import?java.io.DataOutputStream;
import?java.io.File;
import?java.io.FileOutputStream;
import?java.io.IOException;

import?org.apache.commons.httpclient.DefaultHttpMethodRetryHandler;
import?org.apache.commons.httpclient.HttpClient;
import?org.apache.commons.httpclient.HttpException;
import?org.apache.commons.httpclient.HttpStatus;
import?org.apache.commons.httpclient.methods.GetMethod;
import?org.apache.commons.httpclient.params.HttpMethodParams;

public?class?DownLoadFile?{
	/**
	?*?根據(jù)?url?和網(wǎng)頁類型生成需要保存的網(wǎng)頁的文件名?去除掉?url?中非文件名字符
	?*/
	public??String?getFileNameByUrl（String?urlString?contentType）
	{
		//remove?http://
		url=url.substring（7）;
		//text/html類型
		if（contentType.indexOf（“html“）!=-1）
		{
			url=?url.replaceAll（“[\\?/:*|<>\“]“?“_“）+“.html“;
			return?url;
		}
		//如application/pdf類型
		else
		{
??????????return?url.replaceAll（“[\\?/:*|<>\“]“?“_“）+“.“+
??????????contentType.substring（contentType.lastIndexOf（“/“）+1）;
		}	
	}

	/**
	?*?保存網(wǎng)頁字節(jié)數(shù)組到本地文件?filePath?為要保存的文件的相對地址
	?*/
	private?void?saveToLocal（byte[]?data?String?filePath）?{
		try?{
			DataOutputStream?out?=?new?DataOutputStream（new?FileOutputStream（
					new?File（filePath）））;
			for?（int?i?=?0;?i?				out.write（data[i]）;
			out.flush（）;
			out.close（）;
		}?catch?（IOException?e）?{
			e.printStackTrace（）;
		}
	}

	/*?下載?url?指向的網(wǎng)頁?*/
	public?String?downloadFile（String?url）?{
		String?filePath?=?null;
		/*?1.生成?HttpClinet?對象并設(shè)置參數(shù)?*/
		HttpClient?httpClient?=?new?HttpClient（）;
		//?設(shè)置?Http?連接超時?5s
		httpClient.getHttpConnectionManager（）.getParams（）.setConnectionTimeout（
				5000）;

		/*?2.生成?GetMethod?對象并設(shè)置參數(shù)?*/
		GetMethod?getMethod?=?new?GetMethod（url）;
		//?設(shè)置?get?請求超時?5s
		getMethod.getParams（）.setParameter（HttpMethodParams.SO_TIMEOUT?5000）;
		//?設(shè)置請求重試處理
		getMethod.getParams（）.setParameter（HttpMethodParams.RETRY_HANDLER
				new?DefaultHttpMethodRetryHandler（））;

		/*?3.執(zhí)行?HTTP?GET?請求?*/
		try?{
			int?statusCode?=?httpClient.executeMethod（getMethod）;
			//?判斷訪問的狀態(tài)碼
			if?（statusCode?!=?HttpStatus.SC_OK）?{
				System.err.println（“Method?failed:?“
						+?getMethod.getStatusLine（））;
				filePath?=?null;
			}

			/*?4.處理?HTTP?響應(yīng)內(nèi)容?*/
			byte[]?responseBody?=?getMethod.getResponseBody（）;//?讀取為字節(jié)數(shù)組
			//?根據(jù)網(wǎng)頁?url?生成保存時的文件名
			filePath?=?“temp\\“
					+?getFileNameByUrl（url?getMethod.getResponseHeader（
							“Content-Type“）.getValue（））;
			saveToLocal（responseBody?filePath）;
		}?catch?（HttpException?e）?{
			//?發(fā)生致命的異常，可能是協(xié)議不對或者返回的內(nèi)容有問題
			System.out.println（“Please?check?your?provided?http?address!“）;
			e.printStackTrace（）;
		}?catch?（IOException?e）?{
			//?發(fā)生網(wǎng)絡(luò)異常
			e.printStackTrace（）;
		}?finally?{
			//?釋放連接
			getMethod.releaseConnection（）;
		}
		return?filePath;
	}
}

