网页抓取内存溢出

[quote][size=small] 最经一直在做网页抓取工作，使用了自己写的HTML解析器。但是有个问题，就是在得到网页的html代码的时候老报内存溢出，结果程序总是停下来。以下是报的异常代码：[/size][/quote]

Exception in thread "main" java.lang.OutOfMemoryError: Java heap space
        at java.util.Arrays.copyOf(Arrays.java:2882)
        at java.lang.AbstractStringBuilder.expandCapacity(AbstractStringBuilder.java:100)
        at java.lang.AbstractStringBuilder.append(AbstractStringBuilder.java:390)
        at java.lang.StringBuffer.append(StringBuffer.java:224)
        at slcx.com.WebPage.<init>(WebPage.java:67)
        at slcx.stock.gegu.baidu.ExtBdGgHisRecInfo.执行百度个股历史抓取(ExtBdGgHisRecInfo.java:199)
        at slcx.stock.gegu.baidu.ExtBdGgHisRecInfo.main(ExtBdGgHisRecInfo.java:272)

[quote]以下是得到网页html代码的类:WebPage[/quote]

import java.io.BufferedReader;
import java.io.InputStreamReader;
import java.net.HttpURLConnection;
import java.net.URL;

public class WebPage {
	private int connectOutTime = 5000;//链接超时时间
	private int readOutTime = 10000; //读取超时时间

	private String encoding = "gb2312";//网页内容编码
	private String webPageCode = "";//网页代码

	private int searchEnd = 0;// 查找最后的 索引
	private int beginIndex = 0;// 开始 索引

	public WebPage(String link) throws Exception {

		URL url = new URL(link);
		HttpURLConnection httpconn = (HttpURLConnection) url.openConnection();

		httpconn.setConnectTimeout(connectOutTime);
		httpconn.setReadTimeout(readOutTime);

		BufferedReader br = new BufferedReader(new InputStreamReader(httpconn.getInputStream(), encoding));

		StringBuffer sb = new StringBuffer();
		String line = br.readLine();

		while (line != null) {
			sb.append(line + "\r\n");
			line = br.readLine();
		}

		webPageCode = sb.toString();
		this.searchEnd=sb.length();
	}

	public String getWebPageCode() {
		return webPageCode;
	}
}

[quote]以下是是用这个WebPage类的方法，就是用这个方法，老报异常。[/quote]

public void 执行百度个股历史抓取() {
		DBConnection dbc = new DBConnection();

		人员基本信息DAO ryjDao = new 人员基本信息DAO(dbc);
		HashMap<String, Integer> splNameIdMap = ryjDao.得到荐股者名称和编号();

		StockDAO stcDao = new StockDAO(dbc);
		HashMap<String, String> stcNameCodeMap = stcDao.得到股票名称代码();

		生成百度链接DAO scbDao = new 生成百度链接DAO(dbc);
		百度股票信息采集DAO bdgDao = new 百度股票信息采集DAO(dbc);
		ExtBdGgLnk bdg = new ExtBdGgLnk();

		String datetimer=Date.get("yyyy-MM-dd HH:mm:ss");

		Iterator<String> iter = splNameIdMap.keySet().iterator(); //iter长度有2100个左右
		while (iter.hasNext()) {

			String name = iter.next();
			int id = splNameIdMap.get(name);

			ArrayList<String> bdLinklist = scbDao.得到指定推荐者百度URL(id, name, 0);
			if (bdLinklist == null || bdLinklist.isEmpty()) {
				log.info("得到 "+name+ " 的百度搜索结果链接为空");
				continue;
			}
			for (String bdLink : bdLinklist) {

				log.info("现在抓取 " + name +" 的百度链接:" + bdLink);

				ArrayList<String> linkList = bdg.getAllSearchLink(bdLink);
				if (linkList == null || linkList.isEmpty()) {
					log.info("该百度链接没有搜索结果：" + bdLink);
					continue;
				}

				for (String link : linkList) {
					log.info("现在开始抓：" + link + "的荐股信息信息");
					int flag = bdgDao.插入链接(name,link,datetimer);
					if (flag == 2) {
						continue;
					}

					if (!isValidateLink(link)){
						log.info("无效链接："+link);
						continue;
					}

					String webPageCode=null;
					try {
						WebPage web=new WebPage(link);
						webPageCode = web.getWebPageCode();
					} catch (Exception e) {
						e.printStackTrace();
					}					
					if (webPageCode == null || webPageCode.equals("")){
						log.info("该链接没有得到网页代码："+link);
						continue;
					}

					String datetime = 得到推荐股票时间(webPageCode);
					if (datetime == null || datetime.equals("")||datetime.compareTo("2008-01-01 00:00:00")<0){
						log.info("该链接的时间无效："+link);
						continue;
					}

					String stockcode = 得到推荐股票代码(webPageCode, name,splNameIdMap, stcNameCodeMap);
					if (stockcode == null || stockcode.equals("")){
						log.info("该链接得到股票代码无效："+link);
						continue;
					}

					if (flag > 0) {
						int row = bdgDao.更新一条荐股信息(stockcode, name, datetime,link, 0);
						if (row > 0) {
							log.info("成功抓取一条荐股:" + name + "\t" + datetime + "\t"+ stockcode + "\t" + link);
						} else {
							log.info("荐股信息插入失败:" + link);
						}
					}
				}
				scbDao.修改运行过的百度URL(bdLink, 1);
			}
		}

		dbc.close();
	}