Android开发从零开始之HTML解析-Jsoup

package com.zxjzzg.myReader.bookcity;

import java.io.BufferedReader;
import java.io.ByteArrayOutputStream;
import java.io.InputStream;
import java.io.InputStreamReader;
import java.net.HttpURLConnection;
import java.net.URL;
import java.util.Vector;

public class getData {
	//@SuppressWarnings("null")
	public static Vector<String []> getData(String urls, String start, String end, String start1, String end1, int a)throws Exception{
		Vector<String []> mydatas = new Vector<String[]>();
		String []data_list = null ;//
		String []data_list1 = null;
		URL url = new URL(urls);
		HttpURLConnection huc = (HttpURLConnection) url.openConnection();
		huc.setRequestMethod("GET");
		huc.setConnectTimeout(10 * 1000);
		InputStream inStream = huc.getInputStream();
		byte[] data = readInputStream(inStream);
		String html = new String(data);
		System.out.println("开始ASC2转换");
//		html = new  String(html.getBytes("UTF-8")); 
		
//		System.out.println(html);
		int b1 = html.indexOf(start) + start.length()+5;
		int b2 = html.indexOf(end,b1);
		System.out.println("b1="+b1+",b2="+b2);
		String filename = html.substring(b1,b2);
		
		if(a == 1){
			
			filename = filename.replace("<br>", "");
			filename = filename.replace("</p>", "");
			//filename = filename.replace("<p>", "");
			System.out.println("filename==="+filename);
			data_list = filename.split("<p>");
			mydatas.add(data_list);
//			for(int i = 0; i < data_list.length; i ++){
//				System.out.println("data_===="+data_list[0][i]);
//			}
		}else if(a == 2){
			
			filename = filename.replace("</p>", "");
			filename = filename.replace("<p>", "");
			filename = filename.replace(" ", "");
			filename = filename.replace("<br>", "");
			filename = filename.replace(" ", "");
			System.out.println("filename=="+filename);
			data_list = filename.split("书名:");
			
			mydatas.add(data_list);
			
			int c1 = html.indexOf(start1) + start1.length()+5;
			int c2 = html.indexOf(end1,c1);
			System.out.println("c1="+c1+",c2="+c2);
			String myurl = html.substring(c1, c2);
			//System.out.println("myurl=="+myurl);
			myurl = myurl.replace("</p>", "");
			myurl = myurl.replace("<p>", "");
			myurl = myurl.replace("<br>", "");
			myurl = myurl.replace(" ", "");
			data_list1 = myurl.split("\n");
			mydatas.add(data_list1);
			
		}
		return mydatas;
		
	}
	
	public static byte[] readInputStream (InputStream inStream)throws Exception{
		ByteArrayOutputStream outStream = new ByteArrayOutputStream();
			byte[] buffer = new byte[1024*1024];
			int len = 0;
			while((len = inStream.read(buffer)) != -1){
				outStream.write(buffer, 0, len);
			}
			inStream.close();
			return outStream.toByteArray();
	}
}

  • 0
    点赞
  • 0
    收藏
    觉得还不错? 一键收藏
  • 0
    评论

“相关推荐”对你有帮助么?

  • 非常没帮助
  • 没帮助
  • 一般
  • 有帮助
  • 非常有帮助
提交
评论
添加红包

请填写红包祝福语或标题

红包个数最小为10个

红包金额最低5元

当前余额3.43前往充值 >
需支付:10.00
成就一亿技术人!
领取后你会自动成为博主和红包主的粉丝 规则
hope_wisdom
发出的红包
实付
使用余额支付
点击重新获取
扫码支付
钱包余额 0

抵扣说明:

1.余额是钱包充值的虚拟货币,按照1:1的比例进行支付金额的抵扣。
2.余额无法直接购买下载,可以购买VIP、付费专栏及课程。

余额充值