Question

说我有3张这样的表：

Table 1

|Id|Date |Data1|
|--|---- |-----|
|1 |24/05|Some1|

Table 2

|Id|Date |Data2|
|--|---- |-----|
|1 |24/05|Some2|
|1 |12/06|Some2|

Table 3

|Id|Date |Data3|
|--|---- |-----|
|1 |24/05|Some3|
|1 |14/06|Some3|

我如何得到如此结果：

|Id|Date |Data1|Data2|Data3|
|--|-----|-----|-----|-----|
|1 |24/05|Some1|Some2|Some3|
|1 |12/06|NULL |Some2|NULL |
|1 |14/06|NULL |NULL |Some3|

使用MS SQL。

我想出的是

SELECT Id, Date, Data1, Data2, Data3 FROM Table1 
LEFT JOIN Table2 ON Table1.Id = Table2.Id AND Table1.Date = Table2.Date
LEFT JOIN Table3 ON Table1.Id = Table3.Id AND Table1.Date = Table3.Date
UNION 
SELECT Id, Date, Data1, Data2, Data3 FROM Table2
LEFT JOIN Table1 ON Table2.Id = Table1.Id AND Table2.Date = Table1.Date
LEFT JOIN Table3 ON Table2.Id = Table3.Id AND Table2.Date = Table3.Date
UNION 
SELECT Id, Date, Data1, Data2, Data3 FROM Table3
LEFT JOIN Table1 ON Table3.Id = Table1.Id AND Table3.Date = Table3.Date
LEFT JOIN Table2 ON Table3.Id = Table2.Id AND Table3.Date = Table2.Date

有没有更好的方法来查询数据？

Answer 1

快速方法是对临时表中的所有日期和ID进行排序，然后对其进行数据处理。

Select T.id,T.date, T1.Data1,T2.Data2,T3.Data3
FROM
    (
        select id,date from Table1
        UNION 
        select id,date from Table2
        UNION 
        select id,date from Table3
    ) T
LEFT JOIN Table1 T1 on T.id=T1.id and T.date=T1.date
LEFT JOIN Table2 T2 on T.id=T2.id and T.date=T2.date
LEFT JOIN Table3 T3 on T.id=T3.id and T.date=T3.date
ORDER BY T.id,T.date

JOIN数量=唯一表数

Answer 2

select 
    id       = coalesce(t1.id,t2.id,t3.id) 
  , [date]   = coalesce(t1.[date],t2.[date],t3.[date]) 
  , t1.Data1
  , t2.Data2
  , t3.Data3
from Table1 t1
  full join Table2 t2
    on t1.id = t2.id 
   and t1.[date]=t2.[date]
  full join Table3 t3
    on (t1.id = t3.id and t1.[date]=t3.[date])
    or (t2.id = t3.id and t2.[date]=t3.[date])

rextester演示：http://rextester.com/ATCJ12705

返回：

+----+-------+-------+-------+-------+
| id | date  | Data1 | Data2 | Data3 |
+----+-------+-------+-------+-------+
|  1 | 24/05 | Some1 | Some2 | Some3 |
|  1 | 12/06 | NULL  | Some2 | NULL  |
|  1 | 14/06 | NULL  | NULL  | Some3 |
+----+-------+-------+-------+-------+

Answer 3

假设每个表的每个id和日期只有1行，将空值放在UNION中并选择MAX值可以达到相同的效果：

create table #t1 (id int, [date] nvarchar(5), data1 nvarchar(10))
create table #t2 (id int, [date] nvarchar(5), data2 nvarchar(10))
create table #t3 (id int, [date] nvarchar(5), data3 nvarchar(10))

insert into #t1 values (1, '24/05', 'Some1')
insert into #t2 values (1, '24/05', 'Some2'),  (1, '12/06', 'Some2')
insert into #t3 values  (1, '24/05', 'Some3'),  (1, '14/06', 'Some3')

select a.id, a.[date], max(a.data1) as [Data1], max(a.data2) as [Data2], max(a.data3) as [Data3]
from 
( 
select id, [date], [data1] as [Data1], Null as [Data2], null as [Data3] from #t1
union
select id, [date], null as [Data1], [data2] as [Data2], null as [Data3] from #t2
union
select id, [date], null as [Data1], Null as [Data2], [Data3] as [Data3] from #t3
) a
group by a.id, a.[date]

查询3个表中所有数据的最快方法

3 个答案: