XmlWriter 包括方法 WriteRaw，使您可以手动写出原始标记。 该方法禁止转义特殊字符。 这与 WriteString 方法相反，后者将某些字符串转义为等效的实体引用。 被转义的字符可从以下位置找到：XML 1.0 建议的第 2.4 节“Character Data and Markup”（字符数据和标记），以及可扩展标记语言 (XML) 1.0（Fifth Edition [第五版]）建议的第 3.3.3 节“Attribute-Value Normalization”（属性值的标准化）。 如果在编写属性值时调用 WriteString 方法，将转义 ' 和 "。 字符值 0x-0x1F 通过  编码为数字字符实体 �，空白字符 0x9、0x10 和 0x13 除外。
因此，何时使用 WriteString 或 WritingRaw 的指导原则是：在需要遍历每个字符以查找实体字符时使用 WriteString，而 WriteRaw 原样编写为其提供的内容。
WriteNode 方法复制当前节点中的所有内容，并将读取器放在写入器的位置。 然后，读取器前进到下一个同级节点进行进一步处理。 writeNode 方法可以快速地将一个文档中的信息准确地提取到另一个文档中。
下表显示 WriteNode 方法支持的 NodeTypes。
节点类型 |
说明 |
|---|---|
Element |
写出元素节点以及所有属性节点。 |
Attribute |
无操作。 使用 WriteStartAttribute 或 WriteAttributeString 来编写属性。 |
Text |
写出文本节点。 |
CDATA |
写出 CDATA 节节点。 |
EntityReference |
写出实体引用节点。 |
ProcessingInstruction |
写出 PI 节点。 |
Comment |
写出注释节点。 |
DocumentType |
写出 DocType 节点。 |
Whitespace |
写出空白节点。 |
SignificantWhitespace |
写出空白节点。 |
EndElement |
无操作。 |
EndEntity |
无操作。 |
下面的示例显示当提供“<”字符时，WriteString 与 WriteRaw 方法之间的差异。 此代码示例使用 WriteString。
w.WriteStartElement("myRoot")
w.WriteString("<")
w.WriteEndElement()
Dim tw As New XmlTextWriter(Console.Out)
tw.WriteDocType(name, pubid, sysid, subset)
w.WriteStartElement("myRoot");
w.WriteString("<");
w.WriteEndElement();
XmlTextWriter tw = new XmlTextWriter(Console.Out);
tw.WriteDocType(name, pubid, sysid, subset);
输出
<myRoot><</myRoot>
此代码示例使用 WriteRaw，输出将非法字符作为元素内容。
w.WriteStartElement("myRoot")
w.WriteRaw("<")
w.WriteEndElement()
w.WriteStartElement("myRoot");
w.WriteRaw("<");
w.WriteEndElement();
输出
<myRoot><</myRoot>
下面的示例显示如何将 XML 文档从以元素为中心的文档转换为以属性为中心的文档。 还可以将 XML 从以属性为中心的文档转换回以元素为中心的文档。 以元素为中心的模式意味着 XML 文档被设计为具有许多元素但具有很少属性。 以属性为中心的设计具有较少的元素，那些在以元素为中心的设计中应该是元素的内容变成了元素的属性。 因此元素较少，而每个元素的属性较多。
如果已经以任何一种模式设计了 XML 数据，该示例将很有用，因为这样可以转换为另一种模式。
下面的 XML 使用以元素为中心的文档。 元素不包含任何属性。
输入 - centric.xml
<?xml version='1.0' encoding='UTF-8'?>
<root>
<Customer>
<firstname>Jerry</firstname>
<lastname>Larson</lastname>
<Order>
<OrderID>Ord-12345</OrderID>
<OrderDetail>
<Quantity>1301</Quantity>
<UnitPrice>$3000</UnitPrice>
<ProductName>Computer</ProductName>
</OrderDetail>
</Order>
</Customer>
</root>
下面的示例应用程序执行转换。
' The program will convert an element-centric document to an
' attribute-centric document or element-centric to attribute-centric.
