Question

TERADATA数据库

当我进行左联接时，我将获得更多的返回行。我的易失性表中的基本人口为559,157行。而我想做的只是从基本人口收益中获得559,157行。但是运行此查询后：

SELECT distinct a.*, b.column1, b.column2, b.column3, b.column4, b.column5, b.column6, c.column3, SUM(c.column3) AS Total_Column3_Profit, AVG(c.column3) AS Column3_Profit_Average
,CASE 
WHEN b.column5 < a.column6 1
      ELSE 0
      END as column_open_flag
,CASE
WHEN a.column4 < 580 THEN 1
     WHEN a.column4 between 580 and 619 THEN 2
     WHEN a.column4 between 620 and 639 THEN 3
     WHEN a.column4 between 640 and 659 THEN 4
     WHEN a.column4 between 660 and 679 THEN 5
     WHEN a.column4 between 680 and 699 THEN 6
     WHEN a.column4 between 700 and 739 THEN 7
     WHEN a.column4 >= 740 THEN 8
     ELSE 0
     END as column4_band
FROM volatile_table1 a
LEFT JOIN database_table1 b
ON a.column1 = b.column0
LEFT JOIN volatile_table2 c
ON a.column1 = c.column2
GROUP BY a.column1
,a.column2
,a.column3
,a.column4
,a.column5
,a.column6
,b.column1
,b.column2
,b.column3
,b.column4
,b.column5
,b.column6
,c.column3

我现在最终返回了超过2,000,000行。所以我想尝试一个UNION子句。这是我更新的查询：

SELECT a.column1, a.column2, a.column3, a.column4, a.column5, a.column6
,CASE 
WHEN b.column5 < a.column6 1
      ELSE 0
      END as column_open_flag,
CASE
WHEN a.column4 < 580 THEN 1
     WHEN a.column4 between 580 and 619 THEN 2
     WHEN a.column4 between 620 and 639 THEN 3
     WHEN a.column4 between 640 and 659 THEN 4
     WHEN a.column4 between 660 and 679 THEN 5
     WHEN a.column4 between 680 and 699 THEN 6
     WHEN a.column4 between 700 and 739 THEN 7
     WHEN a.column4 >= 740 THEN 8
     ELSE 0
     END as column4_band
FROM volatile_table1 a
UNION
SELECT b.column1, b.column2, b.column3, b.column4, b.column5, b.column6
FROM database_table1 b
UNION
SELECT c.column3, SUM(c.column3) AS Total_Column3_Profit, AVG(c.column3) AS Column3_Profit_Average
FROM volatile_table2 c
GROUP BY a.column1
,a.column2
,a.column3
,a.column4
,a.column5
,a.column6
,b.column1
,b.column2
,b.column3
,b.column4
,b.column5
,b.column6
,c.column3

现在，我不断收到3807错误：对象'a'不存在。我跑了

HELP volatile table

查询，它说两个易失性表确实存在。有人可以指出我的错误方向吗？（我还尝试删除了引用b.column5

编辑：

我已从注释中更新到此内容，现在出现错误3653。'所有选择列表不包含相同数量的表达式：

SELECT distinct a.column1, a.column2, a.column3, a.column4, a.column5, a.column6, b.column1, b.column2, b.column3, b.column4, b.column5, b.column6, c.column3, SUM(column3) AS Total_Profit, AVG(column3) AS Profit_Averag
,CASE 
WHEN b.column5 < a.column6 1
      ELSE 0
      END as column_open_flag
,CASE
WHEN a.column4 < 580 THEN 1
     WHEN a.column4 between 580 and 619 THEN 2
     WHEN a.column4 between 620 and 639 THEN 3
     WHEN a.column4 between 640 and 659 THEN 4
     WHEN a.column4 between 660 and 679 THEN 5
     WHEN a.column4 between 680 and 699 THEN 6
     WHEN a.column4 between 700 and 739 THEN 7
     WHEN a.column4 >= 740 THEN 8
     ELSE 0
     END as column4_band
FROM 
(
SELECT a.column1, a.column2, a.column3, a.column4, a.column5, a.column6
FROM volatile_table1 a
UNION ALL
SELECT b.column1, b.column2, b.column3, b.column4, b.column5, b.column6
FROM database_table1 b
UNION ALL
SELECT c.column3, SUM(c.column3) AS Total_Column3_Profit, AVG(c.column3) AS Column3_Profit_Average
FROM volatile_table2 c
GROUP BY 1
) d
GROUP BY a.column1
,a.column2
,a.column3
,a.column4
,a.column5
,a.column6
,b.column1
,b.column2
,b.column3
,b.column4
,b.column5
,b.column6
,c.column3

Answer 1

如果要在全部合并后进行汇总，请使用子查询或CTE。我认为查询看起来像这样：

select . . .
from (select column1, column2, column3, column4, column5, . . .
      union all
      select column1, column2, column3, column4, column5, . . .
      union all
      select column1, column2, column3, column4, column5, . . .
     ) abc
group by GROUP BY a_column1, column2, column3, column4, column5

尝试运行查询时，teradata中出现错误3807

1 个答案: