利用HttpURLConnection抓取网页取名

闲来无事，利用Java 的HttpURLConnection，使用多线程来抓取网页，计算名字的分数。

仅供娱乐。

 

程序如下 

import java.io.BufferedReader; import java.io.IOException; import java.io.InputStream; import java.io.InputStreamReader; import java.net.HttpURLConnection; import java.net.MalformedURLException; import java.net.URL;   import org.apache.log4j.Logger;   public class XingMing {     static final Logger log = Logger.getLogger(XingMing. class );     public static String read(String urlStr) {     try {      URL url = new URL(urlStr);      HttpURLConnection connection = (HttpURLConnection) url          .openConnection();      connection.connect();      InputStream in = connection.getInputStream();      BufferedReader read = new BufferedReader( new InputStreamReader(in));      StringBuffer buf = new StringBuffer();      String line = null ;       while ((line = read.readLine()) != null ) {        buf.append(line);      }       return buf.toString();    } catch (MalformedURLException e) {       return null ;    } catch (IOException e) {       return null ;    }  }     public static String find(String str, String beginStr, String endStr) {     final int length = beginStr.length();     int index = str.indexOf(beginStr);    String result = null ;     if (index != -1) {       int index2 = str.indexOf(endStr, index + length);       if (index2 != -1) {        result = str.substring(index + beginStr.length(), index2);      }    }     return result;  }     public static String findName(String source, String name) {     // value= 我的姓名『 XX 』的分析：     return find(source, "value= 我的姓名『 " , " 』的分析 " );  }     public static String findScore(String source, String name) {     // <font size=3> 姓名评分： </font><font color=0000ff size=5FONT-SIZE: 10pt;">     // BT, 楷体 ">99.5</font>     return find(        source,         "<font size=3> 姓名评分： </font><font color=0000ff size=5 BT, 楷体 /">" ,         "</font>" );  }     public static void main(String[] args) throws IOException {     final char firstChar = ' 一 ' ;     final char lastChar = ' 龥 ' ;     // 最大开启 100 个线程，可以加快查询速度 .     int maxThread = 100;     int step = (lastChar - firstChar) / maxThread;     for ( int i = 0; i < maxThread; i++) {       char start = ( char ) (firstChar + i * step);       char end = ( char ) (firstChar + i * step + step - 1);      System. out .println( " 开启 " + (i + 1) + " 处理 :" + start + "-" + end          + ( char ) (end + 1));       new CallThread(start, end).start();    }  }     static class CallThread extends Thread {     private char start ;     private char end ;     private String info ;      CallThread( char start, char end) {       this . start = start;       this . end = end;       this . info = this . start + "-" + this . end ;    }       public void run() {       // 姓       final char youname1 = ' 赵 ' ;       final String url = "http://www.xingming.net/cmjg-mz.asp?sex= 男 &youname1="          + youname1 + "&youname2=" ;      String youname2;      String webinfo = null ;       for ( char i = start ; i <= end ; i++, webinfo = null ) {         // 名字规则自己取吧 .         // youname2 = " 良 " + i;         // youname2 = "" + i + i;        youname2 = i + " 敏 " ;         for ( int j = 0; j < 5 && webinfo == null ; j++) {          webinfo = XingMing.read(url + youname2);        }         if (webinfo == null ) {           log .warn( " 获取名字 [" + youname1 + youname2 + "] 失败 " );           continue ;        }        String webName = XingMing.findName(webinfo, "[" + youname2            + "]" );        String webScore = XingMing.findScore(webinfo, "[" + youname2            + "]" );         try {           if (Float.parseFloat(webScore) >= 90) {            System. out .println(youname2 + ":" + webName + ":"                + webScore);          }        } catch (Exception e) {        }         log .info( this . info + ":" + webName + ":" + webScore);         if ((i - start ) % 100 == 0) {          System. out .println( this . info + " 处理了 " + (i - start ) + " 个 " );        }      }      System. out .println( this . info + " 结束了 ....." );    }  } }

 

最新程序：

最终版宝宝取名程序,java版，我家宝宝名字已经确定。  

http://blog.csdn.net/z3h/archive/2008/01/16/2047420.aspx