java多執行緒抓取鈴聲多多官網的鈴聲數據

一直想练习下java多线程抓取数据。

有天被我发现，铃声多多的官网（http://www.shoujiduoduo.com/main/）有大量的数据。

通过观察他们前端获取铃声数据的ajax

http://www.shoujiduoduo.com/ringweb/ringweb.php?type=getlist&listid={类别ID}&page={分页页码}

很容易就能发现通过改变 listId和page就能从服务器获取铃声的json数据， 通过解析json数据，

可以看到都带有{"hasmore":1,"curpage":1}这样子的指示，通过判断hasmore的值，决定是否进行下一页的抓取。

但是通过上面这个链接返回的json中不带有铃声的下载地址

很快就可以发现，点击页面的“下载”会看到

通过下面的请求，就可以获取铃声的下载地址了

http://www.shoujiduoduo.com/ringweb/ringweb.php?type=geturl&act=down&rid={铃声ID}

所以，他们的数据是很容易被偷的。于是我就开始...

源码已经发在github上。如果感兴趣的童鞋可以查看

github：https://github.com/yongbo000/DuoduoAudioRobot

上代码：

<pre class="brush:java;">package me.yongbo.DuoduoRingRobot;
import java.io.BufferedReader;
import java.io.File;
import java.io.FileWriter;
import java.io.IOException;
import java.io.InputStream;
import java.io.InputStreamReader;
import java.net.URL;
import java.net.URLConnection;
import java.util.Iterator;
import java.util.regex.Matcher;
import java.util.regex.Pattern;
import com.google.gson.Gson;
import com.google.gson.JsonArray;
import com.google.gson.JsonElement;
import com.google.gson.JsonParser;
/* * @author yongbo_ * @created 2013/4/16 * * */
public class DuoduoRingRobotClient implements Runnable {
public static String GET_RINGINFO_URL = "http://www.shoujiduoduo.com/ringweb/ringweb.php?type=getlist&listid=%1$d&page=%2$d";
public static String GET_DOWN_URL = "http://www.shoujiduoduo.com/ringweb/ringweb.php?type=geturl&act=down&rid=%1$d";
public static String ERROR_MSG = "listId为 %1$d 的Robot发生错误，已自动停止。当前page为 %2$d";public static String STATUS_MSG = "开始抓取数据，当前listId： %1$d，当前page： %2$d";
public static String FILE_DIR = "E:/RingData/";public static String FILE_NAME = "listId=%1$d.txt";private boolean errorFlag = false;private int listId;private int page;
private int endPage = -1;private int hasMore = 1;
private DbHelper dbHelper;
/** * 构造函数 * @param listId 菜单ID * @param page 开始页码 * @param endPage 结束页码 * */
public DuoduoRingRobotClient(int listId, int beginPage, int endPage)
 {this.listId = listId;this.page = beginPage;this.endPage = endPage;this.dbHelper = new DbHelper();}
/** * 构造函数 * @param listId 菜单ID * @param page 开始页码 * */
public DuoduoRingRobotClient(int listId, int page) {this(listId, page, -1);}
/** * 获取铃声 * */public void getRings() {String url = String.format(GET_RINGINFO_URL, listId, page);String responseStr = httpGet(url);hasMore = getHasmore(responseStr);
page = getNextPage(responseStr);
ringParse(responseStr.replaceAll("\\{\"hasmore\":[0-9]*,\"curpage\":[0-9]*\\},", "").replaceAll(",]", "]"));}/** * 发起http请求 * @param webUrl 请求连接地址 * */public String httpGet(String webUrl){URL url;URLConnection conn;StringBuilder sb = new StringBuilder();String resultStr = "";try {url = new URL(webUrl);conn = url.openConnection();conn.connect();InputStream is = conn.getInputStream();InputStreamReader isr = new InputStreamReader(is);BufferedReader bufReader = new BufferedReader(isr);String lineText;while ((lineText = bufReader.readLine()) != null) {sb.append(lineText);}resultStr = sb.toString();} catch (Exception e) {errorFlag = true;//将错误写入txtwriteToFile(String.format(ERROR_MSG, listId, page));}return resultStr;}/** * 将json字符串转化成Ring对象，并存入txt中 * @param json Json字符串 * */public void ringParse(String json) {Ring ring = null;JsonElement element = new JsonParser().parse(json);JsonArray array = element.getAsJsonArray();// 遍历数组Iterator<JsonElement> it = array.iterator();
Gson gson = new Gson();while (it.hasNext() && !errorFlag) {JsonElement e = it.next();// JsonElement转换为JavaBean对象ring = gson.fromJson(e, Ring.class);ring.setDownUrl(getRingDownUrl(ring.getId()));if(isAvailableRing(ring)) {System.out.println(ring.toString());
//可选择写入数据库还是写入文本//writeToFile(ring.toString());writeToDatabase(ring);}}}
/** * 写入txt * @param data 字符串 * */public void writeToFile(String data)
 {String path = FILE_DIR + String.format(FILE_NAME, listId);File dir = new File(FILE_DIR);File file = new File(path);FileWriter fw = null;if(!dir.exists()){dir.mkdirs();
}try {if(!file.exists()){file.createNewFile();}fw = new FileWriter(file, true);
fw.write(data);fw.write("\r\n");fw.flush();} catch (IOException e) {
// TODO Auto-generated catch blocke.printStackTrace();
}finally {try {if(fw != null){fw.close();}} catch (IOException e) {
// TODO Auto-generated catch blocke.printStackTrace();}}}/** * 写入数据库 * @param ring 一个Ring的实例 * */
public void writeToDatabase(Ring ring) {dbHelper.execute("addRing", ring);}
@Overridepublic void run() {while(hasMore == 1 && !errorFlag){if(endPage != -1){if(page > endPage) { break; }}System.out.println(String.format(STATUS_MSG, listId, page));
getRings();System.out.println(String.format("该页数据写入完成"));}System.out.println("ending...");}
private int getHasmore(String resultStr){Pattern p = Pattern.compile("\"hasmore\":([0-9]*),\"curpage\":([0-9]*)"); 
 Matcher match = p.matcher(resultStr);  
 if (match.find()) {  return Integer.parseInt(match.group(1));
  }  return 0;
}
private int getNextPage(String resultStr){Pattern p = Pattern.compile("\"hasmore\":([0-9]*),\"curpage\":([0-9]*)");Matcher match = p.matcher(resultStr);if (match.find()) {return Integer.parseInt(match.group(2));}return 0;}
/** * 判断当前Ring是否满足条件。当Ring的name大于50个字符或是duration为小数则不符合条件，将被剔除。 * @param ring 当前Ring对象实例 * */private boolean isAvailableRing(Ring ring){Pattern p = Pattern.compile("^[1-9][0-9]*$");
Matcher match = p.matcher(ring.getDuration());
if(!match.find()){return false;}if(ring.getName().length() > 50 || ring.getArtist().length() > 50 || ring.getDownUrl().length() == 0){return false;}return true;}
/** * 获取铃声的下载地址 * @param rid 铃声的id * */
public String getRingDownUrl(String rid){String url = String.format(GET_DOWN_URL, rid);
String responseStr = httpGet(url);return responseStr;}}

更多java多執行緒抓取鈴聲多多官網的鈴聲數據相关文章请关注PHP中文网！