比较两个XML文件，并保存到结果文件

Question

我有我需要比较的差异两个XML文件的差异，XML是非常简单的：

文件1：

<?xml version="1.0" encoding="utf-8"?> 
<Feeds zone="my zone"> 
    <Feed name="attribDump.json">ac1f07edc491a3d237cdfb1a17fc4551</Feed> 
    <Feed name="focus_GroupsKV.txt">0f9e0a14a4ffce6ff5065b6e088c1f84</Feed> 
    <Feed name="NAM_FORMATTED.csv">9e875496cdb072b5e54318d51295fdba</Feed> 
    <Feed name="BNP\activityTitles.txt">2d27c0f19b71b4b411bcb00011d3f8b0</Feed> 
</Feeds> 

和文件2 ：

<?xml version="1.0" encoding="utf-8"?> 
<FeedsRequest version="1"> 
<Feeds zone="my zone"> 
    <Feed name="attribDump.json">ac1f07edc491a3d237cdfb1a17fc4551</Feed> 
    <Feed name="focus_GroupsKV.txt">0f9e0a14a4ffce6ff5065b6e088c1f84</Feed> 
    <Feed name="BNP\activityTitles.txt">e54c5b851ee3ff3f43b10d24f2316431</Feed> 
</Feeds> 
</FeedsRequest> 

文件1是对我们的文件共享和文件2个文件是使用断开连接的设备，将需要的清单列表从文件1.刷新我需要有1支票）确保所有的 文件1中的提要位于文件2和2中）确保找到的所有提要具有相同的hashCode（长字符字符串）。一旦检查完成后，我需要创建一个响应文件，其中包含所有提要的列表，然后指定每个提要的属性（文件已找到并匹配），缺少（找不到文件），或更新（文件被发现，但它是一个旧版本）。

所以基本上结果文件将如下这样：

<?xml version="1.0" encoding="utf-8"?> 
<FeedsResponse version="1"> 
<Feeds zone="my zone"> 
    <Feed name="attribDump.json" status="ok">ac1f07edc491a3d237cdfb1a17fc4551</Feed> 
    <Feed name="focus_GroupsKV.txt" status="ok">0f9e0a14a4ffce6ff5065b6e088c1f84</Feed> 
    <Feed name="NAM_FORMATTED.csv" status="missing">afd2c620053ed4f85ab02b4cc5f7a2b2</Feed> 
    <Feed name="BNP\activityTitles.txt" status="updated">90805b851ee3ff3f43b10d24f2316431</Feed> 

我在做什么目前通过所有文件的循环在文件1，然后检查他们对文件2的差异。在我被困住的地方，自从我使用过XML以来，已经有一段时间了，是如何构建响应文档的。

  FileInfo feedList = new FileInfo(_feedList); 
     FileInfo feedRequest = new FileInfo(_feedRequest); 

     // Load the documents 
     XmlDocument feedListXmlDoc = new XmlDocument(); 
     feedListXmlDoc.Load(_feedList); 

     // Load the documents 
     XmlDocument feedRequestXmlDoc = new XmlDocument(); 
     feedRequestXmlDoc.Load(_feedRequest); 

     //create response doc 
     XmlDocument feedResponseXmlDoc = new XmlDocument(); 

     // Define a single node 
     XmlNode feedListNode; 
     XmlNode feedRequestNode; 

     // Get the root Xml element 
     XmlElement feedListRoot = feedListXmlDoc.DocumentElement; 
     XmlElement feedRequestRoot = feedRequestXmlDoc.DocumentElement; 

     // Get a list of all player names 
     XmlNodeList feedListXml = feedListRoot.GetElementsByTagName("Feed"); 
     XmlNodeList feedRequestXml = feedRequestRoot.GetElementsByTagName("Feed"); 

     // Create an XmlWriterSettings object with the correct options. 
     XmlWriter writer = null; 
     XmlWriterSettings settings = new XmlWriterSettings(); 
     settings.Indent = true; 
     settings.IndentChars = (" "); 
     settings.OmitXmlDeclaration = false; 

     // Create the XmlWriter object and write some content. 
     writer = XmlWriter.Create(_resultPath, settings); 
     writer.WriteStartElement("FeedsDiff"); 

     // The compare algorithm 
     bool feedMatch = false; 

     int j = 0; 

     try 
     { 
      // loop through list of current feeds 
      for (int i = 0; i < feedListXml.Count; i++) 
      { 
       feedListNode = feedListXml.Item(i); 

       string feedListName = feedListNode.Attributes["name"].Value.ToString(); 
       string feedListHash = feedListXml.Item(i).InnerText.ToString(); 

       //check feed request list for a match 
       while (j < feedRequestXml.Count && feedMatch == false) 
       { 
        feedRequestNode = feedRequestXml.Item(j); 
        string feedRequestName = feedRequestNode.Attributes["name"].Value.ToString(); 

