XML: 使用XmlDocument 与 XmlReader 类

一. XmlDocument类：
XmlDocument与XmlReader类从.NET 1.0就已经存在了。W3C定义了一个叫做文件对象模型(DOM: Document Object Model)的标准来处理XML文档。支持DOM的类可以自由地定位并修改XML文档。要想使用XmlDocument类，需要添加System.Xml.dll的引用，并且引入System.Xml命名空间。XmlDocument类使用DOM级别1与级别2，在内存中对XML进行表示。这个类可以定位寻找或者修改XML结点。

1.创建XmlDocument对象：
下面这个例子演示了如何使用XmlDocument类。

private void createAndSaveXmlDocumentToolStripMenuItem_Click(
   object sender, EventArgs e)
{
   //Declare and create new XmlDocument
   var xmlDoc = new XmlDocument();

   XmlElement el;
   int childCounter;
   int grandChildCounter;

   //Create the xml declaration first
   xmlDoc.AppendChild(
      xmlDoc.CreateXmlDeclaration("1.0", "utf-8", null));

   //Create the root node and append into doc
   el = xmlDoc.CreateElement("MyRoot");
   xmlDoc.AppendChild(el);

   //Child Loop
   for (childCounter = 1; childCounter <= 4; childCounter++)
   {
      XmlElement childelmt;
      XmlAttribute childattr;

      //Create child with ID attribute
      childelmt = xmlDoc.CreateElement("MyChild");
      childattr = xmlDoc.CreateAttribute("ID");
      childattr.Value = childCounter.ToString();
      childelmt.Attributes.Append(childattr);

      //Append element into the root element
      el.AppendChild(childelmt);
      for (grandChildCounter = 1; grandChildCounter <= 3;
         grandChildCounter++)
      {
         //Create grandchildren
         childelmt.AppendChild(xmlDoc.CreateElement("MyGrandChild"));
      }
   }

   //Save to file
   xmlDoc.Save(getFilePath("XmlDocumentTest.xml"));
   txtLog.AppendText("XmlDocumentTest.xml Created\r\n");

}

private string getFilePath(string fileName)
{
   return Path.Combine(Environment.GetFolderPath(
      Environment.SpecialFolder.Desktop), fileName);
}

      private void createAndSaveXmlDocumentToolStripMenuItem_Click(
         object sender, EventArgs e)
      {
         //Declare and create new XmlDocument
         var xmlDoc = new XmlDocument();

         XmlElement el;
         int childCounter;
         int grandChildCounter;

         //Create the xml declaration first
         xmlDoc.AppendChild(
            xmlDoc.CreateXmlDeclaration("1.0", "utf-8", null));

         //Create the root node and append into doc
         el = xmlDoc.CreateElement("MyRoot");
         xmlDoc.AppendChild(el);

         //Child Loop
         for (childCounter = 1; childCounter <= 4; childCounter++)
         {
            XmlElement childelmt;
            XmlAttribute childattr;

            //Create child with ID attribute
            childelmt = xmlDoc.CreateElement("MyChild");
            childattr = xmlDoc.CreateAttribute("ID");
            childattr.Value = childCounter.ToString();
            childelmt.Attributes.Append(childattr);

            //Append element into the root element
            el.AppendChild(childelmt);
            for (grandChildCounter = 1; grandChildCounter <= 3;
               grandChildCounter++)
            {
               //Create grandchildren 
               childelmt.AppendChild(xmlDoc.CreateElement("MyGrandChild"));
            }
         }

         //Save to file
         xmlDoc.Save(getFilePath("XmlDocumentTest.xml"));
         txtLog.AppendText("XmlDocumentTest.xml Created\r\n");

      }

      private string getFilePath(string fileName)
      {
         return Path.Combine(Environment.GetFolderPath(
            Environment.SpecialFolder.Desktop), fileName);
      }

最后生成的XML文档如下：

<?xml version="1.0" encoding="utf-8"?>
<MyRoot>
  <MyChild ID="1">
    <MyGrandChild />
    <MyGrandChild />
    <MyGrandChild />
  </MyChild>
  <MyChild ID="2">
    <MyGrandChild />
    <MyGrandChild />
    <MyGrandChild />
  </MyChild>
  <MyChild ID="3">
    <MyGrandChild />
    <MyGrandChild />
    <MyGrandChild />
  </MyChild>
  <MyChild ID="4">
    <MyGrandChild />
    <MyGrandChild />
    <MyGrandChild />
  </MyChild>
</MyRoot>

<?xml version="1.0" encoding="utf-8"?>
<MyRoot>
  <MyChild ID="1">
    <MyGrandChild />
    <MyGrandChild />
    <MyGrandChild />
  </MyChild>
  <MyChild ID="2">
    <MyGrandChild />
    <MyGrandChild />
    <MyGrandChild />
  </MyChild>
  <MyChild ID="3">
    <MyGrandChild />
    <MyGrandChild />
    <MyGrandChild />
  </MyChild>
  <MyChild ID="4">
    <MyGrandChild />
    <MyGrandChild />
    <MyGrandChild />
  </MyChild>
</MyRoot>

2.使用DOM来解析XmlDocument对象：
XmlDocument对象可以使用递归的方法来遍历所有元素。请看下面的例子：

private void parsingAndXmlDocumentToolStripMenuItem_Click(object sender, EventArgs e)
{
   XmlDocument xmlDoc = new XmlDocument();
   xmlDoc.Load(getFilePath("XmlDocumentTest.xml"));
   RecurseNodes(xmlDoc.DocumentElement);
}

public void RecurseNodes(XmlNode node)
{
   var sb = new StringBuilder();
   //start recursive loop with level 0
   RecurseNodes(node, 0, sb);
   txtLog.Text = sb.ToString();
}

public void RecurseNodes(XmlNode node, int level, StringBuilder sb)
{
   sb.AppendFormat("{0,2} Type:{1,-9} Name:{2,-13} Attr:",
      level, node.NodeType, node.Name);

   foreach (XmlAttribute attr in node.Attributes)
   {
      sb.AppendFormat("{0}={1} ", attr.Name, attr.Value);
   }
   sb.AppendLine();

   foreach (XmlNode n in node.ChildNodes)
   {
      RecurseNodes(n, level + 1, sb);
   }
}

      private void parsingAndXmlDocumentToolStripMenuItem_Click(object sender, EventArgs e)
      {
         XmlDocument xmlDoc = new XmlDocument();
         xmlDoc.Load(getFilePath("XmlDocumentTest.xml"));
         RecurseNodes(xmlDoc.DocumentElement);
      }

      public void RecurseNodes(XmlNode node)
      {
         var sb = new StringBuilder();
         //start recursive loop with level 0
         RecurseNodes(node, 0, sb);
         txtLog.Text = sb.ToString();
      }

      public void RecurseNodes(XmlNode node, int level, StringBuilder sb)
      {
         sb.AppendFormat("{0,2} Type:{1,-9} Name:{2,-13} Attr:",
            level, node.NodeType, node.Name);

         foreach (XmlAttribute attr in node.Attributes)
         {
            sb.AppendFormat("{0}={1} ", attr.Name, attr.Value);
         }
         sb.AppendLine();

         foreach (XmlNode n in node.ChildNodes)
         {
            RecurseNodes(n, level + 1, sb);
         }
      }

输出的结果如下：

0 Type:Element   Name:MyRoot        Attr
1 Type:Element   Name:MyChild       AttrID=1
2 Type:Element   Name:MyGrandChild  Attr
2 Type:Element   Name:MyGrandChild  Attr
2 Type:Element   Name:MyGrandChild  Attr
1 Type:Element   Name:MyChild       AttrID=2
2 Type:Element   Name:MyGrandChild  Attr
2 Type:Element   Name:MyGrandChild  Attr
2 Type:Element   Name:MyGrandChild  Attr
1 Type:Element   Name:MyChild       AttrID=3
2 Type:Element   Name:MyGrandChild  Attr
2 Type:Element   Name:MyGrandChild  Attr
2 Type:Element   Name:MyGrandChild  Attr
1 Type:Element   Name:MyChild       AttrID=4
2 Type:Element   Name:MyGrandChild  Attr
2 Type:Element   Name:MyGrandChild  Attr
2 Type:Element   Name:MyGrandChild  Attr

 0 Type:Element   Name:MyRoot        Attr
 1 Type:Element   Name:MyChild       AttrID=1
 2 Type:Element   Name:MyGrandChild  Attr
 2 Type:Element   Name:MyGrandChild  Attr
 2 Type:Element   Name:MyGrandChild  Attr
 1 Type:Element   Name:MyChild       AttrID=2
 2 Type:Element   Name:MyGrandChild  Attr
 2 Type:Element   Name:MyGrandChild  Attr
 2 Type:Element   Name:MyGrandChild  Attr
 1 Type:Element   Name:MyChild       AttrID=3
 2 Type:Element   Name:MyGrandChild  Attr
 2 Type:Element   Name:MyGrandChild  Attr
 2 Type:Element   Name:MyGrandChild  Attr
 1 Type:Element   Name:MyChild       AttrID=4
 2 Type:Element   Name:MyGrandChild  Attr
 2 Type:Element   Name:MyGrandChild  Attr
 2 Type:Element   Name:MyGrandChild  Attr

3.搜索XmlDocument对象：
SelectSingleNode方法可以用来定位XML元素；它需要将XPath查询传递给该方法。下面的代码演示了如何通过使用XPath查询来调用SelectSingleNode方法查询到ID为3的MyChild元素。

private void searchingAnXmlDocumentToolStripMenuItem_Click(
   object sender, EventArgs e)
{
   var xmlDoc = new XmlDocument();
   xmlDoc.Load(getFilePath("XmlDocumentTest.xml"));

   var node = xmlDoc.SelectSingleNode("//MyChild[@ID='3']");
   RecurseNodes(node);
}

      private void searchingAnXmlDocumentToolStripMenuItem_Click(
         object sender, EventArgs e)
      {
         var xmlDoc = new XmlDocument();
         xmlDoc.Load(getFilePath("XmlDocumentTest.xml"));

         var node = xmlDoc.SelectSingleNode("//MyChild[@ID='3']");
         RecurseNodes(node);
      }

GetElementsByTagName方法返回一个包含所有符合条件的XmlNode列表。

private void getElementsByTagNameToolStripMenuItem_Click(
   object sender, EventArgs e)
{
   var xmlDoc = new XmlDocument();
   xmlDoc.Load(getFilePath("XmlDocumentTest.xml"));

   var elmts = xmlDoc.GetElementsByTagName("MyGrandChild");

   var sb = new StringBuilder();
   foreach (XmlNode node in elmts)
   {
      RecurseNodes(node, 0, sb);
   }
   txtLog.Text = sb.ToString();
}

      private void getElementsByTagNameToolStripMenuItem_Click(
         object sender, EventArgs e)
      {
         var xmlDoc = new XmlDocument();
         xmlDoc.Load(getFilePath("XmlDocumentTest.xml"));

         var elmts = xmlDoc.GetElementsByTagName("MyGrandChild");

         var sb = new StringBuilder();
         foreach (XmlNode node in elmts)
         {
            RecurseNodes(node, 0, sb);
         }
         txtLog.Text = sb.ToString();
      }

SelectNodes方法也需要使用XPath查询来返回一个XmlNode 列表。这个方法要比GetElementsByTagName更灵活，因为GetElementsByTagName局限于标签名(tag name)。

private void selectNodesToolStripMenuItem_Click(
   object sender, EventArgs e)
{
   var xmlDoc = new XmlDocument();
   xmlDoc.Load(getFilePath("XmlDocumentTest.xml"));

   var elmts = xmlDoc.SelectNodes("//MyGrandChild");

   var sb = new StringBuilder();
   foreach (XmlNode node in elmts)
   {
      RecurseNodes(node, 0, sb);
   }
   txtLog.Text = sb.ToString();
}

      private void selectNodesToolStripMenuItem_Click(
         object sender, EventArgs e)
      {
         var xmlDoc = new XmlDocument();
         xmlDoc.Load(getFilePath("XmlDocumentTest.xml"));

         var elmts = xmlDoc.SelectNodes("//MyGrandChild");

         var sb = new StringBuilder();
         foreach (XmlNode node in elmts)
         {
            RecurseNodes(node, 0, sb);
         }
         txtLog.Text = sb.ToString();
      }

 

二. XmlReader类：
XmlReader是一个抽象类，用来读取和解析XML。一个更常用的子类是XmlTextReader，它可以一个结点一个结点的读取XML。XmlReader类提供最快并且占用内存最少的方法来读取和解析XML数据流，该读取只能向前，并且没有缓存。这个类适用于读取很大的XML文件，并且需要的数据在文件里靠前的位置。下面是一个使用XmlReader的例子：

private void parsingWithXmlReaderToolStripMenuItem_Click(object sender, EventArgs e)
{
   var sb = new StringBuilder();
   var xmlReader = new XmlTextReader(getFilePath("XmlDocumentTest.xml"));

   while (xmlReader.Read())
   {
      switch (xmlReader.NodeType)
      {
         case XmlNodeType.XmlDeclaration:
         case XmlNodeType.Element:
         case XmlNodeType.Comment:
            sb.AppendFormat("{0}: {1} = {2}",
                              xmlReader.NodeType,
                              xmlReader.Name,
                              xmlReader.Value);
            sb.AppendLine();
            break;
         case XmlNodeType.Text:
            sb.AppendFormat(" - Value: {0}", xmlReader.Value);
            sb.AppendLine();
            break;
      }

      if (xmlReader.HasAttributes)
      {
         while (xmlReader.MoveToNextAttribute())
         {
            sb.AppendFormat(" - Attribute: {0} = {1}",
                              xmlReader.Name,
                              xmlReader.Value);
            sb.AppendLine();
         }
      }
   }
   xmlReader.Close();
   txtLog.Text = sb.ToString();
}

      private void parsingWithXmlReaderToolStripMenuItem_Click(object sender, EventArgs e)
      {
         var sb = new StringBuilder();
         var xmlReader = new XmlTextReader(getFilePath("XmlDocumentTest.xml"));

         while (xmlReader.Read())
         {
            switch (xmlReader.NodeType)
            {
               case XmlNodeType.XmlDeclaration:
               case XmlNodeType.Element:
               case XmlNodeType.Comment:
                  sb.AppendFormat("{0}: {1} = {2}",
                                    xmlReader.NodeType,
                                    xmlReader.Name,
                                    xmlReader.Value);
                  sb.AppendLine();
                  break;
               case XmlNodeType.Text:
                  sb.AppendFormat(" - Value: {0}", xmlReader.Value);
                  sb.AppendLine();
                  break;
            }

            if (xmlReader.HasAttributes)
            {
               while (xmlReader.MoveToNextAttribute())
               {
                  sb.AppendFormat(" - Attribute: {0} = {1}",
                                    xmlReader.Name,
                                    xmlReader.Value);
                  sb.AppendLine();
               }
            }
         }
         xmlReader.Close();
         txtLog.Text = sb.ToString();
      }

输出结果如下：

XmlDeclaration: xml=version="1.0" encoding="utf-8"
  Attribute: version=1.0
  Attribute: encoding=utf-8
Element: MyRoot=
Element: MyChild=
  Attribute: ID=1
Element: MyGrandChild=
Element: MyGrandChild=
Element: MyGrandChild=
Element: MyChild=
  Attribute: ID=2
Element: MyGrandChild=
Element: MyGrandChild=
Element: MyGrandChild=
Element: MyChild=
  Attribute: ID=3
Element: MyGrandChild=
Element: MyGrandChild=
Element: MyGrandChild=
Element: MyChild=
  Attribute: ID=4
Element: MyGrandChild=
Element: MyGrandChild=
Element: MyGrandChild=

XmlDeclaration: xml=version="1.0" encoding="utf-8"
  Attribute: version=1.0
  Attribute: encoding=utf-8
Element: MyRoot=
Element: MyChild=
  Attribute: ID=1
Element: MyGrandChild=
Element: MyGrandChild=
Element: MyGrandChild=
Element: MyChild=
  Attribute: ID=2
Element: MyGrandChild=
Element: MyGrandChild=
Element: MyGrandChild=
Element: MyChild=
  Attribute: ID=3
Element: MyGrandChild=
Element: MyGrandChild=
Element: MyGrandChild=
Element: MyChild=
  Attribute: ID=4
Element: MyGrandChild=
Element: MyGrandChild=
Element: MyGrandChild=

 

三. 小节：
• XmlDocument类读取XML结点是全部存在内存中，任意定位，并且可读也可写的。
• XmlReader类读取XML结点使用的是快速数据流(不是一次读取完将数据全部存在内存中)，只能向前，只读的。
• XmlDocument类易于使用，而XmlReader类读取效率很高。
• XmlDocument类可以使用元素名来读取XML结点。
• XmlDocument类可以使用XPath来读取结点。

 

原文出自《Access Data with Microsoft .NET Framework 4》