?屬性????????????大小?????日期????時間???名稱
-----------?---------??----------?-----??----

?????文件???????4478??2010-06-28?13:46??Chap04\ChannellinkDO.java

?????文件??????17812??2010-06-28?13:46??Chap04\ExtractContext.java

?????文件????????825??2010-04-24?11:08??Chap04\HtmlParser\.classpath

?????文件????????386??2010-05-04?08:52??Chap04\HtmlParser\.project

?????文件????????629??2010-04-19?16:37??Chap04\HtmlParser\.settings\org.eclipse.jdt.core.prefs

?????文件??????11334??2010-04-26?15:54??Chap04\HtmlParser\src\com\lietu\htmlParser\HtmlParser.java

?????文件???????8312??2006-09-23?14:25??Chap04\HtmlParser\src\doc-files\building.html

?????文件???????5074??2006-09-17?12:47??Chap04\HtmlParser\src\doc-files\overview.html

?????文件???????4896??2006-09-17?13:00??Chap04\HtmlParser\src\doc-files\using.html

?????文件??????26096??2006-09-17?07:24??Chap04\HtmlParser\src\org\htmlparser\Attribute.java

?????文件??????10617??2006-09-17?07:25??Chap04\HtmlParser\src\org\htmlparser\beans\BeanyBaby.form

?????文件??????13209??2006-09-17?07:25??Chap04\HtmlParser\src\org\htmlparser\beans\BeanyBaby.java

?????文件??????13762??2006-09-17?07:25??Chap04\HtmlParser\src\org\htmlparser\beans\FilterBean.java

?????文件???????6547??2006-09-17?07:25??Chap04\HtmlParser\src\org\htmlparser\beans\HTMLlinkBean.java

?????文件???????9020??2006-09-17?07:25??Chap04\HtmlParser\src\org\htmlparser\beans\HTMLTextBean.java

?????文件????????213??2006-09-17?07:25??Chap04\HtmlParser\src\org\htmlparser\beans\images\Chain16.gif

?????文件????????278??2006-09-17?07:25??Chap04\HtmlParser\src\org\htmlparser\beans\images\Chain32.gif

?????文件????????140??2006-09-17?07:25??Chap04\HtmlParser\src\org\htmlparser\beans\images\Knot16.gif

?????文件????????167??2006-09-17?07:25??Chap04\HtmlParser\src\org\htmlparser\beans\images\Knot32.gif

?????文件???????8602??2006-09-17?07:25??Chap04\HtmlParser\src\org\htmlparser\beans\linkBean.java

?????文件???????2188??2006-09-17?07:25??Chap04\HtmlParser\src\org\htmlparser\beans\package.html

?????文件??????23110??2006-09-17?07:25??Chap04\HtmlParser\src\org\htmlparser\beans\StringBean.java

?????文件???????3485??2006-09-17?07:25??Chap04\HtmlParser\src\org\htmlparser\filters\AndFilter.java

?????文件??????13731??2006-09-22?20:26??Chap04\HtmlParser\src\org\htmlparser\filters\CssSelectorNodeFilter.java

?????文件???????4224??2006-09-17?07:25??Chap04\HtmlParser\src\org\htmlparser\filters\HasAttributeFilter.java

?????文件???????5213??2006-09-17?07:25??Chap04\HtmlParser\src\org\htmlparser\filters\HasChildFilter.java

?????文件???????4821??2006-09-17?07:25??Chap04\HtmlParser\src\org\htmlparser\filters\HasParentFilter.java

?????文件???????3556??2006-09-17?07:25??Chap04\HtmlParser\src\org\htmlparser\filters\HasSiblingFilter.java

?????文件???????1825??2006-09-17?07:25??Chap04\HtmlParser\src\org\htmlparser\filters\IsEqualFilter.java

?????文件???????3184??2006-09-17?07:25??Chap04\HtmlParser\src\org\htmlparser\filters\linkRegexFilter.java

............此處省略2237個文件信息