详解SQL中FOR XML子句的各种用法

FOR XML子句有四种最基本的模式，如上图所示：

1、AUTO模式：返回数据表为起表名的元素，每一列的值返回为属性；

2、RAW模式：返回数据行为元素，每一列的值作为元素的属性；

3、PATH模式：通过简单的XPath语法来允许用户自定义嵌套的XML结构、元素、属性值

4、EXPLICIT模式：通过SELECT语法定义输出XML的结构

具体实例如下：

1、AUTO模式

(1). SQL语句：

   1:  SELECT EmployeeID,FirstName,LastName FROM Employees FOR XML AUTO, XMLSCHEMA

(2). 所生成的XML文件：

返回XML文件的XML Schema

<xsd:schema targetNamespace="urn:schemas-microsoft-com:sql:SqlRowSet1" xmlns:schema="urn:schemas-microsoft-com:sql:SqlRowSet1" xmlns:xsd="http://www.w3.org/2001/XMLSchema" xmlns:sqltypes="http://schemas.microsoft.com/sqlserver/2004/sqltypes" elementFormDefault="qualified">
<xsd:import namespace="http://schemas.microsoft.com/sqlserver/2004/sqltypes" schemaLocation="http://schemas.microsoft.com/sqlserver/2004/sqltypes/sqltypes.xsd" />
<xsd:element name="Employees">
<xsd:complexType>
<xsd:attribute name="EmployeeID" type="sqltypes:int" use="required" />
<xsd:attribute name="FirstName" use="required">
<xsd:simpleType>
<xsd:restriction base="sqltypes:nvarchar" sqltypes:localeId="1033" sqltypes:sqlCompareOptions="IgnoreCase IgnoreKanaType IgnoreWidth" sqltypes:sqlSortId="52">
<xsd:maxLength value="10" />
xsd:restriction>
xsd:simpleType>
xsd:attribute>
<xsd:attribute name="LastName" use="required">
<xsd:simpleType>
<xsd:restriction base="sqltypes:nvarchar" sqltypes:localeId="1033" sqltypes:sqlCompareOptions="IgnoreCase IgnoreKanaType IgnoreWidth" sqltypes:sqlSortId="52">
<xsd:maxLength value="20" />
xsd:restriction>
xsd:simpleType>
xsd:attribute>
xsd:complexType>
xsd:element>
xsd:schema>
<Employees xmlns="urn:schemas-microsoft-com:sql:SqlRowSet1" EmployeeID="1" FirstName="Nancy" LastName="Davolio" />
<Employees xmlns="urn:schemas-microsoft-com:sql:SqlRowSet1" EmployeeID="2" FirstName="Andrew" LastName="Fuller" />
<Employees xmlns="urn:schemas-microsoft-com:sql:SqlRowSet1" EmployeeID="3" FirstName="Janet" LastName="Leverling" />
<Employees xmlns="urn:schemas-microsoft-com:sql:SqlRowSet1" EmployeeID="4" FirstName="Margaret" LastName="Peacock" />
<Employees xmlns="urn:schemas-microsoft-com:sql:SqlRowSet1" EmployeeID="5" FirstName="Steven" LastName="Buchanan" />
<Employees xmlns="urn:schemas-microsoft-com:sql:SqlRowSet1" EmployeeID="6" FirstName="Michael" LastName="Suyama" />
<Employees xmlns="urn:schemas-microsoft-com:sql:SqlRowSet1" EmployeeID="7" FirstName="Robert" LastName="King" />
<Employees xmlns="urn:schemas-microsoft-com:sql:SqlRowSet1" EmployeeID="8" FirstName="Laura" LastName="Callahan" />
<Employees xmlns="urn:schemas-microsoft-com:sql:SqlRowSet1" EmployeeID="9" FirstName="Anne" LastName="Dodsworth" />