Imports System
Imports System.Xml
Imports System.IO
Imports System.Text
Imports System.Collections
Class ModeConverter
Private bufferSize As Integer = 2048
Friend Class ElementNode
Private _name As [String]
Private _prefix As [String]
Private _namespace As [String]
Private _startElement As Boolean
Friend Sub New()
Me._name = Nothing
Me._prefix = Nothing
Me._namespace = Nothing
Me._startElement = False
End Sub 'New
Friend Sub New(prefix As [String], name As [String], [nameSpace] As [String])
Me._name = name
Me._prefix = prefix
Me._namespace = [nameSpace]
End Sub 'New
Public ReadOnly Property name() As [String]
Get
Return _name
End Get
End Property
Public ReadOnly Property prefix() As [String]
Get
Return _prefix
End Get
End Property
Public ReadOnly Property [nameSpace]() As [String]
Get
Return _namespace
End Get
End Property
Public Property startElement() As Boolean
Get
Return _startElement
End Get
Set
_startElement = value
End Set
End Property
End Class 'ElementNode
' Entry point which delegates to C-style main Private Function.
Public Overloads Shared Sub Main()
Main(System.Environment.GetCommandLineArgs())
End Sub
Overloads Public Shared Sub Main(args() As [String])
Dim modeConverter As New ModeConverter()
If args(0) Is Nothing Or args(0) = "?" Or args.Length < 2 Then
modeConverter.Usage()
Return
End If
Dim sourceFile As New FileStream(args(1), FileMode.Open, FileAccess.Read, FileShare.Read)
Dim targetFile As New FileStream(args(2), FileMode.Create, FileAccess.ReadWrite, FileShare.ReadWrite)
If args(0) = "-a" Then
modeConverter.ConertToAttributeCentric(sourceFile, targetFile)
Else
modeConverter.ConertToElementCentric(sourceFile, targetFile)
End If
Return
End Sub 'Main
Public Sub Usage()
Console.WriteLine("? This help message " + ControlChars.Lf)
Console.WriteLine("Convert -mode sourceFile, targetFile " + ControlChars.Lf)
Console.WriteLine(ControlChars.Tab + " mode: e element centric" + ControlChars.Lf)
Console.WriteLine(ControlChars.Tab + " mode: a attribute centric" + ControlChars.Lf)
End Sub 'Usage
Public Sub ConertToAttributeCentric(sourceFile As FileStream, targetFile As FileStream)
' Stack is used to track how many.
Dim stack As New Stack()
Dim reader As New XmlTextReader(sourceFile)
reader.Read()
Dim writer As New XmlTextWriter(targetFile, reader.Encoding)
writer.Formatting = Formatting.Indented
Do
Select Case reader.NodeType
Case XmlNodeType.XmlDeclaration
writer.WriteStartDocument((Nothing = reader.GetAttribute("standalone") Or "yes" = reader.GetAttribute("standalone")))
Case XmlNodeType.Element
Dim element As New ElementNode(reader.Prefix, reader.LocalName, reader.NamespaceURI)
If 0 = stack.Count Then
writer.WriteStartElement(element.prefix, element.name, element.nameSpace)
element.startElement = True
End If
stack.Push(element)
Case XmlNodeType.Attribute
Throw New Exception("We should never been here!")
Case XmlNodeType.Text
Dim attribute As New ElementNode()
attribute = CType(stack.Pop(), ElementNode)
element = CType(stack.Peek(), ElementNode)
If Not element.startElement Then
writer.WriteStartElement(element.prefix, element.name, element.nameSpace)
element.startElement = True
End If
writer.WriteStartAttribute(attribute.prefix, attribute.name, attribute.nameSpace)
writer.WriteRaw(reader.Value)
reader.Read() 'jump over the EndElement
Case XmlNodeType.EndElement
writer.WriteEndElement()
stack.Pop()
Case XmlNodeType.CDATA
writer.WriteCData(reader.Value)
Case XmlNodeType.Comment
writer.WriteComment(reader.Value)
Case XmlNodeType.ProcessingInstruction
writer.WriteProcessingInstruction(reader.Name, reader.Value)
Case XmlNodeType.EntityReference
writer.WriteEntityRef(reader.Name)
Case XmlNodeType.Whitespace
writer.WriteWhitespace(reader.Value);
Case XmlNodeType.None
writer.WriteRaw(reader.Value)
Case XmlNodeType.SignificantWhitespace
writer.WriteWhitespace(reader.Value)
Case XmlNodeType.DocumentType
writer.WriteDocType(reader.Name, reader.GetAttribute("PUBLIC"), reader.GetAttribute("SYSTEM"), reader.Value)
Case XmlNodeType.EndEntity
Case Else
Console.WriteLine(("UNKNOWN Node Type = " + CInt(reader.NodeType)))
End Select
Loop While reader.Read()
writer.WriteEndDocument()
reader.Close()
writer.Flush()
writer.Close()
End Sub 'ConertToAttributeCentric
' Use the WriteNode to simplify the process.
Public Sub ConertToElementCentric(sourceFile As FileStream, targetFile As FileStream)
Dim reader As New XmlTextReader(sourceFile)
reader.Read()
Dim writer As New XmlTextWriter(targetFile, reader.Encoding)
writer.Formatting = Formatting.Indented
Do
Select Case reader.NodeType
Case XmlNodeType.Element
writer.WriteStartElement(reader.Prefix, reader.LocalName, reader.NamespaceURI)
If reader.MoveToFirstAttribute() Then
Do
writer.WriteStartElement(reader.Prefix, reader.LocalName, reader.NamespaceURI)
writer.WriteRaw(reader.Value)
writer.WriteEndElement()
Loop While reader.MoveToNextAttribute()
writer.WriteEndElement()
End If
Case XmlNodeType.Attribute
Throw New Exception("We should never been here!")
Case XmlNodeType.Whitespace
writer.WriteWhitespace(reader.Value)
Case XmlNodeType.EndElement
writer.WriteEndElement()
Case XmlNodeType.Text
Throw New Exception("The input document is not a attribute centric document" + ControlChars.Lf)
Case Else
Console.WriteLine(reader.NodeType)
writer.WriteNode(reader, False)
End Select
Loop While reader.Read()
reader.Close()
writer.Flush()
writer.Close()
End Sub 'ConertToElementCentric
End Class 'ModeConverter
// The program will convert an element-centric document to an
// attribute-centric document or element-centric to attribute-centric.
using System;
using System.Xml;
using System.IO;
using System.Text;
using System.Collections;
class ModeConverter {
private const int bufferSize=2048;
internal class ElementNode {
String _name;
String _prefix;
String _namespace;
bool _startElement;
internal ElementNode() {
this._name = null;
this._prefix = null;
this._namespace = null;
this._startElement = false;
}
internal ElementNode(String prefix, String name, String nameSpace) {
this._name = name;
this._prefix = prefix;
this._namespace = nameSpace;
}
public String name{
get { return _name; }
}
public String prefix{
get { return _prefix; }
}
public String nameSpace{
get { return _namespace; }
}
public bool startElement{
get { return _startElement; }
set { _startElement = value;}
}
}
public static void Main(String[] args) {
ModeConverter modeConverter = new ModeConverter();
if (args[0]== null || args[0]== "?" || args.Length < 2 ) {
modeConverter.Usage();
return;
}
FileStream sourceFile = new FileStream(args[1], FileMode.Open, FileAccess.Read, FileShare.Read);
FileStream targetFile = new FileStream(args[2], FileMode.Create, FileAccess.ReadWrite, FileShare.ReadWrite);
if (args[0] == "-a") {
modeConverter.ConertToAttributeCentric(sourceFile, targetFile);
} else {
modeConverter.ConertToElementCentric(sourceFile, targetFile);
}
return;
}
public void Usage() {
Console.WriteLine("? This help message \n");
Console.WriteLine("Convert -mode sourceFile, targetFile \n");
Console.WriteLine("\t mode: e element centric\n");
Console.WriteLine("\t mode: a attribute centric\n");
}
public void ConertToAttributeCentric(FileStream sourceFile, FileStream targetFile) {
// Stack is used to track how many.
