2

First I have 4 Tables

Table0, Columns: num, desc
Table1, Columns: num, qty1
Table2, Columns: num, qty2
Table3, Columns: num, qty3
Table4, Columns: num, qty4

(not all num have values in qty1 or qty2 or qty3 or qty4, therefore I need a full join) and my query:

SELECT Table0.num, SUM(Table1.qty1 ), SUM(Table2.qty2 ), SUM(Table3.qty3 ), SUM(Table4.qty4)
FROM Table0
FULL OUTER JOIN Table1 ON Table0.num = Table1.num
FULL OUTER JOIN Table2 ON Table0.num = Table2.num
FULL OUTER JOIN Table3 ON Table0.num = Table3.num
FULL OUTER JOIN Table4 ON Table0.num = Table4.num
GROUP BY Table0.num

Somehow its returning just 1 row of data:

num | qty1 | qty2 | qty3 | qty4 |
---------------------------------
    | 100  | 20   |  77  |  969 |

But I was expecting like the example at

http://www.w3schools.com/sql/sql_join_full.asp

like:

num | qty1 | qty2 | qty3 | qty4 |
---------------------------------
1   |   0  |  2   |  3   |   2  |
2   |   1  |  0   |  0   |   0  |
3   |   7  |  0   |  9   |   0  |
4   |   0  |  0   |  0   |  10  |
5   |   0  |  0   |  7   |   0  |
6   |   8  |  2   |  9   |   3  |
7   |   0  |  1   |  0   |   0  |

(I don't know this solves it) However I got similar to the result the box above by changing all the tables to:

Table1, Columns: num, qty1, qty2, qty3, qty4
Table2, Columns: num, qty2, qty1, qty3, qty4
Table3, Columns: num, qty3, qty1, qty2, qty4 
Table4, Columns: num, qty4, qty1, qty2, qty3
4

3 回答 3

2

每个表中的 num 列之间没有匹配项，因此您将获得外部记录。当匹配键上没有匹配项时，记录显示为该列为空。

完全外连接的方式，Table0.num 需要存在于所有其他表中。即，如果 num == 1 仅在 Table0 和 Table1 中，但不在 Table2 和 3 中，则它不会在所有 4 上都匹配，因此是 null num。

你可能想要的是更像

SELECT Table0.num, 
     (Select SUM(Table1.qty1 ) From Table1 Where Table1.num = Table0.num) as one,
     (Select SUM(Table2.qty1 ) From Table2 Where Table2.num = Table0.num) as two,
     ...
From Table0

我的语法可能有点偏离，并且可能有更有效的方法。但一般的想法是你为每个关系做一个子查询，因为它们是独立的。

于 2013-07-15T19:56:07.170 回答
2

您需要做两件事之一（并且这两件事都假设Table0有所有实例num） -

  1. 如果'leaf'表（1-4）的所有行都已经求和，那么一个简单LEFT JOIN的（在选择中带有a COALESCE()）就足够了 - 你甚至不需要GROUP BY.

  2. 如果您需要对行求和，则需要在连接之前对它们求和，否则不同表中每个 num 的多行将导致结果相乘。

像这样的东西：

SELECT Table0.num, COALESCE(Table1.qty, 0), COALESCE(Table2.qty, 0), 
                   COALESCE(Table3.qty, 0), COALESCE(Table4.qty, 0)
FROM Table0
LEFT JOIN (SELECT num, SUM(qty1) as qty
           FROM Table1
           GROUP BY num) Table1
ON Table1.num = Table0.num
LEFT JOIN (SELECT num, SUM(qty2) as qty
           FROM Table2
           GROUP BY num) Table2
ON Table2.num = Table0.num
LEFT JOIN (SELECT num, SUM(qty3) as qty
           FROM Table3
           GROUP BY num) Table3
ON Table3.num = Table0.num
LEFT JOIN (SELECT num, SUM(qty4) as qty
           FROM Table4
           GROUP BY num) Table4
ON Table4.num = Table0.num

（工作SQLFiddle 示例）

于 2013-07-15T20:20:20.253 回答
-2

有一个变通的解决方案。您可以添加一个额外的表来加入您的表。例如：

Tablex; 列：（tablex_id (PK, AI) 新创建的表）

Table0; 列：（tablex_id, num, desc 添加了新字段“tablex_id”）

Table1; 列：（tablex_id, num, qty1 添加了新字段“tablex_id”）

Table2; 列：（tablex_id, num, qty2 添加了新字段“tablex_id”）

Table3; 列：（tablex_id, num, qty3 添加了新字段“tablex_id”）

Table4; 列：（tablex_id, num, qty4添加了新字段“tablex_id”）

每次将记录添加到Table0、Table1、和时Table2，都应首先将记录添加到并获取并相应地将其放入表中。Table3Table4Tablextablex_id

如果您想加入他们，只需执行以下操作：

select t0.desc, t1.qty, t2.qty, t3.qty, t4.qty
from Tablex tx
left join Table0 t0
  on tx.tablex_id = t0.tablex_id
left join Table0 t1
  on tx.tablex_id = t1.tablex_id
left join Table2 t2
  on tx.tablex_id = t2.tablex_id
left join Table3 t3
  on tx.tablex_id = t3.tablex_id
left join Table4 t4
 on tx.tablex_id = t4.tablex_id
于 2017-10-18T13:25:20.057 回答