2、RAW模式

1:  ----将元素命名为自定义的名称Employee
2:  SELECT EmployeeID,FirstName,LastName FROM Employees FOR XML RAW ('Employee') 
(2). 所生成的XML文件：   
1:  <Employee EmployeeID="1" FirstName="Nancy" LastName="Davolio" />   
2:  <Employee EmployeeID="2" FirstName="Andrew" LastName="Fuller" />   
3:  <Employee EmployeeID="3" FirstName="Janet" LastName="Leverling" />   
4:  <Employee EmployeeID="4" FirstName="Margaret" LastName="Peacock" />   
5:  <Employee EmployeeID="5" FirstName="Steven" LastName="Buchanan" />   
6:  <Employee EmployeeID="6" FirstName="Michael" LastName="Suyama" />   
7:  <Employee EmployeeID="7" FirstName="Robert" LastName="King" />   
8:  <Employee EmployeeID="8" FirstName="Laura" LastName="Callahan" />   
9:  <Employee EmployeeID="9" FirstName="Anne" LastName="Dodsworth" />

3、PATH模式：

(1). SQL语句：

SELECT EmployeeID "@ID",FirstName  "Name/FirstName",LastName "Name/LastName"

FROM Employees FOR XML PATH ('Employee')

(2). 所生成的XML文件

<Employee ID="1">
<Name>
<FirstName>NancyFirstName>
<LastName>DavolioLastName>
Name>
Employee>
<Employee ID="2">
<Name>
<FirstName>AndrewFirstName>
<LastName>FullerLastName>
Name>
Employee>
<Employee ID="3">
<Name>
<FirstName>JanetFirstName>
<LastName>LeverlingLastName>
Name>
Employee>
<Employee ID="4">
<Name>
<FirstName>MargaretFirstName>
<LastName>PeacockLastName>
Name>
Employee>
<Employee ID="5">
<Name>
<FirstName>StevenFirstName>
<LastName>BuchananLastName>
Name>
Employee>
<Employee ID="6">
<Name>
<FirstName>MichaelFirstName>
<LastName>SuyamaLastName>
Name>
Employee>
<Employee ID="7">
<Name>
<FirstName>RobertFirstName>
<LastName>KingLastName>
Name>
Employee>
<Employee ID="8">
<Name>
<FirstName>LauraFirstName>
<LastName>CallahanLastName>
Name>
Employee>

4、EXPLICIT模式

问题：加入要生成如下的XML文档该如何操作？

Nancy Davolio EXPLICIT模式解决这个问题的应用分为两个主要步骤

1.定义要输出的XML文档结构；

2.传入实际的数据值；

(1). SQL语句：

--定义输出XML文档的数据结构SELECT  1 AS Tag,
        NULL AS Parent,
        EmployeeID AS [Employee!1!EmpID],
        FirstName AS [Employee!1!FirstName!element],
        LastName AS [Employee!1!LastName!element]FROM
    Employees UNION ALL--传入实际的数据SELECT  1, NULL, 
       EmployeeID,        FirstName, 
       LastNameFROM    EmployeesORDER BY [Employee!1!EmpID],
[Employee!1!FirstName!element],        [Employee!1!LastName!element]
FOR     XML EXPLICIT

语句含义的解释：

先看看定义XML结构的语句输出结果：

Tag栏用来指定生成元素的嵌套水平；1表示嵌套水平为

Parent栏用来指定当前Tag的父级层次；Null值表示该元素为顶级元素；

EmployeeID AS [Employee!1!EmpID],

说明：当前元素或属性的父级元素的名称！元素的标签号！元素或属性的名称
FirstName AS [Employee!1!FirstName!element],

说明：当前元素或属性的父级元素的名称！元素的标签号！元素或属性的名称！指定值作为元素输出

5、为输出的XML文档添加根元素(Root element)

(1). SQL语句：

SELECT EmployeeID,FirstName,LastName FROM Employees FOR XML AUTO, ROOT('MyRoot')

(2). 所生成的XML文件

<MyRoot>
<Employees EmployeeID="1" FirstName="Nancy" LastName="Davolio" />
<Employees EmployeeID="2" FirstName="Andrew" LastName="Fuller" />
<Employees EmployeeID="3" FirstName="Janet" LastName="Leverling" />
<Employees EmployeeID="4" FirstName="Margaret" LastName="Peacock" />
<Employees EmployeeID="5" FirstName="Steven" LastName="Buchanan" />
<Employees EmployeeID="6" FirstName="Michael" LastName="Suyama" />
<Employees EmployeeID="7" FirstName="Robert" LastName="King" />
<Employees EmployeeID="8" FirstName="Laura" LastName="Callahan" />
<Employees EmployeeID="9" FirstName="Anne" LastName="Dodsworth" />
MyRoot>