Stack stack = new Stack();
XmlTextReader reader = new XmlTextReader(sourceFile);
reader.Read();
XmlTextWriter writer = new XmlTextWriter(targetFile, reader.Encoding);
writer.Formatting = Formatting.Indented;
do {
switch (reader.NodeType) {
case XmlNodeType.XmlDeclaration:
writer.WriteStartDocument(null == reader.GetAttribute("standalone") || "yes" == reader.GetAttribute("standalone"));
break;
case XmlNodeType.Element:
ElementNode element = new ElementNode(reader.Prefix, reader.LocalName, reader.NamespaceURI);
if (0 == stack.Count) {
writer.WriteStartElement(element.prefix, element.name, element.nameSpace);
element.startElement=true;
}
stack.Push(element);
break;
case XmlNodeType.Attribute:
throw new Exception("We should never been here!");
case XmlNodeType.Text:
ElementNode attribute = new ElementNode();
attribute = (ElementNode)stack.Pop();
element = (ElementNode)stack.Peek();
if (!element.startElement) {
writer.WriteStartElement(element.prefix, element.name, element.nameSpace);
element.startElement=true;
}
writer.WriteStartAttribute(attribute.prefix, attribute.name, attribute.nameSpace);
writer.WriteRaw(reader.Value);
reader.Read(); //jump over the EndElement
break;
case XmlNodeType.EndElement:
writer.WriteEndElement();
stack.Pop();
break;
case XmlNodeType.CDATA:
writer.WriteCData(reader.Value);
break;
case XmlNodeType.Comment:
writer.WriteComment(reader.Value);
break;
case XmlNodeType.ProcessingInstruction:
writer.WriteProcessingInstruction(reader.Name, reader.Value);
break;
case XmlNodeType.EntityReference:
writer.WriteEntityRef( reader.Name);
break;
case XmlNodeType.Whitespace:
writer.WriteWhitespace(reader.Value);
break;
case XmlNodeType.None:
writer.WriteRaw(reader.Value);
break;
case XmlNodeType.SignificantWhitespace:
writer.WriteWhitespace(reader.Value);
break;
case XmlNodeType.DocumentType:
writer.WriteDocType(reader.Name, reader.GetAttribute("PUBLIC"), reader.GetAttribute("SYSTEM"), reader.Value);
break;
case XmlNodeType.EndEntity:
break;
default:
Console.WriteLine("UNKNOWN Node Type = " + ((int)reader.NodeType));
break;
}
} while (reader.Read());
writer.WriteEndDocument();
reader.Close();
writer.Flush();
writer.Close();
}
// Use the WriteNode to simplify the process.
public void ConertToElementCentric(FileStream sourceFile, FileStream targetFile) {
XmlTextReader reader = new XmlTextReader(sourceFile);
reader.Read();
XmlTextWriter writer = new XmlTextWriter(targetFile, reader.Encoding);
writer.Formatting = Formatting.Indented;
do {
switch (reader.NodeType) {
case XmlNodeType.Element:
writer.WriteStartElement(reader.Prefix, reader.LocalName, reader.NamespaceURI);
if (reader.MoveToFirstAttribute()) {
do {
writer.WriteStartElement(reader.Prefix, reader.LocalName, reader.NamespaceURI);
writer.WriteRaw(reader.Value);
writer.WriteEndElement();
} while(reader.MoveToNextAttribute());
writer.WriteEndElement();
}
break;
case XmlNodeType.Attribute:
throw new Exception("We should never been here!");
case XmlNodeType.Whitespace:
writer.WriteWhitespace(reader.Value);
break;
case XmlNodeType.EndElement:
writer.WriteEndElement();
break;
case XmlNodeType.Text:
throw new Exception("The input document is not a attribute centric document\n");
default:
Console.WriteLine(reader.NodeType);
writer.WriteNode(reader, false);
break;
}
} while (reader.Read());
reader.Close();
writer.Flush();
writer.Close();
}
}
代码编译后，在命令行键入 <compiled name> -a centric.xml <output file name> 运行代码。输出文件必须存在，而且可以是空文本文件。
对于下面的输出，假定 C# 程序已编译到 centric_cs，命令行是 C:\centric_cs -a centric.xml centric_out.xml。
模式 -a 告诉应用程序将输入的 XML 转换为以属性为中心，而模式 -e 将其改变为以元素为中心。 下面的输出是使用 -a 模式生成的新的以属性为中心的输出。 元素现在包含属性而不是嵌套的元素。
输出：centric_out.xml
<?xml version="1.0" encoding="utf-8" standalone="yes" ?>
<root>
<Customer firstname="Jerry" lastname="Larson">
<Order OrderID="Ord-12345">
<OrderDetail Quantity="1301" UnitPrice="$3000" ProductName="Computer" />
</Order>
</Customer>
</root>