        //checks to see if feed names match 
        if (feedListName == feedRequestName) 
        { 
         feedMatch = true; 
         string feedRequestHash = feedRequestXml.Item(j).InnerText.ToString(); 

         //since we found the node, we can remove it from the request list 
         XmlNode node = feedRequestNode.ParentNode; 
         node.RemoveChild(feedRequestNode); 

         //checks to see if hash codes match 
         if (feedListHash == feedRequestHash) 
         { 
          //if name and code match, move to the next one 
          feedMatch = true; 

          //add 'status="ok"' attribute to the node 
          //feedResponseXmlDoc.ImportNode(feedRequestNode,false); 

          Debug.WriteLine(feedListName + " name and hash match"); 

          j = 0; 
         } 
         else 
         { 

          feedMatch = true; 

          //feed has been updated since last device sync 
          //need to add status='update' attribute and append file to response 
          Debug.WriteLine(feedListName + " name matched but hash did not"); 
         } 
        } 
        else 
        { 
         //names didn't match 
         //add status="missing" to the node 
         j++; 
        } 
       } 
       feedMatch = false; 
      } 
      // end Xml document 
      writer.WriteEndElement(); 
      writer.Flush(); 
     } 
     finally 
     { 
      if (writer != null) 
       writer.Close(); 
     } 

现在，我试图实例循环之前的响应文档，然后只需添加的元素，因为他们被发现，但我有一个很难找到一个简洁的方式来做到这一点。任何帮助表示赞赏。

Answer 1

我想通了：

public void CompareXml(string _feedList, string _feedRequest, string _resultPath) 
    { 
     FileInfo feedList = new FileInfo(_feedList); 
     FileInfo feedRequest = new FileInfo(_feedRequest); 

     // Load the documents 
     XmlDocument feedListXmlDoc = new XmlDocument(); 
     feedListXmlDoc.Load(_feedList); 

     // Load the documents 
     XmlDocument feedRequestXmlDoc = new XmlDocument(); 
     feedRequestXmlDoc.Load(_feedRequest); 

     // Define a single node 
     XmlNode feedListNode; 
     XmlNode feedRequestNode; 

     // Get the root Xml element 
     XmlElement feedListRoot = feedListXmlDoc.DocumentElement; 
     XmlElement feedRequestRoot = feedRequestXmlDoc.DocumentElement; 

     // Get a list of feeds for the stored list and the request 
     XmlNodeList feedListXml = feedListRoot.GetElementsByTagName("Feed"); 
     XmlNodeList feedRequestXml = feedRequestRoot.GetElementsByTagName("Feed"); 

     bool feedLocated = false; 
     int j = 0; 

     try 
     { 
      // loop through list of current feeds 
      for (int i = 0; i < feedListXml.Count; i++) 
      { 
       feedListNode = feedListXml.Item(i); 
       //create status attribute 
       XmlAttribute attr = feedListXmlDoc.CreateAttribute("status"); 

       string feedListName = feedListNode.Attributes["name"].Value.ToString(); 
       string feedListHash = feedListXml.Item(i).InnerText.ToString(); 

       //check feed request list for a match 
       while (j < feedRequestXml.Count && feedLocated == false) 
       { 
        feedRequestNode = feedRequestXml.Item(j); 
        string feedRequestName = feedRequestNode.Attributes["name"].Value.ToString(); 

        //checks to see if feed names match 
        if (feedRequestName == feedListName) 
        { 
         string feedRequestHash = feedRequestXml.Item(j).InnerText.ToString(); 

         //checks to see if hashCodes match 
         if (feedListHash == feedRequestHash) 
         { 
          //if name and code match, set status to ok 
          attr.Value = "ok"; 

          Debug.WriteLine(feedListName + " name and hash match. Status: 'ok'"); 
         } 
         else 
         { 
          //if hashCodes don't match, set status attribute to updated 
          attr.Value = "updated"; 

          Debug.WriteLine(feedListName + " name matched but hash did not. Status: 'updated'"); 
         } 
         feedListNode.Attributes.Append(attr); 
         feedLocated = true; 
        } 
        else 
        { 
         //names didn't match, checking to see if we're at the end of the request list 
         if (j + 1 == feedRequestXml.Count) 
         { 
          //file name wasn't found in the request list, set status attribute to missing 
          attr.Value = "missing"; 
          feedListNode.Attributes.Append(attr); 
          feedLocated = true; 
          j = 0; 

          Debug.WriteLine("Reached the end of the file request list without a match. Status: 'missing'"); 
         } 
         //file name wasn't located on this pass, move to next record 
         j++; 
        } 
       } 
       feedLocated = false; 
      } 
     } 
     finally 
     { 
      Debug.WriteLine("Result file has been written out at " + _resultPath); 
     } 

     feedListXmlDoc.Save(_resultPath); 
    } 

Answer 2

看看我的开源项目CodeBlocks在CodePlex上的differ，它是为这样的情况而设计的。它也可在作为的NuGet“differ”