间了，但我最近在SQL Server中遇到了这个问题，其中第一组UNION查询中的null导致第二组中相应列的古怪格式化。 SQL Server的答案是CAST（colFromFirstSet为myDesiredDataType），以便引擎知道如何处理第二组中的相应列。我假设Oracle中类似的显式转换可能会解决不匹配的数据类型错误。

Answer 3

我猜测COLUMN9是一个字符数据类型。 SUM可能正在隐式转换为数值，但MAX将其视为VARCHAR2（即'9'将大于'10'）。

所以我怀疑你想在第一个中使用MAX（TO_NUMBER（COLUMN9）），或者更改ACTY_COL的定义，以便将COLUMN9作为数字返回。

plsql中union中相同列的不同数据类型

3 个答案: