Question

如何在子选择中简单地对多行进行硬编码？

我知道我可以做（How to select several hardcoded SQL rows?）：

SELECT x.id, SUM(ISNULL(OtherTable.count_column,0)) 
FROM (SELECT 12 AS id
   UNION
SELECT 21 AS id
   UNION
SELECT 101 AS id
/*AND so on */
) AS x
LEFT JOIN OtherTable ON x.id = OtherTable.id
Group BY x.id

这样做是否有一种不那么笨拙和冗长的方式？

我真正想要的是：

SELECT id, SUM(ISNULL(count_column,0)) FROM OtherTable
WHERE id IN (12, 21, 101,/*And So On*/)
GROUP BY id

在这种情况下，对于不存在的ID，它不包括0的总和。有没有办法包含未找到的ID？

我注意到PIVOT用于SQL Server，但我不确定这是否会使它变得更简单/更简洁。

我想我只是在问有没有更好的方法？

Answer 1

使用master..spt_values表格尝试这样的事情。

SELECT x.id, SUM(ISNULL(OtherTable.count_column,0)) 
FROM 
(
SELECT DISTINCT number AS id
FROM master..spt_values
WHERE number >= 1 and number <= 10
) AS x
LEFT JOIN OtherTable ON x.id = OtherTable.id
Group BY x.id

OR

SELECT id, SUM(ISNULL(count_column,0)) 
FROM OtherTable
WHERE id IN (
               SELECT DISTINCT number AS id
               FROM master..spt_values
               WHERE number >= 1 and number <= 10
             )
GROUP BY id

测试数据

DECLARE @OtherTable TABLE(ID INT, count_column INT)
INSERT INTO @OtherTable VALUES
(1, 10), (2,20),(3,30),(4,NULL),(5,50)

<强>查询

SELECT x.id, SUM(ISNULL(t.count_column,0)) Total_Sum
FROM 
(
SELECT DISTINCT number AS id
FROM master..spt_values
WHERE number >= 1 and number <= 7
) AS x
LEFT JOIN @OtherTable t ON x.id = t.id
Group BY x.id

结果集

╔════╦═══════════╗
║ id ║ Total_Sum ║
╠════╬═══════════╣
║  1 ║        10 ║
║  2 ║        20 ║
║  3 ║        30 ║
║  4 ║         0 ║
║  5 ║        50 ║
║  6 ║         0 ║
║  7 ║         0 ║
╚════╩═══════════╝

Answer 2

只需声明一个临时表，用于存储您在“IN Clause”中使用的ID。

DECLARE @temp TABLE(ID INT identity(1,1), yourIDs INT)
INSERT INTO @temp VALUES
(10),(20),(300),(400)

正确加入临时表，以针对您想要的所有ID检索count_column的总和

  select t.yourIDs, Sum(isnull(ot.count_column,0)) 
  from OtherTable ot  
  right JOIN  @temp t on t.yourIDs=ot.id
  group by t.yourIDs

使用subselect选择多个硬编码行

2 个答案: