SQL Server 2005与XML的紧密整合

 

FOR XML子句

增强                                                   说明
RAW模式下ELEMENTS                 RAW查询可以返回以元素为中心的XML结果
NULL值支持                                    支持null值，可以在一元素为中心的结果中包含空值元素
Inline XSD schemas                      可以生成inline XSD架构
TPYE指明返回xml数据类型值       对于FOR XML查询，可以返回xml数据类型的值，使XML嵌套查询成为可能
PATH模式                                         可以象XPath表达式一样定义XML结果
ROOT标识                                        为结果集指定根元素
Elements命名                                  为RAW和PATH模式查询指定命名元素

FOR XML子句范例

1、RAW模式下ELEMENTS
SELECT SaleID,ProductID,Quantity
FROM Sales
FOR XML RAW,ELEMENTS

2、NULL值支持
SELECT SaleID,ProductID,Quantity
FROM Sales
FOR XML RAW,ELEMENTS XSINIL

3、Inline XSD schemas
SELECT ProductID,Name,ListPrice
FROM Production.Product Product
FOR XML AUTO,XMLSCHEMA

4、TPYE指明返回xml数据类型值
SELECT ProductID,Name,ListPrice
(SELECT saleid,ProductID,Quantity
FROM Sales
WHERE Sales.ProductID-Products.ProductID
FOR XML RAW,TYPE)
FROM Products FOR XML AUTO

5、PATH模式
SELECT ProductID AS "@ProductID",
Name AS "Details/@Name",
Description AS "Details/text()"
FROM products FOR XML PATH

6、ROOT标识
SELECT ProductID,Name,ListPrice
FROM Products FOR XML AUTO,ROOT('PRODUCTS')

7、Elements命名

OPENXML函数

增强                                                       描述
文档可以是xml数据类型值                sp_xml_preparedocument存储过程支持xml参数
WITH子句支持XML数据类型             在WITH子句中，XML类型数据能够被返回
Batch-level scoping                           文档handle在批级有效，当查询批结束后，文档handle也被释放

OPENXML函数范例

declare @mydoc xml
set @mydoc='
<Products>
  <Product Category="Book">
    <ID>1</ID>
    <Name>Windows 2003</Name>
    <Vendor>Vendor1</Vendor>
  </Product>
  <Product Category="Book">
    <ID>2</ID>
    <Name>VS.NET2003</Name>
    <Vendor>Vendor2</Vendor>
  </Product>
</Products>'

declare @docHandle int
Exec sp_xml_preparedocument @docHandle OUTPUT,@mydoc

SELECT * FROM OPENXML(@docHandle,'/Products/Product',2)
WITH (ID int,Name nvarchar(50),Vendor nvarchar(50))
--------------------------
declare @mydoc xml
set @mydoc='
<Products>
  <Product Category="Book" ID="1" Name="Windows 2003" />
  <Product Category="Book" ID="2" Name="VS.NET  2003" />
</Products>'

declare @docHandle int
Exec sp_xml_preparedocument @docHandle OUTPUT,@mydoc

SELECT * FROM OPENXML(@docHandle,'/Products/Product',1)
WITH (Category nvarchar(50),ID int,Name nvarchar(50))

在数据库中存放XML

优点：
对结构化和非结构化数据实现单一存储
在关系模式中定义可变内容
选择最适合的数据类型

功能：
XML Indexes
基于XQuery的数据检索
基于XQuery的数据修改

XML架构支持：
Typed XML需要架构验证
UnTyped XML需要架构验证

怎样使用Untyped XML

声明xml数据类型
隐式转换字符串
显示转换字符值
使用Convert显示转换字符串
使用well-formed XML

Untyped XML范例

1、声明xml数据类型
CREATE TABLE Invoices
(
 InvoiceID INT,
 SalesDate DateTime,
 CustomerID INT,
 ItemList XML
)
DECLARE @itemDoc xml

2、隐式转换字符串值
SET @itemDoc = '<Items>etc.</Items>'

3、显示转换字符串
SET @itemDoc = CAST('<Items>etc.</Items>') AS XML

4、显示CONVERT显示转换字符串
SET @itemDoc = CONVERT(xml,'<Items>etc.</Items>')

5、使用well-formed XML
SET @itemDoc = CONVERT(XML,'<items>etc.')
ERROR!

怎样管理XML架构

1、建立XML架构集合
CREATE XML SCHEMA COLLECTION SalesSchema
AS
'<?xml version="1.0" standalone="yes"?>
<xs:schema id="Sales" xmlns="http://www.gocean.com.cn" xmlns:xs="http://www.w3.org/2001/XMLSchema" xmlns:msdata="urn:schemas-microsoft-com:xml-msdata">
  <xs:element name="Sales" msdata:IsDataSet="true" msdata:Locale="zh-CN">
    <xs:complexType>
      <xs:choice minOccurs="0" maxOccurs="unbounded">
        <xs:element name="Product">
          <xs:complexType>
            <xs:sequence>
              <xs:element name="ID" type="xs:int" minOccurs="0" />
              <xs:element name="Name" type="xs:string" minOccurs="0" />
              <xs:element name="Qty" type="xs:int" minOccurs="0" />
            </xs:sequence>
          </xs:complexType>
        </xs:element>
      </xs:choice>
    </xs:complexType>
  </xs:element>
</xs:schema>
'

2、查看schema信息
SELECT * FROM sys.xml_schema_collections
SELECT * FROM sys.xml_namespaces

3、修改schema集合
ALTER XML SCHMEMA COLLECTION cvSchemas

4、删除schema集合
DROP XML SCHMEMA COLLECTION cvSchemas

怎样使用Typed XML

1、声明typed列或变量
CREATE TABLE HumanResources.EmployeeResume
(
 Emplyee INT,
 Resume XML (cvSchemas)
)

2、给typed XML赋值
INSERT INTO HumanResources.EmployeeResume
VALUES(1,'<?xml version="1.0" ?>
 <resume xmlns="http://cvSchemas">
 ...</resume>'

3、使用CONTENT或DOCUMET允许/禁止插入片段
CREATE TABLE Orders
(OrderID int IDENTITY(1,1),
 CustomerID int,
 OrderDetail xml (SalesSchema))

insert orders values(1,'<Sales><Product><ID>1</ID><Name>p1</Name><Qty>100</Qty></Product></Sales>')
--------------------------------
CREATE TABLE Orders
(OrderID int IDENTITY(1,1),
 CustomerID int,
 OrderDetail xml (DOCUMENT SalesSchema))

insert orders values(1,'<Sales><Product><ID>1</ID><Name>p1</Name><Qty>100</Qty></Product></Sales>
<Sales><Product><ID>1</ID><Name>p1</Name><Qty>100</Qty></Product></Sales>')

管理XML Indexes

1 建立主 XML index
alter table orders
add constraint pk_orders_orderid
primary key clustered(orderid)

CREATE PRIMARY XML INDEX xidx_item
ON Sales.Invoices(ItemList)

CREATE PRIMARY XML INDEX xidx_details
ON orders(details)

2 建立辅助 PATH XML index
CREATE XML INDEX xidx_ItemPath
ON Sales.Invoices(ItemList)
USING XML INDEX xidx_Item FOR PATH

CREATE XML PATH xidx_details_path
ON orders(details)
USING XML INDEX xidx_details FOR PATH

3 建立辅助 PROPERTY XML index
CREATE XML INDEX xidx_ItemProp
ON Sales.Invoices(ItemList)
USING XML INDEX xidx_Item FOR PROPERTY

CREATE XML INDEX xidx_details_property
ON orders(details)
USING XML INDEX xidx_details FOR PROPERTY

4 建立辅助 VALUE XML Index
CREATE XML INDEX xidx_ItemVal
ON Slaes.Invoices(ItemList)
USING XML INDEX xidx_Item FOR VALUE

CREATE XML INDEX xidx_details_value
ON orders(details)
USING XML INDEX xidx_details FOR VALUE

使用 XQuery 

1 什么是 XQuery
XQuery 是查询XML数据的语言
/InvoiceList/Invoice[@InvoiceNo=1000]

FLOWER 语句(for,let, order by, where,return)
语句      说明
for         循环通过同属节点
where   应用筛选标准
return   指定xml返回值

使用XQuery表达式 - 演示

declare @mydoc xml
set @mydoc='
<AAA>
  <BBB/>
  <BBB/>
  <CCC>
    <DDD/>
    <BBB>
       <EEE/>
    </BBB>
   </CCC>
</AAA>'

select @mydoc.query('//BBB')
select @mydoc.query('//BBB[1]')
select @mydoc.query('/AAA/BBB[1]')
select @mydoc.query('/AAA/BBB[last()]')

declare @mydoc xml
set @mydoc='
<AAA>
  <BBB ID="1"/>
  <BBB ID="2"/>
  <CCC>
    <DDD/>
    <BBB ID="3">
       <EEE/>
    </BBB>
   </CCC>
</AAA>'

select @mydoc.query('/AAA/BBB[@ID="1"]')

select @myDoc.query('/bookstore/book/title')

查询条件可以是attribute, 也可以是element，如下是element示例
select @myDoc.query('/bookstore/book[price>30]')

declare @myDoc xml
set @myDoc = '
<AAA>
  <BBB>HELLO</BBB>
  <BBB>Welcome</BBB>
  <BBB NAME="NAME1"/>
  <CCC ID="1">
    <DDD/>
    <BBB>OK
      <EEE ID="1"/>
    </BBB>
    <BBB/>
   </CCC>
</AAA>
'
select @myDoc.query('
for $id in //BBB
return <result>{data($id)}</result>')

使用XML数据类型的方法 

1 Use the query method
SELECT xmlCol.Query(
'<InvoiceNumbers>
{
 for $i in .InvoiceList.Invoice
 return <InvoiceNo>
   {number($i/@InvoiceNum)}
     </InvoiceNuo>
}
</InvoiceNumbers>'

select @myDoc.query('
for $id in //BBB
return <result>{data($id)}</result>')

2 Use the value method
SELECT xmlCol.value(
'(/InvoiceList/Invoice/@InvoiceNo)[1]','int')

3 Use the exist method
SELECT xmlCol.exist(
'/InvoiceList/Invoice[@InvoiceNo=1000]'
)

4 Bind relational columns and variables
SELECT Invoices.query(
'<Store>
 {sql:column("StoreName")}
</Store>'

使用 Modify 方法修改 XML

1 Use the insert statement
SET @xmlDoc.modify(
'insert element salesperson{"Bill"}
as first
into (/InvoiceList/Invoice)[1]')
------------------------------------------
INSERT
declare @doc xml
set @doc='<Products></Products>'

set @doc.modify(
'insert (<Product><ID>L01</ID><Name>LL01</Name></Product>)
into (/Products)[1]')

set @doc.modify(
'insert (<Product><ID>L02</ID><Name>LL02</Name></Product>)
as first into (/Products)[1]')

set @doc.modify(
'insert (<Product><ID>L03</ID><Name>LL03</Name></Product>)
as last into (/Products)[1]')

set @doc.modify(
'insert attribute Price {"20.50"} into (/Products/Product)[1]')

select @doc
-------------------------------------------

2 Use the replace statement
SET @xmlDoc.modify(
'replace value of
(/InvoiceList/Invoice/SalesPerson/text())[1]
with "Ted"')
-------------------------------------------
set @mydoc.modify('replace value of (/bookstore/book/price/text())[1] with "99.50"')

set @mydoc.modify('replace value of (/bookstore/book/@id)[1] with "10"')

set @mydoc.modify('
replace value of (/bookstore/book/@id)[1]
with(
if(/bookstore/book[@id="1"]) then
  "10"
else
  "100"
)
')
------------------------------------------

