Question

我正在尝试将我的数据规范化，因为它是从Excel工作表中输入的。我拉数据的文件有一堆列sibling1_name，sibling1_age，sibling1_affected等最多4个兄弟，4个孩子，4个亲戚等。我想把它全部输入到一个名字，年龄，受影响和关系的新表中。我找到了正确输入第一个兄弟姐妹的方法（见下文），但我不确定如何添加其他兄弟。有什么建议吗？

INSERT INTO Family
            (ID,
             Name,
             Age,
             Affected,
             Relationship)
SELECT ExcelPatients.id,
       ExcelPatients.sibling1_name     AS Name,
       ExcelPatients.sibling1_age      AS Age,
       ExcelPatients.sibling1_affected AS Affected,
       "Sibling"
FROM   ExcelPatients
WHERE  (( ( ExcelPatients.Sibling1_name ) IS NOT NULL ))
       AND ExcelPatients.id NOT IN (SELECT DISTINCT ID  AND Name
                                    FROM   Family);

Answer 1

INSERT INTO Family
            (ID,
             Name,
             Age,
             Affected,
             Relationship)

SELECoT ExcelPatients.id, ExcelPatients.sibling1_name AS Name, 
ExcelPatients.sibling1_age AS Age, 
ExcelPatients.sibling1_affected AS Affected, "Sibling"
FROM ExcelPatients
WHERE (((ExcelPatients.Sibling1_name) Is Not Null))
AND NOT EXISTS  (SELECT DISTINCT ID FROM Family where family.id =  ExcelPatients.id and Family.name =  ExcelPatients.sibling1_name)

UNION

SELECT ExcelPatients.id, ExcelPatients.sibling2_name AS Name, 
ExcelPatients.sibling2_age AS Age, 
ExcelPatients.sibling2_affected AS Affected, "Sibling"
FROM ExcelPatients
WHERE (((ExcelPatients.Sibling2_name) Is Not Null))
AND NOT EXISTS  (SELECT DISTINCT ID FROM Family where family.id =  ExcelPatients.id and Family.name =  ExcelPatients.sibling2_name)

UNION

SELECT ExcelPatients.id, ExcelPatients.sibling3_name AS Name, 
ExcelPatients.sibling3_age AS Age, 
ExcelPatients.sibling3_affected AS Affected, "Sibling"
FROM ExcelPatients
WHERE (((ExcelPatients.Sibling3_name) Is Not Null))
AND NOT EXISTS  (SELECT DISTINCT ID FROM Family where family.id =  ExcelPatients.id and Family.name =  ExcelPatients.sibling3_name)

UNION 

SELECT ExcelPatients.id, ExcelPatients.sibling4_name AS Name, 
ExcelPatients.sibling4_age AS Age, 
ExcelPatients.sibling4_affected AS Affected, "Sibling"
FROM ExcelPatients
WHERE (((ExcelPatients.Sibling4_name) Is Not Null))
AND NOT EXISTS  (SELECT DISTINCT ID FROM Family where family.id =  ExcelPatients.id and Family.name =  ExcelPatients.sibling4_name)

如果没有看到数据，我不知道UNION ALL或UNION是否是正确的选择。如果名称只能在4个兄弟列中的一个中，则使用UNION ALL，如果可以重复，则使用UNION UNION。由于您正在清理来自其他来源的数据，因此UNION可能更安全但选择更慢。 NOT EXISTS往往是SQL Server中最快的比较，这就是我选择它的原因。

组合多个SELECT函数，其中两个都具有WHERE

1 个答案: