HtmlAgilityPack 加载中文页面时出现在乱码的最终解决方案

      前几天我在博客园转载了一篇"net HtmlParser初步使用研究"的文章,之后看过文章评论才知道原来现在已经有HtmlAgilityPack这个类库可以更方便地对HTML内容进行分析和提取。因此今天特别学习和实践了一下HtmlAgilityPack和XPath,并作下笔记。

 

 HtmlAgilityPack是用C#写的开源Html Parser。不过可能在某些方面设计不尽完善,或者是对中文的支持并不是很好,例如,抓取中文页面时就出现乱码.

我的代码如下:

 

            HtmlWeb hw = new HtmlWeb();
            HtmlDocument doc 
= hw.Load("http://www.xinlg.com");
            //HtmlNodeCollection nodes 
= doc.DocumentNode.SelectNodes("//a");
        Response.Write(doc);
        

页面输出时,全部是乱码,不过有些页面又是正常的,cnblogs是正常,新浪,QQ,新华网等都是乱码,这可能跟编码有关.

具体的开源代码我就没有去看了,我的解决方案就是自己写一个获取Html的方法.

代码如下:

 

        /// <summary>
        
/// 获取指定URL的HTML源代码
        
/// </summary>
        
/// <param name="url"></param>
        
/// <param name="encoding">如果为NULL 则自动识别</param>
        
/// <returns></returns>
        public static string GetWebHtml(string url, Encoding encoding)
        {
            
try
            {
                HttpWebRequest hwr 
= (HttpWebRequest)HttpWebRequest.Create(url);
                HttpWebResponse res;

                
try
                {
                    res 
= (HttpWebResponse)hwr.GetResponse();
                }
                
catch
                {
                    
return string.Empty;
                }

                
if (res.StatusCode == HttpStatusCode.OK)
                {
                    
using (Stream mystream = res.GetResponseStream())
                    {
                        
//没有指定编码,
                        if (encoding == null)
                        {
                            
return DecodeData(mystream, res);
                        }
                        
//指定了编码
                        else
                        {
                            
using (StreamReader reader = new StreamReader(mystream, encoding))
                            {
                                
return reader.ReadToEnd();
                            }
                        }
                    }
                }

                
return null;
            }
            
catch
            {
                
return null;
            }
        }


        
private static string DecodeData(Stream responseStream, HttpWebResponse response)
        {
            
string name = null;
            
string text2 = response.Headers["content-type"];
            
if (text2 != null)
            {
                
int index = text2.IndexOf("charset=");
                
if (index != -1)
                {
                    name 
= text2.Substring(index + 8);
                }
            }
            MemoryStream stream 
= new MemoryStream();
            
byte[] buffer = new byte[0x400];
            
for (int i = responseStream.Read(buffer, 0, buffer.Length); i > 0; i = responseStream.Read(buffer, 0, buffer.Length))
            {
                stream.Write(buffer, 
0, i);
            }
            responseStream.Close();
            
if (name == null)
            {
                MemoryStream stream3 
= stream;
                stream3.Seek((
long)0, SeekOrigin.Begin);
                
string text3 = new StreamReader(stream3, Encoding.ASCII).ReadToEnd();
                
if (text3 != null)
                {
                    
int startIndex = text3.IndexOf("charset=");
                    
int num4 = -1;
                    
if (startIndex != -1)
                    {
                        num4 
= text3.IndexOf("\"", startIndex);
                        
if (num4 != -1)
                        {
                            
int num5 = startIndex + 8;
                            name 
= text3.Substring(num5, (num4 - num5) + 1).TrimEnd(new char[] { '>''"' });
                        }
                    }
                }
            }
            Encoding aSCII 
= null;
            
if (name == null)
            {
                aSCII 
= Encoding.GetEncoding("gb2312");
            }
            
else
            {
                
try
                {
                    
if (name == "GBK")
                    {
                        name 
= "GB2312";
                    }
                    aSCII 
= Encoding.GetEncoding(name);
                }
                
catch
                {
                    aSCII 
= Encoding.GetEncoding("gb2312");
                }
            }
            stream.Seek((
long)0, SeekOrigin.Begin);
            StreamReader reader2 
= new StreamReader(stream, aSCII);
            
return reader2.ReadToEnd();
        }

 

我测试过了,大型网站都没有问题.

最后再调用时,代码如下:

 

 

            string Html = XINLG.Labs.Utils.NetUtil.GetWebHtml("http://www.cnblogs.com/pick/"null);
            HtmlDocument doc 
= new HtmlDocument();
            doc.LoadHtml(Html);

 

在上面的代码中,如果你不知道指定的url编码是什么,直接写NULL,自动判断.

另外,网上还介绍了方法,可以去 HtmlAgilityPack的官网下载源代码,打开项目,自己改一下,代码我就不提供了,大家可以自己搜索一下.

 

 

 

posted on 2011-06-23 18:34  cestarme  阅读(4518)  评论(9编辑  收藏  举报