3 Use the delete statement
SET @xmlDoc.modify(
'delete
(/invoiceList/Invoice/SalesPerson)[1]')
-----------------------------------------
declare @myDoc xml
set @myDoc = '
<bookstore>
  <book category="COOKING" id="1" >
    <title>Everyday</title>
    <author>Giade De</author>
    <price>30.00</price>
  </book>
  <book category="COMPUTER" id="2" >
    <title>Windows 2003</title>
    <author>Mike</author>
    <price>50.00</price>
  </book>
  <book category="SOFTWARE" id="3" >
    <title>VS.NET2003</title>
    <author>Mike</author>
    <price>90.00</price>
  </book>
</bookstore>
'
set @mydoc.modify('delete (/bookstore/book[@id="1"])')
set @mydoc.modify('delete (/bookstore/book[@id="1"])[1]')
set @mydoc.modify('delete (/bookstore/book/price)[1]')
set @mydoc.modify('delete (/bookstore/book/price/text())[1]')
----------------------------------------------------------

使用nodes方法转换XML输出

1 使用query, value和exist方法带xml变量
SELECT nCol.value('@ProductID','int') Product,
  nCol.valus('@Quantity','int') Qty
FROM @xmlOrder.nodes('/Order/ListItem')
AS nTable(nCol)
----------------------------------------------------------
declare @myDoc xml
set @myDoc = '
<bookstore>
  <book category="COOKING">
    <title>Everyday</title>
    <author>Giade De</author>
    <price>30.00</price>
  </book>
  <book category="COMPUTER">
    <title>Windows 2003</title>
    <author>Mike</author>
    <price>50.00</price>
  </book>
  <book category="SOFTWARE">
    <title>VS.NET2003</title>
    <author>Mike</author>
    <price>90.00</price>
  </book>
</bookstore>
'
select @myDoc.query('/bookstore/book/title')
-----------------------------
select @myDoc.query('/bookstore/book[price>30]')
select @myDoc.query('for $x in /bookstore/book
where $x/price>30
return $x/title')

select @myDoc.query('for $x in /bookstore/book/title
order by $x
return $x')

select @myDoc.query('for $x in /bookstore/book/title
return <li>{data($x)}</li>')

select @myDoc.query('for $x in /bookstore/book/title
order by $x
return <li>{data($x)}</li>')
----------------------------------------------------------
declare @myDoc xml
set @myDoc = '
<bookstore>
  <book category="COOKING" id="1" >
    <title>Everyday</title>
    <author>Giade De</author>
    <price>30.00</price>
  </book>
  <book category="COMPUTER" id="2" >
    <title>Windows 2003</title>
    <author>Mike</author>
    <price>50.00</price>
  </book>
  <book category="SOFTWARE" id="3" >
    <title>VS.NET2003</title>
    <author>Mike</author>
    <price>90.00</price>
  </book>
</bookstore>
'
value查询

select @myDoc.value('(/bookstore/book/@id)[1]','int')

exist查询

select @myDoc.exist('/bookstore/book/title="VS.NET2003"')
select @myDoc.exist('/bookstore/book[@id=1]')
---------------------------------------------------------

结果集中绑定表中列

select orderid,'L01' as ProductID,Customer,
Details.query('
<OrderDetails>
<Customer>{sql:column("Customer")}</Customer>
    {
  for $x in //row
  return $x
}
</OrderDetails>
')
from orders

2 使用APPLY运算符
SELECT nCol.value('../@OrderID[1]','int') ID,
  nCol.valus('@ProductID[1]','int') Prod
FROM Sales.Orders
CROSS APPLY OrderDoc.nodes('/Order/ListItem')
AS nTable(nCol)