wIndows phone 7 解析Html数据

 

在我的上一篇文章中我介绍了windows phone 7的gb2312解码,

/kf/201111/112551.html

解决了下载的html乱码问题,这一篇,我将介绍关于windows phone 7解析html数据，以便我们获得想要的数据.

 

这里,我先介绍一个类库htmlagilitypack,（上一篇文章也是通过这个工具来解码的）. 类库的dll文件我会随demo一起提供

 

这里,我以新浪新闻为例来解析数据

 

 

 

先看看网页版的新浪新闻

 

http://news.sina.com.cn/w/sd/2011-11-27/070023531646.shtml

 

然后我们看一下他的源文件，

 

发现新闻内容的结构是

 

view sourceprint?<p class="blkcontainersblk"> 

 

                <h1 id="artibodytitle" pid="1" tid="1" did="23531646" fid="1666">title</h1> 

 

                <p class="artinfo"><span id="art_source"><a href="http://www.sina.com.cn">http://www.sina.com.cn</a></span>  <span id="pub_date">pub_date</span>  <span id="media_name"><a href="">media_name</a> <a href=""></a> </span></p> 

 

  

 

                <!-- 正文内容begin --> 

 

                <!-- google_ad_section_start --> 

 

  

 

                <p class="blkcontainersblkcon" id="artibody"></p> 

 

</p>

 

大部分还有id属性,这更适合我们去解析了。

 

接下来我们开始去解析

 

第一： 引用htmlagilitypack.dll文件

 

第二：用webclient或者webrequest类来下载html页面然后处理成字符串。

 

view sourceprint?public  delegate void callbackevent(object sender, downloadeventargs e); 

 

       public  event callbackevent downloadcallbackevent; 

 

       public void httpwebrequestdownloadget(string url) 

 

       { 

 

             

 

           thread _thread = new thread(delegate() 

 

           { 

 

               uri _uri = new uri(url, urikind.relativeorabsolute); 

 

               httpwebrequest _httpwebrequest = (httpwebrequest)webrequest.create(_uri); 

 

                _httpwebrequest.method="get"; 

 

               

 

               _httpwebrequest.begingetresponse(new asynccallback(delegate(iasyncresult result) 

 

               { 

 

                   httpwebrequest _httpwebrequestcallback = (httpwebrequest)result.asyncstate; 

 

                   httpwebresponse _httpwebresponsecallback = (httpwebresponse)_httpwebrequestcallback.endgetresponse(result); 

 

                   stream _streamcallback = _httpwebresponsecallback.getresponsestream(); 

 

 

 

                   streamreader _streamreader = new streamreader(_streamcallback,new htmlagilitypack.gb2312encoding()); 

 

                   string _stringcallback = _streamreader.readtoend(); 

 

                  

 

                   deployment.current.dispatcher.begininvoke(new action(() => 

 

                   { 

 

                       if (downloadcallbackevent != null) 

 

                       { 

 

                           downloadeventargs _downloadeventargs = new downloadeventargs(); 

 

                           _downloadeventargs._downloadstream = _streamcallback; 

 

                           _downloadeventargs._downloadstring = _stringcallback; 

 

                           downloadcallbackevent(this, _downloadeventargs); 

 

 

 

                       } 

 

                   })); 

 

 

 

               }), _httpwebrequest); 

 

           }) ; 

 

           _thread.start(); 

 

       } 

 

      // }

 

o(∩_∩)o! 我这个比较复杂, 总之我们下载了html的数据就行了。 

 

贴一个简单的下载方式吧

 

view sourceprint?webclient webclenet=new webclient();   

 

  

 

         webclenet.encoding = new htmlagilitypack.gb2312encoding(); //加入这句设定编码   

 

  

 

         webclenet.downloadstringasync(new uri("http://news.sina.com.cn/s/2011-11-25/120923524756.shtml", urikind.relativeorabsolute));        

 

  

 

         webclenet.downloadstringcompleted += new downloadstringcompletedeventhandler(webclenet_downloadstringcompleted);

 

 现在处理回调函数的e.result

 

view sourceprint?string _result = e._downloadstring; 

 

 

 

           htmldocument _doc = new htmldocument(); //实例化htmlagilitypack.htmldocument对象 

 

           _doc.loadhtml(_result);         //载入html 

 

 

 

           htmlnode _htmlnode01 = _doc.getelementbyid("artibodytitle");  //新闻标题的div 

 

           string _title = _htmlnode01.innertext; 

 

 

 

           htmlnode _htmlnode02 = _doc.getelementbyid("artibody");     //获取内容的p   

 

           string _content = _htmlnode02.innertext; 

 

          // int _count= _htmlnode02.childnodes.where(new func<htmlnode,bool>("p")); 

 

           int _pindex = _content.indexof(" .blkcomment"); 

 

 

 

           _content= _content.substring(0,_pindex); 

 

 

 

           #region　新浪标签 

 

           htmlnode _htmlnodo03 = _doc.getelementbyid("art_source"); 

 

           string _www = _htmlnodo03.firstchild.innertext; 

 

           string _wwwint = _htmlnodo03.firstchild.attributes[0].value; 

 

           #endregion 

 

           // string _source = _htmlnodo03; 

 

           //_htmlnodo03.childnodes 

 

 

 

           #region 发布时间 

 

           htmlnode _htmlnodo04 = _doc.getelementbyid("pub_date"); 

 

           string _pub_date = _htmlnodo04.innertext; 

 

           #endregion 

 

 

 

 

 

           #region 来源网站信息 

 

           htmlnode _htmlnodo05 = _doc.getelementbyid("media_name"); 

 

           string _media_name = _htmlnodo05.firstchild.innertext; 

 

           string _modia_source = _htmlnodo05.firstchild.attributes[0].value; 

 

           #endregion 

 

 

 

           media_namehyperlinkbutton.content = _pub_date + " " + _media_name; 

 

           media_namehyperlinkbutton.navigateuri = new uri(_modia_source, urikind.relativeorabsolute); 

 

           titletextblock.text = _title; 

 

           contenttextblock.text = _content;

 

 

 

结果如下图所示：

 

 

 

网页的大部分标签是没有id属性的,不过幸运的是htmlagilitypack支持xpath

 

那就需要通过xpath语言来查找匹配所需节点

 

xpath教程：http://www.w3school.com.cn/xpath/index.

 

 

 

案例下载：

 

http://115.com/file/dn87dl2d#

myframework_test.zip

 

作者 青瓷